Statecraft

14 September 2026 · analysis

The law nobody defended

Two years of Dutch asylum legislation as a case of form without an owner

by Jacob Huibers · Lees in het Nederlands →

I. Three votes on a single Tuesday afternoon

On Tuesday 21 April 2026 the Dutch Senate voted on three bills that two years earlier were together to have formed the strictest asylum policy ever. First the amending bill meant to soften the criminalisation of irregular residence, so that anyone helping an undocumented person would not be liable as a party to a crime: rejected, 37 for, 38 against. Then the Asylum Emergency Measures Act itself, the bill into which that criminalisation had found its way by amendment: rejected, 31 for, 44 against. The CDA and SGP, which had made their support for the act conditional on the amending bill, voted against as soon as the amending bill had fallen. Third, the Two-Status System Act: passed, 41 for, 34 against. Five motions were rejected.1

What stands out about that afternoon is who was missing. The coalition that had devised the law had fallen. The minister who had introduced it had gone. The party that had written the amendment on criminalisation voted against the amending bill needed to get its own law through the Senate, in company it meets nowhere else: GroenLinks-PvdA, D66, SP, Volt, ChristenUnie and the Party for the Animals also voted against, because they did not want the criminalisation at all; VVD, CDA, SGP, JA21, BBB and FVD voted in favour.1 The cabinet that had inherited the law had wanted to implement it “in full” in its coalition agreement, but proved unable to organise a majority for it in the Senate. Nobody in the chamber could explain why this was a good law. There were only people who could explain why they had to vote for or against it.

A law that falls because it is bad is a normal event in a state under the rule of law. A law that falls because nobody owns what is in it any more is something else. The second is what happened here, and the track leading to it shows how.

II. The timeline as pattern

The file begins in the autumn of 2024, when the Minister for Asylum and Migration tried to introduce stricter policy through emergency state powers, an instrument for war and natural disaster, while by every authoritative expert account there were no grounds for it. The advice that was never written described, on the basis of documents later released under freedom-of-information law, what the civil servants did and did not advise at the time: they steered the minister away from emergency powers, but the advice to make full use of the existing Aliens Act was never written.2 That is the first moment at which form, an emergency regulation, took precedence over substance, the question of what existing law already allowed.

The second moment is 10 February 2025. At the minister’s request, the Advisory Division of the Council of State publishes its urgent opinions on the Asylum Emergency Measures Act and the Two-Status System Act. The verdict is as clear as an opinion can be: the explanatory memorandum has not made it plausible that the measures contribute to limiting the inflow or to a more efficient procedure, while it is clear that several measures place an extra burden on the Immigration and Naturalisation Service and the judiciary. The Division advises against introducing the bills in this form.3 On 7 March 2025, twenty-five days later, they are introduced in this form.4

The third moment is 3 July 2025. The House of Representatives passes the Asylum Emergency Measures Act by 94 votes to 56, including the Vondeling amendment, which criminalises irregular residence through a new article 108a.4 The amendment came to a vote without the Advisory Division having been heard on it. The guidance the government subsequently requested arrived on 27 August 2025 and was twofold: there is a real possibility that anyone helping a person without residence rights is liable as an accessory, and the questions that arose after the vote should have been answered during the preparation and parliamentary treatment of the amendment; the legislative process before the vote had been careless.5 It fits a pattern. The Temporary Committee on Delivery Organisations established in 2021 that motions and amendments are structurally not tested for workability.6 A parliament that amends without testing adds substance to a form without knowing whether the form can hold it.

The fourth moment is the amending bill, introduced on 12 November 2025 and passed by the House on 18 December 2025 by 91 votes to 59.7 To keep the CDA and SGP on board in the Senate, the criminalisation has to be softened: those who help are not liable. The amending bill is a repair of an amendment that was a repair of a law that, according to its own adviser, did not do what it promised. Each layer repairs the previous one; no layer returns to the question the Advisory Division had received no answer to in February 2025.

The fifth moment is the change of cabinet. The Jetten cabinet, which took office in February 2026, adopts its predecessor’s laws “in full” in the coalition agreement, while municipalities, the immigration service and the judiciary warn against them. The Association of Netherlands Municipalities argues that criminalisation drives undocumented people below the radar; at the immigration service and the Council for the Judiciary there is a fear that the laws will increase workloads and slow procedures, clogging the asylum chain further rather than relieving it.8 A new cabinet that adopts a bill it did not write, has not argued afresh, and whose implementers it has not heard, is adopting a form. The substance stays with the previous owner, who is no longer there.

Six moments, and at each of them the form moves on while the substance stays behind. An emergency regulation instead of the existing law. An introduction despite the opinion. An amendment without a test. An amending bill on an amendment. An adoption without reconsideration. And finally a vote in which the only body still able to review, the Senate, did so, by a margin of one.

This is the pattern from The dissociated legislator at the scale of a single file.9 That paper described how the Advisory Division is ritually intact and practically decoupled: between 2022 and 2025 the heavy opinions on the box 3 wealth tax, the Asylum Dispersal Act, the Affordable Rent Act and the benefits recovery legislation were all delivered, published and ignored without real adjustment. The asylum file adds two more opinions, and one difference: this time the review had a consequence. It came in the Senate, and it turned on the arithmetic, with the substance left untouched.

III. What strictness cannot do

The reflex in this file has been the same for two years: stricter, faster, national on top of European. That reflex confuses a property of the form with a property of delivery. Strictness sits in the text of the law. Effect arises at the immigration service, in the courtroom, in the municipality that no longer sees an undocumented person. The Advisory Division said it in February 2025 in one sentence: not plausible that the inflow will fall, plausible that the chain will be more heavily burdened. A law that is strict and does not work produces a backlog.

On top of that, the national edifice stands on a European one entering into force in the same months. The EU Migration Pact, applicable from 12 June 2026,10 already sharpens the European asylum rules considerably, and the cabinet has, as NRC established in its editorial, not made sufficiently clear what the national laws add to it.8 Anyone who stacks a national law on a European regulation without knowing what the stack produces is making a signal.

Brunsson would recognise this as organised hypocrisy: the talk says strictest ever, the decision says law, the action says queue at the immigration service, and each of the three serves a different audience.11 It is worth being precise here. On 21 April the Senate rejected a form whose substance nobody could vouch for any more; strictness was not the issue. That the Two-Status System Act, on which the Advisory Division had advised in the same terms, was passed, by 41 votes to 34, shows that the Senate was filtering on who still stood behind it.

IV. Four design choices

The first choice is the implementation test on amendments, before the vote. The Temporary Committee on Delivery Organisations recommended it in 2021; the Vondeling amendment is the example of what happens without it. An amendment that introduces a criminal offence touches the police, the public prosecutor, municipalities and care providers; that the question of workability was only put to the Council of State after the vote is an inversion of the order. The Rules of Procedure can make the test mandatory without any need for legislation.

The second choice is a hearing with the Advisory Division on every heavy opinion, before the first round of debate begins. It was proposed in The dissociated legislator as an amendment to the Rules of Procedure; this file is its justification. What the House skipped in 2025, the Senate did in its own way in April 2026, with the uncertainty of a vote that hung on one. A review carried out in the first chamber of treatment does not have to be repaired in the second.

The third choice is an ownership letter on a change of cabinet. A cabinet that adopts a pending bill from its predecessor does so with a letter to both chambers in which it argues the bill afresh, answers the opinions of delivery bodies and of the Advisory Division, and establishes its relation to changed circumstances — here the Migration Pact. Anyone who adopts a form without arguing the substance afresh inherits the problem this paper describes. It is the aikido choice in this file: the political need to show continuity is neither denied nor forced, but redirected to a moment at which continuity only pays if it has been argued again.

The fourth choice is a sequencing rule. National asylum legislation intended to do what a European regulation already does is treated only once that regulation applies and its effect has been measured. The ground is that stacking without measuring is not policy. A period of one year after 12 June 2026 is enough to know what the Pact does and what remains to be done nationally.

V. For interim practice

Every interim assignment begins with inheriting decisions. A predecessor has announced a reorganisation, signed a contract, promised a scheme. The temptation is to carry those decisions out in full, because that brings calm and because reconsidering is read as weakness. In twenty years I have taught myself one rule I have never let go: I do not carry out a predecessor’s decision that I cannot argue afresh to the people who have to implement it. If I cannot, I go back to the client to find the decision’s owner. Bringing a form without an owner into delivery is the fastest way to lose delivery. It notices before the executive does.

In February 2026 the Jetten cabinet did the opposite, and on 21 April received the bill that would have arrived at the same moment in any delivery organisation: the moment someone asks who actually stands behind this.

VI. The open question

The question underneath this file is whether 21 April 2026 is a sign that review works again, or an accident of the arithmetic. If the Senate has regained its role as chambre de réflexion, that is the most important institutional event of the year. If it let a law fall by a single vote because two parliamentary groups needed an amending bill that a third blocked for reasons of its own, then nothing has been restored and only the sum came out differently.

What I suspect, and cannot prove, is the second. The Senate rejected the law, which its adviser had called bad, because its form no longer had an owner. That would mean the system still judges on ownership alone. A law with an owner and no substance gets through; a law without an owner falls, whatever is in it. That is a state under the rule of law waiting for somebody to let go.

The Hague now faces the implementation of the Migration Pact and, if it proves necessary, the drawing board. It would be the first time in this file that substance came first.

Colophon

“The law nobody defended” is a Statecraft analysis applying the dissociation frame from Series I and from The dissociated legislator to a single legislative track, from the autumn of 2024 to the Senate votes of 21 April 2026. It follows on from The advice that was never written (12 May 2026) and takes its occasion from NRC’s editorial of 18 April 2026.

Responses and counter-argument via Statecraft.nl.

Jacob Huibers is an interim manager with over twenty years’ experience in the Dutch public sector. He has worked as cluster manager, cluster director and quartermaster at municipalities ranging from fifty thousand to over two hundred thousand inhabitants and at regional partnerships in the social and physical domains. Statecraft is his platform for strategic reflection on public delivery, pillar IV of House of Viridian.

Footnotes

  1. Senate of the States General, Senaat verwerpt strafbaarstelling illegaliteit, stemt in met tweestatusstelsel, news release of 21 April 2026; files 36.703 (Two-Status System Act), 36.704 (Asylum Emergency Measures Act) and 36.855 (Amending bill on the criminalisation of irregular residence). Plenary treatment 13 and 14 April 2026. ↩ ↩2

  2. Statecraft, Het advies dat niet werd geschreven. Over WOO, noodrecht en de dissociatie van het asielbeleid, 12 May 2026. ↩

  3. Council of State, Advisory Division, opinions W03.24.00364/II (Asylum Emergency Measures Act) and W03.24.00362/II (Two-Status System Act), published 10 February 2025, with the summary Adviezen Asielnoodmaatregelenwet en Wet invoering tweestatusstelsel; Montesquieu Institute, De adviezen van de Raad van State over de Asielnoodmaatregelenwet en de Wet invoering tweestatusstelsel, 10 February 2025. ↩

  4. Senate, file 36.704: introduced in the House of Representatives on 7 March 2025; House vote 3 July 2025, passed by 94 votes to 56. ↩ ↩2

  5. Council of State, Advisory Division, Voorlichting over amendement bij Asielnoodmaatregelenwet over reikwijdte van strafbaarstelling illegaliteit, W03.25.00207/II, adopted 27 August 2025, published 29 August 2025. ↩

  6. Temporary Committee on Delivery Organisations, Klem tussen balie en beleid, parliamentary paper 35387, no. 2, 25 February 2021. ↩

  7. Senate, file 36.855, Novelle aanpassing strafbaarstelling illegaal verblijf. ↩

  8. NRC, editorial Gebrekkige asielwetten zijn symbool van politiek-bestuurlijk gerommel, 18 April 2026. ↩ ↩2

  9. Statecraft, De gedissocieerde wetgever. Hoe het Nederlandse wetgevingsbedrijf zichzelf als wetstoepasser is gaan zien, 5 May 2026, section III. ↩

  10. Dutch government, Nederland dient Nationaal Implementatieplan Europees Asiel- en Migratiepact in bij de Europese Commissie, 12 December 2024; Implementation Act for the Asylum and Migration Pact 2026, file 36.871. The Pact’s regulations apply from 12 June 2026. ↩

  11. Nils Brunsson, The Organization of Hypocrisy, Wiley 1989. ↩