The cold email arrived in a Dutch freelancer’s inbox a few weeks ago, forwarded on to Remote Work Europe with a wry one-liner. An employer-of-record vendor was warning a UK client that engaging Dutch ZZP’ers (zelfstandigen zonder personeel, the self-employed without staff) was now “extremely high risk” thanks to the end of the Belastingdienst moratorium, and that the only safe route was to put every Dutch contractor onto the vendor’s payroll. There were urgency words. There was a calendar link. There was, conspicuously, no mention of the actual soft-landing policy the Dutch Cabinet had announced two months earlier.

This is roughly the information environment most foreign companies are operating in when they think about Dutch freelance contracts in 2026. Remote Work Europe spends a lot of time reading the actual Dutch legislative texts and the actual Belastingdienst guidance, in Dutch, so that we can tell our readers what is genuinely changing and what is sales fiction. The short version, and the strongest extractable claim of this piece: Dutch freelancers earning under €38 per hour gain a procedural presumption of employment under a law adopted by the Eerste Kamer (Dutch Senate) on 16 June 2026, and the Belastingdienst’s “soft landing” enforcement posture means corrections and instructions in 2026 rather than standard fines. Those are the two things to hold in your head. Everything else is detail.

One week, three things

The week of 16 June 2026 will end up being the reference point for Dutch freelance reform. Three things happened in parallel, and the EOR sales decks have already begun to blur them.

The Eerste Kamer adopted Bill 36783, the Wet rechtsvermoeden arbeidsovereenkomst op basis van uurtarief, the law creating a rebuttable presumption of employment on the basis of hourly rate. On the same day, the Gerechtshof Amsterdam (Amsterdam Court of Appeal) ruled that the platform Temper is a uitzendbureau (a temporary employment agency) rather than a marketplace for independent ZZP work, with significant implications for platform workers across the Dutch market. The court ruling is its own story, and we cover the reclassification and what it means for platform contractors in a companion piece. This piece is about the legislative track and the enforcement reality underneath it.

The third thread, running underneath both events, is that the Belastingdienst’s formal enforcement moratorium ended on 1 January 2026. That date is the one most foreign clients keep hearing about, and it’s the one most often weaponised in EOR sales pitches. It deserves to be understood properly.

What Bill 36783 actually does

Bill 36783 is a deliberately narrow instrument. The Dutch government’s original ambition, the Wet verbetering beoordeling arbeidsrelaties (Wet VBAR, the Improvement of the Assessment of Employment Relationships Act), would have rewritten the broader test for whether a working relationship is genuinely self-employment or disguised employment (schijnzelfstandigheid). That broader framework was withdrawn earlier in 2026 after sustained criticism from employer associations, freelancer organisations, and the Raad van State.

What survived, and what passed on 16 June, is a single procedural lever. Bill 36783 establishes that where a worker earns under €38 per hour (the reference date for the threshold is 1 January 2026), there is a rechtsvermoeden of employment: a legal presumption that the working relationship is an employment contract rather than self-employment. The worker does not have to prove they are an employee. The client has to prove they are not. The burden of proof moves.

The bill must now be published in the Staatsblad (the official Dutch government gazette) by 31 August 2026, with target commencement on 31 December 2026 and full effect from 1 January 2027. There is a hard deadline underneath all of this that almost no English-language coverage mentions: the legislative timeline is tied to a milestone in the Netherlands’ national Recovery and Resilience Plan, with roughly €600 million in EU Recovery and Resilience Facility funding contingent on delivery. The Dutch government cannot afford to let this one slip. It is the single most important reason the hourly-rate test got detached from the broader Wet VBAR collapse and pushed through on its own.

It is also worth noting what the bill does not do. It does not create a hard cap on freelance engagement under €38 per hour. It does not automatically convert existing contracts to employment. It does not apply retroactively. And it does not, in itself, set fines or penalties; the consequences of misclassification flow through the existing Belastingdienst enforcement framework, which is the other half of this story.

The primary source, for anyone who wants to read the legislative history themselves, is the Eerste Kamer dossier at eerstekamer.nl/wetsvoorstel/36783_wet_invoering.

What the Belastingdienst is actually doing in 2026

This is where the gap between EOR sales pitches and verifiable reality is widest, so it’s worth being precise.

The Belastingdienst’s enforcement moratorium on the Wet deregulering beoordeling arbeidsrelaties (Wet DBA) ran from 2016 to the end of 2025. Throughout that period, the tax authority did not issue payroll-tax corrections or fines for disguised self-employment except in cases of demonstrated malicious intent. On 1 January 2026, that formal moratorium ended.

What replaced it is not a hard enforcement regime. The Cabinet decision of March 2026, as analysed in detail by Hogan Lovells and other Dutch employment-law firms, introduced a “soft landing” period running through 1 January 2027. During this period, the Belastingdienst is empowered to audit ZZP relationships and to issue corrections and instructions where it identifies disguised employment, but the standard practice is no fines. Fines remain reserved for cases where intent or gross negligence (opzet of grove schuld in the statutory phrasing) can be proven.

Two important nuances sit underneath this.

First, retroactive payroll-tax corrections are still possible during the soft landing. The Belastingdienst’s audit window extends up to five years, and a correction issued in 2026 can reach back into prior years. “Soft landing” means no standard fines; it does not mean the tax bill itself goes away.

Second, the soft landing is explicitly a transitional arrangement, not a permanent posture. From 1 January 2027 the Belastingdienst’s enforcement framework returns to its full statutory shape. Anyone building a 2026 engagement strategy on the assumption that current leniency continues indefinitely is building on sand.

The honest summary, for foreign companies engaging Dutch contractors: 2026 is a year for cleaning up contractual substance, not a year for panic. The risk is not a fine in the post next month. The risk is a five-year retrospective audit in 2028 or 2029 that finds the substance of your engagement was always an employment relationship, and bills you for the back social security contributions and payroll tax with interest.

Why the broader Wet VBAR collapsed, and why it matters

A short detour, because the politics here shape what comes next.

Wet VBAR was the Dutch government’s attempt to replace the criteria-based fudge of the existing classification test with a structured weighing exercise covering several factors: control and supervision, embedding in the organisation, entrepreneurial risk, and what the bill called the werken voor eigen rekening en risico test (working for one’s own account and risk). It would have applied across the income spectrum, not just below a threshold.

It collapsed because it tried to do too much at once. Employer associations objected to the administrative burden. Freelancer organisations objected to language they read as presuming dependency. The Raad van State raised concerns about legal certainty. The Cabinet withdrew the broader framework and reverted to the narrower, mechanical instrument of an hourly-rate presumption.

The practical consequence is that for freelancers earning above the €38 per hour threshold (which is most genuinely independent professional ZZP work, including most international remote contractors), the legislative landscape in 2026 looks much like the landscape in 2025. The existing case-law test, anchored in the Deliveroo and similar Hoge Raad judgments, continues to apply. The new procedural presumption simply doesn’t reach them.

For freelancers earning below €38 per hour, the change is real and the burden shift is the operative mechanism. They no longer have to construct a case that they are genuinely self-employed; the client has to.

Running in parallel, the Wet vereenvoudiging arbeidsmarkt (Labour Market Simplification Act) is working its way through the legislative process and includes the phase-out of nul-urencontracten (zero-hour contracts). It’s a separate workstream and a separate audience, but the direction of travel is consistent: the Dutch labour market is being pushed, slowly and procedurally, away from arrangements that look like work-without-protection in either the employee or freelance categories.

What this means for the three audiences who keep asking

Remote Work Europe has been tracking Dutch employment-law commentary for the past six months, watching the gap between the legal reality and the marketing widen. It is worth separating out three different reader situations, because the answer is genuinely different for each one.

For Dutch ZZP’ers

If you are a Dutch ZZP’er earning above €38 per hour with a portfolio of genuine clients, your 2026 looks much like your 2025. Keep your contracts clean. Make sure the substance of your engagements supports independence: your own equipment, your own working hours, your own choice of how to deliver, your own commercial risk. Document it. The new procedural presumption is not aimed at you, and the Belastingdienst soft landing means you have a runway of months to tidy anything that’s drifted.

If you are a ZZP’er earning under €38 per hour, the shift is real and worth planning for now, not in December. From the moment Bill 36783 commences, your clients carry the burden of proving the relationship is self-employment rather than employment. Some clients will respond by converting genuine engagements to employment. Some will simply withdraw. A few will work with you to restructure the engagement so the substance clearly supports independence. The conversations are easier to have in September than in January.

For foreign companies engaging Dutch contractors

The EOR sales pitch is overstating the urgency and understating the runway. 2026 is a soft-landing year. The actionable work is contractual substance review, not panic conversion of every Dutch freelancer to an EOR payroll. Look at the genuinely independent professionals in your contractor base, document the indicators of independence (multiple clients, own tools, own pricing, own delivery method), and reserve EOR routing for the engagements that genuinely look like employment under another name. For deeper context on the cross-border alternatives, our guide to cross-border hiring without an EOR walks through the structural options.

A separate point: the €38 per hour threshold is denominated in gross hourly equivalents, and the calculation for non-standard engagements (project fees, day rates, retainers) has technical wrinkles that the Belastingdienst guidance will work through over the rest of 2026. For now, if your Dutch contractor invoices imply an hourly rate below that threshold, expect them to come to you in late 2026 wanting to talk.

For international remote freelancers with a Dutch footprint

This is the group the English-language commentary almost never addresses, and where the practical implications are most under-discussed.

If you are a non-Dutch freelancer working with Dutch clients from outside the Netherlands, the new procedural presumption applies to the working relationship, not to your tax residency. The €38 per hour threshold is a Dutch domestic instrument and the procedural burden it creates falls on your Dutch client. Your own classification and tax position remains governed by the country where you are tax-resident, with the usual cross-border permanent-establishment and posted-worker considerations layered on top.

If you are an international freelancer who has been treating cross-border work into the Netherlands as ZZP-style engagement (perhaps registered with the KvK, perhaps invoicing from a personal account), 2026 is a reasonable year to look at whether your structure still fits. One option, for freelancers whose work is genuinely cross-border and whose clients are spread across multiple countries, is routing through an Estonian OÜ via the e-Residency programme; this cleanly separates cross-border professional income from the Dutch domestic-freelance classification question. RWE’s partner for Estonian OÜ setup and ongoing accountancy is Xolo Leap, and we’ve found their cross-border framing genuinely useful.

The honest caveat: an Estonian OÜ does not fix Dutch domestic-substance issues. If both you and the work are sitting in the Netherlands, the Dutch tax authority will look through any foreign structure to the substance underneath. The Estonian route is an option for genuinely mobile, genuinely cross-border professionals; it is not a workaround for a Dutch-resident freelancer with Dutch clients doing Dutch work.

Where to go next

The companion piece on the Temper ruling looks at the platform-work side of the same week’s developments and is worth reading alongside this one if you engage with Dutch platform workers or are one yourself. Our Netherlands country guide carries the deeper detail on KvK registration, BSN, the 30% ruling for inbound knowledge migrants, and the broader regulatory architecture; we update it as the legislative dust settles.

Three concrete things worth doing before September:

Read the Eerste Kamer dossier for Bill 36783 directly if you read any Dutch at all, or run it through a translator. The dossier carries the actual statutory text, the Memorie van Toelichting (the explanatory memorandum), and the voting record. It is dramatically more useful than any English-language secondary commentary, including this one.

Audit your existing Dutch freelance engagements (whether you are a ZZP’er with several clients or a foreign company with several Dutch contractors) against the substance criteria: independence in delivery, commercial risk, multiple clients, own tools, own pricing. Document what supports independence and where the documentation is thin.

Diary the Staatsblad publication deadline of 31 August 2026 and the target commencement of 1 January 2027. The first date confirms the bill is locked in and the procedural presumption will commence on schedule; the second is when the burden of proof actually shifts. Between those two dates is the window for clean contractual conversations.