No Nordic country has ever enacted a statutory national minimum wage. Denmark came closest to being pressured into one in 2022. It sued the European Union instead — and in November 2025, the European Court of Justice ruled in its favor.
The question the ruling raised is worth sitting with: how do workers in five of Europe’s highest-paying labor markets protect their wages without a legal floor?
01 / What a collective agreement does that a law cannot
A minimum wage law sets a floor that applies to every worker in a country, regardless of sector, employer, or union membership. A collective bargaining agreement does something more specific: it sets wage terms between a defined group of employers and the workers who negotiate against them.
In most countries, collective agreements cover a fraction of the workforce. The legal minimum does the heavy lifting for everyone else — the workers at non-union employers, in sectors too fragmented to organize, or in labor markets where employer power is too concentrated to contest.
In Denmark, Sweden, Finland, and Iceland, collective agreements cover between 80 and 90 percent of all workers. In Norway the figure is closer to 50 percent, but Norway supplements its model with a targeted mechanism for industries most exposed to wage undercutting.
At 80-90 percent coverage, the agreement is the floor. There is no separate population of unprotected workers for a statutory minimum to reach.
The practical consequence is visible in the numbers. In 2024, the collectively negotiated minimum for unskilled workers in Denmark’s manufacturing sector reached DKK 136.15 per hour (Industriens Overenskomst, §22(1), effective March 2024). The cleaning sector set a floor of DKK 157.75 per hour under the Collective Service Agreement between DI and 3F/Serviceforbundet. Neither figure is required by law. Both hold because the social partners who set them have been doing so since 1899, and neither side has an interest in breaking the system that produced them.
Sweden’s model works through a similar norm-setting mechanism. The Industriavtalet — the Industrial Agreement between the export-manufacturing sector’s unions and employer confederation — sets the acceptable wage growth rate for the entire Swedish economy each bargaining round. Sectors that overshoot create inflationary pressure. Sectors that undershoot face union resistance. The manufacturing norm functions as a coordination device without statutory authority. Effective minimum wages via collective agreement in Sweden currently range from approximately SEK 24,000 to 27,000 per month for entry-level positions, depending on sector (Medlingsinstitutet, collective agreement wage data).
Fig · 01 — Effective wage floors: Nordic collective bargaining minimums vs EU statutory peers, 2025
02 / Why the model holds
Three structural properties sustain the Nordic wage model without legal compulsion.
High union density gives workers genuine negotiating power. Collective bargaining only works when workers can credibly threaten to withhold labor. In countries where union membership covers a small share of the workforce, employers can route around agreements by shifting work to non-union contractors, temporary agencies, or posted workers.
Union membership in Iceland is approximately 90 percent. In Denmark, approximately 67 percent (including yellow unions; the figure falls to 54 percent when only confederal unions are counted). In Finland, approximately 65 percent. In Sweden, approximately 66 percent. In Norway, approximately 52 percent. (OECD/AIAS ICTWSS dataset; Nordic Economic Policy Review 2025.) Compare that to the United States, at approximately 10 percent, or the United Kingdom, at approximately 23 percent. In those labor markets, union density is too low for collective agreements to function as universal wage-setting mechanisms. A statutory floor is necessary because the social infrastructure to build one organically does not exist.
Employer organizations are as coordinated as unions. Bargaining requires two organized parties. The Nordic model’s durability comes partly from the fact that employer confederations are as consolidated and committed to the system as union confederations — and have been for over a century.
In Denmark, the 1899 Septemberforliget (September Compromise) between the Danish Employers’ Confederation and the Danish Confederation of Trade Unions established the framework that still governs wage-setting today. In Sweden, the equivalent moment was the 1938 Saltsjöbaden Agreement. In Iceland, Act No. 80/1938 formalized collective bargaining rights at the same time the Icelandic labor movement was consolidating. In Finland, the Collective Agreements Act followed in 1946.
These are not recent policy experiments. They are frameworks that have accumulated institutional legitimacy over 80 to 125 years, with both sides having deeply embedded interests in their continuation.
The manufacturing sector anchor holds wages economy-wide. In both Denmark and Sweden, the export-oriented manufacturing sector sets the wage norm — the acceptable increase rate — for the entire economy each bargaining cycle. Because the manufacturing sector competes internationally, its wage increases reflect real productivity gains rather than domestic political pressure. Other sectors follow the norm because deviating from it either invites union action (if too low) or creates inflation that erodes everyone’s real gains (if too high).
This coordination produces a different outcome than a statutory minimum. A legal floor sets a point that every sector can race above, with no mechanism to coordinate the pace. The Nordic norm sets a pace that the whole economy moves together.
Fig · 02 — Union density: Nordic countries vs selected international peers, 2024
03 / Norway: where the model meets its limits
Norway’s situation differs from its neighbors in one important respect. Union density at approximately 52 percent is high by international standards but noticeably lower than Denmark, Sweden, Finland, or Iceland. Lower density means gaps in coverage — workers in non-union environments with no agreement setting their pay.
Norway’s response came in 1993 with the Allmenngjøringsloven, the General Application Act. Under this law, the government can extend the wage rates from a sector’s collective agreement to all workers in that sector, regardless of whether their employer is party to the agreement. A tariff board (Tariffnemnda) evaluates applications and can make the rates legally binding.
Nine sectors are currently extended under this mechanism: construction, electrical work, cleaning, hotel and restaurant, agriculture, fish processing, shipbuilding, freight transport, and passenger transport. For workers in those sectors, the rates are law. For everyone else, the voluntary collective agreement system continues to operate.
The extension approach is specifically targeted at industries where foreign subcontractors and posted workers have historically undercut locally negotiated wages. From June 15, 2025, the extended minimum for unskilled construction workers is NOK 239.61 per hour. For skilled workers, NOK 264.32 per hour. (Source: Arbeidstilsynet allmenngjøring rate schedule, effective 2025-06-15.)
Norway’s hybrid model illustrates that the Nordic objection is not to minimum wages as such. It is to a universal statutory floor that substitutes for collective bargaining across the entire economy. Where collective bargaining cannot reach specific sectors reliably, Norway brings the law in — but only there.
04 / The EU test
In September 2022, the European Parliament and Council adopted Directive 2022/2041 on Adequate Minimum Wages. The directive required member states with collective bargaining coverage below 80 percent to develop national action plans to increase it. For member states already above that threshold, the directive affirmed broad principles of wage adequacy without mandating specific floors.
Denmark and Sweden — both with collective agreement coverage well above 80 percent — voted against adoption. Their objection was constitutional rather than practical: the EU Treaty explicitly excludes pay from the scope of EU social policy coordination. A directive that touched wage-setting mechanisms, even indirectly through coverage thresholds and adequacy criteria, overstepped the EU’s legal competence.
Both countries filed annulment actions before the Court of Justice of the European Union. On November 11, 2025, the Grand Chamber of the CJEU ruled in Case C-19/23 (Kingdom of Denmark v European Parliament and Council). The Court annulled Article 5(2) of the directive — which prescribed mandatory criteria for setting minimum wage levels — along with the cross-reference to Article 5(2) in Article 5(1) and the condition in Article 5(3) preventing decreases under automatic indexation. The Court found those provisions exceeded EU competence by directly interfering with member state wage-setting. All other provisions of the directive were upheld. The ruling confirmed that the directive does not require member states to introduce a statutory minimum wage.
Finland and Iceland, which had followed the proceedings closely, were not parties to the legal challenge but benefited from the clarity it produced. Norway, as an EEA member rather than an EU member state, had monitored developments without being directly bound by the directive.
The ruling did not invalidate the directive’s approach for the 23 EU member states with statutory minimum wages. It confirmed only that the Nordic model, as actually practiced, was sufficient to meet the directive’s underlying aims — and that the EU could not require countries already delivering adequate wages through collective bargaining to build a parallel statutory structure.
05 / What the Nordic model cannot export
The November 2025 ruling closed one question. It opened another: whether the features that make the Nordic model work can be transplanted to labor markets that lack them.
The answer, in most cases, is no — not quickly.
The social partner infrastructure that sustains collective bargaining in Denmark and Sweden accumulated over more than a century. Employer confederations as consolidated as union confederations. Norm-setting mechanisms accepted across sectors without legal compulsion. Institutional memory of what happens when the system breaks down. These properties cannot be legislated into existence.
The relevant question for much of Europe is different: whether countries with declining collective agreement coverage — Germany’s coverage has fallen from approximately 79 percent in the mid-1990s to approximately 50 percent today (IAB Establishment Panel; WSI), Southern Europe’s uneven coverage, Central and Eastern Europe’s weaker social partner institutions — can rebuild the density required for collective bargaining to function as a universal wage floor.
Where that rebuilding is not plausible within the relevant policy horizon, a statutory minimum is not a second-best solution. It is the only mechanism available.
The Nordic model does not argue against statutory minimums in general. It argues only that in countries where the social infrastructure to do without one already exists, the infrastructure tends to produce better outcomes than the law would.
That is the case the November 2025 ruling ultimately confirmed. Whether any other country can make it is a separate question entirely.
Wage figures for Denmark, Sweden, Norway, Finland, and Iceland are sourced from the wage.is country data files, drawing on primary publications from Statistics Denmark (Danmarks Statistik), the Swedish National Mediation Office (Medlingsinstitutet), Statistics Norway (Statistisk sentralbyrå), Statistics Finland (Tilastokeskus), and Statistics Iceland (Hagstofa Íslands). Collective bargaining minimum wage figures cited in Section 1 are verified against sector agreement publications via Arbeidstilsynet (Norway), the Danish Ministry of Employment, and Medlingsinstitutet (Sweden). The CJEU Case C-19/23 reference is verified via the official judgment registry at curia.europa.eu. Union density figures from OECD/AIAS ICTWSS dataset 2024 and Nordic Economic Policy Review 2025. No figures are drawn from LLM training data.