Sword of Damocles hanging over Swiss heads

The devil is in the detail of the treaty package with the EU

by Professor Dr Dr h.c. Carl Baudenbacher*

(24 July 2026) In the official documents on the EU–Switzerland treaty package, the “Explanatory Report” of June 2025 and the dispatch of March 2026, the Federal Council does not stint on praise for its more than seventy negotiators, who between March and December 2024 took part in more than 200 negotiating meetings and rounds across fourteen specialist areas, or tracks.

It is said again and again that all negotiating objectives were met, indeed more than met. Since these texts were written by the federal administration, one may assume that the civil servants awarded themselves the applause. Anyone who experienced the defunct GDR will be reminded, to some extent, of the slogans of the Socialist Unity Party (SED) at party congresses.

A sober assessment of the results, however, gives rather less cause for enthusiasm. I am not speaking here of the institutional one-sidedness of the agreements, with Switzerland’s dynamic adoption of EU law, de facto supervision by the European Commission and the Court of Justice of the European Union’s monopoly on interpretation. These elements of an unequal treaty, combined with the undertaking to make regular “cohesion payments” in favour of the less prosperous EU Member States, had already been accepted by the Federal Council on 15 December 2023.

II. “Gaps” and “exceptions”

In substantive law too, Switzerland’s negotiating successes prove modest, or uncertain. This applies first to the “absence of requirements” and to “exceptions”. As is well known, the switch from a “horizontal” to a “vertical” approach after the interruption of negotiations in May 2021 was justified on the basis that embedding the institutional elements in the individual agreements would make it possible to negotiate exceptions to institutionalisation. The treaty texts do indeed regularly contain a provision under which, where there is an exception to the obligation of dynamic adoption of law and the dispute does not involve the interpretation or application of concepts of Union law, the arbitral tribunal decides the dispute without referring the matter to the CJEU. In reality, this is largely propaganda.

According to the official account from Bern, “absent requirements”, issues “not covered” and “explicit exceptions” show that Switzerland has been able to protect sensitive areas better than under the 2018 draft Institutional Agreement. That is only partly true. The fact that a treaty is silent on a question does not automatically mean that the area concerned has been carved out and left to Swiss autonomy. Silence may equally be understood as a gap to be filled by the competent legal actors, namely the Commission and the CJEU.

Explicit exceptions also offer only limited security. They provide more clarity than the mere absence of requirements, but their scope would likewise not be determined autonomously by Switzerland. The CJEU would have to rule on their reach, and according to its settled case law exceptions are, as a rule, to be interpreted narrowly. Moreover, exceptions are always subject to changing political, economic and legal circumstances. What is granted today as a special arrangement may tomorrow be reassessed and restricted under pressure from the principle of homogeneity, the requirement to create a level playing field, and the desire for market opening. No stable order can be built on exceptions. They remain snapshots within a system that is structurally geared towards uniformity.

III. A serious blunder by our negotiators

There is also a failure by the Swiss negotiators on a central point: the question of the conditions under which the European Commission may initiate a dispute-settlement procedure and thereby, ultimately, procure a binding interpretation of a legal question by its own court.

Article 10(3) of the 2018 draft Institutional Agreement provided in this respect: “Where the dispute raises a question concerning the interpretation or application of a provision […], the arbitral tribunal shall refer the matter to the Court of Justice of the European Union. The judgment of the Court of Justice of the European Union shall be binding on the arbitral tribunal.” By contrast, provisions such as Articles 10(1) IP-FMPA and 32(1) of the Electricity Agreement, under the heading “Procedure in the event of difficulties of interpretation or application”, provide that in the event of difficulties of interpretation or application of the agreement or of a Union legal act referred to therein, the contracting parties shall consult in the Joint Committee with a view to finding a mutually acceptable solution. If the Joint Committee does not find a solution within three months, either contracting party may submit the dispute to an arbitral tribunal. If interpretation or application of relevant Union law is required for dispute settlement, the arbitral tribunal refers the question to the CJEU; if, by contrast, the dispute concerns an exception and does not involve concepts of Union law, the arbitral tribunal decides itself. Where the arbitral tribunal makes a reference, the CJEU’s decision is binding on it.

Most of the other treaties contain provisions in the same terms. Under these provisions, the initiation of a procedure leading to the CJEU is not confined to disputes. The European Commission may also act where difficulties of interpretation or application arise in advance of such a procedure.

IV. Abstract interpretation procedure

Anyone familiar with the professional working methods of the European Commission will understand that this change cannot be accidental. Articles 10(1) IP-FMPA and 32(1) of the Electricity Agreement recall Article 68(3) of the EC Treaty in the Amsterdam version of 1997. That provision read: “The Council, the Commission or a Member State may request the Court of Justice to give a ruling on a question of interpretation of this title or of acts of the institutions of the Community based on this title. The ruling given by the Court of Justice in response to such a request shall not apply to judgments of courts or tribunals of the Member States which have become res judicata.”

Title IV governed the CJEU’s preliminary-ruling jurisdiction in the area of visas, asylum, immigration and other policies relating to the free movement of persons. The procedure was intended to allow clarification of how particular laws were to be understood without there having to be a concrete dispute, with claimant and defendant. Unlike the classic preliminary-ruling procedure under what is now Article 267 TFEU, Article 68(3) EC Amsterdam was not about a dialogue between national courts and the CJEU, but between the political institutions, the Council and the Commission, as well as the Member States, and the CJEU. The interpretation of a particularly sensitive area was thereby concentrated in the CJEU.

The political institutions of the Union were given an independent right of initiative. Article 68(3) EC Amsterdam was therefore not merely a technical jurisdictional rule, but the expression of a system that deliberately concentrated interpretative authority in politically sensitive questions in the hands of the political Union institutions and the CJEU. That came at the expense of the Member States. It may be described as an instance of EU-internal asymmetric dispute settlement.

V. The parallel between Article 68(3) EC and the EU–Switzerland treaties

Article 68(3) EC Amsterdam did not establish an inter-State dispute-settlement procedure, but a special Community-internal mechanism to secure uniform interpretation in the area of the then Title IV EC. The EU–Switzerland treaties provide for a pro forma arbitral tribunal which must involve the CJEU and follow it where questions of interpretation of Union law are “involved”. In both cases, the Commission, a political institution, is given the right to have questions of interpretation clarified by the CJEU at any time, independently of any concrete dispute and thus in abstract terms. In the case of Article 68(3) EC Amsterdam, the asymmetry existed in the relationship between the EU and the Member States. Under the EU–Switzerland treaties, it exists between the parties. The fact that the inter-State procedure under the EU–Switzerland treaties contains a hybrid element in the form of the pro forma arbitral tribunal is due merely to the parties’ desire to maintain the appearance that Switzerland retains a certain degree of sovereignty.

The rule in Article 68(3) EC Amsterdam was never applied. It was later abandoned in the course of treaty development. Its underlying idea is now experiencing a kind of revival in the EU–Switzerland treaty package.

VI. Sword of Damocles

The wording of Articles 10(1) IP-FMPA and 32(1) of the Electricity Agreement means that the Commission may at any time request an authoritative ruling on doubtful questions from its own court, the CJEU, whenever it is politically expedient. Whether a concrete conflict exists is irrelevant. The Commission can, as it were, order binding interpretations from the CJEU in advance. Its negotiating position in the Joint Committee has also been strengthened. It may be assumed that the partial transformation of the concrete dispute-settlement mechanism envisaged in 2018 into an abstract one was brought about by the European Commission itself. Whether the much-praised Swiss negotiators understood what was happening here is open to doubt. Switzerland is thereby placed in the position of the legendary Damocles. In a Greek legend, Damocles envied King Dionysius his wealth and power. To show him how deceptive such fortune is, the king had him sit on the throne above which a sword hung from a single horsehair, liable to fall at any moment. Damocles then recognised that power and prosperity offer no security, and wished to leave the throne.

The Swiss have signed the treaties with the EU. They can now bring themselves to safety only by voting no in a referendum. The fact that, contrary to general expectation, an abstract control procedure has been smuggled into the agreements is a further reason for subjecting them to a mandatory referendum requiring a majority of both the people and the cantons.

* Professor Dr Dr h.c. Carl Baudenbacher, a Swiss jurist and professor emeritus of private, commercial and economic law at the University of St Gallen, was President of the EFTA Court in Luxembourg from 2003 to 2017. Today he is a partner in a law firm and a visiting professor at the London School of Economics.

(First published in “Weltwoche”, No. 25/2026)
(Translation by the author)