21 min readThe question of internal EU reform returned to the Union's agenda after 2022. The abuse of the veto by individual member states slowed the adoption of strategic decisions and exposed the EU's vulnerability in situations that require a rapid response – from the war in Ukraine to the change in US foreign policy. At the same time, granting Ukraine candidate status reset the enlargement process and brought back the prospect of an EU of more than 35 member states, in which reaching unanimity will be even harder.
Today around 80% of decisions in the EU are taken by qualified majority voting (QMV): Under the Lisbon Treaty, such a decision must be supported by at least 55% of member states (15 of the 27 countries) representing 65% of the EU population. A blocking minority must include at least 4 member states representing more than 35% of the population.
Unanimity in the Council of the EU is retained in the most sensitive areas: the Common Foreign and Security Policy (CFSP), enlargement of the EU, taxation and finance. Consensus also underpins the work of the European Council, which sets the Union's general political direction and priorities.
To avoid blockages, the EU has already used both existing and ad hoc mechanisms. In 2022, through the "constructive abstention" mechanism, Austria, Ireland and Malta did not take part in financing arms supplies to Ukraine, while Hungary stayed out of the creation of the European Union Military Assistance Mission in support of Ukraine (EUMAM) and, later, of the Ukraine Assistance Fund (UAF).
Under Article 31(1) of the Treaty on European Union, a state that has abstained from a vote in foreign and security policy is not obliged to implement the decision, but must accept that it commits the Union and refrain from obstructing its implementation. Although this article does not cover enlargement, a procedure similar in substance was applied in 2023 during the decision to open accession negotiations with Ukraine: Hungary's Prime Minister Viktor Orbán left the room and effectively abstained. This created a precedent for using the analogy of "constructive abstention" in the European Council.
To overcome Hungary's "vetocracy", the EU also tried to suspend its voting rights under Article 7 of the Treaty on European Union. However, despite systematic violations of EU principles and interests by Viktor Orbán's government, the procedure produced no result because of the lack of unanimity.
Such piecemeal solutions, however, do not address the problem systemically. In the long run the drivers of EU institutional reform, primarily Germany, seek a full transition of the Council of the EU to QMV across all areas of decision-making before the next enlargement. This requires amendment of the Union's founding treaties – the Treaty on European Union (TEU) and the Treaty on the Functioning of the European Union (TFEU).
This could be done through Article 48 TEU, which provides for convening a European Convention, unanimous agreement of the member states at an intergovernmental conference and subsequent ratification by their national parliaments or through referendums. The process can take years and does not guarantee a result: the adoption of the EU Constitution in 2005, for example, failed because of unsuccessful referendums in France and the Netherlands.
In September 2023, an independent working group convened by the governments of France and Germany presented recommendations on reforming the EU. It proposed several ways to speed up treaty change: a simplified procedure without convening a Convention, which nevertheless does not remove the need for unanimity and national ratification; linking institutional reform to future accession treaties; and a separate framework treaty on enlargement and reform, which could determine in advance the changes needed for an EU of more than 30 members.
To ease the concerns of skeptics, the group proposed safeguards for the move to QMV. In particular, a "sovereignty safety net" modeled on Article 31(2) TEU would allow any state to suspend the adoption of a decision if it threatened its vital interests. Another option is to change the QMV formula from 55% of states and 65% of the population to 60% and 60%. On the questions of greatest importance to member states, the so-called "supermajority" formula could apply – 72% of states and 65% of the population (Article 238 TFEUº) – or the "consensus minus one" model. A separate safeguard could be the possibility of opting out of implementing decisions (opt-out) in areas that are too sensitive for the sovereignty of individual states.
Nevertheless, there is generally no appetite among member states for treaty change. The veto allows them to advance national interests and to demand compensation or concessions on other issues. Positions also differ by area: for Sweden and the Netherlands it is important to keep the veto in tax policy, and for Cyprus and Greece in foreign policy, given their wish to retain leverage over Türkiye.
The most realistic approach for now is therefore a short-term one: the gradual introduction of QMV in individual areas (case by case) without amending the founding treaties. The most relevant areas for Ukraine are EU enlargement and the EU's Common Foreign and Security Policy.
Apart from "constructive abstention", the main route to QMV in the CFSP without treaty change is the passerelle clauses – the transitional provisions of the Lisbon Treaty that allow the Council of the EU to move from unanimity to QMV in a given area (the general Article 48(7) TEU or the specific Article 31(3) TEU for foreign and security policy). Member states must, however, support such a decision unanimously, in effect voluntarily limiting their own veto. No such unanimity exists at present. In addition, in some countries this step requires ratification by their parliaments (Germany, Austria, Denmark, Poland, Ireland, the Czech Republic and Malta) 2.
The transition is supported by the "Group of Friends" on QMV in EU foreign and security policy, created by France and Germany in 2023. It also includes Belgium, Denmark, Finland, Italy, Luxembourg, the Netherlands, Romania, Slovenia, Spain and Sweden. Their position rests on the EU's need to respond to crises faster without individual countries blocking decisions, as Cyprus did over EU sanctions against Belarus in 2020, seeking to increase the Union's pressure on Türkiye, or Hungary over sanctions against Russia in 2022-2024, pressing for the unfreezing of European funds
At the same time, most states are cautious about QMV in the Common Foreign and Security Policy, since these areas bear directly on sovereignty and, for the neutral Austria, Ireland, Malta and Cyprus, also on the foundations of their security and defense policy. Smaller states also fear that giving up the veto would reduce their weight and strengthen the influence of the large countries. The demographic criterion of QMV, in particular, favors Germany and France, whose combined | population accounts for around a third of the EU's. They will be able to influence substantially the formation of a blocking minority, which must include 4 member states representing more than 35% of the population.
Nevertheless, the move to QMV in the CFSP through the transitional provisions comes with safeguards and limits. Article 31(2) contains a safety net mechanism that allows a member state to block a decision taken by QMV for "vital and stated reasons of national policy", Article 31(4), meanwhile, does not permit QMV to be applied to decisions with defence or military implications. In practice it is hard to separate purely foreign policy decisions from security ones, so any real move to QMV in the CFSP will carry certain limits.
If QMV in the CFSP is needed for faster decisions that are critical to the Union's ability to act, QMV in enlargement is meant to adapt the institutions to a larger membership without the EU losing its capacity to function (absorption capacity). It is also meant to limit the blocking of the technical accession process because of bilateral historical or territorial disputes – that is, the "bilateralization" of accession, which undermines candidates' trust in the EU and deepens enlargement fatigue among member states.
The history of EU enlargement illustrates this problem well. France opposed the accession of the United Kingdom in 1963 and 1967, and later of Spain because of the risks to agriculture; Slovenia opposed Croatia in 2008-2009 over a border dispute; Croatia has opposed Serbia since 2016 over questions of history and national minorities; Greece and Bulgaria opposed North Macedonia and Albania for more than 10 years over political, historical and linguistic disputes; and Hungary has opposed Ukraine over the question of national minorities and its wish to secure the unfreezing of EU funds.
The EU enlargement methodology updated in 2020 narrowed the room for manipulation in part: grouping the 35 chapters into 6 clusters reduced the number of opportunities to block progress.
Any intermediate step, however – approving the negotiating framework, for example, or assessing the criteria for opening clusters – still requires the unanimity of all member states.
This unanimity is an established practice, however, rather than a treaty requirement. Article 49 TEU governs only the decision on accession as such and sets no requirements for the intermediate stages: "The applicant State shall address its application to the Council, which shall act unanimously".
A move to QMV in enlargement policy therefore does not require amendment of the founding treaties or use of the passerelle clauses, which do not apply to enlargement. It can instead be effected through a procedurally simpler mechanism – revision of the negotiating framework with Ukraine and/or other candidates in the part concerning negotiating procedures. The decision on such a move would nevertheless still require the unanimous agreement of all member states.
In view of this, the most realistic option is a compromise scenario: applying QMV to the intermediate or technical stages of the negotiations while retaining consensus on accession itself and on ratification of the accession treaty. These are the models currently under discussion in the EU:
Some researchers warn that a move to QMV at the intermediate stages of accession could lead to the Accession Treaty being blocked at the end if member states and candidates fail to settle bilateral issues in good time. Another alternative could therefore be transferring part of the technical decisions on the negotiations to the European Commission, which would reduce the political element of the veto. Member states, however, are skeptical about strengthening the Commission's role in enlargement and security.
All the proposals considered remain at the level of discussion. The European Parliament consistently supports introducing QMV in individual areas, the European Commission is preparing a new enlargement strategy for September 3030, and EU leaders are due to discuss reform of the process at the October summit.
The ambiguous positions of member states and the coming electoral cycles, in particular the strengthening of right-wing parties in France and Germany, may nevertheless narrow the space for reform again. In the near term, therefore, the most realistic path is a gradual, sectoral move to QMV without amendment of the founding treaties in the areas where unanimity does the most damage to the EU's ability to act and to enlargement.
In contrast to the German idea of federalizing the EU and moving fully to qualified majority voting, the French model of a "multi-speed Europe" assumes different rates and depths of integration of states into the EU and into particular formats. This is already partly being realized within the euro area and the Schengen area.
The concept has been discussed in the EU since the 1980s. Under Emmanuel Macron's presidency it again became a central element of the debate on EU reform. In 2017 the Rome Declaration committed member states to "act together, at different paces and intensity" against the backdrop of preparations for the accession of the Western Balkans. The idea of a multi-tier EU gained new impetus in 2023 after the report of the Franco-German working group. The document proposed closer integration of the most ambitious states in the social, financial, industrial and defense spheres and identified four levels of participation:
In 2026 the concept took practical shape, in particular in connection with the need to provide Ukraine with a loan of 90 billion euros. The decision was supported by 24 member states, without the participation of Hungary, Slovakia and the Czech Republic. To avoid their veto, the countries used the enhanced cooperation mechanism provided for by Article 20 of the Treaty on European Union and Articles 326-334 of the Treaty on the Functioning of the European Union.
The mechanism allows at least nine states to deepen cooperation in particular areas and to adopt legally binding decisions that apply only to the participants in the initiative. It is used as a last resort, when EU-wide agreement cannot be reached within a reasonable period. Launching the mechanism requires a request to the European Commission, the support of the Council of the EU (at least 15 states representing 65% of the Union's population) and the consent of the European Parliament.
Such an approach can help avoid institutional paralysis during crises. At the same time, member states fear that excessive differentiation will weaken EU unity and turn the Union into an "exclusive club" run by Berlin and Paris, In practice, the "multi-speed Europe" model is most clearly visible in EU competitiveness, security and enlargement.
At the beginning of 2026, Ursula von der Leyen called on member states to make more active use of the enhanced cooperation mechanism amid pressure from the United States and China and debates about barriers to business created by differences in national rules. In the same period the finance ministers of Germany and France launched the E6 format, which brought together the six largest EU economies: Germany, France, Italy, Spain, Poland and the Netherlands, which together account for around 70% of the Union's population and GDP.
The format aims to strengthen EU defense, create a Savings and Investments Union (SIU), reinforce the euro and secure critical raw materials for the EU. Following a similar logic, the E19 format emerged at an informal meeting of EU leaders in February 2026, focused on completing the single market, regulatory simplification, lower energy prices and reviving trade between the countries.
The first result of the E6 was agreement among the participants on reforming financial markets after almost a decade of fruitless attempts: in May 2026 the "euro six" backed a stronger role for the European Securities and Markets Authority (ESMA) in supervising cross-border infrastructure and crypto-assets. The E6, however, needs the support of at least nine member states to secure a lecision at EU level. In addition, the closed format of the discussions requires greater transparency so as not to deepen the concerns of smaller.
The E6 also aims to accelerate the growth of defense spending and the implementation of industrial policy. The group's countries host the EU's most competitive defense companies, which creates potential for joint projects and for turning the defense industry into an engine of economic growth.
At the same time, states are not always able to agree even on joint defense programs. In June 2026, France and Germany decided to end joint development of the Future Combat Air System (FCAS) fighter and of next-generation tanks* because of disagreements between the contracting companies.
Differences also exist over which spending in the next EU budget should be cut in order to finance defense. This matters particularly because increasing defense spending in the new budget is one of the Union's priorities, yet the participants have so far been unable to reach consensus. At the same time, progress on joint procurement, security integration and coordinated investment could turn the format into a model of cooperation for states that are willing and able.
Differentiated integration also became topical in the context of EU enlargement in the spring and summer of 2026, primarily because of a potential 2027 deadline for Ukraine's accession that the Ukrainian and American sides discussed in the context of possible peace talks with the Russian Federation. The following models were considered:
The member states effectively rejected the first three models because of an unwillingness to support the Commission's initiatives and a wish to retain control over enlargement policy. The idea of associate membership, for its part, drew criticism in Ukraine as a possible form of "second-class" membership. At the same time, another Franco-German proposal based on the idea of "associate membership" continues to be discussed in the EU as a possible interim format for candidate countries.
For now, the EU and Ukraine continue along the path of merit-based accession, which is meant to prevent both the political "deception" of candidates and instant enlargement without the criteria being met. The process is held back by delays in opening negotiating clusters and by Ukraine's uneven progress on reforms.
Europe remains highly vulnerable to security threats, as the United States is increasingly focused on deterring China and on domestic security. Washington's intention to shift most of the responsibility for NATO's conventional defense capabilities onto European allies as early as 2027 does not match the pace of Europe's rearmament.
Despite the differences within the Alliance, Europeans continue to rely on it for defense while trying to close the specific capability gaps created by the reduced role of the United States. In response to the reduction of the American presence on the continent, the German-Netherlands army corps will take over command of NATO land forces in Estonia and Latvia. Italy will assume leadership of the Joint Force Command (JFC) in Naples, the United Kingdom will head the JFC in Norfolk, and Poland and Germany will lead the JFC in Brunssum in turn,
A separate issue remains the security of EU members that are not in NATO and are therefore not covered by its collective defense guarantee. These are Austria, Cyprus, Ireland and Malta. On March 2, 2026, for example, a Shahed drone attacked a British military base in the Republic of Cyprus, most likely launched by the terrorist group Hezbollah. This problem is also relevant in the context of Ukraine's accession: membership in the EU is regarded by Kyiv as a "soft guarantee" of security, but it is important to understand what exactly the EU is capable of and what the mechanism for fulfilling these commitments is.
The Treaty on European Union already contains a collective security mechanism. Article 42(7) obliges all EU members to provide a state that has suffered aggression with "aid and assistance by all the means in their power" – wording that is stronger than in NATO's Article 5 ("such action as it deems necessary"). Likewise, the threshold for applying the EU's Article 42(7) is lower ("armed aggression"), than in NATO's Article 5 ("armed attack").
Although applying Article 42(7) does not require a decision of the Council of the EU, the sequence of actions by member states and EU institutions remains extremely unclear. To address this problem, in May 2026 the EU held an exercise intended to work out a clear algorithm of actions in the event of a threat to a state that is an EU member but not a NATO member, and to establish what role the EU institutions should play in such a crisis. The results of the exercise remain non-public. Either way, the EU's Article 42(7) is complementary to NATO's Article 5 and will not replace it any time soon.
Current EU initiatives aim to improve the existing security architecture rather than to create an alternative to it. Permanent Structured Cooperation (PESCO) is aimed at strengthening cooperation between participants in the military industry. The EU Rapid Deployment Capacity is limited in size (5,000 troops), so it is not sufficient to defend Europe. Because of this, voices are again being heard today about the need to create joint European armed forces.
The main argument of those who support this idea is that a single army would be more effective than the 27 separate armies of EU member states and would not depend on US unpredictability. Yet no country has so far taken practical steps to create a military alliance of European states. Such an alliance would in any case be difficult to realize, because the European Union currently lacks a political center ready to bear responsibility for EU security, an integrated military command und significant rapid reaction forces as an alternative to NATO.
First of all, the European Union does not have the capacity to create and deploy a large army. To replace the American land presence on the eastern border, Europe needs a force of 100,000 troops.
Mobilizing that number will be extremely difficult, however, given the reluctance of populations in many EU countries to join the army and the need to keep part of their forces combat-ready under commitments within NATO.
In addition, the EU has almost no experience of taking part in full-scale combat operations: its 23 missions and operations under the CFSP mandate® are extremely limited in scale and have not been used for such a purpose. Delegating national armed forces to a joint European army would also contradict the constitutions of the neutral member states.
The European Union also has no command structures that would ensure successful management of joint forces. The EU's military strategic headquarters in Brussels is designed for only 60 permanent staff and 94 additional specialists who can be brought in for EU operations - hundreds of times fewer than NATO's equivalent personnel. The unwillingness to duplicate the command structures of the North Atlantic Alliance is one of the main obstacles to creating a European army.
Finally, unlike NATO, where leadership belonged to the United States, no European state is ready to take responsibility for the security of the entire European Union. Placing such responsibility on the European Council, where any state could veto the deployment of troops, would be ineffective.
Leadership could potentially be taken by a European Security Council – a grouping of 10-12 member states rotating in order to discuss the most important defense issues, including support for Ukraine?
Because creating such a structure requires transferring sovereignty on defense matters to a supranational EU body, however, European leaders view this proposal extremely critically. Member states do not trust one another on security, and they are also skeptical about the European Commission's ambitious new plans.
The creation of joint EU armed forces therefore remains a distant prospect. European states are searching for alternative formats to ensure their own security, but none of them is aimed at replacing NATO's role. Small alliances, the coalition of the willing, the "multi-speed Europe", PESCO and other initiatives are limited in nature and aimed at resolving specific issues. At the strategic level, Europe's defense remains institutionally in the same format as it was before the arrival of the second Trump administration, and only an existential threat to the EU could change that.
President Macron's March speech on the Europeanization of French nuclear forces provoked a vigorous reaction from partners and considerable media interest. To assess the scale of the change, however, it is worth recalling the constant elements of French doctrine. At its origins stood President de Gaulle and the four "generals of the apocalypse", who sought independence from alliances and from American nuclear guarantees.
Historically, the doctrine concerned exclusively French vital interests and the territory of France. For this reason it would be incorrect to state that French and British nuclear capabilities are "European" nuclear deterrent forces. French nuclear doctrine, however, defines neither "vital interests" nor the perimeter of weapons use precisely, and ambiguity is its central element.
It is defensive in nature and intended to deter enemies from aggression. Specific scenarios for the use of nuclear weapons are not set out, but French analysts point to extreme cases: the threat of occupation or strikes against the political leadership. Limited attacks, including drone strikes, are not grounds for a nuclear response. At the same time, nuclear deterrence must rest on strong conventional forces.
As a state with a small nuclear capability, France is compelled to rely on strategic ambiguity as an important component of its own nuclear doctrine. Its arsenal is designed to make it possible to inflict irreparable damage on an enemy and deter it from attacking (deterrence of the strong by the weak).
The French nuclear forces have two components: the sea-based one – four Triomphant-class submarines that are permanently at sea – and the air-based one, represented by Rafale strategic aviation. A full closed cycle of nuclear weapons production gives France independence from the United States and other partners".
French nuclear deterrence has thus traditionally been sovereign and has deliberately preserved strategic ambiguity. Its "Europeanization" therefore does not mean turning French weapons into a common European guarantee.
Macron's idea is not new. In 1992, François Mitterrand first announced the possibility of creating a European nuclear doctrine; in 1995, Jacques Chirac mentioned the European dimension of French vital interests; and Macron himself invited partners to a dialogue on the European dimension of French deterrence back in 20208.
Now, however, these statements have provoked keen interest because of the transformation of the European security architecture: Russia's full-scale war against Ukraine, the reorientation of the United States toward domestic security and the Indo-Pacific region, and the growing activity of Russia and China in the nuclear sphere. Europe has again faced existential risks, and nuclear weapons have returned to the center of the debate as a guarantee that conventional means cannot fully replace.
In this context, Macron's 2026 speech is not a departure from the previous doctrine but rather its adaptation to new threats. The president suggested that France's vital interests may encompass a broader context, including the security of allies. There is no clear definition of these interests, however: events beyond France may affect them, but this does not mean a guaranteed nuclear response.
To implement this approach, Macron outlined a strategy of forward deterrence – closer cooperation with a limited number of European partners. This means regular consultations, the possibility for allies to take part in nuclear exercises and the potential deployment of Rafale aircraft on partners' territory (forward deployment). For the first time since 1992, France will stop reducing and will begin building up its nuclear forces, while the number of warheads will be classified.
These changes have clear limits, however – they broaden the political horizon of French deterrence but do not change its institutional architecture. France retains full sovereignty: the decision to use its nuclear forces is taken by the French president alone. There is no question of sharing powers with partners, joint nuclear planning, multinational command structures or the transfer of French nuclear forces to allies.
This creates a certain asymmetry: partners can take part in consultations, exercises and the potential deployment of French capabilities, but do not influence the decision on their use. This approach differs from the American doctrine, under which countries can agree or refuse to allow the use of nuclear weapons from their territory. American B61 thermonuclear bombs are currently deployed in Belgium, Germany, Italy, the Netherlands and Türkiye.
In addition, since the new doctrine does not clearly define vital interests, it contains no binding guarantees of protection for allies or of the use of nuclear forces if they are attacked. Closer cooperation may give rise to inflated expectations that Paris may not meet: "partners may begin to believe that they have a place at a table where in reality there is only one chair". Unpredictability, which strengthens deterrence of an adversary, thus at the same time complicates interaction with partners.
The initiative is also limited by the narrow circle of states involved. Macron outlined a group of only eight countries with which France will deepen cooperation: the United Kingdom, Germany, Poland, the Netherlands, Belgium, Greece, Sweden and Denmark. The partners do not include the Baltic states, Finland or Norway, which border Russia and are under direct threat.
In the absence of joint command structures, the new doctrine deepens France's bilateral cooperation with partners, turning it into a multi-bilateral frameworks. The United Kingdom and France have long cooperated in the nuclear sphere under the Lancaster House Treaty (2010) and the Northwood Declaration (2025), which provide for mutual assistance, intelligence sharing and the coordination of nuclear deterrence through the Nuclear Steering Group. The central role among the new partners will be played by Deutschland: in March, President Macron and Chancellor Merz announced the creation of a Franco-German nuclear steering group. Merz also announced the participation of German forces in French nuclear exercises in 2026.
The French initiative is not, however, an alternative to American deterrence. French nuclear forces remain complementary to the American ones and are not intended to replace the American nuclear umbrella. Paris is preparing rather for scenarios in which US involvement could be delayed or run into political obstacles, but it does not replace the United States or NATO and does not create "European nuclear forces"; it complements the existing architecture.
Tellingly, the United States has responded to the initiative with restraint overall. Since the French approach does not undermine American extended deterrence, it could have been seen as a step toward greater European responsibility for its own security. Support for the French initiative from Germany, Belgium and the Netherlands, where American nuclear warheads are deployed, indicates that dialogue with France does not contradict their commitments within the Alliance.
The logic of complementarity also explains France's reluctance to join NATO's Nuclear Planning Group (NPG). France did not become a member even after returning to NATO's command structures in 2009, which had symbolic significance for preserving nuclear independence. Participation in the NG remains a domestically sensitive subject, since it could be interpreted as a "transfer" of the nuclear role to the EU or to Germany.
In France, Macron's speech was received favorably overall across different parties, since he preserved the Gaullist independence of sole decision-making and ambiguity, whereas a clear definition of vital interests could have provoked political resistance. He also did not name a specific scale for a possible increase in the arsenal, in part so as not to attract criticism from voters who favor banning nuclear weapons. Building up the forces without publishing the number of warheads simultaneously reduces domestic resistance, sends a signal to enemies and denies them a specific figure to work from.
At the same time, support for Macron's speech does not guarantee the durability of the new nuclear policy. After the 1990s, interest in nuclear matters declined and nuclear weapons ceased to be seen as a key security guarantee. After the 2027 elections, a new French president may declare a different doctrine. In addition, the return of nuclear weapons to the center of the debate may reduce the population's readiness for a conventional clash if citizens see them as a sufficient security guarantee.
Changes to French nuclear doctrine thus have above all a communication effect. Greater ambiguity about the scope of nuclear weapons use, a possible increase in the number of warheads and a potential widening of the geography of their deployment are meant to make French deterrence less predictable for enemies. This approach also requires the fewest resources, since it demands no additional defense spending, is acceptable to different political forces in France, preserves its sovereignty over its nuclear forces and gives partners no binding guarantees. At the same time, French nuclear doctrine creates more uncertainty for partners, entrenching two contradictory imperatives: Europeanness and sovereignty.
The publication is prepared under the project “Strengthening the Analytical Capabilities of the Foreign Policy Decision-Making with the Civil Society” of the Centre for International Security with the support of the Konrad-Adenauer-Stiftung Ukraine.
The full text of the document is available for download as a PDF.
