ABSTRACT

The President of the Republic, Emmanuel Macron, will not be able to stand for re-election in 2027. This analysis argues that ‘Macronism’ has never constituted centrism in the doctrinal sense. None of the criteria of historical centrism have been met by the practices observed since 2017; rather, there has been a centralisation of power. Whilst the so-called ‘central’ bloc is disintegrating electorally, this centralisation is shifting towards institutions, enacted laws and European decisions.

Article 6 of the Constitution limits the presidency to two consecutive terms: for the first time under the Fifth Republic, a Head of State is effectively bound by this restriction. The public debate has raised the following question: Who will succeed Emmanuel Macron? However, a change of government alters the holder of executive power but does not, overnight, change those who sit on the supervisory bodies, nor the laws already enacted – provided the other candidates do not appear willing to repeal them – nor the decisions being taken in Brussels. The second question that then arises is this: what will remain of this policy once its architect is no longer president?

A ‘centrism’ that is not really centrism

It seems necessary, first of all, to examine the term used, as ‘centrism’ historically refers to a distinct political tradition: representative democracy, a culture of parliamentary compromise, and a balance between economic freedom and social protection. Let us compare these criteria with the practices observed since Emmanuel Macron came to power.

Representative democracy? Major strategic shifts (support for the conflict in Ukraine, the reorientation of the doctrine of nuclear deterrence) have been announced by the Head of State in speeches, prior to any parliamentary debate. No referendum has been held since 2005, not even on the law on assisted dying, which nevertheless touches on French society’s conception of the human person.

Parliamentary compromise? Article 49(3), which allows a bill to be passed without a vote by MPs, has been used 23 times in twenty months by Élisabeth Borne’s government alone – a frequency unprecedented in the Fifth Republic. Even more telling: the reform of the Labour Code was driven through by executive orders as early as 2017, even though the government held an absolute majority in the National Assembly. Bypassing parliamentary debate was therefore not a matter of necessity, but of choice. The most significant precedent remains that of the health crisis, which from March 2020 was managed by a ‘Health Defence Council’ whose composition was determined solely by the President, whose deliberations were subject to national defence secrecy, and which was under no obligation to report to Parliament, even though Article R. 1122-1 of the Defence Code does not confer any health-related powers on this body.

Economic and social balance? Abolition of the wealth tax (ISF), a single flat-rate levy of 30 per cent on capital income, two rounds of tightening of unemployment benefit rules, and raising the state pension age to 64. Each measure eases the burden on capital or reduces a form of protection. None of them goes in the opposite direction.

The French people themselves have made their choice: in the Ipsos-CEVIPOF election survey of September 2026, 51 per cent classified Renaissance as ‘right-wing’, compared with only 21 per cent who classified it as ‘centrist’. We must therefore distinguish between two concepts that the word conflates: centrism as a doctrine, which implies transferable principles independent of individuals, and centrality as a position of power, which consists of occupying the political space.

The electoral bloc is breaking up

The facts speak for themselves. A study by the Jean-Jaurès Foundation in May 2026 found that, among voters who voted for Emmanuel Macron in 2022, only 35 per cent remain loyal to the so-called ‘centrist’ bloc. Among them, support is split between Édouard Philippe, Gabriel Attal, Raphaël Glucksmann and Bruno Retailleau, whilst 11 per cent now say they intend to vote for Marine Le Pen. When considered in relation to the entire electorate, none of these candidates alone exceeds 17.5 per cent in the first round, whilst Marine Le Pen stands at between 33 per cent and 35 per cent in all the scenarios tested.

The two main successors are, moreover, keen to distance themselves from the record: Gabriel Attal, for example, describes the pension reform spearheaded by Édouard Philippe as “brutal” and “ineffective”. Macronism, which has no doctrine, produces not a successor but rivals.

As for the centre, its focus has shifted

Indeed, the analysis must be approached from a different angle, as public policy does not depend solely on those who decide it; it also depends on those who implement it. The appointments to be made in 2026 therefore deserve our full attention.

Richard Ferrand, former Secretary-General of En Marche (Renaissance), will chair the Constitutional Council until 2034. He will therefore announce the results of the 2027 and 2032 presidential elections. Amélie de Montchalin, the current Minister for Public Accounts, was appointed the first President of the Court of Auditors in February 2026. This is the first time under the Fifth Republic that a budget minister has moved directly from the Ministry of Finance to the institution responsible for auditing public accounts. Marc Guillaume, former Secretary to the Government, became Vice-President of the Council of State in May 2026. Emmanuel Moulin, Secretary-General of the Élysée, has been tipped for the Bank of France until 2032.

None of these appointments is irregular; all comply with constitutional procedures. Nevertheless, the bodies concerned are those responsible for overseeing the exercise of power: ensuring the legality of laws, monitoring public funds, and annulling acts of the executive. Furthermore, their terms of office, by their very nature, extend beyond that of the President who appointed them. When questioned about these appointments, the Presidency justified them on the grounds of a desire to ‘secure the institutions’. The implications of this phrase warrant careful consideration: secure against what, and for whose benefit?

The same logic applies to legislation. To take just one recent example, the End-of-Life Act, passed in July 2026 after three rejections by the Senate and despite the absence of any referendum on a subject that touches on society’s conception of the human person and brings about a paradigm shift in civilisation. As yet, no candidate has committed to repealing such legislation.

Then, with regard to emergency powers, there are two distinct regimes which must not be confused. The first stems from Article 47 of the Military Programming Act of 1 August 2023, which has become Article L. 2212-1 of the Defence Code: it concerns the requisitioning of persons and property and is based on a threat that is merely foreseeable (rather than current or imminent), which can be invoked in advance at the sole discretion of the executive. The second stems from Article 21 of the amending Act promulgated on 16 August 2026, which establishes the ‘state of national security alert’: it requires a ‘serious and present’ threat, but in three scenarios of considerable scope, one of which reproduces the 2023 wording almost verbatim and another which refers to the deployment on national territory of ‘allied forces’, a concept which the text does not define. Its effects are far-reaching: derogation from regulatory standards in six areas and, above all, exemption from all consultation obligations, including those vis-à-vis the Council of State. It is triggered by a decree of the Council of Ministers, without prior notification to Parliament, whose authorisation is required only after a period of more than two months, with no limit on the number of extensions.

As these two regimes are partly based on the same activation mechanism, they can be triggered simultaneously on the basis of a single threat. With the state of emergency, the state of alert and general mobilisation, France now has five states of emergency that can potentially be applied concurrently, two of which can be activated on the basis of the same circumstances. None of these measures will expire in May 2027; they will remain available to the next elected president. Which presidential candidate proposes to repeal these regimes?

The European level, where national control no longer applies

The most sensitive issue is that of nuclear deterrence. On 2 March 2026, speaking from Île Longue, Emmanuel Macron announced a major shift in French doctrine (‘advanced deterrence’), providing for the temporary deployment of strategic air forces to allied countries and a nuclear steering group with Germany, which nine states have agreed to. The President expressly ruled out any sharing of the final decision: it would therefore be inaccurate to speak of a pooling of France’s nuclear arsenal. The constitutional issue remains unresolved nonetheless. A deterrence doctrine commits the very existence of the nation. Yet its modification was decided by a single man, announced in a speech, without a parliamentary vote or prior debate. Stationing strategic assets on the territory of third countries automatically increases the likelihood that France will be drawn into a confrontation it would not have chosen. This decision, which commits future generations, is at the discretion of a Head of State whose term of office ends in less than two years.

A third risk deserves attention, as it concerns the last area reserved for the Member States. Article 4(2) of the Treaty on European Union expressly stipulates that national security remains the sole responsibility of each Member State, and the common defence policy requires unanimity in the Council, precisely because the Member States have never intended to place this area under Community jurisdiction. However, in her State of the Union address on 16 September 2026, Ursula von der Leyen asserted that Europe was capable of defending itself because ‘we have thoroughly rethought the security of its Member States’, before concluding that ‘the only sovereignty that matters is that which Europe offers to Europeans’.

The phrase is not merely a figure of speech. Sovereignty received from a higher authority is not sovereignty at all, but a revocable delegation; the very essence of a sovereign is that he derives his power from no one. Admittedly, a speech has no normative value, and the initiatives announced will have to be adopted unanimously. Nevertheless, transfers of competence in the field of defence no longer take place through a formal treaty subject to ratification: they occur through an accumulation of technical instruments, joint funding, industrial programmes, joint procurement and interoperability standards. Each is presented as a voluntary form of cooperation. Taken together, they create a dependency that renders decision-making autonomy theoretical. A state that no longer controls the funding, production or standards of its armaments retains the formal sovereignty to decide, but loses the means to decide otherwise.

This is the same mechanism observed in the case of digital identification: after being struck down by the Constitutional Council in Paris in August 2026, the measure banning minors from social media was immediately taken to the European level, where French constitutional review no longer applies.

An issue that the French debate has so far failed to address

Finally, the European timetable deserves closer scrutiny. António Costa’s term as President of the European Council ends in late May 2027, a few weeks after the expiry of the French presidential term. However, Article 15(5) of the Treaty on European Union stipulates that this President is elected by the European Council by a qualified majority: there is no universal suffrage, no vote by the European Parliament, nor even a requirement for the candidate to have the support of their national government. Donald Tusk was thus re-elected in 2017 even though his own Polish government voted against him.

The obstacles are real, given that the four successive holders of the post were all former heads of government of relatively small states, and the role was conceived as a mediating presidency. There is no evidence to suggest that such a plan exists, but the possibility is legally open.

The horizon is not 2027

Article 6 of the French Constitution prohibits only two consecutive terms of office; Emmanuel Macron could therefore perfectly well stand for election in 2032, at the age of fifty-four. A former president who can legitimately return to office has no reason to allow the positions he helped to establish to wither away, and most of those appointed this year will still be in office by that date.

A change of government alters who holds presidential power without, however, changing the composition of the Constitutional Council, nor necessarily the laws enacted or the negotiations underway in Brussels. ‘Macronism’ could therefore disappear as a political identity whilst surviving as a culture of government – which would be the most difficult form of legacy to perceive, as it cannot be measured by the number of MPs.

The question that the next political cycle will have to address goes far beyond the person of Emmanuel Macron himself: can we truly turn a new page when those who wrote the previous one remain in the positions that shape public policy?

This analysis summarises the study ‘After Macron: what will survive beyond 2027’, by the same author.