Dear colleagues,

Dear Judge François Biltgen,

Ladies and gentlemen,

It is a real pleasure to be here with you today.

Let me begin by warmly thanking the organisers for the invitation to contribute to this important conference. I am honoured to be here.

The question you have chosen – is EU law fit for purpose? – is vast. I will not pretend to exhaust it. Instead, I hope to offer a view from where I stand, leaving the nuances – and perhaps some of the disagreements – to our discussion.

It is also a question both timely and necessary.

It is timely because our Union is being called upon to respond simultaneously, to war at its borders, migratory pressures, climate change, digital transformation, and geopolitical competition – as well as to the uncertainty brought by powerful new technologies, from artificial intelligence to online platforms whose influence on our democracies and our way of life is already plain.

It is necessary because, in such a context, there is always a temptation to ask more of the law: can it move faster, be simpler, remove friction – and, above all, deliver more quickly on its promises?

These are, of course, legitimate goals.

But if effectiveness is necessary, it is not sufficient. A perfectly effective law could be oppressive. Conversely, a law embodying admirable principles may prove ineffective in practice.

The question, then, is not simply whether a law achieves its intended objective. We must also ask whether it does so in a way that respects rights, the principles of good administration – legality, accountability, transparency, participation, and fairness – and the dignity and agency of the people affected by it.

There is a difference between the values that underpin a legal system and the reality of their implementation. The Ombudsman operates precisely in that space between principle and practice. We do not assess the political wisdom of policy choices, nor do we act as a court. A complaint does not prove that a law is deficient. It is evidence – evidence that obliges us to examine whether the legal system is delivering what it promises. And sometimes we must conclude that an institution fell short of what a principle requires –  that it should have done more. That is maladministration.

In 2025, we opened 492 inquiries, the vast majority of which concerned the European Commission and EU agencies. They ranged from public access to documents, infringement procedures, grants, contracts, and staff matters, to the adoption of binding acts and soft law in fields such as public health, fundamental rights, and artificial intelligence.

In 63% of the cases we closed, the institution concerned settled the complaint. We found maladministration in just over 4%.

Surprisingly few, you might say. But numbers tell only part of the story. The substance lies in the cases themselves. I will share a few that illustrate where the real difficulties lie.

My overall reflection is this: EU law and its principles are, in many important respects, fit for purpose.

But their effectiveness depends decisively on administrative practice. Too often, the principles of good administration are seen as obstacles to effectiveness and efficiency, rather than what they truly are: indispensable safeguards of public trust – never more so than in times of emergency.

And once principles are viewed as obstacles, the temptation to compromise on them is real. There is also a quieter tendency: to honour a principle only where the Court of Justice has already spelt out, in unambiguous detail, how it must be applied – and to invoke the lack of legal certainty, or of case law, as a reason to lower the standard of diligence.

Ladies and gentlemen,

Let me start with a recent set of cases concerning the European Commission’s use of derogations from its Better Regulation rules when preparing legislative proposals it considered to be urgent.

These inquiries concerned three very different fields: corporate sustainability due diligence, countering migrant smuggling, and the Common Agricultural Policy. In each case, the Commission invoked urgency to omit steps ordinarily foreseen in its Better Regulation framework, notably impact assessments and public consultations. In two of the cases, there were also concerns about compliance with the European Climate Law.

Why does this matter for today’s question?

Because although the Better Regulation rules have been adopted by the Commission internally, they are not simply administrative housekeeping. They give effect to core principles of EU law-making recognised in the Treaties: openness, transparency, and citizens’ participation. They are also linked to the principle of proportionality, because they help ensure that EU legislation is necessary, reasoned, evidence-based, and non-arbitrary.

Effectiveness matters – as I said at the outset – and so does efficiency. The Union’s ability to address crises more effectively than Member States acting alone has always been one of the strongest sources of its legitimacy.

But the pursuit of speed cannot hollow out the principles established by the Treaties and developed by the Court of Justice. They, too, sustain EU legitimacy and public trust.

In my inquiries, I found several procedural shortcomings. There was no clear definition of what counted as ‘urgency’, nor of the circumstances in which the Commission can derogate from its own rules. Records of derogations, their justifications and climate consistency assessments were insufficient, making it difficult for citizens to understand and assess whether the Commission had acted in a non-arbitrary manner. The so-called ‘analytical documents’, which replace full impact assessments, were published late. Stakeholder consultations were not sufficiently broad or balanced. In one instance, an inter-service consultation was shortened to less than 24 hours over a weekend. Taken together, those shortcomings amounted to maladministration.

Let me be precise. The Commission must be able to react quickly to crises, emergencies, and novel challenges. Of course it must. I also recognise the discretion of the Commission and of the Council in deciding when urgency exists. But law-making must remain predictable for those who wish to exercise their democratic right to influence it. Urgency may justify a faster procedure. It cannot justify an arbitrary one.

This is the heart of the matter. If principles seem to stand in the way, it does not mean that they are no longer “fit for purpose”. It means that the administrative rules and mechanisms giving effect to them must be improved – and where necessary rebalanced.

That is why I recommended that the Commission define urgency more clearly, properly record and explain its derogations, and ensure that even accelerated law-making preserves the essential features of a transparent, evidence-based, and inclusive process. Where discretion expands, the duty to give reasons must expand with it.

The Commission’s reply was constructive, but its commitments remain too general. I will continue to monitor this area, not least because incoming complaints suggest a broader structural challenge.

Indeed, I have just opened a new inquiry into how the Commission consulted the public on carbon capture and storage technology – including whether it asked about its economic, social, and environmental impacts - and another into its decision to exempt so-called ‘Reality Checks’, an important tool of stakeholder engagement, from its standard transparency measures.

My second example is public access to documents. Regulation 1049 was designed twenty-five years ago for a world of paper trails and physical archives. Today, negotiations may happen by text message, information lives in databases, and decisions are increasingly shaped by algorithms.

The question is simple: can a framework built for the past still deliver accountability in the digital age?

The legal definition has endured. A document is any content, whatever its medium: content decides, not medium. The difficulty lies in practice. A document is not its registration. A document is not its retention. Content exists the moment it is created; registration is merely an internal act of filing; retention is how long that content is kept. When institutions treat registration as a precondition for a document’s existence, they may create accountability-free zones, where decisions leave no trace.

We have seen this in practice. For example, emails underlying Commission reports have been treated as falling outside the Regulation, then deleted under a retention policy, or searches for text messages have been limited to messages that had already been registered. In all of these cases, my Office found maladministration, and in relation to the text messages the General Court already recognised an obligation to search for them.

And one point admits no compromise. Once content has been requested, it must be preserved – whatever the institution later concludes about whether it is disclosable. That conclusion can be debated, appealed, reviewed. What cannot happen is for the content to disappear while the debate is under way.

The same tension goes further. When information sits in a database, access may depend on what the institution chose to programme its systems to retrieve. When AI assists a decision, there may be no document to request at all – and the training data and specifications may be held by contractors, potentially placing them beyond the Regulation’s reach. Indeed, the right of access is quietly migrating from a document-based logic to an information-based one.

This is also why a purely reactive model no longer suffices. How can you request a text message you do not know exists? Proactive transparency shows the way: Commissioners’ travel expenses, trilogue documents made available while negotiations are live.

Yet progress remains fragile. Our recent inquiry found that the Council and the Commission still do not fully comply with clear case law on legislative documents, invoking generic concerns that the Courts have repeatedly rejected. Once case law is settled, failing to apply it is no longer merely a transparency problem. It risks becoming, at least in part, a rule of law problem.

And there is a deeper shift at work. Regulation 1049/2001 rests on a presumption of openness: access is the rule, refusal the exception, and every exception must be justified. In practice, however, requests are increasingly handled through the lens of risk management.

And the burden has shifted with it. Citizens are now expected to find and cite court judgments, Ombudsman decisions, and internal guidelines simply to make their case. Deadlines are missed and documents arrive when they no longer matter. Those who persist must turn to the Ombudsman or the General Court, routes that can be slow, costly, and complex. No one should have to become an EU law expert to exercise a fundamental right. Openness that must be litigated, document by document, is not the openness the Treaties promised.

Regulation 1049/2001 has proved resilient. But resilience is not the same as effectiveness. A framework that depends on case-by-case litigation to decide whether a text message is a document, or an AI output is accessible, is not one citizens can rely on in advance.

The task is not to weaken a right that has served the Union well. It is to make sure that procedures and practice catch up with a reality that has already arrived.

Let me now step back, because these cases share a common thread. The problem is not necessarily the quality of the substantive law. It is the quality of the administrative machinery through which that law is made and applied.

Take Better Regulation: impact assessment, consultation, evidence-based policymaking – the principles are there. In access to documents, transparency is firmly established. Yet in both, difficulties persist. The principle may be transparency; the question is how transparency actually works when a citizen asks for a document. The principle may be better regulation; the question is how it actually shapes the preparation of legislation.

So, I would distinguish three things: the law, the codification of the administrative procedures that give it effect, and their implementation in practice. Often, the principle is sound. What fails, however, is the procedure – insufficiently clear, predictable, or consistent.

And that matters, because administrative procedures are not mere technical details. They are essential mechanisms through which citizens experience the rule of law. If an institution has discretion but the criteria for exercising it are unclear; if procedures exist but are not documented; if citizens cannot understand why a given process was followed (or not) and why a decision was taken – then we have a problem of good administration, even where the legislation is perfectly sound.

Good administration is, in many respects, the infrastructure that allows good law to become reality. The rule of law does not end where the legislation ends; it continues in the procedures through which public power is exercised.

Much of what my Office examines is not hard law, but soft law and administrative practice: Better Regulation guidelines, transparency frameworks, ethical codes, conflict-of-interest rules. They are not binding in the strict sense, yet they shape how power is exercised. At best, they embed a culture of justification and rules-based administration. At worst, they become vehicles for flexibility but without accountability.

What we witness, then, is rarely a formal rejection or betrayal of law. It is more subtle: it reflects a gradual relaxation of procedural discipline; the normalisation of urgency; the indiscriminate use of flexibility.

Good administration procedures cannot rest on soft law alone. Yet, the Union administrative law has grown by accretion – rule by rule, sector by sector, agency by agency, case by case. As actors have multiplied and procedures have become more complex, no comprehensive, horizontal legislation has emerged. Thus, the result is a patchwork: rich in principles, poor in common procedure.

And it is the citizen who pays the price. The rights guaranteed by Article 41 of the Charter – to be heard, to reasoned decisions, to access to one’s file, and to have a decision within a reasonable time – can be experienced differently depending on which institution, agency or sector one happens to encounter.

The remedy is not new. Article 298 of the Treaty provides the legal basis for an open, efficient, and independent European administration. When the European Parliament endorsed the European Ombudsman’s Code of Good Administrative Behaviour in 2001, it already asked for binding rules. In 2013, it called for a law of administrative procedure, and in 2016, it put forward a full draft regulation. Scholars – some of them here in Luxembourg – gave us the ReNEUAL Model Rules, showing that codification is feasible, even if discussable.

I believe the time has come again for us to imagine what a single, horizontal code of EU administrative procedure could do. Not as a monument to bureaucracy, nor a straitjacket for sectoral specificity, but as a common floor.

Clear rules on the right to be heard and the duty to give reasons. On time limits. On record-keeping, so that no decision escapes review because nothing was written down. On what urgency permits – and what it never excuses. On the use of automated and AI-assisted systems. Procedures that are clear, accessible, and predictable – for those working inside the institutions and, crucially, for the citizens affected by their decisions.

Such a code would not slow the Union down. Predictable procedures are efficient procedures. Resilience and legality are not competing projects. They are interdependent.

This is where I turn to you. Codification of this kind has always advanced on the strength of academic work - of scholars willing to map the fragments and imagine the whole. The project is old. The need to debate it remains. I dare to invite you to help bring it back to the table.

Dear colleagues,

So, is EU law fit for purpose? In terms of its principles, largely yes. The real question is whether our legal and administrative frameworks continue to discipline discretion – requiring reasons, transparency and proportionality – or whether they become managerial tools subordinated to expediency.

If flexibility becomes the underlying principle, then democracy and legality suffer together. The remedy lies in clearer rules, sound record-keeping, proper training, and above all a public service culture that sees legality and good administration not as constraints on efficiency, but as indispensable safeguards of public trust.

Overall, institutions engage seriously with our recommendations and many use them to improve their practices. Yet compliance remains slower and more contested in some areas – transparency above all. What matters most is whether recommendations are treated as criticism to be managed, or as part of a shared effort to strengthen EU administration.

I am committed to that effort, as European Ombudswoman. But no administration can be the sole judge of its own conduct. That is why your voice matters.

And perhaps the real question is not just: is EU law fit for purpose? But is it fit for whose purpose? A law may be highly effective from the perspective of an institution pursuing a given objective, but be experienced very differently by the citizen subject to it.

From where I stand, a law is fit for purpose when its legitimate aim, its legal design, its administration and its effects remain aligned with the rights and interests of the people it governs.

The citizen should not be the afterthought of legislation. The citizen should be the measure against which we continually test whether the law is working.

I look forward to our discussion today. I count on your engagement tomorrow.

Thank you for your attention.