The revised Deposit Guarantee Schemes Directive and the quiet expansion of soft law

This piece examines how the European Banking Authority, one of the European Union’s financial supervisory agencies, is using its soft-law powers to harmonise national deposit guarantee schemes under the newly revised Deposit Guarantee Schemes Directive. Its aim is to show that the Authority’s expanding role rests less on binding technical standards than on formally non-binding Guidelines, and that this is where the harmonising work is being done. The point reaches well beyond deposit insurance: it is a live illustration of how European Union agencies shape the single market through instruments that are, on paper, not binding at all, yet in practice prove decisive.

A common promise, an uneven framework

Every deposit guarantee scheme rests on a promise that most savers never think about until the day it matters: if their bank fails, their money, up to one hundred thousand euro, is safe. That promise is what stops trouble at a single bank from turning into a run on many, and it is one of the quiet foundations of financial stability. The collapse of confidence that spread through European banking after 2008 showed how quickly that foundation can crack, and how unevenly national safety nets were built to withstand the strain.

For more than a decade the European Union has tried to make deposit protection work the same way in every Member State. The current framework, Directive 2014/49/EU, took a large step in that direction, harmonising the coverage level, funding obligations and repayment deadlines. Yet it left much to national discretion: the way schemes are organised, how they are governed, how they interact between each other, and above all how they may intervene. The result is a framework that is common in its headline guarantees but still fragmented in its machinery.

The revised Directive, Directive (EU) 2026/804 of 30 March 2026, published in the Official Journal on 20 April 2026 and applicable from 11 May 2028, is the latest attempt to close that gap. It forms part of the European Commission’s broader reform of bank crisis management and deposit insurance, whose headline ambition is to let deposit guarantee funds be used not only to repay depositors but to support the resolution of failing banks and to prevent failures in the first place. Less remarked upon, but no less significant, is what the reform does to the role of the European Banking Authority, and the quiet instrument through which that role expands: soft law.

Harmonising through soft law

The European Banking Authority’s regulatory toolkit is worth closer examination, because the Authority acts in more than one way. Before turning to the revised Directive, it is useful to distinguish the two types of power it holds: its hard law powers and its soft law powers. Under its hard law powers it drafts regulatory and implementing technical standards, detailed rules that become binding across the Union. Crucially, it only drafts them; it is the European Commission that formally adopts them, as delegated or implementing acts. Although the Authority formulates the technical standards, the power of formal adoption rests with the Commission

Under its soft law powers, by contrast, the Authority acts in its own name. Article 16 of its founding Regulation, Regulation (EU) No 1093/2010, empowers it to issue Guidelines addressed directly to national authorities and to the schemes themselves. These Guidelines are, formally, not binding. But they carry real weight through what is known as the “comply-or-explain” mechanism: an authority that does not intend to follow a Guideline must say so publicly and give its reasons. That duty is reinforced by the reputational pressure of being named as non-compliant, a mild but effective form of peer discipline. It is, in short, a soft power designed to produce hard results.

With that distinction in mind, the shape of the revised Directive comes into focus and what is striking about the revised Directive is how heavily it leans on the soft law channel. The Directive confers eleven distinct mandates on the Authority, and the parallel revision of the Bank Recovery and Resolution Directive adds one more: twelve in total, to be delivered in three batches ahead of the Directive’s application in 2028. According to the Authority’s own roadmap for delivering these mandates, eight of the twelve take the form of Guidelines; only four are binding technical standards. Two-thirds of the Authority’s new workload, in other words, is soft law.

This points to an important characteristic of the harmonisation process. The choice of instrument appears to depend on the nature of the subject matter: where a single, uniform rule is required, such as the content of the depositor information sheet or the treatment of funds held in client accounts, the legislator relies on binding technical standards; where the subject matter is more variable, more dependent on national arrangements, or more reliant on operational judgement, it is left to Guidelines. Soft law becomes the tool of choice precisely in the areas where the political negotiations behind the Directive could not themselves reach a common solution, and the task of building convergence is, in effect, delegated to the Authority to pursue afterwards through softer means.

This reliance on soft law is neither an accident nor a novelty. Under the 2014 Directive the Authority had already made Guidelines its principal harmonising instrument, issuing them on deposit protection matters as varied as stress testing, payment commitments, the calculation of contributions, cooperation between national schemes, and the delineation of a scheme’s available financial means. Several of these were issued on the Authority’s own initiative, without any express instruction in the Directive to do so.

The revised Directive gathers these threads and gives them firmer footing. What were often self-standing initiatives now becoming explicit legal mandates: stress testing flows from Article 4(13), the calculation of contributions from Article 13(3), and cooperation between schemes from Article 14(9). The change is subtle but telling. Soft law that the Authority once volunteered, the legislator now requires, lending the instrument greater legitimacy while confirming that it remains the preferred vehicle for convergence.

If those mandates consolidate familiar ground, others break new ground entirely, and here the link between soft law and the reform’s central aim is clearest. Because the revised Directive opens deposit guarantee funds to use in resolution and prevention, the Authority must now write Guidelines on the conditions under which the cap on a scheme’s contribution to resolution can be lifted, on the use of scheme funds in preventive measures (Article 11a), on the business reorganisation plans that accompany such support (Article 11c), and on how schemes should invest their money prudently in the meantime (Article 10(13)).

Each of these areas involves difficult trade-offs between protecting depositors, preserving financial stability and avoiding the use of public money. In each, the legislator has laid down a principle and left Member States some room to manoeuvre. And in each of them, it is soft law that is trusted to draw them together, a striking vote of confidence in an instrument that carries no formal binding force.

Soft law in action and its limits

These new responsibilities are not distant prospects: the method is already on display. On 29 June 2026 the Authority published its roadmap; on 23 July it opened the first four public consultations, running until 23 October, with a public hearing held on 24 September. Three of the four consultations refer to binding technical standards, on depositor information (Article 16), on information exchange between banks, schemes and authorities (Article 16a), and on the treatment of client funds (Article 8b). The fourth, revealingly, is a set of Guidelines on the investment of available financial means (Article 10(13)).

That fourth document shows the technique in miniature. The draft Guidelines fix common standards for diversification, low risk and liquidity, so that a scheme’s funds can be mobilised quickly when a bank fails. Yet they are expressly designed, in the Authority’s own words, to build on existing practices among European Union schemes and to apply a proportionate approach, preserving flexibility to align the new rules with what schemes already do. This reflects a recurrent trade-off in soft law between the degree of harmonization required to align national frameworks and the flexibility needed to accommodate arrangements already in place This is the same trade-off that produced a high level of compliance under the 2014 Directive, notwithstanding the non-binding character of the Guidelines.
Whether this reliance on soft law instruments proves effective is borne out by the compliance record rather than assumed, and the experience of the past decade suggests that it genuinely works. Across the Guidelines issued under the 2014 Directive, the large majority of national authorities either complied or declared their intention to comply, and the Authority repeatedly revised its Guidelines in response to peer-review findings, a sign of an instrument that learns and improves. Where binding legislation left space for national discretion, soft law nonetheless achieved meaningful convergence.

The mechanism may not be without weaknesses, however. It arguably works best where reputational stakes are high and administrative capacity is plentiful, conditions that are perhaps more likely to hold for large, well-resourced authorities than for smaller ones. One pattern that can be observed in the compliance record is not open refusal but simple silence: an authority that neither complies nor explains, possibly because the reputational cost of non-response is felt to be lower than the administrative cost of a reasoned justification. A further, subtler tension might arise from the fact that Guidelines are addressed both to national authorities and to the schemes themselves, so that a scheme could in practice follow a Guideline that its supervising authority has formally declined. Were that to happen, the compliance record would arguably capture the authority’s position rather than the reality on the ground. This divergence between an authority’s formal position and supervised entities’ actual practice is already observable in banking supervision more generally.

From fragmentation to convergence

Taken together, these features may point toward a broader reading. The European Banking Authority’s growing authority over deposit protection appears to rest less on new binding rules than on an enlarged mandate to persuade. Comply-or-explain and reputational pressure have, on the evidence so far, shown some capacity to pull divergent national schemes toward convergence, and the revised Directive might be read as wagering that they can do so across a far wider field, from investment strategy to the use of funds in resolution. It is a concrete test of a much broader phenomenon: the reliance of European Union agencies on soft law to shape the single market from within, through instruments that are formally non-binding yet practically decisive. If the Authority succeeds, the revised Directive will stand as further evidence that, in the architecture of European financial regulation, soft law has quietly become the instrument of choice for turning fragmentation into convergence.

Gaia Chiarini is a PhD candidate in EU law at the Paris Lodron University of Salzburg. She studied at the University of Bologna, obtaining a BSc in Economics and Finance and an MSc in Law and Economics of Insurance and Finance. She also completed a professional Master’s in Law and Practice of the Banking Union at the same university. During her PhD, she undertook traineeships at the European Central Bank. She is a member of the EBI Young Researchers’ Group. Her research focuses on the Banking Union, financial-markets regulation, and cross-border bank consolidation.

Research for this blogpost was conducted as part of EUCHALLENGES, a Jean Monnet Centre of Excellence co-funded by the European Commission under grant agreement no. 101127539