4.3 Proportionality
The principle of proportionality, as enshrined in Article 5(4) TEU,(1) See also Protocol (No 2), cit. above. postulates that any actions of the EU institutions shall not exceed what is necessary to achieve the objectives of the Treaties and the relevant test established in the landmark judgment in Fedesa.(2) Case C-331/88 Fedesa and Others EU:C:1990:391. Under the “Fedesa test”, measures must be appropriate and necessary for achieving their objective and must not impose burdens disproportionate to the aims pursued.
While expanding the mutual recognition scheme might be a tool that could contribute to enhancing competitiveness, its suitability for maintaining or improving safety standards – especially for safety-critical items – remains open to debate.
Indeed, the current mutual recognition scheme, developed by the EU ROs, appears to be achieving the objectives of Regulation (EC) No 391/2009 effectively,(3) Report from the Commission to Parliament and Council, COM(2015) 382 final, cit. above. as recognised by the European Commission in its 2015 report and subsequent reviews.(4) Most recently, see Technical Review Meeting Report 21 November 2024, https://www.euromr.org/stakeholder-relations.
The potential benefits of a broadening mutual recognition scheme to items currently not deemed “appropriate”, such as cost efficiency or reduced administrative burden, must be weighed against the risks: inconsistent implementation, heightened liability complexities, and compromised safety.
The distinctive characteristics of the maritime industry make the application of mutual recognition a particularly delicate balancing act, in needing to ensure that no measure goes beyond what is strictly necessary, consistent with the CJEU’s proportionality test. The maritime industry is characterised by the custom-built nature of ships and the complex interplay of their components within unique systems.
Ships are often custom-designed and built either as unique vessels or in small series, tailored to specific requirements and incorporating particular design features, technologies, and operational capabilities. It is therefore common that each new ship (especially complex vessels such as cruise ships, offshore service vessels etc.) is designed to a unique specification set by its owner, and tailored to their specific trade, route, and cargo. Variations in port infrastructure, canal dimensions, climate conditions, and owner preferences (e.g. fuel efficiency vs speed) mean that a generic design must often be modified. Even in more standardized market segments (e.g. bulk carriers or tankers), an order might consist of only a handful of sister ships. This high degree of customisation means that, on a system level, ships differ significantly from each other and even “standard” ship designs undergo evolution and custom tweaks from ship to ship. The components and equipment installed vary with the specific design and operational needs of that vessel, and their interaction with the overall system is, in many cases, unique to each vessel.
The economics of shipbuilding also do not encourage large production of single designs – each newbuild project must essentially be profitable on its own or within a small-scale production. Unlike cars or planes, which are manufactured for mass markets, ships are usually built only after a buyer is secured. In addition, a ship’s lifespan is 25-30 years or more and does not renew as frequently as for example automobiles. Projects are therefore often tailored closely to the ship owner’s needs and there is little opportunity to amortise significant R&D spend over large production runs. In practice, and especially in Europe, shipyards excel in one-off or highly specialised vessels – e.g. each cruise ship has a unique layout and features – while it is only in some high-volume lines, often built in Asian yards, that we see moderate standardisation (such as with certain oil tankers or bulk cargo ships). The shipbuilding industry’s structure further contributes to design diversity. There is no equivalent of a “model year” in shipbuilding; each contract may spawn a new design iteration.
The custom and decentralised nature of ships introduces several challenges for mutual recognition:
First, the interaction of components within the unique system of each ship means that certifying components in isolation, without considering their integration into the specific ship's systems, poses significant risks. A component that is suitable and safe in one context may not be appropriate in another, due to differences in system design, operating conditions, or other factors. The bespoke nature of ships requires a case-by-case approach to certification, ensuring that every component is assessed in the context of the specific vessel.
Second, the potential consequences of a component failure in the maritime context are severe. Each ship represents a significant investment and carries substantial safety and environmental risks. A failure of a critical component can lead to catastrophic outcomes, including loss of life, environmental disasters, and significant economic losses. The catastrophic potential of maritime accidents necessitates a rigorous and cautious approach to certification and mutual recognition.
Third, the certification process for each ship is complex and comprehensive, reflecting the higher risks, diverse operational profiles, and longer operational lifespan of vessels. Certifying components requires a detailed understanding of how they will function within the specific ship's systems and operating conditions. This complexity limits the potential efficiency gains from expanded mutual recognition and increases the risks associated with certifying components without thorough, ship-specific evaluation.
Given these challenges, the potential benefits of expanding mutual recognition to safety-critical items in the maritime sector are limited. The unique and customised nature of ships, the critical importance of system integration, and the severe potential consequences of failures, make a cautious approach necessary. Mutual recognition in this context must be carefully considered to ensure that safety and environmental protection are not compromised.
From a practical perspective, third-party acceptance – e.g. non-EU flag States or insurance underwriters – also constrains the appropriateness of an expanded mutual recognition, particularly for critical components. For instance, IUMI warns that: “Should any RO be allowed to certify and approve components and equipment for a vessel at all safety levels and regardless of which society will be responsible for classing the vessels, neither the classification society nor owners or underwriters will really know what quality of vessels they have or what quality of components have gone into them.”(5) IUMI, cit. above. If such stakeholders refuse to accept a certificate not issued by the RO classing the vessel, requiring re-tests instead, the benefit to the industry may be minimal.
In light of these considerations, the principle of proportionality favours a more cautious approach to mutual recognition in the maritime sector, particularly for safety-critical items.
For lower-risk and/or non-safety critical items, mutual recognition remains justified and widely accepted by OEMs, ROs, and many financiers or underwriters seeking consistent standards. The current system developed by EU ROs, on the other hand, appears to strike a balance between the objectives of Regulation (EC) No 391/2009, without compromising safety.
However, broadening its scope could be disproportionate, if this means introducing significant new risks or burdens without clear offsetting benefits. The unique characteristics of the maritime industry suggest that risks associated with expanded mutual recognition may outweigh the potential benefits, supporting the argument for maintaining the current scope of mutual recognition. The growing interconnectivity of maritime systems that are susceptible for emergent risks further complicates this assessment. If expanded mutual recognition in the maritime industry led to a situation where no single EU RO then had comprehensive oversight of critical systems, the long-term costs in terms of safety risks and potential accidents would outweigh any short-term competitive gains.