Introduction

After more than two decades of negotiations, the World Trade Organisation’s (WTO) Agreement on Fisheries Subsidies (“AFS”) entered into force in September 2025. It applies to marine wild capture fishing and fishing-related activities at sea and prohibits harmful subsidies that contribute to overfishing, illegal, unreported and unregulated (IUU) fishing, and fishing on the high seas.

Article 4 prohibits Members from granting or maintaining subsidies linked to overfished stocks. Under Article 4.2, a stock is considered ‘overfished’ if it is recognised as such by the relevant coastal Member or Regional Fisheries Management Organisations (RFMOs) on the basis of best scientific evidence “available to it”. Thus, the provision delegates the scientific determination of fish stock status to specialized authorities. However, the Agreement is silent on the standard of review to be applied by the WTO when such determinations are challenged in dispute settlement.

This silence is significant as recognition under Article 4.2 operates as a legal trigger. In case a stock is recognised as ‘overfished’, it invites subsidy prohibitions under Article 4.1, and the dispute settlement mechanism under the SCM framework may be invoked. So far, the agreement doesn’t clarify whether the panels should accord complete deference to scientific institutions, subject their determinations to strict scrutiny, or adopt an intermediate approach. This lack of guidance may lead to inconsistency in the resolution of disputes concerning overfished stocks. This article submits that the “best scientific evidence” phrase in Article 4.2 emphasizes the need for a deferential and reasonableness-based standard of review, which will enable the panels to review scientific validity without overreach, which is crucial to consistent dispute resolution.

Understanding Article 4 within the WTO Subsidy Framework

AFS has integrated the core legal framework of the Agreement on Subsidies and Countervailing Measures (“SCM Agreement”) to determine its scope and operation. Article 1 of AFS restricts its application to measures that qualify as subsidies under SCM Article 1.1 and that meet the specificity requirement under SCM Article 2. Article 4 acts like a specialized prohibition on subsidies for overfished stocks. Under Article 4.2, the “overfished” status must be recognised by the relevant coastal member or RFMO in accordance with the best scientific evidence available to it. 

The Agreement thus creates a two-part system in which the biological status of fish stocks is entrusted to authorities with technical expertise, and the legal consequences are governed by the WTO regime. By designating the coastal Members and RFMOs as recognising entities, the WTO avoids overreach into specialized environmental management, while retaining jurisdiction over trade discipline. This respects the institutional autonomy of environmental institutions that are better positioned to assess complex marine data. At the same time, it ensures that environmental findings are used to enforce trade disciplines rather than to create arbitrary trade barriers. 

Likewise, the link between the Fisheries Agreement and the Agreement on the Application of Sanitary and Phytosanitary Measures (“SPS Agreement”) can primarily be found in their shared reliance on scientific justification as a prerequisite for trade-restrictive measures. Similar to AFS, the SPS agreement also requires actions to be based on scientific principles and risk assessments. However, the AFS introduces a unique qualifier by the critical interpretative pivot “available to it”. Given that data collection at sea is complex and resource-intensive, this qualifier gives a member state contextual flexibility to base its determination of “overfished stocks” on the best science within its practical reach. In other words, the evidentiary standard is not absolute but relative to the specific recognising authority.

By contextualizing the science as discretionary, the treaty acknowledges that data availability is context-specific to each Member, while still requiring them to use the highest quality data they possess. However, by delegating the determination of a stock’s biological health to specialized authorities, the WTO relies on the technical expertise of external or national bodies to provide the factual foundation for applying trade disciplines, thereby creating room for interpretative inconsistency. During the negotiations, members considered whether national authorities should rely on information ‘publicly available’ when determining ‘best scientific evidence’. Nevertheless, the final text settled on ‘available to it’, reflecting a deliberate negotiating choice to preserve contextual flexibility for Members with asymmetric scientific capacities.

Understanding Data Asymmetry

When information about subsidies and fish stocks is held exclusively by the subsidizing government, it raises practical concerns about data asymmetry and transparency. Although Article 8 has sought to maintain transparency by mandating member notifications on subsidies, vessels, fish stocks, and conservation-related data, much of the underlying scientific material remains within national control. Moreover, the collection of such biological data at sea can be expensive or technically complex. In such circumstances, Members can opt to use outdated data or claim the lack of availability of relevant scientific information to avoid the “overfished” status.

At the same time, there is a contrast between Article 3 (IUU Fishing) and Article 4 (Overfished Stocks) of the Fisheries Agreement. The IUU determinations have explicit procedural safeguards in place, like notification, information exchange, and formal determinations, which are similar to the due process rights emphasized in the broader SCM and DSU frameworks. On the other hand, Article 4.2 operates through a lean delegation of biological recognition without comparable procedural safeguards found in other trade defence investigations. The absence of safeguards makes the substantive review of scientific determinations more consequential in dispute settlement. For example, Country A, a developing coastal nation with limited resources, relies on a five-year-old fisheries survey as the ‘best scientific evidence available to it’ to claim that its stock of mackerel is healthy, continuing to subsidize its fishing fleet. Country B, a large fishing nation, challenges this subsidy before the WTO, citing new satellite data and catch reports indicating that the stock has collapsed. The WTO panel seized of this dispute would need to decide how to review Country A’s initial determination. Was it reasonable for Country A to rely on the older data? Did it ignore the new data that was ‘available to it’? The answer will depend on the standard of review the panel chooses to apply.  However, the adopted standard of review must distinguish between ‘malicious non-disclosure’ and ‘genuine lack of capacity’. The AFS incorporated the ‘available to it’ qualifier and SDT provisions to acknowledge the resource gap and capacity constraints among developing countries and LDCs, ensuring that Members are not penalized for data they genuinely do not possess.

Why WTO Panels Cannot Avoid Review

The recognition under Article 4.2 is a legal condition precedent for the validity of the prohibition under Article 4.1. If the underlying scientific determination is scientifically incorrect because of methodological error, then the prohibition has no legal foundation. In the absence of a specific standard under AFS, the general Article 11 of the Dispute Settlement Understanding (DSU) applies. This requires an objective assessment of the matter, including the facts, to be made by the Panels. In other words, they must affirm that the Member has complied with the requirements of the treaty. This prevents the Panel from conducting a de novo review of the underlying science, meaning that they evaluate the reasonableness and adequacy of the recognising authority’s scientific assessment. Thus, the Panels would need to engage with the underlying science to determine whether a Member has acted on a rational and reasoned scientific basis.

Complete deference could shield arbitrary determinations from challenge, and complete substitution of panel judgment would undermine the delegation of authority contained in Article 4.2. Without a coherent review standard, disputes risk inconsistent decisions that could undermine the credibility of the WTO’s environmental efforts.

Competing Standards of Review under Article 4.2

The determination of whether the fish stock in question is “overfished”, as provided in Article 4.2, poses a structural challenge for WTO adjudicators, as they must decide how much weight to assign to the scientific determinations made by the recognising authority. This question was actively debated during the AFS negotiations themselves, where the members considered various approaches of review based on WTO jurisprudence:

Complete Deference Model

Based on institutional autonomy, this model entails automatic recognition of assessments done by coastal Members and RFMOs, which are better positioned than trade panels to evaluate biological data and make factual determinations. However, this approach was actively debated but not adopted during AFS negotiations due to the risk of shielding biased or methodologically weak findings from scrutiny. Members could rely on outdated or selectively interpretative data to maintain otherwise prohibited subsidies, undermining the SCM requirement of positive evidentiary support.

De Novo Scientific Scrutiny Model

A de novo approach would require Panels to independently reassess the biological data, thereby conducting their own scientific trial, irrespective of the recognising authority’s judgment, authorising maximum evidentiary control over the “overfished” label. However, most members agreed that this risks institutional overreach, as WTO Panels lack the technical expertise to conduct complex biological assessments. This concern reflects the Appellate Body’s reasoning in the EC-Hormones case, in which it clarified that the mandate of DSU Art 11 limits panels’ authority to conduct fact-finding. Essentially, the applicable standard thus adopted is neither de novo review nor total deference, but rather an objective evaluation. Moreover, a de novo approach would negate the “available to it” qualifier by imposing an absolute evidentiary benchmark. This would undermine the rights of developing countries and LDCs to utilise the specific technological capabilities available to them, thereby violating the principles of treaty interpretation outlined in Articles 3.2 and 19.2 of the DSU.

Deferential Reasonableness Standard Model

The deferential reasonableness standard is built on the existing SPS risk assessment model, as developed in the EC-Hormones cases (DS26, DS48 and DS321), where the Appellate Body developed a two-step process for conducting risk assessment. First, the panel assesses the scientific basis for an SPS measure and verifies that it is derived from a respected and legitimate source, in accordance with the relevant scientific community’s standards. Second, it reviews whether the risk assessor’s reasoning is “objective and coherent” to ensure that the conclusions have sufficient support in the scientific evidence relied upon.

Similarly, the deferential test of reasonableness is concerned with the adequacy and reasonableness of the process employed by the recognising authority. Under this approach, a Panel ensures that the determination is grounded in positive evidence and reflects an objective and reasoned evaluation of the available data, without substituting its own scientific conclusions. This is underpinned by transparency obligations under Art 8 of AFS, which converts a Member’s internal data into a reviewable record of their due diligence. This approach is not only legally sound but also essential to the operational viability of the Fisheries Agreement. It aligns with DSU Article 11 and preserves the institutional balance of Art 4.2. It ensures that biological determinations are reviewed for methodological integrity without undermining the contextual flexibility embedded in the “available to it” qualifier. If a Member ignores high-quality data that is readily available to it, or provides a scientifically unsupported determination, a panel can conclude that the evidentiary standard hasn’t been satisfied.

In case a Member fails to provide necessary information or withholds information, then panels may apply the “facts available” doctrine, analogous to Article 12.7 of SCM. Consequently, Panels have the authority to draw appropriate adverse inferences if a Member fails to cooperate by not providing information. The panel must distinguish between a genuine lack of data and strategic non-disclosure in light of the “available to it” standard and applicable SDT provisions. Here, Article 13 of the DSU can serve as a practical mechanism to address strategic non-disclosure. It grants panels the right to seek information from any relevant source and to request an advisory report on scientific or technical matters. If the advisory report contradicts the member’s determination, the panel may conclude that the contrary data was ‘available to it’ and draw adverse inferences accordingly. This approach reflects the broader principle of Common But Differentiated Responsibilities (CBDR), in that the panel must refrain from penalising cases of genuine capacity constraints, ensuring that the AFS’s flexibility is not misused while its core objective remains intact.  

The AFS aims to curb harmful subsidies that contribute to overcapacity and overfishing, thereby supporting sustainable fisheries. The reasonableness standard will promote predictability and ensure that subsidies are disciplined when a stock is biologically vulnerable, while avoiding imposing an impossible evidentiary burden on Members with limited information. By limiting the panel to review objective reasoning, methodological assessment, and evidentiary support, it shall ensure that if members conduct a scientifically substantiated and objective assessment, their recognition will withstand scrutiny.

Conclusion

The major challenge for AFS is striking a balance between environmental governance and trade adjudication. Its credibility depends on disciplining harmful subsidies without being perceived as a tool for protectionism. As the prohibition relies upon biological brute facts to discipline harmful subsidies, it requires a delicate balance between specialized knowledge in fisheries and WTO adjudication.

The term “available to it” plays a significant role in maintaining the credibility and legitimacy of AFS among developing and LDC Members. It acknowledges that each Member has different institutional capacities and ensures it doesn’t penalize Members for lacking unavailable sophisticated data. However, this flexibility should not make recognitions immune from review. The credibility of the Agreement shall be maintained only if ‘overfished’ recognitions are based upon rational and substantiated scientific assessment. The deferential reasonableness review effectively reconciles the biological considerations of the Agreement with the institutional limits of WTO adjudication. By focusing on the adequacy and objectivity of a Member’s assessment, Panels can uphold the treaty’s mandate for sustainable fisheries while respecting the delegation of expertise to coastal states and RFMOs.

---

Ayushi Mishra is a final-year undergraduate student at Dr Ram Manohar Lohiya National Law University, Lucknow.  

---

Picture Credit: Deccan Herrald