Not a September Vote: How the WTO's "Fish Week" Puts Taiwan's Distant-Water Fishing Fleet on the Transparency Exam Table

The first-tranche Agreement has three most easily identifiable core elements. Members shall not grant subsidies to vessels or operators that have been found to be engaged in illegal, unreported and unregulated (IUU) fishing; shall not grant subsidies to fishing activity targeting an overfished stock, unless the subsidy or related measures are implemented to rebuild the stock to a biologically sustainable level; nor shall members grant subsidies for fishing on the unregulated high seas — that is, fishing outside the jurisdiction of a coastal state and outside the competence of a relevant regional fisheries management organization. ^[3]

These three provisions do not mean "all fisheries support is unlawful." The Agreement preserves precise boundaries: for example, disaster relief may be handled within the conditions of Article 11 if it is limited to a specific disaster, a specific area, and a set time period, and if any resulting rebuilding only restores the stock to its pre-disaster level. ^[3] Judging any subsidy to be in violation simply because of its name is therefore just as careless as dressing up every form of support as livelihood policy.

The harder problem that the second tranche must address goes deeper: when a subsidy is not given directly to an IUU vessel, and the fishing area is within a management framework, but the subsidy nonetheless lowers fuel, shipbuilding, or operating costs over the long term, causing fleet size and fishing effort to exceed what the resource can bear — how should that be constrained? How should wealthy economies and developing members share the burden? How much policy space should small-scale fisheries, low-income fishers, and remote communities retain? These are precisely the core issues that members have failed to agree on for years, and that MC14 has again handed back to Geneva. ^[1][2]

Fish Week's political significance is to put these divides back on the table; its legal significance, however, is not yet settled. Any sentence written as "the WTO will ban fuel subsidies in September" or "Taiwan has already obtained a certain permanent exemption" lacks sufficient primary-source support as of August 24, and should be flagged as unverified rather than treated as a headline answer.

The part of the current Agreement most likely to be underestimated is not the prohibition provisions, but the notification and transparency requirements of Article 8.

Beyond existing subsidy notifications, members must indicate the type of fishing activity for which the subsidy provides support; and, to the extent possible, provide relevant information on the status of the stock concerned, conservation and management measures, fleet capacity, the names and identification numbers of vessels receiving the subsidy, and catch data broken down by species or groups of species. Members must also notify annually the vessels and operators affirmatively determined to have engaged in IUU fishing, and, within one year of the Agreement's entry into force, describe their domestic implementing measures and fisheries regime. ^[3]

This is not a bookkeeping sheet of "how much budget was allocated this year." It requires the government to link data systems that are otherwise scattered: the budget and subsidy system knows who receives the money; the vessel system knows where a vessel operates; electronic catch reporting knows what was caught; scientific assessments know the status of the stock; and audit and enforcement systems know who has previously violated the rules. Without a common identifier and a clear methodology, even a complete aggregate total cannot answer the follow-up questions that other members may raise under Article 8.5. ^[3]

The WTO convened the first meeting of the Committee on Fisheries Subsidies in December 2025, confirming that the Committee will review members' notifications, monitor the operation of the Agreement, review its implementation at least annually, and, in principle, meet no less than twice a year. ^[5] In other words, transparency is not a one-time report to be filed and forgotten — it enters an institution that can be continuously questioned, compared, and revised.

For Taiwan, this is not a castle built on nothing. The Distant Water Fisheries Act has already written marine resource conservation, combating IUU fishing, catch traceability, and sustainable management into its purposes; institutionally, its tools include fishing licenses, vessel position monitoring, electronic catch reporting, port landing inspections, and observers. ^[7] The Fisheries Agency's audit strategy for exporters also requires transaction and supporting documents to be retained for at least five years, and requires that the links from catching, transshipment, landing, and transport through to processing and sale be effectively connected. ^[8]

But "already having traceability" does not automatically mean "already meeting all WTO notification requirements." Supply-chain traceability primarily answers where a shipment came from and whether it can lawfully enter the market; WTO subsidy transparency must additionally answer whether a given piece of government support is linked to a particular vessel, a particular stock, and a particular fishing incentive. The two can share data, but their purposes are not entirely the same.

Imagine two expenditures both called "equipment upgrades."

The first replaces an old fishing vessel's engine with a more powerful one and increases its freezer capacity, letting the vessel stay at sea longer and bring back more catch. The second installs vessel position monitoring, electronic reporting, and safer life-saving equipment, making it easier for the government to manage operations and reduce accidents and violations. They may sit next to each other in a budget document, yet their effects on fishing capacity and on the resource are entirely different.

Now imagine two expenditures both called "fisher welfare." One increases payments based on actual catch volume or days at sea, which may lower the cost of fishing. The other is decoupled from output and provides occupational-injury, retirement, or career-transition protection, which may create less of a fishing incentive. These differences cannot be judged by policy title alone — they must be judged by eligibility criteria, calculation method, duration, and exit conditions.

Taiwan's first step, therefore, should not be to draw up a blacklist of "subsidies likely to be banned," but to build six fields for every piece of support: policy purpose, beneficiary vessel or personnel, fishing area, the stock and catch concerned, the direction of its effect on fishing capacity, and its exit or review conditions. Where the data is insufficient, it should be flagged as unverified rather than filled in with political slogans.

This also helps avoid a different policy risk: cutting all fisheries support indiscriminately for the sake of safety. If monitoring, scientific surveys, low-carbon retrofits, crew safety, career transition, and time-limited disaster recovery are all reduced together, the government may end up weakening precisely the capacity it needs to implement the Agreement and to maintain social resilience.

Viewed through the lens of development rights and the Global South, the second-tranche negotiations carry a historical question that cannot be skipped over: some wealthy economies have already built large fleets over many years of subsidies, technology, financing, and market access; once resource pressure becomes impossible to ignore, if new rules only restrict latecomers from expanding while failing to address existing capacity and historical responsibility, sustainability could become a tool for freezing the existing market ranking in place.

Small-scale fishers are also not the same as large industrial fleets. Fuel-price spikes, extreme weather, and port disruptions can be a matter of survival for cash-poor communities. If transparency requirements demand expensive systems, scientific assessment, and heavy administrative manpower, members with weaker capacity may end up bearing the cost first simply because they "cannot fill out the forms."

The strength of this counter-argument is its reminder to negotiators that subsidy rules simultaneously allocate ocean resources and development space, and should not moralize every policy difference. Its weakness is that historical unfairness cannot automatically justify any and all fuel, shipbuilding, or operating support today; if such support continues to raise fishing pressure, it is often the small-scale fishers most dependent on coastal waters and healthy stocks who lose their livelihoods first.

Viewed through the lens of rules-based order and environmental governance, subsidies keep fishing trips that would otherwise be economically unviable going out to sea anyway, converting private cost into public depletion of a shared resource. When a vessel extends its voyage on the strength of cheap fuel, financing, or shipbuilding support, the market price no longer fully reflects the cost of stock decline, monitoring, and enforcement. Fishers and states without subsidy capacity find it even harder to compete.

The estimate cited in the WTO's entry-into-force announcement is that global marine capture fisheries subsidies run to roughly USD 35 billion a year, of which about USD 22 billion are considered harmful; the same announcement also notes that 35.5 percent of the world's fish stocks were overfished in 2021. ^[6] These are global estimates, not Taiwan-specific figures, and they cannot be used to directly judge any particular Taiwanese program; but they explain why members are demanding that subsidies be linked to stock and fleet data.

The strength of this position is that it demands every exception be verifiable: who receives it, why, when it stops, and whether the stock has improved. Its weakness is that if only data is demanded without acknowledging differences in scientific assessment capacity and administrative capacity, the rules could end up making it easier for members with more resources to demonstrate their own compliance.

Both sides have grasped half the truth. Fisheries subsidies can protect livelihoods, improve safety, and support transition; they can also sustain overcapacity and prolong fishing pressure. The most useful dividing line is not "all subsidies are good" or "all subsidies are bad," but four questions.

First, does the support directly or indirectly increase fishing capacity and effort? Second, is it targeted at prohibited activities or an already-overfished stock? Third, is it tied to verifiable conditions such as monitoring, stock recovery, fleet reduction, or career transition? Fourth, does it have a clear time limit and exit mechanism?

If the answers remain unclear, a policy need not be scrapped immediately, but it should first be placed under yellow-light review. Conversely, an "equipment subsidy" labeled as green cannot be given a green light on the strength of its label alone if it actually expands range and freezer capacity.

As of the data cutoff date, only a low-confidence, conditional inference can be made about the final second-tranche rules: if members can build consensus on overcapacity and overfishing support, policies with higher transparency that can demonstrate a link between the subsidy and sustainability goals are more likely to retain room for adjustment; support with disconnected data and no exit conditions will face greater negotiating and market risk. This is not a prediction of a future WTO text.

There is work the government can do now, without waiting for September. First, build a common data model so that budget, licensing, vessel position, catch, stock assessment, audit, and penalty data can all be linked using a consistent vessel and fisheries identifier. The publicly disclosed layer can protect commercial confidentiality, but the audit layer must still retain the level of detail needed to answer WTO questions.

Second, break policy support down into categories: capacity-enhancing, management-oriented, social-protection, disaster relief, and transition support. Set evidence requirements, time limits, and review indicators separately for each category, rather than forcing a single set of rules onto every use.

Third, run a dry-run of Article 8 questions and answers in advance. Have staff outside the original handling unit play the role of other members, pressing item by item on stock status, beneficiary vessels, fishing areas, catch, and management measures, to find out where the data cannot be linked together. Taiwan is already an accepting member of the Agreement, and transparency capacity is itself a form of negotiating credibility.

For fishermen's associations, trade bodies, certification agencies, and financial institutions, the biggest risk is that every small operator ends up filling out a different form in a different format. Industry intermediaries should build a shared dictionary and documentation package: where subsidy eligibility, vessel identification, electronic catch data, transshipment and landing records, sales records, and stock and RFMO/A management status are each kept, and who is responsible for updating them.

Finance and insurance can also make data quality a condition, but the responsibility cannot be pushed entirely onto small operators. If government and intermediaries provide a reliable interface, compliant fleets can obtain financing, insurance, and buyer trust at lower verification cost; if all that changes is more forms to fill out, transparency will simply become a barrier of scale.

Intermediaries should also prepare operators for "transition rather than a cliff edge." If a given form of support faces higher risk in the future, resources can be redirected toward monitoring, safety, and energy-saving equipment that does not increase fishing capacity, toward crew skills, fleet reduction, and career transition — rather than waiting for the rules to be finalized before stopping abruptly.

For processors, exporters, cold-chain operators, and small vessel owners, the most practical thing to do right now is three things.

First, list every piece of government support received over the past three years, keeping the application conditions, calculation method, beneficiary, purpose, validity period, and closure documentation. Second, link product traceability data back to the vessel, the source of the catch or aquaculture, transshipment, landing, and processing batch; the Fisheries Agency's existing audits already require at least five years of records and effective linkage across these stages — this is not new work invented solely for the WTO. ^[8] Third, check whether equipment investment actually increases fishing capacity; if the main benefit is safety, monitoring, or energy savings, there should be specifications and performance data to prove it, rather than relying on the project's name alone.

Businesses do not need to correctly guess every bracketed clause of the second tranche's final text by August 24. What they need is to be sure that when a buyer, a bank, or a regulator asks in the future, they can answer with documents — not with a verbal claim of "we've always been compliant."

First, whether the chair's report from the September Fish Week lays out concrete text that can serve as a basis for continued negotiation, rather than merely a summary of positions. Second, whether the scope, duration, and conditions of special and differential treatment converge. Third, how the determination of overcapacity and overfishing links scientific stock assessment, fleet capacity, and subsidy design. Fourth, whether member notification and follow-up-question practices converge into a common format. Fifth, whether Taiwan publishes an account of its institutions and data that can interface with Article 8 of the WTO Agreement.

Before these five signals appear, any claim that "a comprehensive ban has already been finalized" or that "Taiwan's subsidies are entirely unaffected" is premature. What can be said with confidence is that the first tranche has already made transparency a routine institutional requirement; it can also reasonably be inferred that the closer the second tranche gets to convergence, the higher the cost of disconnected data will become.

September will not automatically produce a globally agreed final set of rules just because the meeting's name carries the word "Week." The WTO chair has already made clear that this is not a decision-making moment. ^[1] For news editors, that is a qualifier that belongs in the headline; for Taiwan's policy, however, it is not a reason to delay.

Taiwan has already accepted, and is already bound by, the first-tranche Agreement. From now on, discussing fisheries support cannot rest only on how much budget cares for how many people, or how much output value has grown, nor can it lean solely on the word "sustainability." Every piece of support must be traceable to a vessel, a fishing area, a stock, a set of catch data, and an exit condition.

This capability is not built solely to defend against the WTO. It can also make it easier for genuinely law-abiding, traceable fishing operators willing to transition to prove their value to buyers, banks, and society. The worst outcome is not that some subsidy gets adjusted — it is that Taiwan, having genuinely invested in management and transition, ends up unable to prove that the money reduced risk rather than increased fishing pressure, simply because the data does not connect.

Fish Week's bell has not yet rung. What Taiwan needs to do now is not guess at a vote, but bring its ledger up to a standard that can withstand the world's questions.