Hostname: page-component-74656f9884-qr5l4 Total loading time: 0 Render date: 2026-11-05T04:16:34.146Z Has data issue: false hasContentIssue false
  • English
  • Français

Beyond the European Model: China’s Developmental Precaution in Cultivated Meat Regulation

Published online by Cambridge University Press:  04 November 2026

Yi Lu Open the ORCID record for Yi Lu [Opens in a new window]
Yi Lu*
Affiliation:
Faculty of Law, Boğaziçi University , Istanbul, Türkiye
*
Email: luyiclaudia@gmail.com
Rights & Permissions [Opens in a new window]

Abstract

The precautionary principle is widely regarded as a distinctively European legal construct, grounded in science-conditioned risk assessment, provisional measures and a health-protective orientation. This article challenges that assumption by examining how the principle works within China’s own regulatory tradition in the case of cultivated meat, a novel food technology that existing frameworks in both jurisdictions were not designed to address. Drawing on the treatment of cultivated meat under China’s Three New Foods regime and its national biomanufacturing policy, it identifies three ways in which China departs from the European model. The trigger for precaution shifts from scientific uncertainty established through assessment to a default exclusion that presupposes it, precautionary measures become open-ended rather than provisional, and the purpose of precaution widens from protecting health to steering industrial development. These departures form a coherent regulatory logic, termed here developmental precaution. Under this logic, the pace of regulatory opening tracks the readiness of a strategic industry rather than safety alone. China’s case shows that features long treated as intrinsic to the precautionary principle are in fact particular to the European tradition. This supports a more pluralist account of precautionary regulation.

Keywords

cultivated meatdevelopmental precautionprecautionary principle

Information

Type
Articles
Information
European Journal of Risk Regulation , First View , pp. 1 - 15
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2026. Published by Cambridge University Press

I. Introduction

Cultivated meat occupies an awkward position in food law.Footnote 1 Produced by extracting animal cells and proliferating them in bioreactor environments, it is recognisable as meat but not the product of agriculture,Footnote 2 involves no slaughter, and has no established record of long-term human consumption.Footnote 3 These features place it outside the categories on which food law conventionally relies. When regulatory authorities confront such a product, the precautionary principle offers what appears to be a clear mandate. The principle at issue here is the one that operates within food law, not the broader environmental or public health versions from which it descends. A comparison between the European Union and China, however, reveals that this apparently shared mandate produces radically different governance outcomes. The EU has channelled cultivated meat into its Novel Food Regulation,Footnote 4 subjecting applications to a science-intensive pre-market safety assessment by the European Food Safety Authority (EFSA) under the General Food Law’s precautionary clause.Footnote 5 Member States and market actors have nonetheless responded in sharply divergent ways. Italy enacted a national prohibition grounded in precautionary reasoning,Footnote 6 while French and Dutch companies pioneered novel food authorisation applications for cultivated meat products to the Commission.Footnote 7 China has instead absorbed the product into its Three New Foods regime, where market access is suspended by default, scientific uncertainty is the assumed background condition rather than the trigger for action, and the National Health Commission exercises broad discretion within a policy environment that frames biotechnology development as a national strategic priority.

China’s scientific and industrial capacity in cultivated meat has advanced rapidly, with pilot-scale bioreactor production demonstrated and a domestic industry of multiple firms now operational.Footnote 8 In response to a 2020 legislative proposal on cultivated meat, the NHC answered by reference to the new food ingredient framework, indicating that cultivated meat falls within that pathway,Footnote 9 although no dedicated assessment guidelines have been published.Footnote 10 China’s treatment of cultivated meat thus produces a form of precaution that operates through mechanisms the European model does not share, namely default market exclusion without affirmative regulatory action, open-ended timelines without statutory review obligations, and integration with national industrial policy.

This divergence poses a question that comparative risk regulation literature has not adequately addressed. When the dominant European analytical lens is applied beyond the transatlantic context, regulatory systems that diverge from the European model tend to be read as incomplete approximations of it rather than as differently organised.Footnote 11 This article argues that the precautionary principle, as it operates within China’s administrative law tradition, takes a form that is functionally coherent on its own terms but structurally distinct from the European model. The divergence runs across three critical dimensions, namely the epistemic trigger for precautionary action, the temporal character of precautionary measures, and the normative purpose precaution is understood to serve. These dimensions reflect a different underlying philosophy of regulatory governance, one in which the state’s role in managing technological uncertainty is inseparable from its role in steering economic development. This article proposes the concept of developmental precaution to capture this distinct regulatory logic.

Part II reconstructs the precautionary principle as a European construct, identifying its philosophical foundations and three structural features that define its operation in EU food safety law. Part III examines China’s institutional response to cultivated meat, tracing the shift from a preventive framework designed to enforce compliance against known harms to a precautionary gatekeeping mechanism operating in the absence of any identified risk. Part IV develops the comparative analysis across the three dimensions of divergence and introduces the concept of developmental precaution. Part V concludes.

II. The precautionary principle as a European construct

Part II identifies three structural features of the EU’s precautionary model that are particular to its institutional architecture rather than universal.

1. Origins and philosophical foundations

The precautionary principle did not originate in food law. It emerged at the international level in the field of environmental protection, most prominently in Principle 15 of the 1992 Rio Declaration,Footnote 12 and only later migrated into regional and domestic legal systems, expanding from environmental regulation into food safety governance.Footnote 13 In European law, it entered through the environmental policy traditions of West Germany, where the concept of Vorsorgeprinzip was developed in the 1970s as a basis for proactive environmental protection.Footnote 14 Its transposition into food governance came later still, and the features it acquired in that context are products of this migration across regulatory domains rather than intrinsic attributes of the principle. Within EU law, its most legally significant codification for food governance came in Article 7 of the General Food Law, which gave the principle a form that both reflected and shaped its subsequent development in European regulatory theory and practice. Article 7 provides for precautionary measures but does not itself define the principle, and a working definition appears only later, in the Commission’s 2000 Communication on the Precautionary Principle.Footnote 15 Philosophically, the precautionary principle is a response to the problem of decision-making under uncertainty. It holds that when the available scientific evidence is insufficient to assess a risk with confidence, but there is a plausible basis for concern that harmful effects may materialise, decision-makers are not required to await further scientific confirmation before acting.Footnote 16

The philosophical foundations of the principle alone, however, do not explain its specific institutional form in EU food law. That form is the product of regulatory history. The BSE crisis of the 1990s was the defining formative episode for the European model of precautionary governance in food safety.Footnote 17 The institutional failures exposed by that crisis, including the subordination of scientific advice to political interests and the absence of transparent risk communication, generated comprehensive reform that placed scientific independence and precautionary caution at its centre. The creation of EFSA, the codification of the precautionary principle in the General Food Law, and the separation of risk assessment from risk management were responses to a crisis that had made visible the cost of inadequate precaution.Footnote 18 The European model is therefore not simply a rational design but the product of a particular historical experience of regulatory failure.

2. Three structural features of the EU model

The first structural feature of the EU’s precautionary approach is the centrality of scientific assessment as both a prerequisite and a constraint. The General Food Law permits precautionary measures only where, following an assessment of available scientific information, uncertainty over the possibility of harmful effects persists. Precaution does not substitute for scientific assessment. It is the response to an assessment that has been completed and yet remains inconclusive.Footnote 19 This science-first logic reflects a deliberate choice to anchor risk governance in empirical analysis rather than political intuition, reinforced by the institutional separation of risk assessment (EFSA) from risk management (Commission and Member States),Footnote 20 which constrains political authority over risk decisions through independent scientific judgment.Footnote 21 In practice, this logic translates into a procedurally elaborate novel food assessment process. An applicant submits a dossier to the Commission, which refers it to EFSA for a scientific opinion. EFSA validates the application and then has a statutory nine-month period within which to produce that opinion. On the basis of the opinion, the Commission adopts an implementing act authorising or refusing the product, subject to comitology approval by the Member States, and an authorised product is added to the Union list.Footnote 22 In practice, assessments have taken considerably longer than the nine-month period, reflecting the complexity of the scientific questions raised by products like cultivated meat for which no established safety database exists. Two applications for cultivated meat products, submitted in July 2024 and January 2025 respectively, are currently proceeding through this process as of writing, with no published EFSA opinion in either case.Footnote 23

The second structural feature follows from the first, namely the explicit requirement of temporariness. Precautionary measures adopted under Article 7 must be reviewed within a reasonable time, in light of developments in the scientific knowledge necessary to complete a more comprehensive risk assessment.Footnote 24 This provision encodes a philosophy of regulatory dynamism. Precaution is a holding position rather than a permanent verdict, subject to revision as evidence accumulates. The assumption is that the scientific uncertainty that justifies precautionary action is, in principle, resolvable, and that the regulatory response must track the evolution of the evidence base rather than crystallising around an initial judgment of risk. The EU’s experience with GMO regulation illustrates both the aspirational character of this requirement and the difficulty of enforcing it in practice. For example, the de facto moratorium on new GMO approvals that operated through much of the 2000s attracted a WTO challenge for constituting undue delay in the completion of approval procedures.Footnote 25

The third structural feature is the orientation of the principle toward the protection of diffuse, collective interests rather than the advancement of specific economic or industrial objectives. The EU’s food safety framework is formally premised on the institutional separation of risk management from the promotion of industrial competitiveness,Footnote 26 though this separation is contested in practice, as the influence of industry lobbying on regulatory outcomes demonstrates.Footnote 27 Public participation is institutionalised through defined channels. EFSA conducts public consultations on scientific matters at the risk assessment stage, and the framework’s general objectives orient risk governance toward consumer and public health interests, formally prioritising consumer protection over commercial considerations.Footnote 28 When Member States have sought to use precautionary reasoning to protect domestic agricultural industries, as Italy’s cultivated meat banFootnote 29 most visibly demonstrates, they have done so in ways that create tension with the internal market and are subject to challenge as restrictions on the free movement of goods. The normative purpose of precaution, in the European model, is therefore framed in formal separation from industrial policy, whatever pressures political practice brings to bear on that separation. These three features, namely science conditionality, temporariness, and health-protective orientation, form the structural benchmark against which China’s approach is examined in Part IV.

This account describes the precautionary principle in its consolidated post-BSE form. Its centrality in EU food and biotechnology governance is, however, increasingly contested. Three recent developments illustrate the point. The proposed Food and Feed Safety Omnibus would remove the automatic reapproval requirement for certain pesticides after ten years. The New Genomic Techniques Regulation, adopted in 2026, removes category 1 NGT plants from the GMO framework altogether.Footnote 30 For category 1 plants, it replaces case-by-case risk assessment with criteria of equivalence to conventionally bred varieties.Footnote 31 The Novel Food Regulation is itself under review with a view to streamlining the authorisation of novel foods, including cultivated meat. These developments do not displace the structural features identified here, but they show that the European model is itself in flux, which makes its comparison with China’s trajectory all the more pertinent.

III. Regulating the unknown: China’s institutional response to cultivated meat

Part III examines the institutional architecture through which China has responded to cultivated meat, from preventive enforcement to precautionary gatekeeping.

1. The 2015 food safety law and the limits of established precaution

China’s Food Safety Law, as comprehensively revised in 2015, identifies four governing principles for food safety work, namely prevention as priority (预防为主), risk management (风险管理), whole-process control (全程控制), and social co-governance (社会共治), with prevention listed first as the foundational orientation.Footnote 32 The revision responded to a series of high-profile scandals that had severely damaged public confidence in the safety of the food supply.Footnote 33 In practice, however, this preventive framework was designed for known risk categories, addressing harms whose character was already understood from prior incidents involving adulteration, regulatory evasion, and supply chain failures.Footnote 34

The institutional architecture established by the 2015 revision reflects this enforcement-centred orientation. Three agencies share primary responsibility. State Administration for Market Regulation (SAMR) exercises enforcement authority over food safety, the National Health Commission (NHC) sets the scientific standards governing permissible ingredients and additives, and the China National Center for Food Safety Risk Assessment (CFSA) provides technical support through risk assessment. This structure was built to police compliance with known food safety standards across a vast and fragmented supply chain, not to resolve epistemic uncertainty about products whose risks have never been characterised.

The same orientation shapes the specific instrument through which novel products are screened. The Food Safety Law requires pre-market safety assessment for new food ingredients, new food additive varieties, and new food-related product varieties, collectively administered as the Three New Foods regime (三新食品).Footnote 35 This requirement was built to screen conventional novel ingredients such as plant extracts or newly used microorganisms, whose risks can be measured against known comparators. Cultivated meat has no such comparator, and its potential hazards have no precedent in the regulatory system.

A further complication arises from the distribution of regulatory authority across agencies. Although the NHC administers the new food ingredient approval on which market entry depends, the production of cultivated meat may also engage the parallel authority of SAMR, the Ministry of Agriculture and Rural Affairs (MARA), and the General Administration of Customs, depending on the production process.Footnote 36 The interaction among these administrative actors has not been formally coordinated through any published interagency guidance.Footnote 37 The result is that although the approval pathway determining market entry is settled, responsibilities across production, certification and import remain uncoordinated.Footnote 38

2. The Three New Foods regime and the structure of default precaution

The Measures for the Administration of New Food Ingredients define new food ingredients as products with no history of traditional consumption in China, classified into four categories: animals, plants and microorganisms; components isolated from them; food components with altered original structure; and other newly developed food ingredients.Footnote 39 The NHC indicated that cultivated meat falls within the new food ingredient frameworkFootnote 40 when it answered a 2020 legislative proposal by reference to that pathway,Footnote 41 and CFSA’s engagement through its dedicated working group proceeds on the same basis. Its jurisdictional home is therefore effectively settled. What remains unsettled is the method of assessment, because cultivated meat poses a qualitatively different challenge from the products the framework was built around. The character of the risk itself is undefined, the categories of potential harm have no precedent in the regulatory system, and no established methodology exists for assessing them.

The defining structural feature of the Three New Foods regime is its default prohibition, whereby no product within the relevant categories may be produced or marketed without prior NHC approval, regardless of whether any specific safety concern has been identified.Footnote 42 Where scientific uncertainty about a novel food cannot be resolved, the product cannot be approved. Unresolved uncertainty operates as a ground for continued exclusion rather than as a basis for conditional authorisation. Applicants must submit a comprehensive dossier reviewed by CFSA,Footnote 43 with the requirements most recently elaborated through the 2024 supplementary requirements for genetically modified microorganisms (GMMs).Footnote 44 This structural gatekeeping mechanism constitutes what this article terms default precaution, a form of precaution embedded in institutional design rather than triggered by specific scientific uncertainty, distinguishing it from the precautionary principle as applied in European regulatory discourse. The practical operation of this default precaution is visible in cultivated meat’s current regulatory position. As of writing, no cultivated meat product has been approved.Footnote 45 Neither the first nor the second batch of the 2025 Annual Plan for National Food Safety Standards includes a dedicated cultivated meat standard,Footnote 46 and the draft guideline reported under preparation since early 2024 has not been released for public comment.Footnote 47

Some institutional groundwork has been laid, however. CFSA co-hosted a November 2023 seminar with the FAO on cell-based food and precision fermentation, and a dedicated regulatory working group was concurrently established as an industry-academic initiative.Footnote 48 Of these developments, the 2024 GMM requirements are the most concrete output, though they address only the GMM subset of novel food products. The 2024 requirements also addressed a previously ambiguous dual-track procedure. The Measures on Agricultural GMOs mandate prior certification for products containing agricultural GMOs, including additives, but whether food-processing microorganisms fall within that “agricultural” scope was never formally resolved. Applicants in practice obtained biosafety certificates from MARA by reference rather than strict statutory obligation.Footnote 49 The 2024 requirements eliminated this ambiguity by establishing a single submission pathway through the NHC.Footnote 50

Yet the consolidation into a single NHC track raises an unresolved question for cultivated meat, namely whether its pathway will follow the same model or require inter-ministerial coordination, given that animal cell cultivation potentially engages livestock and veterinary competences.Footnote 51 This jurisdictional uncertainty points to a deeper problem. CFSA operates under scientific norms requiring comparator data and reproducible risk characterisation methodologies,Footnote 52 neither of which exists for cultivated meat in settled form.Footnote 53 The result is an institution caught between scientific rigour it cannot yet satisfy and the developmental priorities the state attaches to the sector. This tension, however, never has to be resolved because the assessment is subject to no enforceable overall deadline. The Chinese procedure is subject to two nominal time limits, neither of which constrains the total duration. The twenty-working-day period for the licensing decision excludes the time taken by expert review, and the sixty-working-day period for expert review applies only to each individual round.Footnote 54 The review panel may return a “deferred re-examination” conclusion that restarts the cycle with a fresh request for supplementary material, and the number of such cycles is not capped,Footnote 55 so these nominal limits place no ceiling on the total duration of assessment. The institution need not choose between approval, which it cannot scientifically justify, and refusal, which it has no mandate to issue.

China’s regulation of cultivated meat also exposes the limitations of substantial equivalence. The concept, originally developed for GMO assessment, compares a novel product against its conventional counterpart.Footnote 56 It once stood at the centre of the EU’s earlier Novel Food Regulation, but under the current Regulation it survives only in a residual role for the labelling of novel foods. In China, it operates as a statutory threshold under which a product found substantially equivalent to an existing food or an approved new food ingredient triggers termination of the review.Footnote 57 China’s Measures define substantial equivalence as identity with an existing food in “species, source, biological characteristics, principal components,” and other criteria, with “substantially consistent” production processes and quality requirements.Footnote 58 Cultivated meat satisfies none of these criteria in settled form, and the difficulty is compounded where the cell line has been immortalised through gene editing, which may bring the product within the separate scope of agricultural GMO certification. The framework’s own statutory language thus reveals its inadequacy for cultivated meat.Footnote 59 CFSA has not yet published an alternative methodology for products that lack a conventional counterpart. The result is what this article terms discretionary paralysis. The institution possesses formal authority to act but lacks the methodological basis to exercise it, producing deferral subject to no institutional discipline.

IV. Three structural differences and their theoretical implications

Part IV develops the comparative analysis across three structural dimensions and introduces the concept of developmental precaution.

1. Epistemic difference: scientific uncertainty versus default gatekeeping

The most consequential structural difference between the European and Chinese approaches concerns the epistemic condition that activates precautionary governance. The European system operates on two levels. The first is a default prohibition that excludes unapproved novel foods from the market,Footnote 60 and the second is a precautionary principle that governs how residual uncertainty is handled once scientific assessment is underway.Footnote 61 It is the second level, the precautionary principle proper, that reveals the structural contrast. In the European model, precaution is activated by a specific finding, namely a completed risk assessment, evaluated scientific evidence and persisting uncertainty about the possibility of harmful effects. The precautionary principle does not operate in the absence of prior scientific inquiry. Rather, it operates where that inquiry has been completed but leaves residual doubt. The EU Courts have confirmed this understanding on multiple occasions.Footnote 62

This science-first logic both constrains and legitimates precautionary action. It requires uncertainty to be demonstrated through assessment rather than assumed, and grounds the resulting measure in the authority of scientific evaluation. The result is a system in which the precautionary principle operates as a rule about how to respond to assessment that is present but inconclusive, rather than as a general licence for protective action in the face of novelty.Footnote 63 The institutional architecture of EFSA reinforces this logic by separating risk assessment from risk management, structuring precautionary decision-making around a scientific process that both enables and disciplines the exercise of regulatory authority.Footnote 64

China’s Three New Foods regime also imposes a default prohibition, so the distinction does not lie in default exclusion as such, but in the role that scientific uncertainty plays within each system. In the European model, uncertainty is what the assessment process is designed to resolve. In China, it is the standing condition that the default prohibition presupposes. The default prohibition is a structural feature applicable to all products within scope, irrespective of whether any specific risk has been identified or any assessment conducted. Scientific uncertainty is not the trigger for the default gatekeeping mechanism but the assumed background condition that the regime addresses by requiring positive approval before market access is granted. As a result, the applicant bears the entire burden of demonstrating safety from the outset, in the absence of any prior determination by the regulator that a specific uncertainty needs to be addressed.Footnote 65

One might object that the default prohibition is merely ordinary administrative licensing rather than precaution. The objection proves too much, since it would apply equally to any pre-market approval system, including the EU’s Novel Food Regulation. What distinguishes precautionary governance from routine licensing is the relationship between the prohibition and uncertainty. Routine licensing verifies compliance with known and specifiable conditions, whereas precautionary measures respond to the possibility of harm that cannot yet be assessed with confidence. China’s Three New Foods regime falls into the latter category. The default prohibition applies precisely to products whose safety profile is unknown, and the dossier requirements are structured around the demonstration of safety in the face of that unknown. The precautionary character of the system lies in the epistemic structure of the burden it imposes, whatever its formal label may be.

2. Temporal difference: provisional measures versus open-ended deferral

If the epistemic difference concerns what activates precaution, the temporal difference concerns how long it lasts. In the European model, precautionary measures must be provisional. The General Food Law requires that precautionary measures be reviewed within a reasonable time as scientific knowledge develops, on the assumption that the uncertainty justifying them is tractable and the regulatory position will evolve as evidence accumulates.Footnote 66 This provisional character is not merely a formality but performs a disciplining function, reflecting a liberal regulatory philosophy in which state restrictions on market freedom are treated as exceptions requiring continuous justification rather than as normal instruments of governance. It pressures regulators to maintain and update their assessments and prevents precautionary measures from hardening into de facto permanent prohibitions. For example, the EC–Hormones case established that the precautionary principle does not override the SPS Agreement’s requirement for risk assessment.Footnote 67 The GMO moratorium case further demonstrated that approval procedures left incomplete are legally vulnerable in the WTO system.Footnote 68 Together these examples create an external discipline that reinforces the internal temporariness requirement of EU food law.

China’s approach contains no equivalent temporal discipline. Whereas the General Food Law requires precautionary measures to be reviewed as scientific knowledge develops,Footnote 69 the Three New Foods regime imposes no obligation to revisit a product’s continued exclusion over time.Footnote 70 The Measures provide for reassessment only of approved products whose safety later comes into question.Footnote 71 The default prohibition, once in place, is not subject to any scheduled reassessment of whether it remains justified. The NHC conducts case-by-case reviews at its own pace, without published timelines or procedural accountability. The experience of 2’-fucosyllactose, another product within the Three New Foods regime, illustrates the point. Its approval took seven years longer in China than in the EU. The system accommodated a multi-year review without any procedural mechanism requiring its conclusion.Footnote 72

Cultivated meat presents a harder case. The product category has been acknowledged in official planning documents as a strategic priority,Footnote 73 yet the regime imposes no procedural pressure of the kind a binding statutory timeframe would create. Cultivated meat faces the absence of any settled basis for assessment. The holding position characteristic of precautionary governance can therefore persist without structural constraint on its duration. Precautionary governance of cultivated meat is therefore not provisional in practice but an open-ended administrative posture, insulated from the disciplines that give the European model’s temporariness requirement its operational significance.

This open-ended temporality raises a question under the SPS Agreement. Article 5.7 permits a member to adopt provisional measures where relevant scientific evidence is insufficient, provided it seeks the additional information necessary for a more objective risk assessment and reviews the measure within a reasonable period.Footnote 74 A regime that suspends market access with no obligation to complete assessment within a defined period fits awkwardly within this discipline. The tension is not presently justiciable, since China has adopted no specific SPS measure against imported cultivated meat that a trading partner could challenge, and the default prohibition applies equally to domestic and imported products. Were China’s treatment of imported cultivated meat to harden into a measure affecting trade, its open-ended character would be hard to square with the review obligation in Article 5.7.

3. Normative difference: health protection versus developmental governance

Beyond epistemic conditionality and temporariness, the two systems diverge in the purpose precaution serves. In the European model, the precautionary principle is anchored to the protection of consumer health and environmental integrity, and industrial or national strategic interests are formally separated from the precautionary framework rather than integrated into it.Footnote 75 Member States have nonetheless invoked precaution to protect domestic interests, as Italy’s cultivated meat ban demonstrates,Footnote 76 and the European Parliament has conflated safety concerns with broader social and economic objections in its resolution on cultivated meat.Footnote 77 The Italian ban, however, was subordinated to procedural discipline by the EU’s institutional architecture.Footnote 78 These examples show that the EU’s legal architecture provides mechanisms to enforce the separation when safety and industrial interests threaten to merge.Footnote 79

China’s regulatory approach to cultivated meat is structured differently. The Three New Foods regime is administered by the NHC, whose primary mandate is public health protection, and the formal criteria for novel food approval are safety-based. In this respect, the institutional architecture superficially resembles the European model, but the policy context assigns the regulatory system an explicit developmental function. For example, the 14th Five-Year Bioeconomy Plan identifies the development of synthetic biology-based novel protein products and the building of biosecurity risk governance capacity as co-equal strategic objectives, integrating regulatory infrastructure into the industrial development agenda rather than treating it as an external constraint.Footnote 80 The regulatory system is therefore positioned not as a gatekeeper defending consumers against industry but as a strategic instrument enabling the state to govern the pace and direction of a priority industry’s development.

More broadly, the concept of new quality productive forces (新质生产力) reinforces this framing. The framework evaluates institutional arrangements, including regulatory systems, by their capacity to support technology-driven economic development.Footnote 81 The precautionary principle, in China’s adaptation, is consequently expected to serve both purposes simultaneously, namely to protect consumers and to create the conditions for strategic industries to develop safely. This dual function does not exist in the European model as a formal legal matter. Even where it appears in practice, it does so in tension with the model’s design. The difference between the two models lies elsewhere. The European model is not free of industrial or political influence, as the role of lobbying in areas such as the New Genomic Techniques Regulation makes plain. What sets it apart is that its institutional architecture locates industrial promotion outside the precautionary framework and provides mechanisms to challenge its intrusion. China’s framework instead integrates industrial development into the regulatory system as a legitimate governance objective.

The theoretical implication is that the normative purpose of precautionary governance is not fixed but contestable. The European model’s insistence on the separation of health protection from industrial policy is a choice embedded in a particular institutional architecture, not a logical requirement of precautionary reasoning as such.Footnote 82 The EU’s distinctive precautionary culture reflects specific historical, institutional and political factorsFootnote 83 rather than any universally shared understanding of what precaution requires.Footnote 84 China’s adaptation demonstrates that precaution can be redirected toward developmental ends without ceasing to perform a precautionary function. Market access remains conditioned on safety demonstration, but the framework assigns the regulatory system a different overall purpose.

4. Developmental precaution as theoretical synthesis

The divergences in epistemic conditionality, temporariness and normative purpose examined above point toward a distinct concept. Developmental precaution integrates precautionary gatekeeping into the state’s industrial steering apparatus, so that safety assessment becomes an instrument of industrial sequencing rather than an independent constraint upon it. Though the concept shares intellectual terrain with the developmental state literature, the two differ in one critical respect. That literature focuses on how states steer industrial development through investment, procurement and targeted policy,Footnote 85 but treats regulation as an external variable. Developmental precaution identifies regulation as endogenous to the developmental logic.Footnote 86

This account may invite an obvious objection. China’s regulatory response to cultivated meat is plainly still taking shape, and the absence of a cultivated meat-specific methodology, the possibility of indefinite review and the discretionary character of NHC review could all be read as signs of a system still under construction rather than a distinct regulatory model. This objection is ultimately unpersuasive for two reasons. The first concerns the policy environment. The 14th Five-Year Bioeconomy PlanFootnote 87 and the new quality productive forces frameworkFootnote 88 expect regulatory and industrial development to proceed in tandem, shaping the pace at which cultivated meat regulation advances. The Chinese state has demonstrated rapid regulatory development in pharmaceuticals and medical devices when political priorities demand it. The relative slowness of cultivated meat regulation is consistent with a pattern of allowing the industry to develop at a pace aligned with the state’s strategic planning horizon. This selectivity, with its systematic acceleration where strategic interests require it and deliberate deferral where they do not, is functionally equivalent to state-steered sequencing, whether or not consciously designed as such.Footnote 89 The second, and more fundamental, reason concerns the default gatekeeping mechanism itself. The presumption of market exclusion built into the Three New Foods regime is, in fact, a design feature established in 2013,Footnote 90 years before cultivated meat emerged as a regulatory challenge in China around 2020, and therefore cannot be explained as a by-product of incomplete development.

What then distinguishes the deferral effect of developmental precaution from an ordinary trade barrier? A trade barrier is static and exclusionary by design, whereas developmental precaution is calibrated to open when the conditions for safe market entry are established. This distinction between developmental precaution and a mere trade barrier is reinforced, but also complicated, by the legal framework within which the NHC operates. The Food Safety LawFootnote 91 assigns the NHC a public health mandate, and the principle of reasonableness (合理性原则) in Chinese administrative lawFootnote 92 requires that the NHC’s review discretion bears a rational relationship to that mandate. Delay is therefore defensible, in principle, only insofar as it serves the accumulation of safety evidence, not the maturation of domestic industry. But unlike the EU, where EFSA’s independence and the temporariness requirement safeguard this separation, the Chinese system provides no equivalent enforcement mechanism. The problem is compounded by the fact that the two justifications, safety and industrial maturation, are in practice indistinguishable to an outside observer, since the regime discloses neither the reasoning nor the timeline of its deferral. That the state has funded upstream cellular agriculture research since 2020Footnote 93 while withholding downstream market access confirms the pattern this article terms developmental precaution. In this pattern, regulatory opening tracks industrial readiness rather than the resolution of scientific uncertainty alone.

V. Conclusion

This article challenges any account of the precautionary principle that treats it as a universal legal standard. China’s approach to cultivated meat is precautionary in a substantive sense, conditioning market access on safety demonstration and maintaining that condition until an adequate scientific basis for approval exists. The logic through which this precaution is expressed, however, departs from the European model in ways that are structural rather than incidental. In the European model, the pace of regulatory opening is set by the progress of scientific assessment. In China’s model, it tracks the readiness of a strategic industry beyond safety readiness alone. Developmental precaution, the concept proposed here, captures this distinct logic. It describes a regulatory form in which precautionary gatekeeping is integrated into the state’s industrial steering apparatus. The pace of regulatory opening then tracks the readiness of a strategic industry rather than the resolution of scientific uncertainty alone.

What the China case reveals most clearly is that the European model’s structural features are not the inevitable expression of precautionary logic. They are particular choices, reflecting particular institutional arrangements, political commitments and historical experiences. The science conditionality of EU precaution reflects its post-BSE commitment to grounding regulatory legitimacy in independent scientific assessment. The temporariness requirement reflects a liberal regulatory philosophy in which provisional measures must be continuously justified. The health-protective orientation reflects the constitutional architecture of EU food law, in which consumer protection is formally prioritised over market integration and industrial development, even though, as Part II noted, this separation is contested in political practice. Yet none of these features is necessary to a precautionary system, as China’s approach demonstrates. China’s developmental precaution conditions market access on safety while simultaneously aligning the pace of regulatory opening with the state’s industrial priorities.

This article does not argue that either model is superior to the other. The argument is more modest and, it is hoped, more durable. The precautionary principle has no single correct form. Its structural features are design choices rather than logical necessities, and China’s case exposes assumptions in the European model that are rarely made explicit. It follows that a more adequate theory of the precautionary principle in global food governance must account for this plurality, developing a framework broad enough to analyse different versions of precaution on their own terms. Because cultivated meat is a tradable product that will cross borders, this divergence is not confined to domestic regulation but will surface in the institutions of international food trade. How these two models will interact in Codex standard-setting, WTO dispute settlement under the SPS Agreement,Footnote 94 and bilateral dialogue is a question the rapid commercial development of cultivated meat will make urgent. Although this interaction has not yet materialised, building the governance architecture to manage it should begin with recognising that other models of precaution are not just politically different but built on different foundations.

Competing interests

The author declares none.

References

1 See L. Petetin, “Frankenburgers, Risks and Approval” (2014) 5 European Journal of Risk Regulation 168; N. Stephens and others, “Bringing Cultured Meat to Market: Technical, Socio-Political, and Regulatory Challenges in Cellular Agriculture” (2018) 78 Trends in Food Science and Technology 155.

2 L. Manning, “Responsible Innovation: Mitigating the Food Safety Aspects of Cultured Meat Production” (2024) 89 Journal of Food Science 4638.

3 M. J. Post and others, “Scientific, Sustainability and Regulatory Challenges of Cultured Meat” (2020) 1 Nature Food 403, at 410–11; S. Chriki and J.-F. Hocquette, “The Myth of Cultured Meat: A Review” (2020) 7 Frontiers in Nutrition 7.

4 Regulation (EU) 2015/2283 of the European Parliament and of the Council of 25 November 2015 on novel foods [2015] OJ L 327/1, Art. 3(2)(a)(vi).

5 Regulation (EC) No 178/2002 of the European Parliament and of the Council of 28 January 2002 laying down the general principles and requirements of food law, establishing the European Food Safety Authority and laying down procedures in matters of food safety [2002] OJ L 31/1, Art. 7.

6 Legge 1 dicembre 2023, n 172 (GU n 281, 1 December 2023), Art. 2.

7 Gourmey, “Gourmey Files the First EU Novel Food Submission for Cultivated Meat” (EIT Food, 26 July 2024) <https://www.eitfood.eu/news/gourmey-files-the-first-eu-novel-food-submission-for-cultured-meat> (last accessed 15 September 2026); Mosa Meat, “Submitting Our First EU Market Authorisation Request” (Mosa Meat, 22 January 2025) <https://mosameat.com/blog/submitting-our-first-eu-market-authorisation-request> (last accessed 15 September 2026).

8 “Zhou Zi Wei Lai Completes China’s Largest Cultivated Meat Pilot Factory” (周子未来建成国内最大细胞培养肉中试工厂) Interface News (界面新闻) (15 December 2025) <https://m.jiemian.com/article/13761662.html> (last accessed 15 September 2026); L Chen, “Global Forum on Cultured Meat Held in Nanjing” (2026 国际细胞培养肉产业论坛在南京举行) China Science Daily (中国科学报) (27 April 2026) <https://news.sciencenet.cn/htmlnews/2026/4/563735.shtm> (last accessed 15 September 2026); Z. Jiulong, “Cultivated Meat: Growing Meat in the Laboratory” (细胞培养肉：在实验室里“种肉”) Xinhua (13 June 2024) <https://www.news.cn/tech/20240613/2768a8d80ead48c1abb7221797f761d8/c.html> (last accessed 15 September 2026).

9 NHC, Reply to Proposal No. 3471 (Science and Technology Category No. 187) of the Third Session of the 13th National Committee of the CPPCC (关于政协十三届全国委员会第三次会议第3471号（科学技术类187号)提案答复的函), 27 July 2020, published 22 January 2021.

10 J. Li and J. Peng, “Construction of Regulatory Framework for the Food Safety of Cultured Meat from the Perspective of Systemic Risk Governance” (系统性风险治理视域下细胞培养肉食品安全监管框架建构) (2026) 47 Food Science (食品科学) 26.

11 J. B. Wiener and others (eds), The Reality of Precaution: Comparing Risk Regulation in the United States and Europe (Routledge 2011); Y. Lu, “Critical Thinking About the Precautionary Principle in China’s Food Safety Law” (2016) 11 Frontiers of Law in China 692.

12 Rio Declaration on Environment and Development (13 June 1992) UN Doc A/CONF.151/26 (vol I), princ. 15.

13 On the principle’s migration across legal regimes and its expansion beyond environmental law, see N. de Sadeleer, Environmental Law Principles: From Regulation to Ecosystemic Governance (2nd edn, Oxford University Press 2020), at 137–53.

14 On the German origins of Vorsorgeprinzip, see K. von Moltke, “The Vorsorgeprinzip in West German Environmental Policy” in Royal Commission on Environmental Pollution, Twelfth Report (HMSO 1988); de Sadeleer, supra, note 13, at 91–6.

15 European Commission, “Communication on the Precautionary Principle” COM(2000) 1 final, 2 February 2000, s. 3.

16 Case C-157/96 R v Ministry of Agriculture, Fisheries and Food, ex parte National Farmers’ Union [1998] ECR I-2211, para. 63; Communication on the Precautionary Principle, supra, note 15, ss. 5.1–5.2.

17 E. Vos, “EU Food Safety Regulation in the Aftermath of the BSE Crisis” (2000) 23 Journal of Consumer Policy 227, at 231.

18 European Commission, “White Paper on Food Safety” COM(1999) 719 final, 12 January 2000; Regulation (EC) No 178/2002, supra, note 5, Arts. 6–7, 22–3; Vos, supra, note 17, at 233–4, 245–6.

19 Regulation (EC) No 178/2002, supra, note 5, Art. 7(1).

20 Ibid., Arts. 6, 22–3; White Paper on Food Safety, supra, note 18, ss. 4–5.

21 E. Fisher, Risk Regulation and Administrative Constitutionalism (Hart Publishing 2007) 211; O. Renn and A. Klinke, “Risk Governance: Concept and Application to Institutional Risk Management” in A. Alemanno and others (eds), Better Business Regulation in a Risk Society (Springer 2012) 17.

22 Regulation (EU) 2015/2283, supra, note 4, Arts. 10–12; EFSA Panel on Nutrition, Novel Foods and Food Allergens (NDA), “Guidance on the Preparation and Submission of an Application for Authorisation of a Novel Food in the Context of Regulation (EU) 2015/2283 (Revision 1)” (2021) 19 EFSA Journal 6555, ss. 3.2–3.4.

23 Gourmey and Mosa Meat, supra, note 7.

24 Regulation (EC) No 178/2002, supra, note 5, Art. 7(2).

25 Panel Report, EC–Measures Affecting the Approval and Marketing of Biotech Products WT/DS291/R, WT/DS292/R, WT/DS293/R (29 September 2006) paras. 7.1781–7.1783; R. Lofstedt, “The Precautionary Principle in the EU: Why a Formal Review is Long Overdue” (2014) 16 Risk Management 137, at 154.

26 G. Majone, “Regulatory Legitimacy” in G. Majone (ed.), Regulating Europe (Routledge 1996) 284; A. Alemanno, “Introduction: Foundations of EU Food Law and Policy” in A. Alemanno and S. Gabbi (eds), Foundations of EU Food Law and Policy: Ten Years of the European Food Safety Authority (Routledge 2016) p 1.

27 On the influence of industry lobbying on EU biotechnology regulation, see G. Winter, “The European Union’s Deregulation of Plants Obtained from New Genomic Techniques: A Critique and an Alternative Option” (2024) 36 Environmental Sciences Europe 47.

28 Regulation (EC) No 178/2002, supra, note 5, Art. 5; Vos, supra, note 17, at 234, 244; Good Food Institute Europe, “Explaining the EU Regulatory Framework for Cultivated Meat” (2024).

29 Legge n 172/2023, supra, note 6, Art. 2.

30 Regulation (EU) 2026/1388 of the European Parliament and of the Council of 17 June 2026 on plants obtained by certain new genomic techniques and their products, and amending Regulation (EU) 2017/625 [2026] OJ L 2026/1388.

31 Ibid., Arts. 5, 6; Winter, supra, note 27.

32 Food Safety Law of the People’s Republic of China (中华人民共和国食品安全法), adopted 28 February 2009, revised 24 April 2015, amended 29 December 2018, further amended 29 April 2021, Art. 3.

33 F. Snyder and Y. S. Kim, “China’s 2015 Food Safety Law: Crossing the River but Feeling the Stones and Avoiding Low Branches?” (2018) 6 Chinese Journal of Comparative Law 1.

34 Food Safety Law, supra, note 32, Arts. 34, 53–4; Snyder and Kim, supra, note 33, at 1–10.

35 Food Safety Law, supra, note 32, Art. 37.

36 Ibid., Arts. 2, 5, 91.

37 X Guan and others, “Technological Trend and Suggestions for Cultured Meat Industry” (细胞培养肉产业技术发展态势及建议) (2023) 25(3) Strategic Study of CAE [中国工程科学] 251.

38 C. Scott, “Analysing Regulatory Space: Fragmented Resources and Institutional Design” (2001) Public Law 329.

39 Measures for the Administration of Safety Review of New Food Ingredients (新食品原料安全性审查管理办法), National Health and Family Planning Commission Order No. 1, 31 May 2013, amended 26 December 2017, Art. 2.

40 NHC Reply, supra, note 9.

41 S. Ding, C. Li and G. Zhou, “Progress and Challenges in Cultured Meat Technology and Industrialisation” (细胞培养肉技术及产业化进展与挑战) (2022) 22(12) Journal of the Chinese Institute of Food Science and Technology [中国食品学报] 33.

42 Food Safety Law, supra, note 32, Arts. 37, 124(8).

43 Measures on New Food Ingredients, supra, note 32, Arts. 5–6; NHC, Provisions on the Declaration and Acceptance of New Food Ingredients (新食品原料申报与受理规定) (2021), Art. 7.

44 CFSA, Notice on Improving the Safety Assessment Material Requirements for “Three New Foods” (关于完善“三新食品”安全性评价资料要求的通知) (13 September 2024), with annexed Requirements for Safety Assessment Application Materials for Genetically Modified Microorganisms Used in Food Processing (Trial) (食品加工用遗传修饰微生物安全性评价申报材料要求（试行)).

45 D. J. Ettinger, J. X. Li and S. (Sharon) Tian, “From the Year of the Snake to the Year of the Horse: China’s Food Law Year-in-Review 2025” (18 February 2026) <https://www.khlaw.com/insights/year-snake-year-horse-chinas-food-law-year-review-2025> (last accessed 15 September 2026).

46 NHC, Notice of the General Office of the National Health Commission on Issuing the 2025 Annual Plan for the Formulation and Revision of National Food Safety Standards (国家卫生健康委办公厅关于印发 2025 年度食品安全国家标准立项计划的通知), Guo Wei Ban Shi Pin Han [2025] No. 319 (国卫办食品函 (2025) 319号), 25 July 2025.

47 Ettinger, Li and Tian, supra, note 45.

48 “FAO and CFSA Hold Seminar as Cell-Cultured Food Regulation Takes a New Step” (联合国粮农组织与国家食品风评中心召开研讨会，细胞培养食品法规迈出新的一步), Eastday.com (东方网) (26 December 2023) <https://ex.chinadaily.com.cn/exchange/partners/82/rss/channel/cn/columns/sz8srm/stories/WS658a6f85a310c2083e4149e0.html> (last accessed 15 September 2026).

49 Ministry of Agriculture of the People’s Republic of China (now MARA), Measures for the Administration of the Safety Assessment of Agricultural Genetically Modified Organisms (农业转基因生物安全评价管理办法), Order No. 8, 5 January 2002, as amended 2004, 2016 and 2017 and 2022, Arts. 3, 8.

50 CFSA, “Three New Foods” Notice, supra, note 44.

51 MOA, Measures, supra, note 49, Art. 8; CFSA, “Three New Foods” Notice, supra, note 44.

52 Measures on New Food Ingredients, supra, note 39, Arts. 3, 6, 23.

53 FAO and WHO, Food Safety Aspects of Cell-Based Food (28 March 2023).

54 Standing Committee of the National People’s Congress, Administrative Licensing Law of the People’s Republic of China (中华人民共和国行政许可法), adopted 27 August 2003, amended 23 April 2019, Arts. 42, 45; Measures on New Food Ingredients, supra, note 39, Art. 11; NHC, Implementation Specifications of Administrative Licensing Matters for Approval of New Food Raw Materials (新食品原料审批行政许可事项实施规范) (25 September 2023), pt. 7.

55 Implementation Specifications, supra, note 54, pt. 7.

56 OECD, Safety Evaluation of Foods Derived by Modern Biotechnology: Concepts and Principles (OECD 1993). On substantial equivalence under the earlier Novel Food Regulation and its application to cultured meat, see Petetin, supra, note 1; and L Petetin, “Precaution and Equivalence: The Critical Interplay in EU Biotech Foods” (2017) 42 European Law Review 831.

57 Measures on New Food Ingredients, supra, note 39, Art. 16; NHC, Guide to the Administrative Approval of New Food Ingredients (新食品原料审批办事指南) 2023.

58 Measures on New Food Ingredients, supra, note 39, Art. 23.

59 Y-C Lin and others, “Cultivated Meat for Sustainable Food Security and Environmental Resilience” (2026) 13 Future Foods 100984, ss. 3.1–3.5.

60 Regulation (EU) 2015/2283, supra, note 4, Art. 6.

61 Regulation (EC) No 178/2002, supra, note 5, Art. 7.

62 Case T-13/99 Pfizer Animal Health SA v Council [2002] ECR II-3305, paras. 139–144; Case C-236/01 Monsanto Agricoltura Italia SpA v Presidenza del Consiglio dei Ministri [2003] ECR I-8105, para. 111.

63 Commission, Precautionary Principle Communication, supra, note 16, paras. 4–5.

64 Regulation (EC) No 178/2002, supra, note 5, Arts. 22–3; Fisher, supra, note 21, at 211.

65 Measures on New Food Ingredients, supra, note 39, Art. 6.

66 Regulation (EC) No 178/2002, supra, note 5, Art. 7(2).

67 Appellate Body Report, EC Measures Concerning Meat and Meat Products (Hormones) (16 January 1998) WT/DS26/AB/R, WT/DS48/AB/R, para. 124; Agreement on the Application of Sanitary and Phytosanitary Measures, Marrakesh Agreement Establishing the World Trade Organization, Annex 1A, 1867 UNTS 493 (1994) (SPS Agreement).

68 EC–Biotech, supra, note 25.

69 Regulation (EC) No 178/2002, supra, note 5, Art. 7(2).

70 Measures on New Food Ingredients, supra, note 39, Arts. 11, 16; Administrative Licensing Law, supra, note 47, Arts. 42, 45.

71 Measures on New Food Ingredients, supra, note 39, Art. 18.

72 2’-fucosyllactose was authorised in the EU as a novel food ingredient in 2016 and approved in China as a nutritional fortifier in 2023. Commission Implementing Decision (EU) 2016/376 of 11 March 2016 [2016] OJ L70/27; NHC, Announcement on 15 “Three New Foods” Including Peach Gum (关于桃胶等 15 种“三新食品”的公告) No. 8 of 2023.

73 MARA, 14th Five-Year Plan for National Agricultural and Rural Science and Technology Development (“十四五”全国农业农村科技发展规划), Nong Ke Jiao Fa [2022] No. 1, March 2022, Section on “Future Food Manufacturing.”

74 SPS Agreement, supra, note 67, Art. 5.7.

75 Regulation (EC) No 178/2002, supra, note 5, Arts. 5, 7, 22–3.

76 Legge n 172/2023, supra, note 6, Art. 2; M A Fino and others, “Cultivated Meat beyond Bans: Ten Remarks from the Italian Case toward a Reasoned Decision-Making Process” (2024) 7(12) One Earth 2108.

77 European Parliament Resolution of 19 October 2023 on European Protein Strategy (2023/2015(INI)) P9_TA(2023)0375 [2024] OJ C/2024/2657.

78 European Commission, Communication from the Commission – TRIS/(2023) 3362, Notification 2023/0675/IT, 29 January 2024.

79 B. van der Meulen, “Concluding Observations” in G. Formici, M. C. Mancini and L. Scaffardi (eds), Cell-Based Meat in the European Union and Beyond (Springer 2025) p. 289.

80 National Development and Reform Commission of the People’s Republic of China, 14th Five-Year Plan for the Development of the Bioeconomy (国家发展改革委⟪“十四五”生物经济发展规划⟫), issued 10 May 2022, pt. IV, s. 10 and pt. VII, s. 18.

81 Xi Jinping, remarks at Heilongjiang Province inspection (September 2023), People’s Daily (12 September 2023) <https://www.news.cn/politics/leaders/2023-09/09/c_1129854063.htm> (last accessed 16 June 2026); Li Qiang, Government Work Report (政府工作报告), delivered at the Second Session of the 14th National People’s Congress, 5 March 2024, s. III(1).

82 Majone, “Regulatory Legitimacy,” supra, note 27; C. Johnson, “The Developmental State: Odyssey of a Concept” in M. Woo-Cumings (ed.), The Developmental State (Cornell University Press 1999), ch. 2.

83 L. Reins and T. Ala-Lahti, “Innovation, Precaution, and Sustainable Development in EU Environmental Law–A False Trichotomy?” (2026) 38 Journal of Environmental Law 1, at 6, 8–9.

84 D. Vogel, The Politics of Precaution: Regulating Health, Safety, and Environmental Risks in Europe and the United States (Princeton University Press 2012), at 1–20.

85 Johnson, “The Developmental State,” supra, note 82, ch. 3; Y. Y. Ang, How China Escaped the Poverty Trap (Cornell University Press 2016), ch. 2.

86 J. Jordana and D. Levi-Faur, “The Politics of Regulation in the Age of Governance” in J. Jordana and D. Levi-Faur (eds), The Politics of Regulation (Edward Elgar 2004).

87 14th Five-Year Bioeconomy Plan, supra, note 80, pts. IV and VII.

88 Xi, Remarks, supra, note 81.

89 State Council of the People’s Republic of China, Opinions on Reforming the Review and Approval System for Drugs and Medical Devices (国务院关于改革药品医疗器械审评审批制度的意见), Guo Fa [2015] No. 44, 18 August 2015.

90 Measures on New Food Ingredients, supra, note 39.

91 Food Safety Law, supra, note 32, Art. 3.

92 National People’s Congress, Administrative Litigation Law of the People’s Republic of China (中华人民共和国行政诉讼法), adopted 4 April 1989, amended 1 November 2014, further amended 27 June 2017, Art. 70(6).

93 Ministry of Science and Technology, Notice on the Publication of Application Guidelines for the 2020 Annual Projects of National Key R&D Programme Special Projects including Green Biomanufacturing (18th Batch) (科技部关于发布国家重点研发计划“绿色生物制造”重点专项 2020 年度项目申报指南的通知（第十八批)), 25 September 2020.

94 SPS Agreement, supra, note 67, Art. 5.7.