D3.4 Quantified ex-post impacts of trade in sea biomass on biodiversity

A significant share of sea biomass – particularly fishmeal and fish oil – feeds the aquaculture sector. The production of both fisheries and aquaculture is highly concentrated, creating complex international trade interdependencies that link aquaculture expansion to biodiversity threats in both marine and terrestrial ecosystems. As the aquaculture sector remains relevant to global food systems and its demand for sea biomass continues to grow, promoting sustainable aquaculture production (and expansion) becomes essential to mitigating its associated biodiversity risks. Achieving this requires the participation of both governments and non-government actors. Governments should continue to support their domestic aquaculture industries while ensuring that the aquatic products available in their markets meet robust social and environmental standards. At the same time, non-government actors can provide incentives to aquaculture producers to adopt sustainable production practices. This report thus explores the roles of two trade-related measures, regulatory standard-like non-tariff measures (NTMs) and voluntary market-based aquaculture sustainability certification. Together, these two measures capture how the imports and exports of aquaculture-related biomass are governed, with NTMs regulating access to domestic markets and sustainability certification influencing production practices in export-oriented supply chains. We first assess standard-like NTMs at the country level. To protect consumer health and safety, importing countries tend to impose regulations on aquatic foods from overseas, such as testing, labelling, and certification requirements that include detailed information on ingredients, net quantity, production methods, and antimicrobial residue levels. Some NTMs also address environmental risks (e.g., invasive species) or safeguard global common resources (e.g., protected endangered species of wild fauna and flora). These measures may influence domestic producers by altering domestic market dynamics via changing consumer demand for higher-quality and safer products, or as unintentional side effects of protectionist policies. We ask whether these NTMs implemented by importing countries hinder their domestic aquaculture producers, thus reinforcing the existing geographical unevenness in global aquatic product trade. We then borrow the aquaculture development archetype framework by Partelow et al. (2025) to contextualize how these impacts vary across countries’ differences in economic, environmental, and governance dimensions. As these country groups have various aquatic farmed biodiversity types (e.g., brackish, marine, freshwater), the findings offer implications for farmed species resource management. The analysis here utilizes the output of Deliverable 3.1 Database on biomass trade, production, demand, and biodiversity proxies. For this analysis, we use a panel of 99 aquaculture producer countries during 2010-2020 and apply two-way fixed-effect regressions. Next, we turn to Vietnam’s Mekong Delta as a case study to explore the landscape-level impacts of aquaculture sustainability certification. The mangrove ecosystems in Vietnam support a wide range of biodiversity, including 521 species from fish to birds, mammals, reptiles, bivalves, sea cucumbers, insects, and plants. However, 148 species are reported to experience population declines. Between 1973 and 2020, Vietnam lost approximately 2150 hectares of mangrove annually, primarily due to aquaculture expansion. We ask whether aquaculture sustainability certification can help protect mangrove habitats. To answer this, we rely on a grid-based panel data, where each grid (cell) represents a mangrove-aquaculture landscape. We then evaluate the impact of cells’ exposure to certified shrimp farms on two mangrove outcomes: absolute mangrove area and relative share within each landscape. We establish credible counterfactuals using matching techniques that account for the geographical traits and historical land-use changes in the local and adjacent landscapes. We then quantify the average and dynamic effects of certification with two-way fixed effects and event-study analysis. This analysis draws on spatial data from Milestone MS6 Identification of aquaculture production sites. Our findings highlight the nuanced impacts of the two studied governance measures across scales. At the country level, NTMs’ impact on domestic aquaculture production is shaped by four aquaculture archetypes that represent diverse contexts of aquaculture development. Emerging aquaculture producers (mainly Central Asia and West Africa countries) benefit most from the positive impacts of NTMs. In contrast, the inherent characteristics of the other archetypes hint at systemic barriers against the implementation and compliance with NTMs, even in the short term. When compliance requires substantial fixed-cost investments, developing aquaculture producers (Southeast Asia and some Latin American countries) are most negatively affected. But when compliance involves marginal costs associated with inspections or certification, wealthy aquaculture producers (Western Europe, the U.S., Canada, and Japan) are most negatively affected. For aquaculture producers with limited aquatic food engagement (Eastern European and a few African, Latin American, and Asian countries), NTMs show consistently insignificant effects, reinforcing their categorization as having limited aquaculture growth. These findings underscore the importance of incorporating aquaculture archetypes into the designs of trade policy instruments. At the local level, aquaculture sustainability certification shows a positive association with mangrove outcomes, signaling positive news for mangrove conservation through market mechanisms. However, its minimal and localized effect suggests the need to revisit the targets of certification schemes and to complement conservation with other policies.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.