Logistics Service Providers as Compliance Intermediaries: Reallocating Traceability Obligations in Vietnam’s Agro-Forestry Export Supply Chains

to relationships between actors; and documents published before 2015, with the exception of two foundational theoretical works retained for their role in constructing the analytical framework. The analytical procedure comprised three steps. Step one was deductive coding according to four obligation groups drawn directly from the regulatory text: supplying geolocation information, ensuring the integrity and separability of consignments, retaining due diligence records for the statutory period, and supplying information upon request by the competent authority. Step two was cross-tabulation: for each obligation group, the study determined simultaneously two attributes, namely the actor bearing legal responsibility and the actor holding the data in practice, each attribute coded on a three-point scale of full, partial and none. The difference between the two attributes constitutes the indicator measuring the degree of misalignment. Step three was cross-verification between source tiers. Coding was performed independently by two persons, points of disagreement were resolved through discussion, and the initial level of agreement was reported. Rules for handling conflicts were specified in advance to ensure replicability. An item of information was admitted to the results only when confirmed by at least one tier-one or tier-two source, or by two mutually independent tier-three sources. Where tiers conflicted, information from the higher tier was given priority. Conflicts that could not be resolved by this rule are stated openly in the results section rather than discarded silently; a typical case is the inconsistency among secondary sources regarding Vietnam’s country risk classification status. The research design has three limitations that must be acknowledged. First, the results are conceptual and are stated as propositions requiring empirical testing after the regulation enters into force. Second, the documentary base is systematically biased towards formal actors; the informal collector segment leaves few documentary traces, so the role of this group may be reflected incompletely. Third, the legal framework is still evolving, and the findings are therefore tied to the regulatory position as at October 2026. 5. Analytical Results Analysis of the original texts allows the due diligence obligations under the European Union deforestation regulation to be separated into four groups of differing legal character. The first group comprises information collection obligations, in which the key requirement is the geolocation coordinates of all plots of land where the commodity was produced, together with description, quantity, country of production, time of production and information on the supplier. The second group comprises risk assessment and risk mitigation obligations. The third group comprises the obligation to establish a due diligence system and retain records for a period of five years. The fourth group comprises the obligation to submit a due diligence statement and to supply information when the competent authority of a Member State conducts a check. A legal point requiring clarification is the scope of the simplified due diligence mechanism applicable to commodities originating from a country in the low-risk group. This mechanism exempts the importing operator only from the second obligation group. The remaining three groups are retained in full. The one per cent check rate reflects only the probability of inspection, not the volume of data that must be prepared. Misreading this point is the source of the widespread but inaccurate claim that low-risk status allows Vietnamese firms to enjoy relaxed traceability requirements. A second point frequently misunderstood is that the regulation does not impose an obligation of physical separation between consignments. What the regulation requires is that the due diligence statement list all plots corresponding to the specific consignment. The logical consequence of this requirement is stricter than a ban on mixing: mixing is not prohibited, but it expands the list of plots to be declared and means that a single non-compliant plot can invalidate the entire consignment. This is precisely the mechanism that turns traditional aggregation practices into a systematic source of legal risk. Table 1 presents the matrix contrasting the four obligation groups with the two attributes coded according to the procedure set out in section four. Table 1. Matrix contrasting legal obligations under the European Union deforestation regulation with actors holding the data in practice in Vietnam’s agro-forestry export supply chains Obligation group Legal basis Actor bearing legal Actor holding the Typical failure mode in the absence Degree of misalignmen t responsibilit y data in practice of a compliance intermediary Collecting geolocation information on plots of production land Article 9 Importer in the European Union Smallholder s generate it; collectors and procurement agents hold it in fragmented form; processing firms hold it in aggregated form Coordinates exist but cannot be linked to the specific consignment; the declared list of plots is incomplete or inaccurate High Risk assessment and risk mitigation Articles 10 and 11, exempted under the simplified due diligence mechanis m in Article 13 for low- risk origin Importer in the European Union Importer; providers of remote sensing data No significant failure arises under Vietnam’s current conditions Low Ensuring consignmen t traceability and retaining records for five years Articles 9 and 12 Importer in the European Union; the exporting firm is bound only by contract The logistics service provider is the sole party continuously holding transport, warehousing and customs documents for each consignment The consignment’s journey cannot be reconstructed when queried retrospectively ; records are dispersed across multiple parties with no one responsible for retention High Submitting the due diligence statement and supplying Article 4 and the chapter on checks by competent authorities Importer in the European Union Data are scattered upstream; the importer holds only the portion Response time exceeds the required deadline; the consignment is detained Medium information upon inspection already declared pending supplementary evidence Overall - Responsibilit y is concentrated almost entirely in a single actor located outside Vietnamese territory Data are dispersed across three groups of domestic actors, none of which bears a statutory obligation Risk is funnelled towards Vietnamese exporting firms through contractual clauses rather than through law High Note: The degree of misalignment is determined on a three-point scale of low, medium and high, on the basis of the difference between the codes in the two columns for legal responsibility and control over data. Read vertically, Table 1 reveals a pronounced asymmetry. The legal responsibility column contains essentially a single actor, whereas the data control column extends across at least three groups of actors, none of which bears an obligation towards the regulatory authority. Read horizontally, high misalignment is concentrated in the first and third obligation groups, that is, precisely the two groups from which low-risk status grants no exemption. This result reinforces the paradox set out in the introduction and shows that it is not a rhetorical claim but a structural feature of the regulation’s design. The cause of the high degree of misalignment lies in the procurement structure of the chain. Vietnam’s coffee and rubber industries operate on a base of very numerous small-scale smallholder producers connected to processing firms through multiple layers of collectors and agents. Relationships at these intermediary layers rest largely on cash, trust and local custom, and rarely leave verifiable documentation. A survey of 286 smallholders across three continents found that the greatest barrier lies not in awareness but in the ability to demonstrate legality and supply coordinates (Mwangi et al., 2025), while a survey of 333 smallholders in 16 countries recorded that 74 per cent feared being pushed out of tightly regulated markets (International Institute for Sustainable Development, 2026). At the collection point, coffee from dozens of households is poured into a single mass. From that moment, the one-to-one relationship between plot coordinates and physical volume no longer exists in reality; it exists only if reconstructed through documentation. This is the point of data rupture represented by the dashed arrow in Figure 1. It is precisely at this point of rupture that four compliance functions emerge at the logistics service provider. The first function is generating and retaining documents attached to the consignment. Bills of lading, warehouse receipts, delivery notes, tally records and customs declarations are the only documents in the chain drawn up for each consignment, bearing a time stamp and possessing independent legal value. The second function is maintaining physical separation between volumes of goods with different records. This is not mandated by the regulation but required by the logic of risk, and it can be performed only at the warehousing stage. The third function is linking commercial data with customs procedure data, where freight forwarders play a familiar technical coordinating role. The fourth function is supplying retrospective evidence when a consignment is queried months later. Transaction cost theory explains why transferring these four functions is the efficient option. Freight forwarders already operate consignment-level documentary infrastructure for business reasons, independently of any legal requirement; the marginal cost of adding traceability data fields to existing infrastructure is far lower than the cost for each exporting firm to build its own collection system for thousands of suppliers. This observation is consistent with the conclusion that the midstream of value chains in developing economies plays a decisive role yet has been neglected in research (Reardon et al., 2021), and it extends the research stream on the governance role of midstream actors to a group of actors that do not take ownership of the goods (Grabs & Carodenuto, 2021). Whether the compliance intermediary can in fact operate, however, depends on four conditions that Vietnam’s current situation does not fully meet. The first is data capability. The domestic logistics service market consists mainly of small-scale firms, and a systematic review of this group of actors shows that information technology remains the least developed of the five thematic areas surveyed (Evangelista et al., 2018). The second is the legal basis. Current law on logistics service business does not define the responsibilities, powers and retention periods applicable to traceability data, leaving this function outside the scope of default contractual obligations. The third is interoperability standards. Growing-area data, enterprise data and electronic customs data currently lack a common identification standard permitting automatic matching at consignment level. The fourth is the cost allocation mechanism. When legal obligation is concentrated in the importer while implementation costs arise upstream, costs tend to be pushed down to the weakest link through contractual clauses, reproducing precisely the power asymmetry mechanism described in research on multi-tier chains (Wilhelm et al., 2016). Taken together, the analytical results show that the logistics service provider has become a compliance intermediary in practice without being recognised as such institutionally. The distance between its actual role and its legal status constitutes the systemic risk facing Vietnam’s agro-forestry export chains during the current transitional period. 6. Discussion The analytical results raise three issues requiring discussion at the theoretical level, the policy level and the level of generalisability. Theoretically, the study’s findings compel a reconsideration of how the supplier capability variable is defined in the global value chain governance framework. Gereffi et al. (2005) understand supplier capability primarily as production capability, that is, the ability to meet a product’s technical requirements at a given level of quality and cost. Under a regulatory regime based on the obligation of proof, the capability that determines a link’s position is instead data capability, that is, the ability to generate, retain and transmit evidence attached to each consignment. These two types of capability do not coincide and do not necessarily reside in the same actor. A processing firm may meet the highest production standards yet still fail in compliance terms if it cannot access data residing at another link. This is the point at which the present study supplements the original theoretical framework. A second theoretical consequence concerns the double agency model. Wilhelm et al. (2016) describe the first-tier supplier as an actor that both complies and transmits requirements downwards, on the assumption that responsibility propagates along the ownership of goods. The concept of the compliance intermediary proposed here describes a mechanism that differs in nature. This actor does not own the goods and has no contractual relationship with the lead firm, yet it holds the information infrastructure on which another party’s obligation depends. In other words, power here derives not from a position in the flow of ownership but from a position in the flow of information. This finding extends the research stream on the governance role of midstream actors to service-providing actors (Grabs & Carodenuto, 2021; Reardon et al., 2021), and at the same time reveals the limits of the bilateral buyer-supplier approach. In policy terms, the most important finding is the dual character of low-risk status. Administratively, this status is a considerable negotiating achievement, reducing the check rate and removing one obligation group from the importing side. In terms of incentives, however, it may produce a perverse effect. When the public signal is low risk, firms readily infer that their existing level of preparation is sufficient, whereas three of the four obligation groups remain intact and are precisely the three with the highest degree of misalignment. It must further be emphasised that the acceptance standard in practice is set not by the regulatory authority but by the buyer. The importer, bearing full legal responsibility, will tend to demand a level of proof above the statutory minimum in order to protect itself, and the cost of that above-standard requirement will be transferred back upstream through contractual clauses. The distributional consequence is that risk accumulates at the weakest link (International Institute for Sustainable Development, 2026). One plausible counter-argument must be addressed directly. It may be argued that the market will adjust by itself: if data has value, exporting firms will purchase the service and logistics service providers will upgrade their capabilities accordingly. This argument is correct in principle but overlooks three practical obstacles. The value of traceability data is realised only when the whole chain participates, creating a classic coordination problem. The benefits of investment accrue to downstream links while the costs arise upstream, creating a dispersion of investment returns. Most importantly, without a legal basis establishing retention responsibility, the service provider has no incentive to bear legal risk for an obligation that is not its own. These three obstacles explain why institutional intervention is necessary rather than waiting for the market. With respect to generalisability, the proposed analytical framework is not tied to any single regulation. It can be applied to any regulatory regime sharing three features: obligations evidenced by data rather than by product attributes, the consignment rather than the firm as the unit of evidence, and dispersed supply. Mechanisms on corporate sustainability due diligence, on carbon border adjustment and on digital product passports all satisfy these three features. Similarly, the framework can be transferred to exporting countries with dispersed smallholder structures. The limit of generalisability lies in the third condition set out in the theoretical framework: a central actor already possessing an operational data flow must exist. In chains where the logistics stage is itself fragmented and informal, a compliance intermediary will not emerge spontaneously, and the problem then becomes one of building public infrastructure rather than one of allocating roles. 7. Conclusions And Policy Implications The study proceeds from a paradox: Vietnam enjoys low-risk status with a check rate of only one per cent, yet the volume of data to be prepared for each consignment is not reduced at all. Three answers emerge. First, traceability obligations are allocated highly asymmetrically: legal responsibility is concentrated almost entirely in the importer within the European Union, whereas the underlying data are scattered across smallholders, collectors and logistics service providers, none of which bears a statutory obligation. Second, the logistics service provider becomes a compliance intermediary because it is the only link already operating consignment- level documentary infrastructure for business reasons, and therefore faces the lowest marginal cost in taking on the function of generating and retaining evidence. Third, the effectiveness of the compliance intermediary depends on four conditions that are currently unmet: the data capability of service firms, a legal basis establishing retention responsibility, data interoperability standards and a cost allocation mechanism. For State regulatory agencies Add to the law on logistics service business provisions on the service provider’s responsibility to retain and supply traceability data, together with a minimum retention period corresponding to the requirements of the importing market. Issue a common consignment identification standard for the electronic customs system and growing-area databases. Incorporate geolocation data fields into the set of electronic export documents for the three commodity groups of coffee, rubber and timber. For industry associations Develop model documentation linking plot coordinate lists to consignment codes. Organise data capability training for small-scale freight forwarders. Establish a cost-sharing mechanism between exporting firms and suppliers rather than pushing the entire burden onto smallholders. For exporting firms Incorporate data obligation clauses into logistics service contracts, specifying clearly the scope, format and retention period. Do not treat low-risk status as grounds for delaying investment, because the acceptance standard in practice is determined by the buyer and not by the check rate. For logistics service firms Position compliance assurance services as a new revenue stream rather than an incidental cost. Prioritise investment in consignment-level data matching capability before investing in complex technology. The study has two main limitations. First, the results are conceptual and are stated as propositions requiring empirical testing once the regulation enters into application. Second, the documentary base is biased towards formal actors, while the informal collector segment leaves few documentary traces. Two directions for further research are proposed: a quantitative survey of the data capability and actual compliance costs of freight forwarders after the December 2026 milestone; and a comparative case study of the coffee chain and the timber chain in order to test the conditions for the emergence of a compliance intermediary in two different chain structures. References Arthur Cox. (2025). EUDR: One year postponement agreed. https://www.arthurcox.com/insights/eudr-one-year-postponement-agreed/ Astill, J., Dara, R. A., Campbell, M., Farber, J. M., Fraser, E. D., Sharif, S., & Yada, R. Y. (2019). Transparency in food supply chains: A review of enabling technology solutions. Trends in Food Science & Technology, 91, 240–247. https://doi.org/10.1016/j.tifs.2019.07.024 Bộ Nông nghiệp và Môi trường [Ministry of Agriculture and Environment]. (2026). Tình hình xuất nhập khẩu nông lâm thủy sản Việt Nam sang thị trường EU năm 2025 [Vietnam’s agro-forestry-fishery exports and imports to the EU market in 2025]. https://agro.gov.vn/images/files/B%c3%a1o%20c%c3%a1o%20n%c4%83m%202025 _XK%20NLTS_EU_VN.pdf Commission Implementing Regulation (EU) 2025/1093 establishing the list of countries or parts thereof classified by risk level under Regulation (EU) 2023/1115. Evangelista, P., Santoro, L., & Thomas, A. (2018). Environmental sustainability in third-party logistics service providers: A systematic literature review from 2000–2016. Sustainability, 10(5), 1627. https://doi.org/10.3390/su10051627 Gereffi, G., Humphrey, J., & Sturgeon, T. (2005). The governance of global value chains. Review of International Political Economy, 12(1), 78–104. https://doi.org/10.1080/09692290500049805 Grabs, J., & Carodenuto, S. L. (2021). Traders as sustainability governance actors in global food supply chains: A research agenda. Business Strategy and the Environment, 30(2), 1314–1332. https://doi.org/10.1002/bse.2686 Hiệp hội Gỗ và Lâm sản Việt Nam [Vietnam Timber and Forest Product Association]. (2026). Doanh nghiệp gỗ xuất khẩu sang châu Âu tăng cao nhất từ trước đến nay [Timber enterprises’ exports to Europe record the highest increase ever]. https://trungtamwto.vn/tin-tuc/31760-doanh-nghiep-go-xuat-khau-sang-chau-au-tang- cao-nhat-tu-truoc-den-nay International Institute for Sustainable Development. (2026). Smallholders and the European Union Deforestation Regulation: Challenges, implications, and pathways to compliance. https://www.iisd.org/publications/report/smallholders-european-union- deforestation-regulation Mwangi, S., Sorsby, N., & Wagner, K. (2025). Smallholder EUDR perceptions. International Institute for Environment and Development. https://www.iied.org/22673g Reardon, T., Liverpool-Tasie, L. S. O., & Minten, B. (2021). Quiet Revolution by SMEs in the midstream of value chains in developing regions: Wholesale markets, wholesalers, logistics, and processing. Food Security, 13(6), 1577–1594. https://doi.org/10.1007/s12571-021-01224-1 Regulation (EU) 2023/1115 of the European Parliament and of the Council of 31 May 2023 on the making available on the Union market and the export from the Union of certain commodities and products associated with deforestation and forest degradation. Snyder, H. (2019). Literature review as a research methodology: An overview and guidelines. Journal of Business Research, 104, 333–339. https://doi.org/10.1016/j.jbusres.2019.07.039 Wilhelm, M. M., Blome, C., Bhakoo, V., & Paulraj, A. (2016). Sustainability in multi-tier supply chains: Understanding the double agency role of the first-tier supplier. Journal of Operations Management, 41, 42–60. https://doi.org/10.1016/j.jom.2015.11.001 Williamson, O. E. (1985). The economic institutions of capitalism: Firms, markets, relational contracting. Free Press.