In the Netherlands, the Customs Administration supervises the flow of goods that enter or leave the European Union (EU) via the country’s external borders. “Supervision” is to be understood in a broad sense, and includes all acts aimed at establishing compliance with the wide range of legislation that governs international trade in goods with the EU.

When it comes to ensuring the compliance of imported products with EU safety requirements – an activity known as “market surveillance” – Dutch Customs cooperates with several specialized agencies. For these market surveillance agencies, the supervision of imported goods is only one among many tasks. They must also ensure that the goods available on the market, whether or not they are imported, comply with the complex system of rules and legal requirements covering their production, processing, storage, transport and sale.

Among these market surveillance agencies is the Netherlands Food and Consumer Product Safety Authority (NVWA), which monitors the safety of food, feed and consumer products and enforces regulations related to animal welfare and nature. In the area of food and feed safety, supervision extends throughout the food and feed chain, from farm to fork, and includes imports of agricultural raw commodities such as soy. Controlling imported goods is a particular challenge for the NVWA, considering the volumes of non-EU goods that enter the Netherlands daily, mainly via the Port of Rotterdam and Amsterdam Airport (Schiphol).

Cooperation between the NVWA and Dutch Customs is facilitated by covenants signed at ministerial level. Although enforcing regulations at the border is not the NVWA’s core business, we would like, in this article, to open the discussion on how to improve this cooperation, especially in the area of joint risk analysis, taking the example of food and feed safety.

The EU food and feed safety regulations distinguish between two types of controls at the EU’s external borders:

• mandatory controls on goods from certain third countries, due to a known or emerging risk or because there is evidence that serious non-compliance with the relevant rules might be taking place. This type of control is referred to in this article as Type 1 controls. Products subject to Type 1 controls are listed in various regulations. The list of products and their designated origins is updated frequently.[1] Examples of identified risks are pesticide residues in goji berries, salmonella in black pepper, aflatoxins in peanuts, palm oil containing Sudan dye, or sulphites in dried apricots.
• risk analysis-based controls to ascertain that known risks are dealt with, and to shed light on new risks. This type of control is performed on the basis of the National Control Plan[2] and is referred to in this article as Type 2 controls.

In the Netherlands all consignments must be presented to Customs, which carries out the risk assessment. Close cooperation has been established between the NVWA and Customs in order to conduct Type 1 and Type 2 controls efficiently, in line with EU regulations which require the implementation of “a common integrated system of official controls”.[3]

Let’s take Type 1 “integrated (non-)veterinary control” as an example. The EU regulations stipulate that consignments of products subject to Type 1 controls must be notified to the competent authority prior to their arrival at the external EU border. The authority has to check the documentation for all consignments, and a prescribed percentage of them will undergo a physical inspection. If the Dutch Customs officers reviewing the documentation suspect non-compliance in the case of a consignment of products subject to Type 1 controls, they will hand the case over to NVWA officers who are stationed next door, at the same location, in order for them to handle the consignment.

Implementing Type 2 controls also requires close cooperation, as the NVWA lacks information about the consignments that enter the EU, but does have information about the risks posed by certain products, countries of origin, exporters or importers, as well as information on seasonal patterns. Customs has information about the supply chain, and can identify anomalies in trade flows. Establishing the risk management strategy is therefore a joint effort. By combining information, both agencies can monitor and manage risks. Based on the joint risk analysis, Customs knows which consignments to draw to the NVWA’s attention if they enter the EU. The NVWA is duly informed, and decides on the optimal moment for supervision in order to have the minimum impact on trade flows – this may be at the time of entry into the EU, during transport or temporary storage awaiting the definitive Customs procedure, during a Customs procedure, or after release for free circulation in the internal EU market.

Until the early 2000s, the decision to inspect consignments subject to Type 2 controls occurred after a Customs declaration for release for free circulation was lodged. Customs would suspend the release for free circulation when it had reason to believe that the consignment might present a risk, awaiting instructions from the NVWA. This caused unforeseeable delays and frustration to importers.

The NVWA started looking for a better way to proceed, and decided to change the timing of the decision to perform an inspection, so that it would occur before the goods arrived in the EU. Traders would be notified before the goods arrived. This approach would enable them to anticipate a foreseeable delay.

Although controls are organized to be as seamless as possible for trade, the joint risk assessment process does have scope for improvement. Even though they share a mutual ambition to address risks efficiently and effectively, Customs and the NVWA do not share the same nomenclature of goods or the same system of identifying companies.

When classifying goods, the two authorities speak a different language. Customs uses the Combined Nomenclature (CN), which is based on the WCO Harmonized System and covers all goods. Other agencies use different coding systems targeting a specific sector, or targeting goods with certain attributes. Most of the time, these classification systems have different levels of granularity.

Let’s take the example of feed additives. The EU lists in a publicly available register all the feed additives that can be imported. The products vary from basic commodities (e.g. ferric oxide) to very complex preparations (e.g. “preparation 6-phytase produced by Komagataella pastoris”). When comparing the 1,557 entries listed in the register in July 2019, around half of the feed additives (48%) match a single CN code. The rest combine two or three products in the CN codes – for instance, one additive can be made of an oil, an essence and an extract. In the most extreme example, one code (2102.1090.10) represents 122 additives (they are all different enzymes).

The same issue exists with wildlife commodities. A considerable proportion of these are classified into broad categories in the CN, instead of being placed in smaller, well-defined taxonomic units. This lack of trade code granularity limits the ability to monitor wildlife trade.[4]

Also, the sphere of use of the goods is not always detailed in the goods description in the Customs nomenclature. Sphere of use can sometimes be a classification criterion in the CN, but not always. Although such information is often not important to Customs, it can be critical to national authorities. For example, in the Netherlands the intended use of facemasks – medical, professional or domestic – determines which of the three national authorities is in charge of controlling the product’s compliance.

Sometimes, reviewing the documentation submitted with the Customs declaration will be enough to clarify the matter, and determine whether an inspection should take place, but not always. This gives rise to uncertainties and divergent interpretations as to the nature of the goods being imported, and sometimes leads to discussions about which authority is competent.

Customs and market surveillance authorities also use different codes to identify companies. Businesses and people wishing to trade must use the Economic Operators Registration and Identification (EORI) number as their identification number in all Customs procedures when exchanging information with Customs administrations. Market surveillance authorities, as they focus mainly on the internal market, rely on their national identification system, which does not necessarily match the EORI numbers. To overcome this difference for the purpose of joint risk analysis, other data elements (such as addresses) must be reconciled. However this does not always work: sometimes the chosen data element is not available, e.g. because it is not mandatory to report it, and sometimes typing errors impede the matching of data.

Although practical obstacles exist, it goes without saying that for market surveillance authorities, close cooperation with Customs is worthwhile assuming that the legal preconditions for exchanging data are met.

Undertaking a joint risk analysis with Customs can provide insights into domains and risks that a market surveillance agency would not gain by itself, based solely on the analysis of its own data. For instance, Customs data can reveal that importers unknown to the surveillance agency are trading in risky products which fall under its supervision. Equally, thanks to joint risk analysis unnecessary controls can be avoided by showing that an importer does not trade in risky products. This results in an efficient use of resources and less hassle for traders.

Inspections can therefore be better informed and tailored. The inspector knows what to look for, and what not to look for! Compliance can then be established in the least intrusive manner possible. Traders, who are also tax-payers, can see that public resources are well-managed and that controls are well-prepared and justified.

Joint risk management also enables authorities to detect fraud cases and new offenders. Research in the field of food safety has proven that simply performing a larger number of random inspections is not effective in detecting fraud.[5] By pooling data and expertise, agencies can identify inconsistencies and anomalies that may be indicative of fraudulent actions[6] (Scherpenisse, Schram, & van Twist, 2017) which would otherwise remain unnoticed.

The duty of cooperation is formally enshrined in the EU regulations on the border controls performed to ensure that imported products meet regulatory requirements. For many years Customs has been given, and has played, a leading role in coordinating the actions of regulatory agencies at the border. Improving cooperation between the agencies responsible for matters of safety and security, human and animal health, and the economy and the environment, is one of the key goals of the Netherlands’ strategic Customs policy plan.[7]

In practice, policies supporting the concept of Coordinated Border Management have given rise to initiatives such as the Dutch Single Window for Trade and Transport. Nevertheless, besides enabling the use of a legal framework and technological tools, there is more to be done to improve joint risk management. In particular, a uniform system for the identification of goods and companies should be developed. This would require efforts from both sides, and would necessitate the involvement of the WCO and relevant EU institutions. This represents an ambitious undertaking, but the expected benefits are huge in terms of enhanced supervision capacities and trade facilitation. Now that the work related to the implementation of new rules and procedures with the United Kingdom has been completed, it may be time to start looking into this matter.

More information
e.f.enning@NVWA.nl

[1] For example, Regulation (EU) No. 2019/1793, Regulation (EU) No. 284/2011.

[2] Based on regulation (EU) No. 2017/625, Article 44(1), first sentence https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:32017R0625&from=EN#d1e4939-1-1.

[3]idem, (56).

[4] Chan, H., Fischer, G., Yang, F., & Zhang, H. (2015). Improve Customs Systems to monitor Global Wildlife Trade – Widely used trade codes lack taxonomic granularity. Science, Vol. 348 Issue 6232, 291-292: https://www.researchgate.net/publication/275037458_Improve_customs_systems_to_monitor_global_wildlife_trade.

[5] Gussow, K. (2020). Finding food fraud: Explaining the detection of food fraud in the Netherlands. Amsterdam: Vrije Universiteit Amsterdam.

[6] Scherpenisse, J., Schram, J., and van Twist, M. (2017). Tijd, toezicht en techniek: Temporele uitdagingen van digitalisering voor de NVWA. Den Haag: NSOB.

[7] Heijmann, Ensing, van’t Veld, and Neggers (2015). Coordinated border management in the Netherlands. WCO News 76.