1.Introduction

Digitalisation is one of the buzzwords of European governance. The hype around the term has quieted down somewhat precisely when it is starting to make a concrete and visible impact in the administration of the European integration project. One of the most advanced of such initiatives at the EU level is the e-customs initiative, which is intended to create a fully digital environment for customs and extra-EU trade.

Digitalisation in general describes the process in which formerly paper-based material is digitised and procedures are moved to digital communication technology. Digitalisation constitutes a transformation process, whose most concrete manifestation is the ‘paperlessness’ of interactions with the administration. Less concrete manifestations include the analysis of large quantities of data to inform public steering efforts, as well as algorithm-based or algorithm assisted decision-making. A digital environment does not necessarily include all of these aspects, but it establishes the opportunity to make use of such techniques. The move to digital communication technology in virtually all cases leads to the generation of the digital data necessary for data scienceWhile not a science in the academic sense, ‘data science’ is the term usually employed by practitioners as evidenced by a myriad of job descriptions in the digital industry. It is used to denote the craft of extracting useful information from data. and algorithm-based decision-making.

This paper aims at connecting the norms and values that structure the academic legal discussion of administrative processes in the EU to the concrete manifestations of one of the most relevant efforts for administrative reform today, namely the process of digitalisation.

Digitalisation carries the hope of solving some of the most difficult challenges of this century, ranging from a revolution of the work environmentFor a variety of approaches to this phenomenon, see for example Any Hines & Chris Carbone, ‘The Future of Knowledge Work’, Employment Relations Today 40(1) (2013), 1-17; Sabine Pfeiffer, ‘Digital Labour and the Use-value of Human Work. On the Importance of Labouring Capacity for understanding Digital Capitalism’, Communication, Capitalism & Critique 12(2) (2014), 599; Karsten Gareis, Stefan Lilischkis & Alenxander Mentrup, ‘Mapping the Mobile eWorkforce in Europe’ in J.H.E. Andriessen and M. Vartiainen (eds.), Mobile Virtual Work (Springer, 2006), 45-69. to reinventing democracy.An example for this is the ‘liquid democracy’ propagated by the ‘Pirate Party’ for example in Germany. Cf. https://wiki.piratenpartei.de/Liquid_Democracy (last accessed on 30 April 2018, in German). For more projects cf. also see https://liqd.net/en/ (last accessed on 30 April 2018). Environmentalists and economists place faith in the possibility of an economy that is not based on the use of natural resources,Carmen Nadia Ciocoiu, ‘Integrating Digital Economy and Green Economy: Opportunities for Sustainable Development’, Theoretical and Empirical Researches in Urban Management 6(1) (2011), 33-43. and some governance experts see the possibility of a public authority that is both more knowledgeable and less intrusive.Patrick Dunleavy et al, ‘New Public Management Is Dead – Long Live Digital-Era Governance’, Journal of Public Administration Research & Theory 16 (2006), 467-494; Hans Scholl & Margit Scholl, ‘Smart Governance: A Roadmap for Research and Practice’, iConference Papers (2014), 163-176. In customs administration, digitalisation is seen as the most pressing task for ‘modernisation’, as witnessed by its pride of place in the ‘Modernised Customs Code’,Regulation (EC) No 450/2008 of the European Parliament and of the Council of 23 April 2008 laying down the Community Customs Code (Modernised Customs Code), OJ L145/1 (2008). and after its recast in the Union Customs Code (UCC).Regulation (EU) No 952/2013 of the European Parliament and of the Council of 9 October 2013 laying down the Union Customs Code, OJ L269/1 (2013).

The EU Commission gives voice to some of these hopes in its statements regarding e-government and the digital economy.See https://ec.europa.eu/digital-single-market/en (last accessed on 30 April 2018) and Communication from the Commission, ‘A Digital Single Market Strategy for Europe’, COM(2015) 192 final. It is its aim to advance digitalisation of both the private and the public sectors within the EU in order to secure Europe’s welfare. There are already a number of regulatory initiatives which seek to realise the aspirations of a quick digitalisation. One of the forerunners in this regard is the EU customs regime. Customs as a policy field is characterised by the large percentage of professional actors who are willing and able to critically accompany the construction process. At the same time, the large number of individual but oftentimes repeated procedures promise efficiency gains through digitalisation. After all, automation is most effective in the case of repetitive processes.

However, it has become clear to anyone but the staunchest optimists that digitalisation of global markets is a double-edged sword: as we recognise the new opportunities information technologies offer for public authorities,Wolfgang Hoffmann-Riem, ‘Verwaltungsrecht in der Informationsgesellschaft – Einleitende Problemskizze’ in Wolfgang Hoffmann-Riem & Eberhard Schmidt-Aßman (eds.), Verwaltungsrecht in der Informationsgesellschaft (Nomos, 2000), 28 et seq. we also need to realise that the global structure of the digital economy arguably has reduced public steering capacity.Klaus Lenk, ‘Ausserrechtliche Grundlagen für das Verwaltungsrecht in der Informationsgesellschaft: Zur Bedeutung von Information und Kommunikation in der Verwaltung’ in Wolfgang Hoffmann-Riem & Eberhard Schmidt-Aßmann (eds.), Verwaltungsrecht in der Informationsgesellschaft (Nomos, 2000), 59. To both realise the new potential and adapt to a changed world, administrations are required to modernise around the use of new information technologies.

Technology is naturally a tool, and its potential depends on the ends it is used for and the care with which it is employed. In this light, the normative conditions under which the process of digitalisation takes place take centre stage when preventing this bright new opportunity from turning into a brave new world. Digitalisation famously also supplies opportunities for restricting care, increasing bureaucratic burdens and implementing surveillance mechanisms.

Even though the EU and particularly the Commission can be described as at the forefront of digitalisation, pushing the Member States to change their administrative technologies and providing digital access and digital information, the present normative framework within which this change should be conducted remains worryingly unclear. As a result, the normative sign posts that are supposed to guide implementation remain hidden.

2.Research Approach

This paper applies a normative legal perspective to the ‘technical’ process of customs digitalisation. The conceptual starting point of the inquiry is the differentiation between ends and means, between goals and the tools and procedures used to reach them. The central assumption is that legally defined ends serve an important normative function in structuring and shaping administrative reform and technological development, such as digitalisation. They serve the additional function of making this digitalisation process intelligible to those outside the process itself.

This paper takes the Union Customs regime as a case study for the wider process of digitalisation of governance processes. Customs Administration is an area of EU policy which is already highly digitalised. A digital customs administration requires digital capacities in the private as well as in the public sectors, as both public institutions and private parties are required to use the digital system. As a result, it is a case in which the full variety of administrative norms comes to bear: those that structure the relationship within the public sector, as well as those that are aimed at the interaction between the ‘state’ and the individual.

As a side note, it should be mentioned that the terminology of customs administration is already suitable to the perspective of the information society. The tool of ‘binding tariff information’,Austin Valentine, ‘European binding tariff information, a brief explanation’, ERA Forum 9(3) (2008), 413. for example, exemplifies the dual nature of administrative decisions as information and binding legal decisions. This terminology recognises the normative potential of information. Where information is generated automatically (through information technology), this may lead to the exercise of authority through IT.

The main research question of this paper is: Which norms or values structure the digitalisation of the customs regime and how do they relate to the normative framework of democratic administrations, which is spanned by equality, transparency and participation? The question here does not concern hypothetical or philosophical norms and values, but those that are encapsulated in the legal basis and public policy documents describing the digitalisation of the customs regime. This is the normative framework that the Commission and Member State officials have to be assumed to be working with and which can be used to exercise accountability over the process of digitalisation by whichever accountability forum is employed to oversee this process.On the structure of accountability as oversight over an actor by a forum according to set standards, see Madalina Busuioc, The Accountability and European Agencies (Eburon, 2010), Ch. 3.

Normative guidance comes in the form of legal principles, goals and objectives. In the present context, they are defined as ‘normative’ when they guide action towards a desired end, and go beyond the requirement of prescribed means. Such a reference to ends puts technological innovation in a value-oriented context. By, for example, making clear that the use of IT is intended to enhance the participation of interested parties in administrative processes, digitalisation is given a normative context according to which it can be shaped and against which it can be evaluated. Without such ends, digitalisation has no direction and technological change becomes an end in itself. Without a picture of which ends e-customs serves, there is little guidance on how to choose between different technological options or to develop implementation priorities.

This inquiry into ends is framed as an inquiry into the norms or values. ‘Norms or values’ thus stand for the guiding effects of ends, regardless of whether these are expressed as principles, goals or objectives. Their guiding effect is developed on the basis of their aspirational character.

This research studies the E-Customs Decision,Decision No 70/2008/EC of the European Parliament and the Council of 15 January 2008 on a paperless environment for customs and trade, OJ L23/21 (2008). the Union Customs Code,Regulation No 952/2013 (n. 7). and the subsequent delegatedCommission Delegated Regulation (EU) 2015/2446 of 28 July 2015 supplementing Regulation (EU) No 952/2013 of the European Parliament and of the Council as regards detailed rules concerning certain provisions of the Union Customs Code, OJ L343/1 (2015). and implementing actsCommission Implementing Regulation (EU) 2015/2447 of 24 November 2015 laying down detailed rules for implementing certain provisions of Regulation (EU) No 952/2013 of the European Parliament and of the Council laying down the Union Customs Code, OJ L343/558 (2015). in order to find a description, or at least the implications, of the ends and aspirations that govern the construction and implementation of a digital customs administration. I first examine the legal principles explicitly referred to or sufficiently evident in the established texts. This part lays a focus on the proclaimed goals of the digitalisation initiative. Hereby, for a stated goal or principle to exert guiding force on the digitalisation process, its aspirations must lie beyond the process itself.

As a second step, this paper turns to three widely recognised principles of democratic governance and inquires into their mark on the e-customs digitalisation process. These are the principles of equality, transparency and participation. They relate to the democratic constitution of the EU and can be found in arts 9-12 TEU. In addition, the paper relates these findings to the necessity of data protection and takes a look at how the e-customs regime provides for the protection of the data of individuals and undertakings. The research in these parts becomes more critical, inquiring into whether these principles influence customs digitalisation.

Throughout this paper, focus is placed on the process at the EU level. Significant parts of the development of digital customs technology are taking place at the national level. However, these are steered and coordinated at the EU level. In customs, Member States’ authorities act as EU officials in that they directly implement EU law. Consequently, it is incumbent on the EU legal acts to define all relevant norms and values by which national implementation efforts must abide.

Having laid out my research approach here, I will now turn to describing the customs regime. The process of EU customs digitalisation will be briefly retold (section 3) to introduce the case studied here. I will then draw out the values that can be found in the foundational acts of the e-customs initiative, most prominently the e-customs decision and the Union Customs Code (section 4 and 5), before turning to the question of whether the democratic values of equality, transparency and participation can be found in customs digitalisation (section 6). The implementation of data protection, data security and confidentiality in the e-customs regime will then be discussed in brief (section 7). I will conclude in section 8, arguing for the relevance of including equality, transparency, participation and trustworthy data management in e-customs as a fundamental part of the process.

3.The Digitalisation of the EU Customs Regime

3.1.Historic development

Instead of asking for paper application forms and binders full of documents as evidence, the EU seeks to shift procedures to digital means of communication. The Commission gives as the aspired benefits of such an undertaking the creation of a more efficient and more modern customs environment, which enhances security at the EU’s borders and facilitates trade.European Commission, ‘Electronic Customs’, http://ec.europa.eu/taxation_customs/general-information-customs/electronic-customs_en (last accessed on 30 April 2018).

The EU started the process of digitalisation of the customs regime already more than two decades ago. The construction of a Union-wide database with information pertaining to traded goods, modalities of transport, trading firms and such was first agreed in 1997.Cf Reg 515/97. See also Armin von Bogdandy, ‘Information und Kommunikation in der Europäischen Union: föderale Strukturen in supranationalem Umfeld’ in Wolfgang Hoffmann-Riem & Eberhard Schmidt-Aßmann (eds.), Verwaltungsrecht in der Informationsgesellschaft (Nomos, 2000), 186. An important milestone in the digitalisation process had been reached in 2003 with the adoption of a Council resolution on constructing an entirely paperless customs environment.Council Resolution of 5 December 2003 on creating a simple and paperless environment for customs and trade, OJ C305/1 (2003). For a recount of that history see Kathrin Limbach, Uniformity of Customs Administration in the European Union (Hart Publishing, 2015), Ch. 4. This resolution was further operationalised in 2008 with the E-customs decision.See above (n. 13). This decision remains in force despite the introduction of a new Customs Code in 2013: Regulation (EU) No 952/2013 (n. 7). In the beginning, much of the effort was undertaken by the Member States,See Commission, ‘2008 E-customs Progress Report’ (2008), Taxud c.1 (2009) 172599. and many Member States established their own national e-customs system. As per the decisions, a common portal which referred to all electronic system needed to be operational by 15 February 2011.Art. 4(2) E-customs Decisions (n. 13).

Apparently, by the mid 2000s it was felt that in this reform process the customs regime needed a more wholistic overhaul. Digitalisation was one of the main concerns leading to the modernisation of the customs code in 2008.Kathrin Limbach, Uniformity of Customs Administration in the European Union (Hart Publishing, 2015), 83 et seq. The Modernised Customs Code is based on the establishment of a Europe-wide e-customs system with both Member State and EU components.

Throughout this development, the establishment of an e-customs environment was planned as imminent within a few years. However, by 1 January 2016, this system still could not enter into operation as planned. A new time frame envisages implementation of a digital customs environment by 2025.On 2 March 2018, the European Commission proposed an amendment to the UCC so that customs authorities and economic operators be allowed to continue using transitional arrangements (i.e. existing IT systems or paper-based arrangements) for some of the customs procedures until 2025. See Commission, ‘Proposal for a Regulation of the European Parliament and of the Council amending Regulation (EU) No 952/2013 to prolong the transitional use of means other than the electronic data-processing techniques provided for in the Union Customs Code’ (2018), COM(2018) 85 final.

Before the new customs code was adopted, the EU’s e-government services in this area extended mainly to the creation of the abovementioned web portal, which supplied links to all the national information websites and electronic systems.http://ec.europa.eu/ecip/information_resources/links/index_en.htm#eu (last accessed on 30 April 2018). This portal stopped being maintained in 2014 and presently serves as an archive.

While the complete e-government system is not yet operational and will not be operational for some time, the Commission and the Member States have started with the implementation of certain aspects of this system. This extends to some databases already being operable, as well as to the adaptation of administrative processes to the requirements of a Union wide digital system.

The new unified e-customs system has to be able to be integrated into diverse national IT systems as well as the systems of the economic operators which are sometimes allowed to automatize the transfer of relevant data.See Commission Delegated Regulation (EU) 2015/2446 of 28 July 2015, supplementing Regulation (EU) No 952/2013 of the European Parliament and of the Council as regards detailed rules concerning certain provisions of the Union Customs Code, Ch. 3 of Annex B-01. It is this integration into local systems of economic operators which unlocks the greatest potential efficiency gain. See Julian Krumeich & Dirk Werth, ‘Unterstützung von kleinen und mittleren Unternehmen bei der Durchführung von Exportprozessen’, Mittelstand Digital 2 (2014), 93-100. This is one of the main technical challenges to a unified approach.Steffen Bernius & Donovan Pfaff, ‘Mythen der eRechnung – Wie wissenschaftliche Erkenntnisse den Weg zur Umsetzung des elektronischen Rechnungsverkehrs zeigen’, Mittelstand Digital 2 (2014), 70-80.

3.2.Technical Aspects

The e-customs initiative is an attempt to introduce a digital environment to customs operations. To be functional, this environment needs to be able to accommodate different users and different uses, such as, for example, the import of large quantities of unspun cotton for industrial use, a single bear hide by a hobby hunter, and the decision on the classification of new chemical compounds; furthermore, this extends to gathering information on the activities of specific exporting firms throughout the EU. It might be helpful to take the environment analogy further and to imagine the customs operations as a natural habitat where the individuals of several species (such as customs authorities, large trading forms, SMEs and so forth) interact. What the e-customs initiative is actually building is the natural environment, the landscape on which interactions can take place.

For the initiative to be successful, this landscape needs to be populated. This has been the role of third parties which develop e-customs programmes for end-users.For example, see https://www.ibm.com/industries/government/borders-immigration-customs (last accessed on 30 May 2018); https://help.sap.com/doc/saphelp_gts101/10.1/en-US/47/28020f97574678e10000000a155369/content.htm?no_cache=true (last accessed on 30 May 2018) or https://cloud.oracle.com/opc/saas/datasheets/oracle-customs-mgmt-ds.pdf (last accessed on 30 May 2018). While there are companies that provide IT solutions that enable businesses to enter the digital landscape and thus conduct their customs, this landscape also needs to remain open for the individual traveller who is obliged to declare the import of a single crate of vodka. Such a traveller cannot be expected to buy a professional customs-software suite at the border. At the same time, this landscape must be protected from erosion through the exploitation of structural weaknesses and from abuse, for example, in the form of data theft. Taking into account that traditional paper-based administrative processes enabled a great variety of actors to use the system, the baseline standards set for customs administration by the expectation that a digital environment improves on this system are actually already considerable.

Nevertheless, paper-based processes incorporate phenomena which have come to be seen as significant disadvantages. These are related to the effort required for communication and data gathering. Paper-based systems require the exchange of paper files for communication between custom officials and between customs and traders. This exchange costs time and effort, including the effort to integrate the information thus transmitted into internal procedures. These difficulties in transmitting information further hinder the creation of uniformity. Digitalisation promises interconnectivity and the easy transfer of information, as well as better access to (usable) data. At the same time, a digital system should maintain the flexibility and stability of paper-based customs administration.

The introduction of information technology connects to communication needs of administrations and individuals. As a result, the e-customs initiative structures the required software in terms of information exchanges. This results in a number of interconnected but separable programmes. A first category of communication relates to trading procedures. These are  programmes digitising communication pertaining to exports (Automated Export System, AES) and a programme pertaining to imports (Automated Import System, AIS), as well as a programme relating to Transit (New Computerised Transit System, NCTS). A second form of communication pertains to including the individuals and trade firms into the digital environment. The relevant programmes include the Operator’s registration on the digital landscape (Economic Operator’s Registration and Identifications System, EORI) and the facilitation of repeat interactions with trusted individuals or firms (Authorised Economic Operator System, AEO). Finally, there are the communication needs of the customs authorities amongst each other, which are realised through a risk management framework.Commission, ‘The measures: Customs Risk Management Framework (CRMF)’,

https://ec.europa.eu/taxation_customs/general-information-customs/customs-risk-management/measures-customs-risk-management-framework-crmf_en (last accessed on 30 April 2018). These components have been settled on already since 2008,Commission (n. 21). although not all are operational at the time of writing. See Kathrin Limbach, ‘E-Government in der Zollverwaltung der Europäischen Union’, Das Deutsche Verwaltungsblatt 9 (2015), 551.

These projects are complemented by further initiatives, not all of which have fully been agreed upon yet. These proposed initiatives relate to the expansion of the digitalised customs environment to also include less frequent uses, for example, by non-professionals (through the Single Authorisations for Simplified Procedures) or exporters established in third countries. Moreover, these proposals aim at a  further simplification of the procedure for individuals and firms by establishing a Single Electronic Access Point, building an Integrated Tariff Environment, creating an EU Customs Information Portal and establishing a Single Window Interface.For the entire list see http://ec.europa.eu/taxation_customs/general-information-customs/electronic-customs/individual-projects_en#node-135auth_eco (last accessed on 30 April 2018). Here, it needs to be kept in mind that such simplification for customs ‘customers’ requires more sophistication in the digital environment to connect customers’ requests and information to the relevant administrative process. The progress of the digitalisation of customs administration is reported yearly in an Annex to the Multiannual Strategic Plan (MASP), the central steering document of the multilevel and multiagent digitalisation effort. The Annexes, together with the various MASPs, are listed on https://ec.europa.eu/taxation_customs/general-information-customs/electronic-customs_en (last accessed on 30 April 2018).

As might have been suspected from its slow realisation, the implementation of an e-customs system is not as simple as it might at first seem. Not only is it necessary for the custom forms to be compatible with each other so that the customs official in Poland can use the information that the customs authorities in the Netherlands entered, but the electronic system has to be secure, stable and sufficiently easy to use. Moreover, Member States and trading firms have to equally access and use this system. Their respective requirements regarding information to be provided as well as individual rights and data protection to be safeguarded present the pre-conditions needed for the broad usability of the e-government system.

The establishment of a digital customs system constitutes a large infrastructure project. Not only because it necessitates the technological infrastructure on which customs formalities can take place to be set up, but also in a financial sense. Large infrastructure projects tend to cost more than initially expected and to take longer to be realised. This is very much true for the EU e-customs system, which has experienced several deadline extensions and undergone re-scheduling because the original time frame had proven unattainable.Carsten Weerth, ‘Arbeitsprogramm zum UZK 2016 – Beschluss (EU) 2016/578’, Außenwirtschaftliche Praxis 6 (2016), 187-192. In fact, the current time frame might again have to be adjusted and prolonged.Thus, as per Annex II to the MASP 2017, the current estimated completion date of much of the Automated Export System already lies in 2010, with deployment scheduled for 2020-2021. See Commission, ‘Consolidated Project Fiches’ (2017), Taxud a.3 (2017) 6498377, 36, https://ec.europa.eu/taxation_customs/sites/taxation/files/resources/documents/customs/policy_issues/e-customs_initiative/masp_annex2_en.pdf (last accessed on 30 April 2018). The deployment window of the ECC Transit System is even scheduled to extend to 2025. So far, deadlines are extended virtually every MASP cycle.

The technical aspects of the e-customs Union further include the differentiation between information that is shared automatically and information shared only after human intervention, and determination of the format to be used for information exchanges. They include how operators and goods are identified and for which actions what kind of authentication is required. It would go too far to describe all these aspects for all, or even just the main, components of the e-customs system. However, it should be kept in mind that the choices made here are not irrelevant. The architecture of the digital system shapes the interactions that take place in it.Lawrence Lessig, Code and other Laws of Cyberspace (Basic Books, 1997). They enable or restrict data exchange, user access, and communication between the different actors on the system. These architectural choices are influenced by the purpose the IT architects are designing a system for and the framework conditions they have to comply with.

3.3.Digital Customs vs Paperless Customs

It is noticeable that even though the EU Commission embraces digitalisation, the legal acts concerning the digitalisation of the customs regime refer merely to a ‘paperless customs environment’Such as the E-Customs Decision (n. 13). See also: Commission, ‘Communication from the Commission to the Council, the European Parliament and the European Economic and Social Committee – A simple and paperless environment for Customs and Trade’ (2003), COM(2003) 452 final. or ‘electronic customs’.Commission, ‘Electronic Customs’, https://ec.europa.eu/taxation_customs/general-information-customs/electronic-customs_en (last accessed on 30 April 2018). This choice of words implies that only the physical matter on which customs administration operates will be exchanged (from paper to electronic), suggesting that the substantive governance aspects of customs administration remain unchanged.This tendency to use e-government to digitize existing processes, instead of modernizing administration is also criticised by Clara Centeno, Rene van Bavel & Jean-Claude Burgelman, ‘A Prospective View of e-Government in the European Union’, Electronic Journal of e-Government 3(2) (2005), 59-66.

The objective of discontinuing the use of paper largely refers to the automatingChristian Sørbye Friis, ‘Knowledge in Public Administration’ (Keynote) in Maria A. Wimmer (ed.), Knowledge Management in e-Government (Universitätverlag Rudolf Trauner, 2002), 3-14. or reproductiveThomas Zwahr et al, ‘More than Digitisation – The Transformative Potential of E-Governance: An Exploratory Case Study’, System Sciences (2005), 131. capacity of digitalisation, by transferring processes that already do exist, or are at least possible to conduct in paper, to electronic data processing. Such processes promise efficiency gains without requiring much prima facie institutional reorganisation. This is not to say that the new infrastructure could not lead in effect to deeper transformation. But such a transformation is in any case not included in the ambit of the e-customs initiative.

Even though not every effort in digitalisation necessarily requires making use of all the opportunities it offers,Indeed, it might at times be a conscious choice to forego some of these opportunities for ethical or political reasons. In particular the surveillance that is technically possibly will be subject to such restrictions. literature is clear in that the switch from paper to a digital environment implies more than exchanging filing cabinets for hard drives (or whichever storage medium is employed). At the very minimum, the functions that paper automatically guaranteed – to make it detectable when a file was manipulated at the same as allowing for repeatedly opening the same file, for example – must be safeguarded also in the case of electronic information.Wolfgang Hoffmann-Riem, ‘Verwaltungsrecht in der Informationsgesellschaft – Einleitende Problemskizze’, in Wolfgang Hoffmann-Riem & Eberhard Schmidt-Asßman (eds.), Verwaltungsrecht in der Informationsgesellschaft (Nomos, 2000), 18; Alexander Roßnagel, ‘Möglichkeiten für Transparenz und Öffentlichkeit im Verwaltungshandeln – unter besonderer Berücksichtigung des Internets als Instrument der Staatskommunikation’ in Wolfgang Hoffmann-Riem & Eberhard Schmidt-Asßman (eds.), Verwaltungsrecht in der Informationsgesellschaft (Nomos, 2000), 273, 288 et seq. To effectively meet the demands of governing in the current economy, administrations need to make use of at least some of the opportunities that digitalisation offers. However, this requires a re-organisation of administrative processes,Klaus Lenk, ‘Ausserrechtliche Grundlagen für das Verwaltungsrecht in der Informationsgesellschaft: Zur Bedeutung von Information und Kommunikation in der Verwaltung’ in Wolfgang Hoffmann-Riem & Eberhard Schmidt-Aßmann, Verwaltungsrecht in der Informationsgesellschaft (Nomos, 2000), 60. including a re-organisation of its interactions with individuals.

Digitalisation is a more encompassing process. Only one of the components of this process is that digital technology, instead of paper, is used as a means to carry and store information (even if this is the component that is most obvious). Further aspects of digitalisation concern compatibility standards, as well as shaping the communication between the administration and its subjects as will also be elaborated below. In many cases, the information stored and communicated should take a form that machines can process. This requires the collection of information in a way, which has been made independent of the contextual knowledge of human officers.Ibid., 72.

Already now, there are common (paper or electronic) forms to be used for customs declarations.See, for example, the Annexes to Commission Delegated Regulation (EU) 2016/341 supplementing Regulation (EU) No 952/2013 of the European Parliament and the Council, regarding to transitional rules for certain provisions of the Union Customs Code where the relevant electronic systems are not yet operational and amending Delegated Regulation (EU) 2015/2446, OJ L69/1 (2016). These common forms make the move to a common electronic system easier, which could simply present the forms which are now on paper in digital format. Moreover, the Commission and Member States have worked out more specific guidance documents on how the forms are to be interpreted and used.SAD Guidance During the UCC transitional period, Taxud a.3 (2015) 5707081, DIH 15/008 – FINAL. These documents, for example,Further guidance documents can be found at https://ec.europa.eu/taxation_customs/business/union-customs-code/ucc-guidance-documents_en (last accessed on 30 April 2018). describe specific data points (so called ‘boxes’) and how they are to be filled in. They also describe possibilities of dealing with common problems or ambiguities and give examples of how the boxes are to be filled in in more complex cases. The 81 pages of this document are evidence of the ongoing need to communication and coordination even in standardised environments.

This standardisation is in itself a process which navigates the tension between, on the one hand, the flexibility necessary to account for the variety of individual situations with which the customs administration deals and, on the other hand, adherence to the strict forms necessary to make the information gathered ‘processable’ by digital systems.

The EU tackles this tension by making incremental technological steps which are each tested in real life situations and revised if necessary. This provides the time for administrative culture to adapt to the use of information technology. However, this incrementalism can have the effect of obscuring the value choices inherent in building a new administrative infrastructure.

4.Norms and Values Employed in the Customs Regime

The Union Customs Code states that it follows the ‘legal principle’ that “all customs and trade transactions are to be handled electronically”.Recital 17 UCC. Even though this is its only express reference to a legal principle, it is unclear why this rule should be elevated to the status of principle, given the fact that it contains so little fundamental value judgment, and indeed rather little normative guidance beyond its obvious literal application.On the relation of rules and principles, see Cass R. Sunstein, ‘Rules and Rulelessness’ in John M. Olin, Law & Economics Working Paper No. 27 (1994), 3 et seq; Ronald M. Dworkin, ‘The Model or Rules’, University of Chicago Law Review 35 (1967), 22 et seq. The aspirational aspect is entirely missing from this ‘principle’, which instead elevates the means of governance (electronic transactions) to an end in itself.

This ‘principle’ thus exerts no guiding force. It does not help in setting priorities or shaping the way that digitalisation is implemented. In this case, digitalisation would be a process without a goal. However, even though there are no further ‘principles’ referred to explicitly, further normative guidance can nevertheless be gleaned from the legal documents. Indeed, it appears that the principles governing the e- customs administration are not referred to by that name, making them more implicit to the process. Nevertheless, an inquiry into the objective and goals of digitalisation reveals several guiding norms.

Before entering into the discussion of which these ends are, it will be useful to reflect on their normative force. As mentioned, these normative ‘signposts’ are not highlighted as guiding principles in the legal documents (or anywhere else). While it is possible to deduct them from the goals and objectives put forward, or from the structure employed, these normative signposts remain hidden in the sense that they need to be excavated before it is clear which way they point.

Of course, such excavation exercises are open to contestation. This fact reduces the guiding force of normative signposts. Yet, even if another commentator might choose to group these goals slightly differently or phrase the principle they represent in somewhat different terms, their main thrust appears clear, namely as principles aimed at the improved functioning of the internal market. It will become clear that the principles embodied in these goals differ markedly from the principles of equality, transparency and participation discussed in the next section. The discussion below will show that the described goals are difficult to understand as aspirational signposts since their normative side remains underdeveloped. As a result, the direction they point to remains unclear.

4.1.Efficiency

The most ubiquitous principle guiding the digitalisation process is that of efficiency, coupled closely with effectiveness. The search for efficiency leads like Ariadne’s thread through the labyrinth of e-customs regulations and official communications.

Reference to this norm or value can be found frequently in the relevant documents. The Commission describes it on its website as the aspired benefits of digitalisation and the creation of a more efficient and more modern customs environment, which enhances security at the EU’s borders and facilitates trade.http://ec.europa.eu/taxation_customs/general-information-customs/electronic-customs_en. This is a somewhat shortened version of the benefits cited in the e-customs decision.

In this decision, which is the document most squarely aimed at customs digitalisation, the preamble as well as the operative part make reference to efficiency. The preamble gives expression to the aspiration that digitalisation will help to achieve the goals of customs control, such as the effective leveraging of tariffs and a better control of the goods entering the internal market.These are listed as (numbered for ease of reference): 1) the efficiency of the organisation of customs controls, 2) the seamless flow of data in order to make customs clearance more efficient, 3) reducing administrative burdens, 4) a contribution to combating fraud, organised crime and terrorism, 5) fiscal benefits, 6) the protection of intellectual property and cultural heritage, 7) increasing the safety of goods and the security of international trade and 8) enhancing health and environmental protection. See Decision No 70/2008/EC (n. 13), Preamble (2). In the operative part,Ibid., Art. 2. the objectives are defined more concretely.They are a) to facilitate import and export procedures, b) to reduce compliance and administrative costs and to improve clearance times, c) to coordinate a common approach to the control of goods d) to help ensure the proper collection of all customs duties and other charges, e) to ensure the rapid provision and receipt of relevant information with regard to the international supply chain and f) to enable the seamless flow of data between the administrations of exporting and importing countries, as well as between customs authorities and economic operators, allowing data entered in the system to be re-used. Here, too, the provision speaks of the facilitation of import and expert procedures, and adds the improvement of clearance times and reducing costs on the side of operators, as well as the administration. These cost reductions are expected to materialise through harmonisation, faster dissemination of information and better data flows, including between administrations. On the benefit side, the institutions hope for greater compliance and thus a greater amount of tariffs payed. Similar reasoning is found in the UCC.UCC (n. 7), Recital 17.

It is noticeable that the goals of the e-customs decision refer mainly to the goals of customs administration more generally, with digitalisation aiming at an increase in efficiency and effectiveness, as is also expressed in the objectives. Digitalisation is thus a means, not an end, and the objectives of digitalisation are to support the goals of customs control, whichever these may be. The only objectives mentioned which are genuinely connected to digitalisation are those of creating a seamless flow of data and the reduction of the administrative burden.

‘Technical’ goals, such as efficiency, have the potential advantage that their success is objectively measurable. Cost savings and increases in output or revenue can often be measured numerically. However, even as digitalisation seeks to generate the data which enables this measuring exercise, the metric to be used remains undefined.

The perceived ease of measuring goal attainment in case of the goal of efficiency remains elusive. It remains undefined which costs are relevant factors, and against which outputs. Similarly, there are several potential units of measurement, such as money or time, where no choice has been defined. For example, which metric determines the quality of import or export procedures (which are to be facilitated as to art. 2 79/2008/EC) remains unclear. In addition, to use this advantage, someone would have to be actually measuring the customs processes.

The pride of place for the norm of value of efficiency should not surprise. Efficiency and effectiveness are central also more broadly to the communications from the Commission, which seek to reduce (unnecessary) bureaucracy and so-called ‘red tape’.See the High Level Group on Administrative Burdens, 'Cutting Red Tape in Europe: Legacy and Outlook' (Brussels: 2014), http://ec.europa.eu/smart-regulation/refit/admin_burden/docs/08-10web_ce-brocuttingredtape_en.pdf (last accessed on 2 May 2018), but also the ‘Better Regulation Initiative’, the documents pertaining to which are retrievable at https://ec.europa.eu/info/law/law-making-process/planning-and-proposing-law/better-regulation-why-and-how/better-regulation-guidelines-and-toolbox_en (last accessed on 2 May 2018).

Efficiency as a normative goal is most explicit in the New Public Management (NPM) approach and its managerial style reasoning.Rhys Andrews, ‘NPM and the Search for Efficiency’ in Tom Christensen & Per Laegreid (eds.), The Ashgate Research Companion to New Public Management (Ashgate, 2011), 281-294. The concept of efficiency employed here is that of ‘productive efficiency’,Ibid. 284. i.e. a maximisation of the output/input differential, which exhibits a focus on resources (instead of, for example, maximising needs satisfaction or optimising allocation). This is evidenced by the fact that the minimisation of input (i.e. ‘administrative burden’, compliance and administrative costs) and the maximisation of output (fiscal gains) are mentioned explicitly as goals of digitalisation.

However, the shortcomings of NPM are widely known.Janet Newmann, ‘Serving the Public? Users Consumers and the Limits of NPM’ in Tom Christensen & Per Laegreid (eds.), The Ashgate Research Companion to New Public Management (Ashgate, 2011), 349-359. Newer approaches emerge from the need to include a greater variety of values and democratic norms in the conceptualisation of quality in administration.John M. Bryson et al, ‘Public Value Governance: Moving beyond Traditional Public Administration and the New Public Management’, Public Administration Review 74(4) (2014), 445-456. Indeed, it appears that efficiency is in itself normatively empty as it simply seeks to enhance the effect of those processes which are to be handled more efficiently. Efficiency requires the reference point of which goal is supposed to be delivered for it to be able to unfold guiding force. There is no such thing as stand-alone efficiency; the term always requires a conceptualisation of what the desired outcome is and which inputs are regarded as relevant. In this, the adherence to efficiency of a regime is a reflection of the relevance of the goals embodied by this regime. As such, efficiency is ill fitted to serve as a value or norm in its own right.

In customs digitalisation, the desired outcomes are mainly the traditional goals of customs administration, namely the control of cross-border trade and the generation of public revenue from these trade flows. This a rather technical conceptualisation of the role of customs, focusing on control and revenues. These functions are as universal as they are limited in their normative reach.

As a result, it is unclear how the quest for efficiency will guide customs digitalisation specifically. While such structures as the Automated Import and Export Systems can be seen as an implementation of the aspiration for efficiency, they are also simply an expression of digitalisation itself. Automation is usually simply an aspect of digitalisation which may or may not, depending its form and context, lead to efficiency gains. In as far as digitalisation is often in and of itself thought to lead to an increase in efficiency, it is unclear what this norm or value would add to the development of IT infrastructure. In as far as this link should not be assumed automatically to exist, ‘efficiency’ in and of itself cannot exert guiding force as long as its normative context remains undefined.

4.2.Service Orientation

Accompanying the pursuit of efficiency is the goal of service orientation and reducing not only the resources needed by the administration, but also those to be expended by its subjects. While relevant communications refer to this goal more explicitly,Communication from the Commission to the Council, the European Parliament and the European Economic and Social Committee of 24 April 2003 – A simple and paperless environment for Customs and Trade, COM(2003) 452 final, OJ C96 (2004), s. 2.2. the legal documents founding the digitalisation process refer in this regard mainly to the reduction of administrative burdens and the decrease of the time that procedures take.E-Customs Decision (above n. 13), Recital 2 and Art. 2(1)(a) and (b); UCC (above n. 7), Recital 11.

The 2003 communication on customs digitalisation describes the idea of reducing administrative burdens, stating that “customs administration should facilitate trade transactions”Communication from the Commission to the Council, the European Parliament and the European Economic and Social Committee of 24 April 2003 - A simple and paperless environment for Customs and Trade, COM(2003) 452 final, OJ C96 (2004). Also see COM(2008) 169 final, which maintains that it is a function of customs to ‘support legitimate trade’. (Ibid. 4). and describing the ways in which the digitalisation of customs administration will aid businesses in saving time and administrative costs. The view that customs administration should facilitate trade is remarkable, given that customs are usually regarded as trade restrictions, including in art. 28 TFEU. Here, the UCC presents a more balanced picture than the Commission communications and refines the above-mentioned service orientation of customs digitalisation somewhat by making clear that digitalisation of the customs regime is intended to facilitate business, while at the same time providing for control of goods taken into and out of the EU.UCC (n. 7), Recital 20. This latter formulation acknowledges the tension between the effective control of trade and trade facilitation.

Statements like the above show the changed function of the customs administration, which today is less focused on the generation of revenue, and more on providing securityIntegrated border management, COM(2003) 452 final (02). and combatting fraud.See Kathrin Limbach, ‘E-Government in der Zollverwaltung der Europäischen Union’, Das Deutsche Verwaltungsblatt 9 (2015), 549. Nevertheless, this gives expression to a certain normative context, namely that of digitalisation (if not customs themselves) constituting a service to the market.‘Administrations, public bodies, businesses and users know themselves best what they need. The choice of systems and technologies, of distributed or centralised designs should be entirely according to their choice and needs but need to fully respect agreed interoperability requirements’, COM(2016) 179 final, 4. By, for example, prohibiting the import of products which violate trademarks, this is meant in turn to stimulate the trade with those products that carry a trademark legally.

Service orientation flows from the conviction that the EU as a regional market is in competition with other global regions for trade and productivity.COM(2003) 452 final, COM(2010) 245 final, 3. The need for service orientation is justified with reference to the necessity for the EU itself to be competitive globally.Competition can be described as a necessary corollary of service orientation. See Jane E. Fountain, ‘Paradoxes of Public Sector Customer Service’, Governance 14(1) (2001), 59 et seq. Digitalisation here appears to be viewed as a tool to raise ‘customer’ satisfaction by decreasing the cost of compliance without actually lowering tariffs, thus decreasing fiscal revenues. In economic terms, digitalisation is valued as a mechanism for reducing transaction costs. Yet, here, as in the case of efficiency, it is relevant to question how customer satisfaction is assessed.

Service Orientation has been introduced to public administrations with the rise of NPM approaches, which modelled public administrations on private businesses,For an example arguing the relevance of service orientation in managerial terms, see John Stewart and Michael Clarke, ‘The Public Service Orientation: Issues and Dilemmas’, Public Administration 9 (1987), 161-177. For a survey on the role of service orientation in public administration, see Richard M. Walker et al, ‘Market Orientation and Public Service Performance: New Public Management Gone Mad?’, Public Administration Review 71(5) (2011), 707-717. even though it is more the term than the concept that belongs to managerialism.Especially in the form of ‘customer’ service orientation: Richard M. Walker et al (n. 70), 707-717. What management approaches introduced was the focus on the ‘client’ or ‘customer’, i.e. an individual with which an administration interacts directly. This focal point is different from the focus on ‘public service’, which refers to the public at large.See Jane E. Fountain, ‘Paradoxes of Public Sector Customer Service’, Governance 14(1) (2001), 62. Even though there is some overlap in terms of the persons served and the public interests taken into account, ‘customer’ service is liable at times to differ substantially from public service.

This dichotomy between conceptualising the administration as oriented towards serving the ‘customers’ or towards serving the general public is also one that is relevant in the EU customs regime. To put it simply, customs is generally not a service to the trader but to the general public. Digitalisation might be serving trade in as much as it makes customs less cumbersome, but it will not detract from the necessity of customs clearance and the payment of tariffs. These latter aspects are grounded on public interests – such as the necessity of a well-funded ‘government’ and the interest in controlling dangerous or otherwise problematic goods. The e-customs initiative is indeed striving for both efficiency and lower administrative barriers to trade (the term ‘facilitation’ appears inappropriate here), even though only the latter is conceptualised as a service. As a result, efficiency and service orientation may provide for mutual limits to their pursuit.

Some of the initiatives being implemented show this aspiration towards service orientation in the digitalisation of customs. Thus, the Authorised Economic Operator System, which will allow for faster and somewhat simpler procedures for a group of professional and trustworthy operators, can be understood in the light of service orientation. Similarly, initiatives like a single electronic access point for all customs formalities seek to make it easier to conclude customs procedures.

In addition, the norm or value of service orientation should play a role in the design of national electronic systems in that it brings such issues as accessibility and ease of use to the fore. In this, it could be seen to link to equality, which will later be discussed. However, service orientation does not necessarily work to exclude bias or promote inclusivity and thus differs from equality, as within this concept the focus on frequent or even most profitable users is completely legitimate.

4.3.Uniformity and Integration

To the objectives named above, the UCC adds another one. In its recital 19, the UCC states that digitalisation is intended to “ensure an equivalent level of customs control throughout the Union so as not to give rise to anti-competitive behaviour at the various Union entry and exit points”, and refers to the need for harmonisation.UCC (n. 7), Recitals 20, 21. It thereby explicitly spells out what had been implicit in earlier decisions, namely the goal of customs control uniformity. This uniformity is to inhibit “anti-competitive” behaviour; i.e. behaviour which hinders free market competition.

In the EU, centripetal normative forces have a special place of honour, given that this polity is marked by the ongoing need to create an integrated society. Integration and its implicit strive for greater uniformity are built into the EU project – as characterised by the famous aim of constructing an ‘ever closer union’.For the history of this phrase, see Vaughne Miller, ‘“Ever Closer Union” in the EU Treaties and Court of Justice case law’, House of Commons Briefing Paper 07230 (2015). Even as commentators assure that this uniformity is gradually making place for greater flexibility and cooperation,Grainne De Burca & Joanne Scott (eds.), Constitutional Change in the EU. From Uniformity to Flexibility? (Hart Publishing, 2000), 2; Jan Zielonka, ‘Plurilateral Governance in an Enlarged European Union’, Journal of Common Market Studies 45(1) (2007), 187-209. the customs regime would appear to counteract such assertions. Here, where Member State authorities act as EU institutions and are enforcing EU law, uniformity is still very much sought after. Efforts at coordination have also become widespread, using concepts such as interoperability to structure Member State interactions. The concept of interoperability expresses the idea that flexible coordination simply seeks uniformity at a greater level of abstraction, namely at the level of operability standards and interfaces.

Uniformity – or rather, integration – is one of the core promises of digitalisation,Albert Meijers, ‘Why don’t they listen to us? Reasserting the role of ICT in Public Administration’, Information Polity 12 (2007), 235. which is lauded for its potential to break up ‘silos’, i.e. small organisational units which do not share information. In the case of the EU customs regime, these ‘silos’ are equated with the Member States which also form rather separate organisational entities across which information does not flow freely. Digitalisation here is valued for sustaining increased information flows, enabling local diversity at the same time as increased administrative harmonisation. Accordingly, the formulation of administrative integration seeks to do justice to the potential of greater local differentiation as well as the potential for greater macro-level harmonisation. The UCC falls short of calling for identical customs procedures throughout the EU Member States, instead speaking of an ‘equivalent level of control’ and of mutual recognition of customs decisions and of harmonisation of custom controls and information requirements.UCC (above n. 7), Recitals 20-22.

In this context, it is necessary to remember that the Customs Union is an exclusive EU competence according to art. 3(1) TFEU. It is a fundamental feature of the internal market and one of the areas in which the attribution of competences to the Union is least critical. This is not only caused by the clear language of the Lisbon Treaty, but also follows a functional logic. With the establishment of a free trade area without border controls inside the Union, the control of trade at the outer borders of Union territory has become a genuinely common concern. This extends not only to the principles and general legal rules concerning customs, but also to the practicalities of border control. In order to maintain a Common Market, products must be treated the same irrespective of where they enter the Union, so as not to create price distortions.

The e-customs initiative here offers a tool to achieve greater uniformity, firstly, because as a process of administrative reform, it offers the potential to easily create common standards and practices and, secondly, because the greater expected flows of information between different national customs, administrations have a likely integrative effect. In addition, a certain degree of uniformity is technically necessary for e-customs programmes to interconnect. This is expressed by the requirement of interoperability. As a result, uniformity is both a prerequisite and result of customs digitalisation. As soon as a certain level of uniformity is reached, positive feedback loops are likely to push for greater harmonisation and integration.

Uniformity as an aspiration relates to both, service orientation and efficiency in that it can function to support both. Uniformity supports service orientation in as much as this relieves the administrative burden created by nearly thirty different customs systems. Uniformity supports efficiency in as much as it enables the smoother processing of customs applications where this concerns more than one Member State.

Of the goals mentioned in the EU customs legislation, integration is probably the best conceptualised normatively. The creation not only of a common market, but of a Union, a ‘coming together’ has been EU parlance since the beginning.See the Schuman Declaration of 9 May 1950, made by the French foreign minister at the time, available in English at https://europa.eu/european-union/about-eu/symbols/europe-day/schuman-declaration_en (last accessed on 30 April 2018). Indeed, any step towards a stronger and more effective EU polity and policy should be seen as guided by the goal or value of integration. In the EU Customs Union more specifically, the implementation of an electronic customs system requires the sharing of responsibilities among the customs authorities of different Member States.UCC (n. 7), Recital 23.

More specifically, the norm or value of integration and uniformity finds expression in the standardisation that accompanies the digitalisation process. Here, administrative procedures and forms are standardised so that customs formalities in one Member State can be mapped exactly onto customs formalities in another Member State. This goes to the extent that customs officials in different Member States are instructed in how to account for specific real-life phenomena (i.e. differentiated ownership, as is common in trade procedures) on the customs form so that these are reported in a standardised and uniform way.

Because of ongoing political and academic debates about what integration pertains to and how far it can be justified in relation to other goals, implementing agencies should be able to make informed decisions about what kind of uniformity is required in specific circumstances. However, uniformity has a strong counterweight in the ‘federal’ structure of the Union.

5.No escape from Federalism

The quest for uniformity is complemented by the recognition of the federal composition of the EU. This federalism is not only a legal fact, but constitutes also a normative principle in that it provides for differentiation and diversity. Federalism as a value is expressed in the e-customs decision and the UCC, though only in weak terms. The most notable expression is that a positive opinion of the relevant committee is required to set up a work programme.Ibid., Recital 6. Yet, the modus operandi of the implementation of the e-customs initiative makes clear that as centralised as customs administration may be, federalism remains an important cornerstone of its operation. Member States are responsible for significant parts of the software development and deployment, and remain able to implement differing systems.

This situation is evidence for the fact that the digitalisation project is subject to sometimes contradicting desires and aspirations. First, this is the desire to control the process at the EU level. Because e-customs concerns the EU’s own resources, the EU has a high incentive of making the procedure work better and more effectively. In theory, this would speak for a centralised development of the digital infrastructure; however, the Commission does not have the resources or the capacity to deliver such a massive project. As a result, the Commission is responsible for the development of a number of central software components, and processes at the EU level take care of providing the technical and operational standards that can steer development at the national level.

Secondly, Member States have the desire to keep control of the implementation of EU law and not to give up on their procedural autonomy.For a discussion of the concept of procedural autonomy, see Diana-Urania Galetta, Procedural autonomy of EU member states: paradise lost? A study on the ‘functionalized procedural competence’ of EU member states (Springer, 2011). For the proposition that safeguarding this procedural autonomy is an interest the Member States pursue, cf the discussion on their interest in Gijs Jan Brandsma & Jens Blom-Hansen, ‘Controlling delegated powers in the post-Lisbon European Union’, Journal of Public Policy 23(4) (2016), 531-549. The customs procedures have to fit within the national system of administrative law, and national law guarantees of individual or procedural rights have to be effective also in the customs administration. As a consequence, each Member State develops their own e-customs programmes, although the standards defined at the EU level restrict which solutions are open to the national digitalisation effort.

The question of supranationalism under the condition of localised enforcement (or vice versa) generally arises in EU law,For an overview over the issue see the introduction to Reinhard Slepcevic, ‘The judicial enforcement of EU law through national courts: possibilities and limits’, Journal of European Public Policy 16(3) (2009), 378-394. although it appears more pronounced in areas of exclusive competence. The situation in customs is one that is determined by the general characteristics of the enforcement and application of EU law; however, the tension created by these general characteristics is more acute in the context of digitalisation. Here, the national room for manoeuvre is all but eliminated, despite the necessity of enforcement through national authorities.

As a consequence, procedural autonomy in a digital EU customs environment does not refer to localised decision-making.On the issue of procedural determination or freedom see also François Lafarge, ‘Quelle coopération douanière européenne dans un context de concurrence?’, Cahiers de la sécurité et de la justice 38 (2016), 132-141. Customs decisions are, if still possible, even more standardised within the digital environment than they had been before. This also applies to decision-making, as relevant procedures are prescribed at the EU level. Instead, autonomy, and with it the necessity for localised development of components of the customs environment, relates to form. In countries where administrative decisions require the signature of an official to become valid, the system has to make these signatures possible, and in countries where applications require signatures by an applicant, the same applies. In countries where individuals have the right to consult their file, information needs to be structured so as to allow for the retrieval of such files, and in countries where applicants have the right to receive a decision in the form of an official (paper) document, the system needs to provide for this as well. As a result, the technical challenges of digitalising customs may well differ, as the requirements on the several systems differ as well.

In sum, customs digitalisation is characterised by an interesting mix between the restriction (if not elimination) of substantive discretion and the requirement of procedural diversity. This latter requirement, together with a lack of Union resources, necessitates the distributed development of customs instruments. Whether this form of ‘federalism’ is a model for other areas of EU law remains to be seen, especially as it should be kept in mind that in the area of customs, integrative forces may be untypically strong.

6.Equality, Transparency and Participation

The above-mentioned values are those that can be extracted from the legal and policy instruments pertaining to the e-customs initiative. While these values are not in themselves critical, they hardly already present a balanced and democratic normative system. Indeed, values which are specifically mentioned in legislation are complemented by values which, through their constitutional or cross cutting status, apply to all administrative actions. Such values do not need to be specifically mentioned in the customs regime to be formally applicable.

Administrative regimes throughout Europe recognise general values of human dignity and agency, which are enshrined in administrative principles such as the right to be heard, the administration’s duty to give information, the right to a fair procedure and the right to non-discriminatory treatment. Within the EU, these values are enshrined at constitutional level in arts 9-12 TEU, which founded the duty of equal treatment of all citizens by European bodies, as well as obligations of transparency and participation which apply to non-representative administrative bodies. These values are applicable and relevant to the present situation. The general relevance of these values for administrative procedures is well established, yet their specific place in the digitalisation process requires some elaboration. Also, in the context of e-customs as elsewhere, “[t]ransparent, participative, or collaborative actions taken by government may, when operationalized effectively, have the effect of enabling a citizen to derive substantive economic, social, political or strategic values and/or intrinsic value related to government itself.”MTM Harrison et al, ‘Open government and e-government: Democratic challenges from a public value perspective’, Information Polity 17 (2012), 91.

The values of efficiency, uniformity and, to some extent, service orientation, are already instantiated in the digitalisation project. This not only means that customs officials need to apply them, but also that they are expressed in the customs regime more broadly. The technological landscape itself is or will be structured by them. This landscape, at least in part, takes the place of what otherwise would be effectuated through legal obligations. For example, in a paper-based customs environment, there might a legal duty of all customs authorities to provide information on specific customs applications to their counterpart from other EU-Member States in order to further uniform procedures; however, the digitalised customs system allows for the immediate and automatic distribution of the information on all customs applications to all customs authorities without any specific customs officer even having the opportunity to keep this information within closer local limits. The legal obligation of the official is thereby made obsolete by the technical infrastructure. This also means that where there is a legal norm that could justify the customs official retaining information in specific cases, this rule has limited effect where it is not part of the architecture. The customs official on whom such an obligation to protect information would be imposed does not have the capacity to retain information if the customs enforcement architecture does not support it.

Customs digitalisation restricts the individual discretion of customs officials. This is certainly intentional, as this restricted discretion is necessary to lower variance and increase uniformity and, to a certain extent, the time taken for decisions, thus increasing efficiency. However, it is important to recognise the extent to which the customs system does not rely on human beings, i.e. customs officials applying the law anymore. Rather, it creates a technological landscape in which it is impossible to not apply the law; this technological system needs to support the adherence to values which were previously the responsibility of customs officials.

As a result, in customs digitalisation, such legal constructs acting as cross-cutting principles of administrative law will likely lose traction. Values that are not explicitly incorporated in the structure of the digital environment can be extremely hard to substantiate later on at the point of application. In digitalised governance, governance principles must be explicitly incorporated to remain effective.

In addition, even where the system still operates through the decisions of individual officials, the greater standardisation of these decisions and their pre-determination through European rules leaves an increasingly small margin of discretion to officials, and consequently to national norms and values which could enshrine equality, transparency and participation. Especially in light of the above comments on procedural autonomy and the very formalistic and restricted nature of this autonomy within customs administration, it is not feasible to lay the realisation of equality, transparency and participation at the door of the Member State’s administration only. The Member States increasingly do not have the leeway to ensure these values so long as they are not already the basis of European rules and structures.

Admittedly, arts 9-12 TEU are not the only norms that could be referenced here. The principle of good administration could certainly be relevant to the discussion, and possibly other values such as fairness. However, literature has not settled on what the principle of good administration amounts to in the context of setting general administrative rules.Through its inclusion in the charter, the right to good administration is often phrased with a focus on the protection of the individual – a setting somewhat different from the creation of generally applicable rules. Yet, even in this context, its content remains vague. On the right to, or principle of good administration, see Hans-Peter Nehl, ‘Good Administration as a Procedural Right and/or General Principle?’ in H. Hofmann & A. Türk (eds.), Legal Challenges in EU Administrative Law (Edward Elgar Publishing, 2009), 322-351. In addition, it is doubtful how far the EU law right to good administration binds the Member States. See Michalina Szpyrka, ‘Good Governance and Human Rights. The Right to Good Administration’ in R. Grzeszczak (ed.), Challenges of Good Governance in the European Union (Nomos, 2016), 63-78. Further, while the duty of fairness has a pride of place in UK administrative law,Denis J. Galligan, Due Process and Fair Procedures (OUP, 1997). it does not have equal standing in continental European or EU law doctrine. Equality, transparency and participation contrastingly present a consensus canon of values relevant for the exercise of public authority, as can also be seen by their inclusion in the TEU.

Nevertheless, the elaboration of these principles in this paper and the evaluation of the digitalisation effort in light of them should not be seen as aiming to  exclude similar concerns. Instead, they should function as an exemplification of the problems of the digitalisation effort, which overlooks the relevance of including fundamental normative concerns in structuring the interactions between individuals and the administration. 

6.1.Equality

Equality is a complex concept with a long tradition in legal thinking.Jürgen Schwarze, European Administrative Law (Sweet & Maxwell, 1992), 545. As a concept, it combines the normative recognition of sameness with recognition of variance. The meaning of equality is generally agreed as a norm that describes, at minimum, the prohibition of inappropriate differentiation (bias).Paul Craig, EU Administrative Law 2nd ed. (OUP, 2012), 496. For the present purpose, equality can be approached from two sides – the side of access to the administration and the side of treatment by the administration. Thus, the principle of equality requires that subjects can equally draw attention to their needs and that the administration gives this attention without bias.

Digitalisation concurrently enhances and limits access. The interaction between the subjects of an administration and administrative bodies is shaped by both the specific procedures established and the infrastructure by which these procedures take place. Access is enhanced where the use of digital communication makes, for example, costly travel unnecessary thereby lowering the burden on subjects who would like to interact with the administration. However, the use of digital technology also makes access to customs more difficult.

Where administrative processes require communication on paper and in writing, the availability of paper forms and literacy rates shape the factual access of a subject to the administration, even though the low price of paper and near universal literacy make this fact near invisible in Europe. The same dynamic applies to a digital infrastructure — where administrative processes require communication via information technology, the access to administrative services is predicated on availability of computers, internet connections and digital literacy,So also Mario Martini, ‘Transformation der Verwaltung durch Digitalisierung’, Die öffentliche Verwaltung (2017), 450. as well as technical access to the customs environment. These are much less evenly distributed across Europe than access to paper and the ability to write on it. The problem is then that these preconditions of accessing the customs administration shape and distribute the ability to deal with customs successfully.

Digitalisation also has effects on the equal treatment of subjects by the administration. Aspects such as standardisation and the loss of local and personal discretion can work towards equality before the law.Mark Bovens & Stavros Zouridis, ‘From Street-Level to System-Level Bureaucracies: How Information and Communication Technology is Transforming Administrative Discretion and Constitutional Control’, Public Administration Review 62(2) (2002), 174-184. This is the case where the ‘human’ element brought to the process by the role of the official in the proceedings introduced inequality into the process, i.e. through treating members of certain social groups  or specific individuals better or worse than others. Here, digitalisation promises more equal treatment, in as far as greater standardisation pre-determines outcomes to a greater extent, and in as far as the machines used in machine-assisted decision-making are not subject to the same prejudices as humans. Yet, the question of whether greater generalisation contributes to more unbiased decision-making, or whether such generalisations get in the way of taking individually relevant circumstances into account, thereby increasing inequality by treating unequal situations equally, is an open question.Frans Jorna & Pieter Wagenaar, ‘The “Iron Cage” Strengthened? Discretion and Digital Discipline’, Public Administration 85 (2007), 189-214. In the field of education, standardised testing is even considered to provide a significant contribution to perpetuating inequalities. See Oscar Espinoza, ‘Solving the equity – equality conceptual dilemma: a new model for analysis of the educational process’, Educational Research 49. (2007), 343-363. Similar arguments can be put forward from a restorative justice perspective: Juan Marcellus Tauri, ‘An Indigenous Perspective on the Standardisation of Restorative Justice in New Zealand and Canada’, Indigenous Policy Journal 1 20 (2009), 1-24.

Such considerations are relevant in the present context as EU customs digitalisation relies on standardisation across the EU. However, such standardisation does not happen only in the context of a digitalisation project, nor is it a necessary accompaniment of digitalisation. On the contrary, digital systems are also valued for their capacity to integrate variety and to enable transparency and control in circumstances with low standardization.See Patrick Dunleavy et al, ‘New Public Management Is Dead – Long Live Digital-Era Governance’, Journal of Public Administration Research and Theory 16 (2007), 467-494 or Helen Margetts & Patrick Dunleavy, ‘The second wave of digital-era governance: a quasi-paradigm for government on the Web’, Philosophical Transactions of the Royal Society A 18 (2013), 371. Of course, under conditions of variance, it is nevertheless necessary to define interfaces and thus equivalence and functional fit of varying administrative procedures in order to construct a unified system. However, such ‘meta-standardisation’ is much less constraining that then ‘direct standardisation’ which has been chosen. From the official documents pertaining to customs digitalisation, it is unclear whether the focus on standardisation was a conscious policy choice or whether this was a tool for controlling paper-based administration, which has been transported into the digitalisation effort.

In addition, computer-based systems have been known to exhibit prejudices towards the background, skin colour or other objectively irrelevant characteristics of individuals in contact with the system.Mario Martini‚ ‘Transformation’ (n. 89), 453. Such prejudices do not only work directly, but also indirectly, hiding behind factors with more objective appearance, such as income thresholds, place of residence or occupation. Thus, while computer-based systems can be forcibly unaware of certain characteristics by not being made to compute some information, they can also serve to hide bias where ‘objective’ criteria have the effect of structurally disadvantaging specific groups.Matthew Smith, Merel Noorman & Aaron Martin, ‘Automating the Public Sector and Organizing Accountabilities’, Communications of the Association for Information Systems 1 26 (2010), 4. This bias is harder to detect and overcome as computers have little self-reflexive capacity, and there are very few persons which have enough insight into and knowledge of the algorithms to detect such structural aspects.

Digitalisation of customs administration as such is thus in and of itself neither necessarily equality enhancing, nor necessarily working in discriminatory ways. Instead, digitalisation changes the interactions between subjects and administration, creating more equality in some instances and less in others. Given that the principle of equality would prohibit the deterioration of standards of equality, it is first and foremost necessary to evaluate the effect of digitalisation on the equality of the attention that the administrations give to subjects. The first step for this would be to legally recognise the relevance of the principle of equality, so that it will be inserted into the process of technical guidance and standard setting and ultimately have an effect on decisions regarding the technical design of the e-customs systems.

The UCC makes reference to the requirement of equal treatment and appears to see the guarantee of equal treatment, first and foremost, as a task for the Commission. At three points, the preamble mentions that a certain task is delegated to the Commission under art. 290 TFEU in order to ensure equal treatment of the persons concerned.UCC (n. 7), Recitals 29, 46, 51. Interestingly, the resultant delegated act does not make reference to this goal of equality.The UCC Delegated Act (n. 15), makes no reference to the terms ‘equality’ or ‘equal’, or any permutation of ‘discrimination’ or ‘accessibility’. Instead, even where it accepts derogations from a digitalised procedure, it does so where the customs authorities did not have access to the relevant computer systems, not where the importing or exporting persons suffer from a lack of access.For example in Art. 129d(5)(c) UCC Delegated Act (n. 15), as amended in December 2015.

While the e-customs decision does refer to the question of accessibility, it is not always clear whose access is supposed to be guaranteed.Art. 1 E-Customs Decision (above n. 13). Where specific reference is made, the decision refers to accessibility for customs authoritiesArt. 4(1)(c) E-Customs Decision (n. 13). or economic operators from third countries.Art. 2(3) E-Customs Decision (n. 13). A possible exception is the Single Electronic Access Point, which seeks to simplify access to customs administration. However, as it is necessary to pre-register for access, the barriers to reaching the administration need to be judged by these prior systems, as well as by the Access Point. How difficult or easy this pre-registration is, depends on the national systems in place. These differ in terms of accessibility, as well as in the level of digitalization.

At the same time, the UCC has provisions in place to create avenues for data exchange which are in practice only accessible to more sophisticated operators. At several instances, the UCC conditions the alleviation of procedural burdens on access of the customs authorities to an economic operator’s computer system.For example, arts 13(1), 127(8), 271(4), 274(4) UCC (n. 7). However, such access presupposes relatively sophisticated internal IT management. While this is a form of preferential treatment which might well be justified by the higher frequency of interactions with the customs authorities as expected by large scale professional traders, it is also noticeable that this connection between a justified criterion for discrimination and the preferential treatment is incidental rather than causative. Thus, the sophistication of the IT systems might differ in enterprises of the same size and with the same frequency of interactions with customs authorities.

Overall, there appears to be little awareness of the effects on (in)equality in access generated by the different choices of the design of the administrative and technological process. As a result, while digitalisation offers the potential for greater equality in treatment, there appears to be little conscious effort to realise this potential. In the light of the art. 9 TEU obligation towards equality, this omission is regrettable.

6.2.Transparency

Similar to the issue of equality, digitalisation offers new solutions to problems of (in)transparency. The call for transparency in administrations refers to the requirement that the (democratic) subjects of a public authority should know about the workings of the administration, not only in the abstract, but also in respect of specific processes. To acquire knowledge about the exercise of authority within an administration, this institution needs to disseminate information about its actions. Transparency here refers to the dissemination of this information.

Generally, transparency can apply to both the information that the administrations produce in the course of administrative processes and information about how administrative processes are structured. In the context of customs administration, the former refers, for example, to data on which goods entered or left the EU; on the other hand, the latter refers, for example, to information on internal procedures and practices for processing customs requests.

Information technology makes the transmission of information significantly easier, including the disclosure or publication of information to the public, as well as to affected individuals. To find and transmit electronic data via electronic means requires virtually no transaction costs, once infrastructure investments have been discounted.

However, it should be kept in mind that data is not yet (understandable) information, and that, only under the right circumstances, will it be able to create knowledge about the processes that produced it. The transmission of data is not yet effective for transparency without the recipients corresponding ability to process the data: “[For transparency, data sets] must enable citizens to do something they find valuable and important.”Teresa Harrison et al, ‘Open government and e-government: Democratic challenges from a public value Perspective’, Information Polity 17 (2012), 95.

Several ways appear possible and reasonable to tap into this democratic potential of digitised processes. The EU has flagship projects such as the INSPIREINSPIRE stands for Infrastructure for Spatial Information in Europe and is available at https://inspire.ec.europa.eu/ (last accessed on 3 May 2018). that make data available and support the development of third-party applications that make the information within this data publicly accessible. Yet, INSPIRE is limited to geodata and does not include economic data, such as the data describing the flow of goods through customs.

For e-customs, the Commission has created a portal which provides access to Binding Tariff Information (BTI) decisions.http://ec.europa.eu/taxation_customs/dds2/ebti/ebti_consultation.jsp?Lang=en (last accessed on 24 May 2018). Also see Art. 19(2) UCC Delegated Act (n. 15) and s. III/4 of its Annex A. All information created within a BTI process that is not expressly declared to be confidential is supposed to be included in this portal. Here, an interested party can research all BTIs which have been issued. For an economic operator seeking to find out more about the tariffs his goods might be subject to, this is certainly valuable information. From a more political citizenship perspective, such a portal might be able to provide information on the exercise of public authority, but it is not clear whether this is a very effective way of doing so. The information that would be of interest from this more political perspective might be how different categories of goods (i.e. from certain global regions, or imported by a specific kind of trader) are treated overall. Such information could be drawn from the aggregated BTIs, but not from a list of single BTIs. Quite possibly, a citizen would have to research extensively through this portal and would have to deduce structural information themselves as such information is not readily available. Moreover, BTI decisions are only a fraction of what customs authorities decide and enforce.

Whichever form a more transparent e-customs system would, in fact, take, it is less relevant for the EU to adopt any specific solution to the dissemination of information, than for the EU to make use of this potential for increased transparency at all. It is relevant that the EU institutions and the Member States are aware of this potential and that they consciously interact with this potential in either (partly) fulfilling it or intentionally foregoing it (for other public goods).

The UCC makes reference to the potential of increased transparency more generally by providing that national authorities should make national customs legislation, ‘general administrative rulings’ and applications forms freely available, possibly over the Internet.Art. 14(2) UCC (n. 7). In addition, any person may request information on the application of customs legislation from their customs authorities; however, such a request has to be granted only if it refers to ‘an activity pertaining to international trade in goods that is actually envisaged’.Art. 14(1) UCC (n. 7). As a result, this creates rights which are much weaker than, for example, the EU regulation on access to documents,Regulation (EC) 1049/2001 of the European Parliament and the Council of 30 May 2001 regarding public access to European Parliament, Council and Commission document, OJ L145/43 (2001). where information may be requested without having to justify the request.How regulation 1049/2001 can be applied in this case, where there are no ‘documents’ to be requested is an interesting question, yet this discussion would go beyond the scope of the present paper. The E-customs decision refers only to the provision of information to the customs authorities or the Commission, not of these bodies providing information to the general public or affected persons.

As to the data gathered in the e-customs databases (besides BTIs and the information available through them), there is no mention of any attempt to make such information publicly accessible, in ordered or aggregated form, such as a report. Even though all the relevant data is already collected in electronic form in the e-customs environment, there is no mention of any possibility for the interested public to make use of this data.

Similarly, there is no requirement or even encouragement to disseminate information on internal processes themselves. While the MASPs certainly provide for some of this information, they are geared towards steering, not towards informing. Yet, through progressive digitalisation, these internal processes are increasingly contained in the e-customs software. It would be possible to publish the program code and structure, making it publicly reviewable. Indeed, such ‘open source’ development is practised for most critical security infrastructure.

In sum, while the e-customs initiative goes quite far with making information available in some restricted fields and forms (most notably through the BTI portal), other information is not available at all. There appears to be little conscious accord on how to approach gathered information and the information embedded in the e-customs system. Transparency, as a principle, should also push for the implementation of the transposition of the practices apparent in ‘light houses’ of transparency to other areas of the customs regime.

6.3.Participation

Much of what has been said about transparency is also true for participation. Participation refers to the demand that the public be able to influence the exercise of public authority, thereby creating a say for the subjects over whom authority is exercised.

Participation consists, to a large extent, simply of the reverse of transparency, i.e. transmitting information from the public to the Commission. As digitalisation makes the transmission and integration of this information into already existing knowledge easier, digitalisation offers the potential for enhanced participation. Similar to transparency, there is also the added requirement that the administration needs to be able (and willing) to make use of the information received. However, it is reasonable to assume that the Commission is a sophisticated enough actor that it already has the ability or is able to create the ability to deal with communications from the public, especially since electronic data processing can be helpful here as well.

The potential of digitalisation for participatory processes is described in buzz words such as ‘liquid democracy’ (often heard in Germany).Liquid democracy describes a system with instant feedback between those representing and those represented. This instant feedback is made possible through information technology. Frieder Vogelmann, ‘Flüssige Betriebssysteme. Liquid democracy als demokratische Machttechnologie’, APuZ 48, (2012), 40-46. Margrit Seckelmann & Christian Bauer, ‘Open Government, Liquid Democracy, e-Democracy und Legitimation: Zur politischen Willensbildung im Zeichen des Web 2.0’ in Hermann Hilland Utz Schliesky (eds.), Die Vermessung des virtuellen Raumes (Nomos, 2012), 325-343. This idea of liquid democracy, which has also been called a ‘participative technocracy’,Frieder Vogelmann, ‘Flüssige Betriebssysteme. Liquid democracy als demokratische Machttechnologie’, APuZ 48 (2012), 41. could well be adapted to more administrative processes, with the relevant transformation of representative participation to a non-representative setting.

Questions of participation are raised mostly where the administrative decision-making process requires significant fact finding and/or balancing of several interests. The decisions taken by customs authorities might, for the most part, not show these characteristics; EU customs regulation is highly detailed and leaves rather little room for epistemic or decisional discretion. Here, digitalisation has the likely effect of restricting any further discretion by providing for the potential to monitor more extensively and due to the high level of standardisation of procedures required as described above. However, the digitalisation process itself is one which requires exactly this establishment of a common factual basis and the balancing of the interests of a diverse group of stakeholders, as well as the general public, for whose benefit customs duties are collected. As a result, the digitalisation process itself, and further developments of the digital architecture, should be subject to participation.

While the UCC makes no reference to either consultations or participation of the affected public, the e-customs decision obliges Commission and Member States to regularly consult with economic operators ‘at all stages of the preparation, development and deployment’Art. 13 E-Customs Decision (n. 13). of the e-customs system. Given the context, this appears to refer to a one-time process of the initial creation and roll-out of the e-customs software. However, one can well argue that, in some measure, the development can also be understood as a continuous process. Thus, the participation of economic operators through their consultation is mandated by the e-customs decision.

However, these economic operators hardly approximate the EU as a polity. While they might be the most immediate subjects to customs procedures, in as much as these procedures instantiate trade policies, the subjects can hardly be restricted to the group of economic operators. However, there is no mention of mechanisms for broader participation, such as through watchdog NGOs, fair trade initiatives, local producers and consumers or representatives from third countries.

In addition, it is noticeable that the e-customs decision makes no reference to the potential for ongoing communication between the administration and its subjects that a digitalised system offers. Consultations are a staple of EU policy making and can take the form of a meeting behind closed doors at Brussels just as well as the form of online questionnaires. In theory, the digital infrastructure necessary for the e-customs environment could well support inbuilt communication channels, enabling continuous and instant communications between users and the administration. However, if this is part of the specifications for any component of this environment, this was not visible in the public documents.

It is also clear that the lack of transparency obstructs any more engaged participation. Without information on the e-customs system, an economic operator can only report on their individual experiences, which inherently constitutes anecdotal evidence. Economic operators, lacking information on the e-customs system, will not be able to flag bias in the system or otherwise to contribute to the customs administration’s processes.

All things considered, the e-customs initiative does not make use of the opportunities for better transparency and participation that a digital system offers. Likewise, the EU institutions do not appear to be aware of these opportunities – a situation which is an unwelcome surprise given the relevance that the EU itself places on its digital agenda.See the various initiatives which are listed on https://ec.europa.eu/digital-single-market/ (last accessed on 9 May 2018).

7.Data Subject’s Rights, Confidentiality and Data Security

The values mentioned above are general values of administrative law under the democratic rule of law. Yet, there are also values which are specific to the context of digital data exchange, first and foremost of which is the protection of information. As digitalisation has made sharing and publishing information much easier, keeping information secret has become more important as well.

The most important EU law instrument on the topic, the new Data Protection Regulation (GDPR)Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation), OJ L119/1 (2016). stresses the relevance of transparency and accountability for meaningful individual autonomy. This regulation has introduced significant information rights and accountability obligations of data controllers and processors. However, its reach is restricted to personal information.

There are different aspects to the safeguarding of information, namely data subject’s rights, data security and confidentiality. The protection of data subjects refers to the duty to ensure the integrity of persons in an IT network and to safeguard the collection of data points that represent individual action and expression in communication over digital infrastructure.Also to be found in Art. 8 Charter of Fundamental Rights. Confidentiality prescribes that non-personal data can also be subject to restrictions, when it is sensitive for legal persons. Data security involves the duty to take good care of data and to make unauthorised access technically difficult.

Data subjects’ rights give expression to the need to protect individuals against a negligent or overreaching public authority and thus seek to serve and protect individuals in their relations with public institutions, similar to more traditional democratic principles. Data protection has gained some attention in the context of e-government,For example: Mario Martini, ‘Transformation der Verwaltung durch Digitalisierung‘, Die öffentliche Verwaltung (2017), 443-455. In addition, the reverse is also true: e-government has become a topic in the discussion on data protection. Thus, the General Data Protection Regulation also applies to public authorities. See GDPR (n. 113), Recital 5. not least because of the different operationalisations of this principle in the Member States which has made the transnational implementation of digital governance difficult at times. The GDPR is aimed at improving this situation.

However, data protection, as commonly understood, refers only to the protection of natural persons.See GDPR (n. 113), which even carries this restriction in the title. In the case of customs, most of the data collected and shared refers to legal persons and companies, thereby not falling under the ambit of the GDPR. While there are cases where personal data is involved (i.e. contact data) and in which data protection law will be applicable, these do not concern the bulk of customs data. In such cases, art. 6(1)(e) of the GDPR is relevant, which allows the storage and use of data for reasons of public interest. The GDPR acknowledges the custom authorities’ use of personal data by expressly recognising custom authorities’ obligations to handle personal data.GDPR (n. 113), Recitals 31, 115. In particular, national data protection officers are also authorised to supervise the customs administration’s use of personal data.GDPR (n. 113), Recital 122. The provision is relevant as customs authorities, acting as they are according to direct implementation of EU law, are liable to fall in a grey zone between EU and national administrative restrictions.

The GDPR regulates the use of personal data by and within Member States; however, the use of data by the EU institutions is subject to a different and much older regime, namely Regulation 45/2001.Regulation (EC) No 45/2001 of the European Parliament and the Council of 18 December 2000 on the protection of individuals with regard to the processing of personal data by the Community institutions and bodies and on the free movement of such data, OJ L8 (2001). The synergies and tensions between these two legislative instruments are not yet entirely clear, and the point might even be moot as there is an ongoing legislative process for recasting this instrument.See Legislative Procedure 2017/0002(COD). For an overview of the status of this procedure and related documents, see https://eur-lex.europa.eu/procedure/EN/2017_2 (last accessed on 30 April 2018). While this adaptation is envisaged to bring the regime applicable to the EU institutions more in line with the GDPR, a more in depth analysis of the instrument would be necessary to evaluate this claim.

The second aspect – confidentiality – refers to the fact that in the process of customs applications, companies will share sensitive data with customs authorities, i.e. data which, if generally known, might (negatively) impact the market position of a company. So understood, confidentiality for an undertaking’s operation is what data protection is for natural persons. Undertakings are ensured protection rights, similarly to the rights of data subjects. The UCC provides that this information shall be protected by obligations to professional secrecy.Art. 12(1) UCC (n. 7). While this does not exclude its transfer in the course of customs operations, it has repercussion, for example, for the possibility of public access to documentsArt. 4(2) Regulation 1049/2001. and access through the BTI portal mentioned above.See s. 6.2 on transparency. The classification of information as confidential rests firmly in the hands of economic operators themselves.See Art. 19(2) of the UCC Delegated Act (n. 15). Thus, undertakings have rights to the protection of their sensitive data, which appear to be easy to use. In addition, some information rights are provided for, particularly regarding breaches of confidentiality. Thus, Member States have the duty to inform concerned undertakings where their data has (potentially) been subject to a breach of security.

Lastly, data security, refers to the possibility that data is being accessed without the consent or knowledge of the authority to which it is entrusted. In so far as this concerns personal data, data security is also an object of the GDPR.Art. 32 GDPR (n. 113). However, data security, i.e. the technical infrastructure that ensures secrecy, integrity and availability of the data, is not specific to personal information. Any technical or organisational measures which seek to safeguard data would be useful for both cases.

Here, in order to effectively implement data subjects’ rights, as well as confidentiality, authorities have a duty to secure data against unauthorised access. The difficulty in this is not the obligation to protect in and of itself, but the question to what length an authority must go (and which costs it can be expected to incur) in order to secure data and to guard its integrity and availability.

The UCC and E-customs Decision are conspicuously silent on the topic of data security. This silence is surprising, given the relevance that also technical experts ascribe to the topic and its high profile already for some time. The UCC implementing act does create an obligation to data security;See Art. 3 UCC Implementing Act (n. 16). This provision is based on Art. 16(1) UCC (n. 7), i.e. the duty to implement a system for the exchange of information between authorities. however this obligation is much less specific than the obligations described, for example, in the GDPR. All in all, the UCC appears much more concerned with the security of the shipping and handling operations than with data security.

All these aspects of data management connect fundamentally with citizen trust.William Dutton et al, ‘The cyber trust tension in E-government: Balancing identity, privacy, security’, Information Polity 10/1-2 (2005), 13-23. Taking proper care of the data and using it, non-intimidatingly, in the citizen’s best interest is a precursor to citizens and companies trusting public authorities with their data and thus being willing to use e-government applications. While such use can be mandated, being forced to provide public authorities with sensitive data without trusting their willingness and capacity to safeguard it would likely erode that authority’s legitimacy.

Both, the e-customs decision and the UCC, make explicit reference to data restriction, usually defining it as a limiting condition to digitalisation efforts, and data sharing between Member State authorities.Recital 20, Art. 12(3) UCC (n. 7) and Art. 3(2) E-Customs Decision (n. 13). However, they do not present its constructive, enabling function for building a digital market which enjoys social acceptance and furthers social inclusion. At the least, data protection and confidentiality should be linked constructively with goals, such as customer service orientation, as well as equality.

However, the e-customs initiative treats data management too often as an external restriction, instead of as an internal value for its own operation. The success of the digitalisation process depends, at least in part, on economic operators adopting these new procedures and not seeking to avoid them. While economic operators in principle favour digitalisation, this endorsement is not unconditional.Bradford Rohmer et al, ‘Evaluation of the electronic customs implementation in the EU: Final report’ (2015), https://ec.europa.eu/taxation_customs/sites/taxation/files/resources/documents/customs/policy_issues/electronic_customs/ecust_evaluation_final_en.pdf, (last accessed on 23 May 2018). Likewise, as a pilot project for EU wide e-government, e-customs should show the potential for the positive social and integrative development of the digital market.

Data management interacts with the above described democratic values. At the moment, data protection appears as the most effective tool for transparency and accountability in the E-customs regime. Thus, the GDPR information rights are the furthest reaching information rights applicable to the E-Customs regime, and a GDPR information request promises the greatest insights into data flows and system structures.Under Art. 13-15 GDPR (n. 113), data subjects have the right to request information on which data a controller has on them, how it is used and who it has been handed on to. However, the usefulness of such requests is restricted, since they can only pertain to personal data. Yet personal data presents the marginal case in the E-Customs Regime, where much more data refers to economic processes and undertakings.

8.Conclusions

The pursuit of efficiency, service orientation and uniformity is a laudable endeavour. The reality of a fragmented Union makes striving for these values already challenging, though not impossible. Thus, any criticism to be voiced here is not aimed at the fact that the EU pursues any of the values mentioned. Instead, criticism should be aimed at the fact that efficiency and service orientation are insufficiently conceptualised to provide much normative guidance. Throughout the legal framework, it remains unclear how they are to be assessed.

The tension between integration / uniformity and federalism as values is dealt with more clearly, since it is already known from many other areas of EU law. In the customs region, there is an established balance in which the Member States continue to shoulder much of the technical development, while their discretion is restricted to very formal choices which brings customs administration in line with national procedural requirements. This balance is probably unique to the area of customs, where, as customs is an exclusive EU competence, the EU has established significant legitimate interests in controlling administrative processes.

Most importantly, criticism has to be voiced with an eye on the values of which the digitalisation process remains largely oblivious. Equality, transparency and participation are fundamental values in administrative processes; they are enshrined in the Treaties and should be a core component of any effort towards administrative reform. Careful data management is a building block of any socially acceptable digital market. The fact that customs digitalisation is apparently able to largely neglect these values does not bode well for the democratic credentials of this EU led digital transformation.

In addition, the values presented here interact in a web of interrelations. This goes not only for the well-established parings of transparency and participation or efficiency and service orientation, but also for less discussed interlinkages. Thus, it has been shown that giving voice to service recipients is highly relevant for high service quality.Jane E. Fountain, ‘Paradoxes of Public Sector Customer Service’, Governance 14 (2001), 64 et seq. This way, participatory mechanisms do not only have democratic value (although ‘only’ might be the wrong choice of words here), but they are also necessary to establish the proclaimed goal of orientation towards economic operators’ needs. This is all the more necessary as there are no realistic possibilities for exit from the e-customs system for EU residents. The shallow way in which the goals of the e-customs initiative have been defined makes it difficult to tap into potential synergies. This does not bode well for their guiding force and their effectiveness as ‘signposts’.

Customs is a forerunner of digitalisation at the European level. This fact might have contributed to the rather pedestrian aspirations expressed in the documents establishing the e-customs initiative; with this project, Europe tests the waters and it is thus counterproductive to be overambitious. At the same time, it is important that such a pilot project at least gives an idea of the benefits to be expected. The unimaginative and possibly ignorant way in which the e-customs initiative does not provide for securing and improving equality, transparency, participation and responsible and wholistic data management is no reason for optimism regarding the EU’s digital agenda in the future.