The Institutional Environment

The previous chapter discussed at length how the dominance of services can be seen as one of the most striking aspects of changes in the world economy over recent decades. Today, services account for around 75 per cent of all jobs and GDP in OECD economies – and over 50 per cent in developing countries and emerging economies. While total trade in services has remained constant for the last two decades, developing countries have almost doubled their share in the world trade in services to reach more than 30 per cent in recent years. The significance of services goes beyond their growing share in the economy and close connection to technology and knowledge. It is also intimately related to an expected surge in their internationalisation resulting from durable regulatory reforms. An institutional environment enabling the internationalisation of services has gradually emerged with the application of the General Agreement on Trade in Services (GATS) in 1995, negotiations underway at the World Trade Organization (WTO), and the adoption in 2006 of a new EU directive (2006/123/EC) on services in the internal market. Moreover, as we will see in further detail in this chapter, preferential trade agreements, including the new – and highly controversial – generation of mega-trade deals such as the Canada-European Union Comprehensive Economic and Trade Agreement (CETA), the aborted Transatlantic Trade and Investment Partnership (TTIP) between the United States and the European Union or the Comprehensive and Progressive Trans-Pacific Partnership (CPTPP) among Asia-Pacific countries, specifically target convergence in regulatory approaches, harmonisation of standards, and growth of investment and trade in services as crucial issues.

In the past, technical specifications were largely the preserve of the regulatory framework of law, company standards set by managers, and, to a marginal degree, national standards institutions. Today, the regulatory framework of law has yielded ground to voluntary standards drafted by a raft of international or regional public and private sector bodies. The creation of the WTO in 1995 was a crucial threshold. Unlike the GATT, whose provisions in terms of technical regulations were not very restrictive, the Agreement on Technical Barriers to Trade (TBT), the Agreement on Government Procurement (GPA), the review of the Agreement on Sanitary and Phytosanitary Measures (SPS), and the General Agreement on Trade in Services (GATS) grant international standards a major role in the harmonisation of technical specifications applicable to goods and services. State regulation in this domain must comply with ‘legitimate objectives’. With regard to goods, such concerns are related to health, safety, and environmental issues. In contrast, as we have seen, conflicting understandings of market uncertainties about quality and security are the major issues in the sphere of services; they encompass a wide range of expectations regarding, in particular, competence and professional skills, the capacity to deliver business continuity, data protection and privacy, and consumer protection and information, as well as larger societal and environmental concerns. As the WTO is not a standard-setting body, its promotion of regulatory convergence is made by prompting its members to use international standards. GATS article VI:4 thus assigns to the Council for Trade and Services (through its Working Party on Domestic Regulation) the largely market-inspired task of developing ‘any necessary discipline’ to ensure that regulation by states is not ‘more burdensome than necessary to ensure the quality of the services’. Article VI: 5b specifies that in this respect, ‘account shall be taken of international standards of relevant international organisations’. According to the WTO, regulatory cooperation in services would have much to gain from improving ‘regulators’ understanding of, and confidence in, standards and requirements with which they may not be familiar’ (World Trade Organization, 2012: 186). Similarly, in the wake of earlier guidance, the OECD published in 2012 a new Recommendation on regulatory policy and governance suggesting that members, ‘In developing regulatory measures, give consideration to all relevant international standards and frameworks for co-operation in the same field and, where appropriate, their likely effects on parties outside the jurisdiction’(OECD, 2012, recommendation # 12). Yet, existing provisions still grant a wide range of international bodies the ability to define on their own terms standards affecting the internationalisation of services.

In the United States, standardisation is usually presented as fragmented and organised on a sectoral basis. A variety of competing standards organisations (formal and informal) set market-driven standards exempt from state intervention. The system follows a so-called model of direct participation, where companies have direct access to standard-setting activities with international claims. In contrast, the European standardisation system is coordinated and centralised, and operates under a higher degree of government control. The European standardisation bodiesFootnote ² follow a so-called model of national participation, where a national body holds the voting rights within umbrella standardisation bodies such as the CEN (except for the United States, the system is similar for non-European standardisation bodies members of the ISO). In spite of their differences, the European and American standardisation systems have common characteristics. Both rely on private organisations to shape standards on a voluntary basis. They follow a due process open to all interested parties and their deliberations are based on the ‘state of the art’. The draft standards are subject to public consultation and the general interest is supposed to prevail over particular interests. Finally, their standard-setting bodies recognise the primacy of international standards, even though the understanding of what ‘international’ means remains controversial. Despite these similarities, several conflicts remain between ISO/European and American standards developing organisations (SDOs).

From the American point of view, the national participation model in the European standardisation bodies gives them a substantial advantage at the international level (Zuckerman, Reference Zuckerman1996: 40; Czaya and Hesser, Reference Czaya and Hesser2001: 32). The Vienna and Dresden agreements between the ISO and CEN, respectively with the International Electrotechnical Commission (IEC) and the European Committee for Electrotechnical Standardization (CENELEC), can indeed be seen as benefiting European actors, as they grant provisions for a simultaneous recognition of standards at the European and international levels (with CEN potentially leading the work) and have ensured a coordination of the standardisation work between those organisations. Moreover, with about 4,000 European standards indirectly referenced through 30 directives, the New Approach allows for a presumption of conformity with essential requirements for all firms that claim to be using such standards; but clearly, there will be more European than American firms doing so (ASTM International, n.d.)!

In reverse, from a European point of view, the decentralised and fragmented standard-setting procedures in the United States represent a barrier to the US market. Moreover, American SDOs’ claims to serve the public interest often hinder strong commercial interests and contending regulatory competition. Finally, the international reach of standards developed in the United States tends to undermine the authority of formal standardisation arenas such as the ISO and CEN.

Unsurprisingly, scholars have discussed such transatlantic divergences on the most appropriate institutional foundation of international standards at great length (Schmidt and Werle, Reference Schmidt and Werle1998; Abbott and Snidal, Reference Abbott and Snidal2001; Czaya and Hesser, Reference Czaya and Hesser2001; Egan, Reference Egan2001; Nicolaïdis and Egan, Reference Nicolaïdis and Egan2001; Spruyt, Reference Spruyt2001; Tate, Reference Tate, Hall and Soskice2001; Werle, Reference Werle2001; Mattli and Büthe, Reference Mattli and Büthe2003; Vogel, Reference Vogel, Mattli and Woods2009; Winn, Reference Winn2009: 21; Mattli and Büthe, Reference Mattli and Büthe2011). Yet Egyedi questions such a clear-cut transatlantic divide in standardisation (Egyedi, Reference Egyedi and Jacobs2005). She stresses that this tends to underestimate the opening of most industry consortia and overestimate the democratic institutional pledge of formal organisations. While committees in both cases are formally open and work on a consensus-oriented basis, stakeholders with few resources, whether in civil society organisations or small and medium-sized enterprises, continue to take pains to participate in standard-setting practices undertaken in technical committees. Thus, it is important to overcome the conventional caricatures opposing the American and ISO/European models.Footnote ³ Making any a priori assumption about the role of public authorities in constructing the authority of standards is of little use overall, as it depends on evolutionary variations regarding the political economy of state– market relations as much as on preferences regarding the issues concerned (Dudouet et al., Reference Dudouet, Mercier and Vion2006: 389). This is noteworthy with services, which can be highly technical, but at the same time embody contentious political interests and societal values. For instance, all sorts of standards related to information and communication technologies are used in services related to the development of smart global cities for improving transportation, energy efficiency, sustainable planning, and so on, but none of them would be of much use if left in a regulatory vacuum. More generally, the multiplicity of standards surrounding our everyday life has an influence on our health and safety, regardless of their place in regulation. As we saw in the previous chapter, the inclusiveness of standard-setting processes remains an issue whose significance lies beyond mere regulatory public policies. In a nutshell, standards are regulation.

This prompts us to have a closer look at the institutional settings for service standards provided by the activities of formal SDOs within the ISO environment, the European Union, the United States, as well as by the prospects arising from the new generation of trade agreements.

The ISO Setting

The ISO is a major arena for assessing current developments of service standardisation. As the world’s largest developer and publisher of international standards with a membership of 160 or so mixed private and public national standardisation bodies, the ISO represents the wide range of public and private actors involved in services standardisation. The move into standardisation of services began in 1995 with a Consumer Policy Committee (COPOLCO) workshop in Beijing. Lawrence D. Eicher, then ISO Secretary General, emphasised that manufacturing industry was already changing with the move into generic management system standards and, from there on, ‘the emphasis could change even more to take into account the needs of the burgeoning service industries’ (International Organization for Standardization, 1995). Six workshops were held in the following years with various foci, such as tourism, exhibition management, banking and insurance, and engineering consultancy, as well as multi-sectoral methodological issues for developing service standards. In 2001, a new working group was established to draft a guide on the use and development of service standards from a consumer’s perspective (ISO/IEC Guide 76:2008, Development of Service Standards – Recommendations for Addressing Consumer Issues). Since then, service standards not only appear each year as a key priority area of the work programme of the COPOLCO; this also led the ISO to develop its own Strategy for Service Standardization (International Organization for Standardization, 2016a) to increase ISO’s visibility in this domain, help members develop service standards, and better understand market interests.

The number of ISO standards in relations to services is 700. This is still few (approximately 3 per cent) compared to more than 22,000 international standards and standards-type documents in the whole ISO catalogue (International Organization for Standardization, 2017: 5). Moreover, such figures should be taken with caution as they not only refer to specific requirements to be fulfilled by a service but also all sorts of standards that can support service provision.Footnote ⁴ Thus, standards labelled as belonging to services include domains far removed from what is usually understood as services, such as transport infrastructure, lab techniques, and construction engines. The broad inclusiveness of the international classification for standards shows the uncertainties in defining and classifying service standards, which can never be taken for granted. Yet some developments have taken place in domains epitomising core intangible and relational features of services. This is particularly the case for professionals providing personal financial planning such as in pensions per capitalisation (ISO 22222:2005), in the vocabulary and service requirements for market, opinion, and social research (ISO 20252:2012), and in safety requirements for scuba diving (ISO 24801–1:2014), as well as minimum quality requirements for services provided by tourist information offices (ISO 14785:2014). Those distinct sectoral standards remain marginal in terms of the global service economy. Obviously, large parts of this economy, such as finance and insurance, use instruments developed within their own sector, even if their ability to legitimately claim great authority in self-regulation has been seriously challenged in the context of the global economic crisis. Cross-border service providers also rely on more generic standards, which may indifferently be applied in the production and exchange of goods and services. Among the most widely used are the quality, environmental, and information security management system standards ISO 9000, 14000, and 27000 series, as well as the guidance on conformity assessment provided by the ISO 17000 series or the ISO 31000 guidelines and principles of risk management (Guler et al., Reference Guler, Guillén and Macpherson2002; Prakash and Potoski, Reference Prakash and Potoski2006; Lalonde and Boiral, Reference Lalonde and Boiral2012).

Within the ISO setting, the development of service standards raises challenges pertaining to their content and the distinctiveness of services as compared to generic management standards. The relational and immaterial features of services prompt the development of standards that encroach simultaneously upon the intended quality of a service and the business operating procedures to deliver such a service. In the ISO, the latter is understood as a management system standard (MSS) and is kept separate from the former with dedicated procedures.Footnote ⁵ According to this so-called exclusion principle, any light quality management standards is ruled out from the back door. Yet, according to a number of participants in ISO technical committees, this may sometimes hinder the development of services standards or diminish their attractiveness to end users (International Organization for Standardization, 2017: 12). MSS represents a highly sensitive field of standardisation that requires a so-called justification study (formerly known as ISO/IEC Guide 72) and the adoption of a common document structure and terminology.Footnote ⁶ For instance, the distinctiveness of services and the desire of small and medium enterprises to refer to one single standard as a reference has led the ISO technical committee on tourism to send several requests to the ISO governing body (the Technical Management Board-TMB) asking them to reconsider these rules. In 2012, these requests were unequivocally refused by the TMB, leading to substantial modification and adding to the existing delays in the development of service standards in the tourism sector.Footnote ⁷ Such requirements have impeded the development of service standards in many domains. Overcoming this difficulty will only be possible by setting standards according to a very narrow understanding of the procedural and generic aspects of services. This will make it difficult to include more substantial issues related to societal values and cultural contexts affecting the co-production of services.

Almost fifteen years after the 2005 ISO workshop ‘Global Trade in Services – New Challenges for International Standardization’ and twenty-five years after the launch of the institutional process, progress within the ISO has been meagre. Whereas some developments, such as those in risk management (ISO 31000) or energy management systems (ISO 51000), may come to have a major impact on the service sector, so far, maturity in service standardisation remains weak within the ISO environment.Footnote ⁸

The European Approach

More developments take place in Europe with the European Union in the forefront of both service integration and international standardisation.Footnote ⁹ In 1985, Council Resolution 85/C 136/01 on a ‘New Approach’ to technical harmonisation and standardisation instigated a completely new regulatory technique and strategy. The resolution was a response to the growing role of the European Court of Justice in resolving conflicts on regulatory policies in the internal European market, especially since the 1979 Cassis de Dijon case securing the principle of mutual recognition in the absence of harmonised legislation or technical standards. It was also an early move towards the completion of the Single Market by devising procedures to avoid turning technical specifications into structural impediments to trade. Although member states were suspicious about seeing regulation in this domain transferred to the European authorities, they did perceive the threat of a race to the bottom in public purpose standards as market integration progressed. The New Approach provided a framework for the harmonisation of EU public law only on the general and essential requirements of goods traded on the European market, in particular in the fields of health, environment, safety, and consumer protection. Depending on the sectors affected, technical specifications, performance criteria, and quality requirements are either based on mutual recognition of national standards or delegated to European standard-setting bodies upon formal request from the European Commission. In most sectors, the procedure for monitoring standards is a matter of business self-regulation, since products put on the market are granted a presumption of conformity, solely based on the declaration of the manufacturer (CE marking). Thus, the European New Approach has done more than strengthen the ability of companies to rely on voluntary standards rather than mandatory regulation in the Single Market. By avoiding costly third-party testing and certification, and providing the procedural means for a simultaneous adoption of European standards as international ones (through the so-called Dresden and Vienna Agreements), the EU has also included third countries in its standardisation system. The (largely unintended) outcome has been a powerful strategic positioning of European standards in the global market (for more detail, see Vogel, Reference Vogel and Vogel1995; Egan, Reference Egan2001; Borraz, Reference Borraz2007).

The European Commission was well aware that the emergence of an increasingly dense and extensive European standardisation complex with global reach could also support the 2000 Lisbon Agenda. Services were a core feature of the plan ‘to become the most competitive and dynamic knowledge-based economy in the world’. New emphasis on service standards occurred after the 2005 mid-term review of the Lisbon Agenda and adoption of Directive 2006/123/EC on services in the Internal Market, the so-called Bolkestein Directive, eventually agreed to on second reading in December 2006 and fully implemented since the end of 2009. A horizontal approach to regulatory harmonisation supposedly valid for any kind of service provision at the European level lies at the centre of the directive. The controversial ‘country of origin’ principle has been substituted for the formula ‘freedom to provide services’ in order to ensure conformity with regulations of the place of delivery. The Directive emphasises that the promotion of quality is a crucial issue for the unification of the internal market for services. To this end, it explicitly encourages professional independent associations and standard-development and certification bodies (like the CEN, CENELEC, or ETSI) to develop voluntary quality marks and labels (preamble 102 and article 26).

Against this background, the European Commission undertook a series of action to support service standardisation. It addressed in 2003 a first Programming mandate (M 340) to CEN, CENELEC, and ETSI in the field of services to identify priority sectors of intra-community trade in services. Issues could include horizontal cross-sectoral generic standards and vertical sector-specific standards, as well as service providers or end-users. A second programming mandate (M 371) was addressed to CEN in the field of services in 2005 following the transfer of responsibility for business-related services to DG Internal Market and Services. Half a dozen European standardisation bodies developed eleven projects accordingly. It is worth looking at them in some detail, as the result of this whole exercise shaped the new Regulation on European Standardisation (1025/2012), adopted in October 2012.

The CEN Horizontal European Service Standardization Strategy (CHESSS) was the largest project responding to EU Mandate M/371. It included a consortium of national standards bodies led by the British Standards Institution (BSI), with those from Spain (AENOR), Germany (DIN), Denmark (DS), Estonia (EVS), and the Netherlands (NEN), as well as CapGemini, one of the world leaders in IT services consulting and management. Its final report, published in 2009, examined the feasibility of a generic approach to European service standardisation across multiple service sectors, as opposed to following a sector-specific approach (CHESSS Consortium, 2009). The CHESSS project has raised crucial issues on the distinctiveness of service standards, echoing the aforementioned discussion regarding ISO’s ‘exclusion principle’ between quality management standards (i.e. ISO 9000 series) and what can be purposely standardised for the service sector. Indeed, the importance of quality in services inevitably led to questioning their specificity with regard to quality management standards. One module of the project claims that service standards are not about the ‘how’ but about the ‘what’, i.e. a service standard is not about how to achieve a goal, as with management standards, but specifies the goal to be achieved and the means for assessing its achievement (CHESSS Consortium, 2009, modules 4 and 5). In this regard, the proposal to develop a customer satisfaction index is undoubtedly as crucial in framing conformity assessment procedures in services as weights and measures underpinning similar procedures for products. The distinctiveness of service standards is that they extend beyond procedural issues to cover such issues as common writing models and the terminology employed across the entire service sector. The CHESSS project clearly aimed at ensuring that service standards establish their distinctiveness in the realm of standardisation, as management or performance standards did previously.

Besides the distinctiveness of service standards as such, unsurprisingly, the CHESSS report pointed out the difficulty of involving stakeholders in the development of generic standards when most of them lack the necessary awareness and resources. The difficulty of stakeholders’ involvement in service standardisation has not just been rehashed time and again in subsequent reports.Footnote ¹⁰ It also casts doubts on the institutional structures for setting service standards. For some, the current system is as appropriate for services as for products. In contrast, B2B services are seen as a good case for a new system based on a dual representation with stakeholders besides conventional national bodies, such as European organisations representing industry, SMEs, and consumers. According to the CHESSS report, ‘This double representation system ensures a balanced representation of sectors on the one hand and of national interest on the other hand’ (CHESSS Consortium, 2009: 223 (module 7)). Such recommendations have struggled to gain a significant place in subsequent European initiatives in the wake of the ‘Standardisation package’ adopted by the European Commission in 2016. However, the interest in a single horizontal generic standard with a certification scheme is clearly an attempt to promote service standards on a par with the worldwide achievement of the ISO 9000 series. Thus, the important role of the European Commission in supporting standardisation for the service sector may not only reinforce endogenous recognition of service standards. It could also pave the way for the deterritorialisation of regulatory practices through greater reliance on market mechanisms for the diffusion of such standards.

In contrast, the ten other projects responding to EU Mandate M/371 addressed the specificity of distinct service markets. Afnor, the French national standardisation body, a pioneer in setting national standards in well-defined service sectors, initiated those projects in consultation with some European partners, in particular from the Netherlands and Denmark. The recommendations identify a number of service activities likely to be standardised at various levels, whether European Standards per se, or at a lower level, guidance materials and so-called workshop agreements.Footnote ¹¹ The advantage of a vertical and sectoral approach is largely seen in the quality of the deliberation process likely to better address the distinctiveness of services in sectors of highly relational and immaterial activities. According to Pascal Gautier, head of the Management and Services Unit at Afnor, generic standards in services would soon become burdensome and unrealistic as ‘they require phenomenal efforts which would eventually generate opposition’; in his view, ‘it is much better to favour a niche approach in service standards so as to keep a sector-specific proximity, i.e. to choose a so-called Swiss army knife effect where each blade has its distinct use’.Footnote ¹² However, the ambiguous mixture of private and public actors involved in standardisation processes favoured by this approach remains important. Similarly, the issues concerned do not clearly distinguish between societal or more strictly technical objects of reference. A proper differentiation of actors among stakeholders and issues spanning physical measure to societal values, as well as clear-cut incentives to mitigate representation biases, would be necessary to ensure a fair, substantial, and thorough representation in standardisation processes.

In the wake of these early moves, the Commission initiated a reform of the European standardisation system.Footnote ¹³ Faced with the faster development of service standards at the national than at the European level, the potential creation of barriers to intra-EU trade in services, and services increasingly embedded in the delivery of goods, one of the key objectives was to establish a better inclusion of service standards in the regulatory framework. Despite opposing views of what can be standardised in services, the consultation organised in 2010–2011 led to strong support for including service standards and keeping the principle of national delegation in this domain.Footnote ¹⁴ As such, the entry into force in 2013 of the new regulation on European standardisation (1025/2012) extended the New Approach to services and compelled European national standardisation bodies to provide notification of services standardisation activities. Moreover, the new environment reinforces the support granted to European civil society stakeholders and SMEs. Nonetheless, the new regulatory framework has not necessarily diminished the divide opposing supporters of vertical sector-specific standards, such as Afnor, and horizontal cross-sectoral generic standards, such as those promoted by the British Standards Institution (BSI). This probably explains the mid-range strategy pursued by the European Commission in addressing Mandate M/517 in January 2013 to the CEN, CENELEC, and ETSI for the development of ‘horizontal service standards’; while fostering the standardisation of the generic attributes of services, the mandate emphasises the development of ‘“narrower” horizontal service standards for particular aspects/parts of a full service provision’ as opposed to a single, all-inclusive horizontal service standard. As a result, the framework devised by the European Commission for the development of European service standards explicitly includes the option of a ‘hybrid combination: a horizontal service standard with sectoral add-ons, or a pool of parallel sector-specific standards’ (European Commission, 2016c: 9). Regarding horizontal service standards as such, out of six topics identified by the European standardisation organisations as suitable for future developments at the European level, the European Commission retained the following three, explicitly listed in the new technical committee CEN/TC 447 ‘Services - Procurement, contracts and performance assessment’ created for that purpose in 2016.

The United States: A Special Case

With European standardisation processes usually seen as driven by a coherent and centralised institutional framework in opposition to the fragmented and decentralised American system, the overall design of international standardisation remains unsurprisingly disputed. In Mattli’s words,

As seen previously, despite a number of noticeable differences between the American and European systems, several features do contribute to bridging the conventional gap of such a transatlantic divide.

First, the American National Standards Institute (ANSI), a not-for-profit private organisation, plays a significant role as the national standardisation body in centralising standardisation processes. Its mandate explicitly places ANSI in charge of the coordination and representation of US interests at the ISO and IEC. ANSI also plays a crucial role in enhancing the coherence of standard-setting processes both within the United States and amongst US participants in international arenas. Without developing standards, it coordinates and accredits US-based SDOs, which in turn must comply with the ANSI essential requirements for standards development processes. In fact, the Vice President of International Policy of ANSI takes issue with the depiction of the American standardisation processes as fragmented and decentralised: according to him, they take place in an ‘organised distributed system’.Footnote ¹⁵ This particular account of the American setting presents ANSI’s coordination role in a positive light but also reflects the delineated environment in which US standardisation takes place.

While ANSI is responsible for the coordination of over 200 accredited SDOs, the National Institute of Standards and Technology (NIST) is the federal agency that fulfils a similar role at the level of governmental agencies. Over recent decades, the 1996 National Technology Transfer and Advancement Act and successive revisions of the Circular A-119 of the Office of Management and Budget have played a significant role in enhancing NIST’s profile. Those pieces of legislation and regulation entrust NIST with promoting the use of voluntary standards in lieu of government-unique standards within federal agencies. Whenever government-unique standards are used, they must be fully reported and justified. Moreover, staff across federal agencies take part in the development of voluntary standards in over 500 SDOs, with personnel from NIST alone in 114 SDOs in 2016 (National Institute of Standards and Technology, 2017; United States Government Accountability Office, 2018). The direct involvement of governmental agencies is only part of the relationship between public authorities and standardisation. More than 8,600 standards are referred to in US law, and over 10,500 in public procurement procedures. It is also worth noting that the ANSI Steering Committee not only includes representatives of industry and civil society but also a number of government agencies.

More generally, the primacy of international standards is explicitly recognised in the American system, even though the understanding of what ‘international’ means still remains controversial in the United States Standards Strategy (USSS). This remains, despite the substantial USSS revision passed in 2005, specifically to address such needs and more recent suggestions made by the Government Accountability Office under the aegis of the Trump administration that NIST should better ‘respond to circumstances when U.S. representation in international standards activities may be inadequate’ (United States Government Accountability Office, 2018: 53). Last but not least, as in European reforms, the American system has recognised that participation from the weakest stakeholders is in such short supply that it undermines the legitimacy of technical specifications supposedly driven towards the public interest; this is why recent policies on both sides of the Atlantic have taken initiatives supposedly geared towards supporting the participation of civil society organisations. The US standards strategy points out that ‘government should recognize its responsibility to the broader public interest by providing financial and legislative support, and by globally promoting the principles of our standardization system’ (American National Standards Institute, 2016: 12). In brief, American standardisation processes rely on a broader mix of public and private actors than usually acknowledged.

While the American picture is not dissimilar to the European and ISO ones, current developments in the distinct domain of service standards remain sharply dissimilar across the Atlantic. Services are for the most part low on the agenda of American SDOs. Even the largest standard-setters pay scant attention to how services may challenge the future of standardisation. The American Society of Mechanical Engineers (ASME) includes clean energy and robotics among the five core technologies targeted by its latest strategic plan, to which a number of services could potentially be associated.Footnote ¹⁶ Yet none of them specifically focuses on services. As Bernard Hrubala, Vice-President of ASME and Division Manager at TÜV Rheinland, put it when questioned about a distinct service strategy, ‘our ultimate goal at the end of the day is, don’t matter what the standard is in every country, we want their standards to be consistent with the ASME standards’.Footnote ¹⁷ ASTM International (originally known as the American Society for Testing and Materials) shares this claim to play a leadership role at the global level with an active policy of memoranda of understanding signed with more than one hundred national standards bodies, mostly in developing and emerging countries (Saudi Arabia, Columbia, and Turkey being the three countries most referencing those standards). Yet it ignores the issue of service standards and prefers to give prime importance to sustainability. It is from this standpoint that ASTM International has revised most existing standards and charts new activities such as carbon footprint and alternative fuels. Several years ago, Katharine E. Morgan, who is now President of ASTM international, went to great lengths to explain this shift: ‘We are seeing green, from roofing to isolation to degradable plastics, we are seeing that across a lot of our committees’.Footnote ¹⁸ For its part, NIST sees its role in service standardisation as closely related to strategic issues set by the US administration in domains closely related to recent advances in computing, communications, defence technologies, and healthcare (National Institute of Standards and Technology, n.d.). Finally, in 2013 ANSI launched a Services Sector Initiative to help meet the demands of standardisation within the US services sector and identify priority sectors. While the recommendations made in the wake of this initiative repeat claims towards greater awareness, visibility, outreach, and engagement, it also suggests a need to ‘identify common elements that cut across all service sectors not just one or two specific sectors’.Footnote ¹⁹ Even if the initiative has so far merely led to a few conferences and an enhanced dedicated website, this clearly contrasts with earlier views, according to which abiding by its coordination mandate would be at odds with setting any priority at all as long as its members have not done so – which de facto excluded service standards.Footnote ²⁰

Overall, standardisation in services does not lie at the heart of the American landscape. Interestingly, the few service standards dealt with among American SDOs are confined to domestic issues. For instance, the development of the ASTM Environmental Site Assessment Standard (E1527) merely responded to a requirement set by the US Environmental Protection Agency.Footnote ²¹ Officials in charge of standardisation strategy in the major bodies of the American institutional setting invariably explain the weak concern over service standards by a lack of demand. In ASME words: ‘Our scope is essentially mechanical engineering. Those services type things don’t really fall within our area.’Footnote ²² Moreover, service standards raise the issue of certification. American SDOs remain highly critical of standards likely to be used for certification purposes. Taking the example of the ISO 9000, ANSI emphasises the lack of added value brought by certification: ‘It didn’t add value if you are a large company and you already have an excellent quality management system. What does it bring to spend a few more millions of dollars or euros to get certified to something you do better already?’Footnote ²³ Whether it be an aversion towards certifiable standards or merely qualified isolationism, such a view may face renewed challenges in the importance recently taken by regulatory convergence in negotiations to establish far-reaching free trade agreements across the Atlantic and the Pacific.

Towards New Transatlantic and Transpacific Promises?

During the confirmation hearing before the European Parliament for her appointment as Trade Commissioner, Cecilia Malmström repeatedly emphasised the ‘strategic dimension to the regulatory work’. Referring to the contentious negotiations under way between the European Union and the United States for the establishment of a Transatlantic Trade and Investment Partnership (TTIP), she claimed that ‘[i]f the world’s two biggest powers when it comes to trade manage to agree standards, these would be the basis for international cooperation to create global standards’(European Parliament, 2014: 8). Similarly, in a brochure listing what the EU Trade Commissioner saw as ten myths about TTIP, the strategic dimension of setting high standards in global trade was the first point in countering the idea that TTIP would weaken strict EU standards to protect people and the planet (European Commission, 2016b). Whether it be the aborted TTIP, the Comprehensive Economic and Trade Agreement (CETA) provisionally entered into force in September 2017 between Canada and the European Union, the EU-Japan Economic Partnership Agreement entered into force in February 2019, or the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP) signed in March 2018 between eleven countries of the Pacific rim, the authority conferred on technical standards has swiftly become one the most prominent issues along with investor-state dispute settlements mechanisms in negotiations of a new generation of preferential trade agreements.Footnote ²⁴ European fears about the abolition of food safety standards as protection against importation of ‘frankenfood’ such as chlorinated chicken or hormone beef from the United States may be exaggerated. Still, the conflict-ridden negotiations for such mega-trade deals do include provisions for new harmonised standards (including in services), greater mutual recognition of existing standards, and plans to set up joint councils in charge of designing future convergence around the type and scope of technical standards to be recognised in government regulations.

The momentum towards this new generation of free trade agreements is unmistakably facing setbacks in the wake of the strong opposition coming from both the right and left of the political spectrum. Indeed, as soon as Donald Trump acceded to the United States presidency in January 2017, he pulled out of both TTIP and the Trans-Pacific Partnership and demanded a renegotiation of the North American Free Trade Agreement (NAFTA) with Canada and Mexico. A number of demonstrations and election campaigns in Europe have put the case against such agreements at the core of their demands, notably during the 2017 French Presidential elections. Also worthy of note is the tortuous ratification process by as many as thirty-eight national and regional parliaments for CETA. It demonstrated the increasing politicisation of a number of trade-related issues, including standards and regulatory convergence, as if such a thing had ever been in doubt. When Paul Magnette, the minister-president of Wallonia, stood alone in a tough constructive critique of CETA to force his European and Canadian counterparts to take his concerns seriously as a condition for the required ratification by the regional Parliament of Wallonia, he was certainly right in claiming that ‘This is not only about a treaty with Canada; this is about all future bilateral [trade] agreements. The question actually is: which globalisation do we want?’Footnote ²⁵

With CPTPP signed without the United States and its ratification still underway at the time of writing, TTIP in limbo, and EU partership agreements such as those with Canada and Japan only recently entered into force, it is not worth making a detailed study of the outcome of such negotiations with all their sector-specific variations. It is worthwhile, however, to seek a better understanding of how those new preferential trade agreements are likely to confer authority on standards as de facto or de jure regulatory instruments governing further internationalisation of services. According to the few independent and in-depth studies that exist so far, an ambitious harmonisation of standards would no longer be on the agenda (Cai, Reference Cai2016; De Ville and Siles-Brugge, Reference De Ville and Siles-Brugge2016: 38–61; Pitschas, Reference Pitschas2016; Magnette, Reference Magnette2017). Even if negotiations on mega-trade agreements resume in the future, the prospect of setting global standards thus remains unlikely or at least questionable. Any likely outcome would turn on a mutual recognition of existing standards rather than their harmonisation. Four issues are at stake in this regard.

The first concerns the regulatory chill effect that far-reaching preferential trade agreements may have on governments. This chilling effect characterises a situation in which governments become reluctant to adopt new regulations or to strengthen existing standards for fear of scaring off market actors, particularly foreign investors. The risks of a regulatory chill effect at the expense of welfare policies, consumer protection, public health, or environment policies have chiefly been discussed in connection with negotiations on new investor-state dispute mechanisms, rather than the regulatory cooperation chapters under which harmonisation and mutual recognition of standards are negotiated. They are therefore beyond the scope of this study. They may, however, just as well result from a harmonisation or a comprehensive mutual recognition of standards (De Ville and Siles-Brugge, Reference De Ville and Siles-Brugge2016: 79ff). It is plausible to consider that the sovereign right to uphold a level of protection would be seen as useless when lower levels would be just as acceptable by means of mutual recognition. This is particularly the case where regulations are not completely equivalent in terms of outcome. An outcome in levels of protection less ambitious than the status quo could also result from a joint adoption of less stringent international standards, as for instance with the Codex Alimentarius, as compared to a number of provisions included in European sanitary and phytosanitary standards. Finally, the chilling effect can arise from regulatory cooperation procedures devised for setting future standards. For instance, CETA includes provisions to ‘discuss regulatory reform and its effects on the Parties’ relationship’, with guarantees that ‘consultation and exchange should begin as early as possible’ in regulatory development processes (European Commission, 2016a, art. 21.4.a(i), (art. 21.4.b)). The very fact of having the duty to consider the effects of regulatory reform in the Parties can cool down eagerness towards new or more stringent standards. Moreover, in the absence of dedicated mechanisms to support the involvement of civil society organisations irrespective of their resources, such provisions may unduly benefit business organisations with privileged access to this type of consultation mechanisms and often more reticent towards new or more stringent standards. This brings me back to the question seen again and again in the course of this book: who sets the standards?

The second point raised by the new generation of preferential trade agreements is indeed the transfer of authority in standard-setting procedures likely to flow from their mechanisms of regulatory cooperation. Unlike previous treaties, such agreements are designed as ‘living agreements’, where parties can engage in new areas of regulatory cooperation without the need to re-open the initial international agreement or to modify each other’s institutional framework (Alemanno, Reference Alemanno2015: 631–632). The implementation of future regulatory convergence may thus take place outside existing regulatory agencies related to sovereign states or the EU. Be it the Committee on Regulatory Coherence imagined for CPTPP or the Regulatory Cooperation Forum established by the CETA, this raises significant concerns when the time comes to define in more detail the membership, scope, and functioning of the bodies established for the purpose of such on-going regulatory cooperation. Two issues stand out in particular. First, regarding membership, considering the influence that such bodies may have on future regulation, a fair and balanced representation in defence of the public interest is particularly important. At first sight, CETA, as a forerunner of potential future agreements, appears unambiguous in this regard. It holds that the Regulatory Cooperation Forum ‘shall comprise relevant officials of each Party’, i.e. from regulatory authorities (art. 21.6.3). Yet it directly adds that ‘other interested parties to participate in the meetings’ may as well be invited by mutual consent. While this is clearly consistent with consultations with private parties such as representatives from academia, think-tanks, NGOs, businesses, and consumer and other organisations (article 28.8), here again it leaves the door wide open to distorted lobbying practices as long as it does not include provisions and provide public support to make up for the over-representation of well-resourced business organisations (with less detailed language, CPTPP articles 25.6 and 25.8 raise similar concerns). The second issue deals with the mandatory or voluntary nature of such cooperation. Taking again the case of CETA as the only agreement ratified so far in this domain, it holds that regulatory cooperation activities are undertaken ‘on a voluntary basis’, but requires parties to provide explicit explanations in case it refuses to initiate regulatory cooperation or withdraws from cooperation (article 21.2.6). Such a burden of justification against the voluntary principle was seen by the Wallonian Parliament as crucial enough in the course of its contentious ratification process that it insisted on including the following plain language in the Joint Interpretative Instrument added to the signature of the agreement: ‘regulatory authorities can cooperate on a voluntary basis but do not have an obligation to do so, or to apply the outcome of their cooperation’ (Council of the European Union, 2016, §3). Without such plain language, there would indeed be more of a place for imposing ever increasing areas for convergence to become legally binding.

A third and much-discussed issue is the potential outcome of a greater mutual recognition of existing standards. As Vogel (Reference Vogel and Vogel1995) forcefully argued more than two decades ago, increased economic integration is not necessarily incompatible with stronger regulation and standards in domains such as labour, environment, and consumer protection. Yet, as seen earlier, mutual recognition is more likely to lead to a race to the bottom than to the top, as regulations are rarely completely equivalent in terms of outcome. This might also be the case with services such as education and training, engineering, architecture, electronic communications, transport, legal services, and so on. In contrast to goods, regulations and standards for services are often more decentralised and set by sub-state or non-state bodies such as professional association or private entities – and therefore more difficult to compare. Moreover, service regulations and standards rarely concern the service itself. In order to respond to no end of quality and security issues prompting market uncertainties, they are more likely to define conditions that service providers are expected to fulfil (professional qualifications, etc.) or the circumstances of the services’ delivery (opening hours, location, safeguarding of public services, etc.). As a result, they tend to be more diversified and complex than those pertaining to goods, making their equivalence even more difficult to assess in mutual recognition procedures. Think of someone who completed a professional degree in country A and then moves to country B to take up a job, where she finds that authorities of country B don’t recognise her diploma and ask her to pursue two additional years of study before she could apply for the job which she could have directly taken up in her own country.

At the same time, as service regulation and standards concern the process and, as such, are not necessarily reflected in the actual ‘content’ of a service, they may well be seen to have less effect in the importing country. This explains why importing hormone beef raises more concerns than software programmed in hazardous conditions, even though the latter could breach key information security and protection of privacy requirements. As Hatzopoulas points out, service regulation and standards will ‘meet much lower resistance from consumers in the host State – and therefore local […] rules will be under greater pressure from regulatory competition. If the host State is to safeguard its own standards … it needs specifically to legislate acts of an essentially protective nature. Such rules are unlikely to yield to the effect of mutual recognition’ (Hatzopoulos, Reference Hatzopoulos, Lianos and Odudu2012: 63). In brief, even if we accept the oft-repeated discourse that no provision in mega-preferential trade agreements under negotiation would whatsoever lower existing levels of protection, extending the principle of mutual recognition to service regulations and standards such as professional qualifications (e.g. CETA chapter 11), licensing requirement, and approvals procedures (e.g. CETA chapter 12) is at best intricate and at worst might well be detrimental. At the time of writing, the present state of pending or discontinued negotiations and the types of provisions included in the negotiations achieved with CETA do not allow for clear conclusions.

A final point concerns the implications of mutual recognition of existing regulations and standards for third countries not part of the preferential trade agreement. Irrespective of provisions agreed or under negotiation, preferential trade agreements must comply with the rules established by the WTO. Legal scholars concur that, while the WTO framework is not entirely clear, or coherent, it provides a rather open understanding of how recognition agreements should avoid discriminatory implications for third-parties (Trachtman, Reference Trachtman2003; Nicolaidis and Shaffer, Reference Nicolaidis and Shaffer2005). Regarding services, GATS article VII.2 sets out that States ‘shall afford adequate opportunity for other interested [States] to negotiate their accession to such an agreement or arrangement or to negotiate comparable ones with it’. In its attempt to respond to the fundamental principle of most-favoured-nation (MFN), on which the whole architecture of non-discrimination rests, GATS article VII.2 clearly does not prompt any automatic extension of mutual recognition agreements to third parties. Yet, by means of procedures of notification and ensuing participation to negotiations under way, Mathis (Reference Mathis2012: 72) points out that ‘MFN plays at least a ‘conditional’ role to assess the potential participation of third parties.’ Here again, such principles may be more difficult to realise with services than with goods. As recognition in the domain of services mainly concerns professional qualifications, licensing requirements, and approval procedures, they are more likely to be granted to individuals and firms on a one-by-one basis – in contrast to products, whose conformity assessment is more likely to be valid to all the same products put on the market. Accordingly, it does not require in-depth legal expertise of the provisions negotiated in the context of the new generation of preferential trade agreements to realise that parties who will not be part of the mutual recognition provisions designed for services in such agreements should not expect many spill-over effects for their own benefits, notwithstanding the relatively open WTO framework on mutual recognition of regulations and standards towards third-parties. Ultimately, as VanDuzer (Reference VanDuzer2012) points out, the implementation of such intricate provisions eventually depends on local contexts as well, with actors such as domestic bodies, regulators, and sectoral experts being the real players engaged in the process. While this may add additional uncertainty to rules already identified as unclear and not always coherent, the discretionary power of local agents emphasises the ambiguous authority on which the recognition of standards and regulation rests against the background of the new generation of free trade agreements. Such ambiguity applies not just to third countries not part of those new mega-trade deals. As seen previously, it also supports the regulatory chill effect they may have on participating governments, their mechanisms of regulatory cooperation, and a mutual recognition of existing standards based on a loose understanding of equivalence.

Service Standards and Institutional Ambivalences

The following discussion focuses on how the aforementioned developments matter in assessing the authority of international standards in the service sector along the three core dimensions of my analytical framework, i.e. the institutional continuum of the actors involved, the material continuum of the issues concerned, and the spatial continuum along which such standardisation processes are likely to be recognised across jurisdictions.

Unsurprisingly, public and private actors very much overlap in the standardisation arenas on both sides of the Atlantic, as well as on the international plane of the ISO system and preferential trade agreements. There is also strong evidence of significant public support, in particular within European institutions. However, the limited results of initiatives taken over the years shed light on a common feature on both sides of the Atlantic as well as within the ISO and the context of the new generation of preferential trade agreements: the support and expertise of private actors is crucial in the development of standards. The low level of involvement in the field of service standards in the United States mirrors the difficulty of European and ISO projects in convincing stakeholders from the private sector. This suggests that behind labels of ‘direct participation’ in the United States and ‘national delegation’ for the European and ISO setting, actors setting standards are the same: large firms dominate technical committees, with government agencies attempting in some cases to take part in drafting standards, and not-for-profit associations from civil society remain largely under-represented.Footnote ²⁶ The entry into force of EU Regulations 1025/2012 introduced new processes that improved the monitoring and participation of stakeholder organisations representing consumers’, workers’, SMEs’, and environmental interests. Yet, the first evaluation undertaken under those new commitments points out, euphemistically, that such participation ‘is still challenged’, notably because of their weak position and different capacities in terms of stakeholders’ representation at national level, as well as a lack of inclusiveness at the international level when standards are jointly drafted with ISO or IEC in the lead (European Commission, 2016d: 4, 11–12). In contrast to the direct political influence of the European setting, the American system relies on the indirect influence of the legal and regulatory environment supporting and legitimising the output of formal and informal SDOs. Thus, far from mere fragmentation, the US system hinges upon double coordination mechanisms, ensured by ANSI at the level of formal SDOs and by NIST with regard to governmental agencies. The distinction between national delegation and direct participation therefore appears to be more relevant for describing the space of standards recognition outside the United States than the type of actors involved within the United States. Finally, the difficulties experienced by the European attempts to foster the development of standards in the field of services show how the enrolment of private actors can become an important political issue. The lack of distinct service standardisation processes in the United States here echoes the difficulty in encouraging stakeholder involvement in European projects to develop standards in conjunction with the unification of the market for services. Apparently, European officials have greater influence on the issues put on the agenda than do private actors likely to shore up the processes of setting new service standards. Defining the membership, the scope, and the functioning of the bodies established for on-going regulatory cooperation in the new generation of preferential trade agreements raises the same concerns.

Regarding the objects concerned, the potential scope of international standardisation in the domain of services differs greatly across the Atlantic and beyond. The antagonism between horizontal and vertical standards reflects the struggles at stake in defining what should be standardised in services: should it be the functional attributes of technical interfaces supporting the interaction between providers and customers on a horizontal basis for the widest range of services (information requirements, billing, complaint handling, etc.)? Or should technical specifications be more substantial on a narrower sectoral basis, defining how services can be co-produced and used on a reliable basis with shared expectations regarding their quality? Services’ distinctiveness is clearly at stake here, with an assumption that the more intangible and relational the service is, the more difficult to measure, qualify, and standardise. Yet this does not mean impossible. In spite of all their flaws, European initiatives have helped build a coherent framework for the standardisation of services. The ‘hybrid combination’ imagined by the European Commission for the development of horizontal service standards with sectoral add-ons, or for a pool of parallel sector-specific standards, may eventually overcome the controversy between vertical and horizontal service standards. Even the more shallow horizontal approach may gauge the quality of services, Standards on performance measurement, service contracts, and service procurement expected from the Technical Committee (CEN/TC 447) established in 2016 could provide evidence of the positive impact of a standard on consumers. Moreover, it is worth noting that the case of energy and smart metering suggests that societal issues are likely to be greater in Europe than in the United States, where the focus is on narrower technical and market-driven aspects. While both sides demonstrate interest, American stakeholders narrow it down to technical issues associated with the physical characteristics of the resources delivered by such services. In contrast, European initiatives explicitly point out broader concerns of sustainable development, notably in relation to the implementation of the EU Directive on energy efficiency (2012/27/EU). Finally, this concerns how new standards and mutual recognition of existing ones are likely to lead to a race to the top. We saw that transatlantic and transpacific promises to set new global standards are greatly exaggerated. We should rather double-check the implications of extending the principle of mutual recognition to service regulations and standards such as professional qualifications, licensing requirements, and approval procedures. With the new generation of mega-trade deals still a moving target at the time of writing, the analysis can only be tentative. It shows, however, that such effects are at best intricate, but may well be detrimental.

This brings us to the third dimension defining the transnational hybrid authority of standards: the extent of the space in which technical specifications in the domain of services are likely to be defined, distributed, and recognised across sovereign States. International standards compete in terms of their different sources of legitimacy, as well as their various modes of cooperation. Market adoption is the main source of legitimacy for standards developed by American SDOs. This means that the recognition of standards beyond the sovereign space of the United States primarily relies on the exogenous process of market mechanisms – a good entry point into new markets as expressed by one interviewee. The translation of standards into official languages of various countries and the organisation of training workshops tailored to the distinct needs of well-chosen countries are an integral part of this strategy. This does not mean, however, that American SDOs overlook the legitimacy of their standards based on direct participation. The ASTM Memoranda of Understanding signed by more than one hundred national standards bodies strongly echo the principle of national delegation in use at the CEN and ISO, even if they are part of a contractual and bilateral strategy. In contrast, the legitimacy of standards in the ISO setting outside the United States, particularly in Europe, endorses the principle of national delegation. The diffusion and adoption of standards is consistent with the endogenous logic of territorial sovereignty. However, EU plans in the domain of services may lead to a dual model, in which direct participation would complement the national delegation model. This was thoroughly discussed in the consultation process preceding the adoption of the reform of the European standardisation system (Regulation 1025/2012). However, one should be aware that this would rely more upon the involvement of stakeholders within the European context than the broadening of standard recognition beyond the confines of the EU. This falls short of defining a dedicated procedure for setting future service standards. Finally, the new generation of preferential trade agreements clearly impacts on the spatial continuum along which the power of standards is likely to be recognised. Beyond the implausible prospect of new harmonised standards and the extent of mutual recognition of existing standards and regulation confined to states who take part to the agreement, we saw that parties who will not be part of the agreement will have difficulty in gaining such recognition for themselves. This applies even more in the domain of services, despite the relatively open WTO framework on mutual recognition of regulations and standards towards third parties.

* * *

This chapter has examined the authority of standards within the broader regulatory environment of capitalism by studying the case of service standards in the context of the ISO, European, and American standardisation systems, as well as the prospects of the new generation of preferential trade agreements such as CETA and CPTPP. It showed the intricate and manifold ways in which the ambiguity of the world of standards supports its power across institutional specificities. This not only goes against the view of a compelling transatlantic divide. It also calls for mitigating speculations on the prospects of current and future mega-trade agreements against the setbacks initiated by the Trump administration. With or without deals, the ambiguity on which the authority of standards feeds the regulatory environment of capitalism is here to stay. Nevertheless, over the last few years, developments in service standards have been weaker than expected. The special case of services is a first explanation. American practitioners tend to deny the distinctiveness of service standards per se, while in the European and ISO contexts on-going struggles take place to define what exactly this category may mean and why it would need dedicated procedures likely to better support the development of service standards. An alternative explanation may be that inferring a weak development of service standards reflects a fallacy of composition, as many international standards are developed elsewhere, whether or not tagged ‘service-related’. This sets the agenda for examining other ambiguous and neglected aspects of the transnational hybrid authority of standards. I will begin with standards for the insurance industry, which are, as we will soon see, among the farthest from the standardisation system as usually conceived within ISO and European arenas.