The Test of Tertiarisation

The extension of the tertiary sector is one of the most striking aspects of the shift in recent decades towards a so-called knowledge-based global economy. Services now account for approximately 75 per cent of GDP and employment in the advanced economies of the OECD, and more than 50 per cent in developing countries and emerging economies. As we have seen, services raise contested issues pertaining to employment, trade, and foreign investment. While they have exhibited the strongest growth in the global economy for many years, increasing doubts are emerging around the continuity and sustainability of this movement. This leads us to consider with renewed caution assertions made over the last twenty-five years on the importance of this phenomenon and the transformation it implies for the transnational regulation of global capitalism. Karpik (Reference Karpik1989) associated the shift towards services as part of a new ‘economy of quality and singularities’. In the same vein, Castells (Reference Castells2001: 56) identified the advent of a ‘service society’ as a radical shift towards ‘informational capitalism’. And according to Gadrey (Reference Gadrey2010), an ‘economy of quality, service and knowledge’ calls for the construction of a whole new ‘economy of care’. The rise of services directed towards the immediate and tailor-made satisfaction of consumer needs is, the argument goes, most certainly going to give rise to ‘new forms of competition’ (Petit, Reference Petit, Du Tertre and Laurent2008). Finally, some see the valorisation of knowledge in the service economy as marking nothing other than a new accumulation regime: ‘cognitive capitalism’ (Moulier-Boutang, Reference Moulier-Boutang2007).

The first step needed to stand back from the spin generated around the ‘new service economy’ is to look at what the terms ‘services’ and ‘tertiary sector’ really mean. What are services? This is not a trivial question. Even before initial negotiations attempting to establish a regulatory framework within which to liberalise trade in services had begun in the mid-1980s, the renowned British weekly The Economist had come up with the following oft-cited formula: services are ‘things which can be bought and sold but which you cannot drop on your feet’ (The Economist, 1985). Such a definition does not, of course, do full justice to a plethora of activities difficult to pin down on this basis. Still, it catches in a nutshell what bodies responsible for producing and harmonising international statistics have taken years to define in order to reach a comprehensive approach to statistical issues entailing services, guiding negotiations, and supporting implementation of international agreements encompassing services such as the General Agreement on Trade in Services (GATS). This was in particular the task of the new United Nations Manual on Statistics of International Trade in Services, first published in 2010. The manual restates the usual cautionary note on the heterogeneous range of service activities difficult to encapsulate within a simple definition and remains all the more true given the difficulty in separating services from the goods with which they may be associated in varying degrees.Footnote ² That being said, it defines services as follows: ‘the result of a production activity that changes the conditions of the consuming units, or facilitates the exchange of products or financial assets’ (United Nations, 2010: 8).

A prime interest of this apparently simple definition is to extend the meaning of international trade in services to encompass various modes of supply included in the GATS, in particular the supply of services through the commercial presence of affiliates established abroad. More substantially, it allows for flexibility in responding to opposing conceptions of service activities. Most frequently cited characteristics of services are intangibility (or immateriality such as in teaching or health), co-production (or the relational dimension between a customer defining its needs and a consultancy firm offering a response), heterogeneity (or the idiosyncratic dimension of services ever customised according to clients’ needs and thus supposedly avoiding standardisation), and perishability (that is, the impossibility to stock services in an inventory like goods) (Millar and Choi, Reference Millar and Choi2011: 28). The UN definition clearly abandons the criteria of immateriality and intangibility, which for two centuries had stood at the heart of classical and neoclassical economics. Instead of seeing services as what Adam Smith and his followers saw as residual, non-productive, and immaterial activities, characteristic of non-productive labour, the definition echoes the diversity of service activities and their increasing integration with the production of physical goods. It is from this perspective that it draws on Hill’s seminal contribution (Hill, Reference Hill1977). In contrast to neoclassical analyses, Hill put the focus on the processes involved in the transformation of the state of an individual or object and the relational issues involved when the activity implies co-production between the producer and the consumer of the service. From this focus on changes in the condition of a person or an object, the impossibility of stocking services has nothing to do with a physical impossibility supposedly derived from immateriality; it is, rather, a ‘logical impossibility[, since] a stock of changes is a contradiction in terms’ (Hill, Reference Hill1977: 319). It follows that, as services are exchanged without transfer of ownership in contrast to goods, ‘models of pure exchange economics of a Walrasian type in which existing goods are traded between economic units are quite inapplicable and irrelevant to services’ (Hill, Reference Hill1977: 318). This is why a medical service is not some kind of immaterial drug, a training course some kind of immaterial brain, or a life insurance policy some kind of immaterial grief. Hill’s definition has been widely recognised as the most suitable for both research and statistical purposes. It allows us to stick to as simple a definition as possible for our own purpose. Suffice here to mention a last point, made by Gadrey (Reference Gadrey2003: 18) in order to account for a greater variety of demand rationales characterising services. According to this view, services are deeply embedded in social institutions. They span two axes. The first concerns the ‘type of capacities’ with which the user mainly interacts. It opposes technical capacities (such as the automatic teller machine providing you with cash) to human capacities (such as those displayed by a discrete Swiss banker advising you in a cosy meeting room on how to park cash and dodge tax authorities). The second axe is about the ‘mode of request’ chosen by the user to get a service delivered. It opposes interactive requests (i.e. to ‘be served’ by a service provider such as a customer-relations officer based in a call centre in Bangalore) to non-interactive requests (i.e. to ‘serve yourself’, when you use an IT platform, pay for a nice trip on your own, or do your own cleaning).

This said, there is little disagreement about which activities account for services or not, but much more on how to establish an explicit and comprehensive definition (Illeris, Reference Illeris, Bryson and Daniels2007: 24). We accept Bryson and Daniels’ idea that ‘it is important not to become too distracted by the search for a precise definition of services’ (Bryson and Daniels, Reference Bryson, Daniels, Bryson and Daniel2007a: 4). Even more important is the extent to which a service economy is in fact inextricably linked to the manufacturing industry. In his classic work, Daniel Bell (Reference Bell1973) saw the dawning of the post-industrial era as an age in which services would replace manufactured goods as a growth engine. Consolidation of service activities in economies the world over has not, in fact, led to a major change in traditional industrial production. The situation is not that of a closed system in which an increasing share of service activities results in a corresponding decline in manufacturing or agricultural production. On the contrary, production systems are increasingly interdependent between manufacturing industry and services. Two obvious examples present themselves. The smartphone, which has made Apple the world’s most valuable company and the first-ever company to reach a value of $1,000 billion on the US stock market, is as much a product manufactured by the millions with cheap labour in Foxconn’s Chinese factories as a product of branding and design services praising its trendy and friendly way to be connected to the latest ‘killer app’ in town. For far longer, lift companies have based their profits on a similar overlap between manufacturing and services: much of their added value comes not from the production of lifts, but from service contracts, especially when these are mandatory, as is the case in most developed economies. More generally, services play a core support role in integrated production networks, which rely on transport and logistics, and complex financial and insurance products. The symbiosis of all business services – themselves heavily reliant on information and communication technologies supporting knowledge creation, innovation, and timely processing and dissemination of information – with manufactured goods has become a preponderant reality.

The major change in the structure of our economies lies in the fact that service activities now constitute a fundamental, but not exclusive, dimension of economic activity. The debate on the ‘tertiarisation of industry’ and the ‘industrialisation of the tertiary sector’ (Rubalcaba, Reference Rubalcaba2007) has given way to the notion of ‘servicification’ or ‘servitization’, where services and manufactured goods should be approached as mutually interdependent, as they are increasingly traded as a package (Kommerskollegium, 2012; Cernat and Kutlina-Dimitrova, Reference Cernat and Kutlina-Dimitrova2014). With high-value added services such as design, R&D, and architectural and engineering services concerned, there is clear evidence of the difficulty involved in dealing with a phenomenon which involves both interdependency and integration. Just as the production of manufactured goods can no longer be envisaged independently of all the supporting services upstream and downstream the production chain, industrial and commercial logic is contributing to the increased industrialisation of services. Before Apple’s smartphones even existed, Boden and Miles pointed out: ‘The service economy is not merely an economy in which service sectors are quantitatively dominant. It is one where “service” is becoming a guiding principle throughout the economy’ (Boden and Miles, Reference Boden, Miles, Boden and Miles2000a: 258). This is undoubtedly why Bryson and Daniels allude to ‘a complex process of hybridisation whereby the categories of manufacturing and services are becoming increasingly blurred’ (Daniels and Bryson, Reference Daniels and Bryson2002; Bryson and Daniels, Reference Bryson, Daniels, Bryson and Daniel2007a: 7). A decade later, ISO’s presentation of its strategy for services explicitly refers to the ‘“hybridization” of companies’ offering … [as a reason] to look more closely at evolving market needs in the service sector’ (International Organization for Standardization, 2016b: 8).

The integration of service activities into the economy as a whole assumes that this can be achieved in parallel with the upsurge of the international division of labour over the last two decades or so. In a context characterised by a globalisation of production networks and a common use of information and communication technologies (ICTs), the internationalisation of services has thus become a major issue. Traditionally, the phenomenon of internationalisation was confined to physical goods. Now, with sufficient infrastructure, huge quantities of data, information, and digitised documentation can be exchanged instantly, at minimal cost, anywhere in the world. As we will see in further detail in the following section, the development of this infrastructure has profoundly changed the tradability of services across borders whilst imparting major significance to the notion of knowledge within the service economy. Delivering services and coordinating knowledge on a worldwide scale also assumes increasing movements of factors of production, including the circulation of individuals, financial flows, and direct investments in cross-border trade in services. This has now taken on a new dimension with the massive recourse to big data, cloud computing, the Internet of things, and platforms which use hardware and software resources to deliver all sorts of services. To a certain extent, the Holy Grail would be to take the internationalisation of services full circle; it would abolish the factors of time and space hitherto seen at the heart of a service relation, which in turn often justifies a distinct regulatory environment usually viewed as hindering internationalisation.

However revolutionary the technological shift of the cloud is and notwithstanding the localisation and materialisation of data-farms, the global and timely delivery of services still depends on an architecture of legislation, regulations, and standards which plays a determining role, mandatory as well as voluntary, in establishing the conditions of access to the market. The instruments devised to regulate the internationalisation of services are highly diverse, ranging from intergovernmental organisations, and supervisory and standards bodies to bilateral and multilateral framework agreements, as well as ad hoc stakeholder conferences and roundtables, informal institutional platforms, and industrial and/or non-profit consortia. I shall examine this in full detail later in the chapter. For the time being, suffice it here to emphasise that in one way or another, all of them run up against the difficulty raised previously: how to establish commonly accepted criteria to specify the expected characteristics of a service and define a benchmark against which conformity to a promised service may be judged. Despite an undoubtedly fragmented environment across nations and industries, the issue generally remains the same: define the quality of the service.

In classical and neoclassical market theory, quality is not disputed by the agents, who are held to have the same representation of the item being traded – a representation founded on the supposedly complete information provided by the price signal. Scholars of asymmetries of information have developed a fresh view on the notion of quality as an independent and determining variable in the markets, distinct from self-regulatory markets based on price information. In what has become a classic article of economics, Akerlof (Reference Akerlof1970) showed the fundamental information asymmetry characterising a market using the example of used cars, in which the seller has information about the goods which the buyer does not possess. Such asymmetry prevents the market for physical goods working properly; in addition to ‘brand names’ and ‘state licences’, ‘guarantees’ on product quality are identified as ‘counteracting institutions’ likely to help rectify this (Akerlof, Reference Akerlof1970: 549–550).Footnote ³ George Akerlof, Michael Spence, and Joseph Stiglitz would later receive a joint Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel for this demonstration. They show the extremely narrow condition of validity of the fundamental hypothesis of neoclassical Walrasian economics, only valid if all goods and services exchanged have a homogenous quality, perfectly defined and of common knowledge to all agents. A further hypothesis makes inroads into uncertainty about quality, shared by all agents, particularly with regard to products whose past or – more importantly in a context characterised by the emergence of new markets – future is unknown (Hirschman, Reference Hirschman1970; Lupton, Reference Lupton2005).Footnote ⁴ Some management research analyses this hypothesis in greater depth by drawing a distinction between quality based on personalisation and that based on industrialisation. In the first case, the emphasis is on customer satisfaction, while the second addresses issues such as the reliability of processes of production and service provision (Deming, Reference Deming1981; Sundbo, Reference Sundbo2002). Working from the idea that a service-based relationship involves co-production between the provider and the beneficiary, the analysis focuses on the additional uncertainty resulting from the co-incidence of these two types of quality (Parasuraman et al., Reference Parasuraman, Zeithaml and Berry1985; Grönroos, Reference Grönroos1990; Johnson and Nilsson, Reference Johnson and Nilsson2003). The salient point here is the degree to which quality is always a bone of contention, whose ambiguous status lends itself to be tweaked in various ways and subject to controversy.

French régulation theory has provided several analytical tools for these questions. One of them is that quality uncertainty calls for a specific mode of regulation, distinct from price. While Fordism put price and volume at the core of the mass production and consumption of uniform standard goods, the economic focus now considers quality as a prime form of competition. In order to respond to this puzzle, quality can be conceived as an institution in the sense of the institutional economics of John Commons (Allaire and Lemeilleur, Reference Allaire and Lemeilleur2014). Cautious as ever on the ‘uncertainty of meaning of the word institution’, Commons’ definition is as simple as it is far-reaching: an institution is ‘Collective Action in Control of Individual Action’ (Commons, Reference Commons1934: 69). In sharp contrast to neoclassical economics focused on rational individuals isolated in a state of nature, the individual with whom Commons is dealing is thus an ‘Institutionalized Mind’ (Commons, Reference Commons1934: 73). The quality of a good or a service therefore can neither derive from a price signal nor from any intrinsic attribute of such good or service. On the contrary, it should be viewed as a social construct stemming from power relations between private and public actors who pursue their ever-evolving interests. This view of quality as an institution with explicit and codified procedures at the crossroads of power and interests calls for conventions fixing implicit anticipations and coordination expectations. This is clearly more than just providing information, which can later be passed on to the consumer via a nice label. It stems from complex negotiations, through which a series of institutional forms, regulation agreements, conventions, and standards constitute and situate the qualitative attributes of a given product or service. The actors involved in this process struggle to impose a concept of quality that, following Commons’ definition, allows for collective action to control the individual action of agents involved in economic transactions. This prompts Chanteau (Reference Chanteau2011: §24) to describe standardisation in the field of quality as a ‘total social fact, in which exchanges of goods and signs, as well as ways to control individual behaviours play out in such a way as economic and political facts become indivisible’.

This discussion of quality is not specifically focused on services, though. In contrast, Callon et al. (Reference Callon, Méadel and Rabeharisoa2002) put forward a semantic shift from quality to qualification that opens up major implications for analysing service activities. While régulation scholars draw on Commons to emphasise the power relations involved in the institution of quality implemented by third-party certification or participatory guarantee systems, the concept of qualification describes power issues following the idea of product differentiation first developed by Chamberlin (Reference Chamberlin1925) in his Theory of Monopolistic Competition. Chamberlin emphasised the growing importance of the ability of sellers to differentiate products from one another; furthermore, he considered that the ability to modify the list of qualities of a product should be seen as a strategic resource for firms. According to Callon et al., this process of qualification/requalification is at the heart of the practices involved in the service industries. As service activities are viewed as the basis of the economy of qualities, they depend on differentiated modes of regulation. Markets for services are often based on the expectation that consumers actively participate in the re/qualification of products. Their regulation should accordingly counterbalance the power position given to the seller by the possibility of monitoring users, observing their preferences, and targeting the products offered to them. According to the authors, nowhere is this biased collaboration between supply and demand more obvious than on the Internet. Whilst the authors wrote at the beginning of ICT-supported singularisation of products, Internet-supported services are now part of our daily life. Data ownership and privacy protection have become prominent issues for all sorts of applications used on smartphones, such as Google map, Uber, Booking.com, Amazon, and the like – all adapting their offer to your own, and unlike any other consumer’s history. This new mediation of the consumer in the trust and judgement-based tools of services has clearly demystified the notion of ‘quality of service’ and made it possible to envisage changes in the many and various mechanisms used to judge, evaluate, sanction, and grant trust. The cognitive resources of the users are mobilised in such a way as to engage debates previously monopolised by scientists in economics and, to some extent, other social sciences. As Callon et al. (Reference Callon, Méadel and Rabeharisoa2002: 96–97) note, qualifying products in a service economy is likely to provoke ‘a profound transformation of the rules by which markets function … The organisation of markets becomes a collective issue and the economy becomes (again) political’. As qualification practices are assumed to generate disputes of all kinds as to the nature and level of trust required in the market transactions of services, they regard ‘hybrid fora’ as promising avenues for debates on the organisation of such markets. As we saw in the previous chapter, hybrid fora are arena that mix expert and lay knowledge for debating technical choices involving the wider public (Callon et al., Reference Callon, Lascoumes and Barthe2001); here the technology is less about ‘quasi-objects’ that bracket nature with culture, than the qualification devices designed to build trust between service providers and consumers. While this might be conceivable at the local level of communities, it is much harder to imagine at the global level that service offshoring has reached today. Before discussing the relationship among qualification, standards, and regulation, this prompts us to examine in more detail the issues at stake with the internationalisation of services.

The 75/25 Puzzle

In the introduction to this chapter we saw how forecasts of the overall number of jobs threatened by the ability to relocate services abroad caused widespread alarm in the early years of the 2000s. More than a decade has passed since the heyday of globalisation narratives that fed such expectations. In the meantime, the economy has been struck by a global crisis whose extent was only rivalled in the 1930s. Job losses for the less skilled and, increasingly, the middle classes in rich countries have remained a core issue. While part of the onus falls on manufacturing, services are seen as bearing the brunt of the expense. Big firms led the way, using the Internet to offshore information technology, back-office work, and even much more complex tasks to places such as India, Morocco, the Philippines, and Fiji. Smaller firms soon followed suit. The scope of service activities identified as ‘tradable’ across borders has widened as well. The Economist (2013), widely known for its liberal views (in the UK sense), recognises that ‘offshoring has brought huge economic benefits, but at a heavy political price’. At the same time, services are still seen as remaining inherently difficult to internationalise. Companies have begun to reconsider what many saw as common sense. In contrast to offshoring – a relocation of activities abroad, mainly for cost purposes – the new game in town, in particular in the United States, is reshoring – a relocation of activities at home, not only for quality purposes but also with increasingly competitive costs. Against this background, let us take a closer look at what is at stake, first by examining existing data on the internationalisation of services.

Services now account for approximately 75 per cent of GDP and employment in the advanced economies of the OECD, and more than 50 per cent in developing countries and emerging economies. Nevertheless, we are still far from a globally integrated supply of services. Sectorial coverage remains narrow and apparently no upsurge of total trade in services has taken place in the last two decades. As Tables 3.1 and 3.2 show, services continue to represent less than 25 per cent of world trade. What has happened is a significant ‘shift of composition’ between developed and developing countries. Over twenty years, developing countries have almost doubled their share to reach around 30 per cent since 2012. This share of services in world trade, under 25 per cent, remains in stark contrast to the 75 per cent of GDP and employment in the advanced economies of the OECD. This is what I call the 75/25 puzzle of the internationalisation of services.

Only a portion of those activities involves international trade transactions per se, with tasks outsourced according to an arm’s length principle. A significant number take place within the network of multinational firms, between parent companies and their affiliates, described in the jargon as ‘captives’. This prompts us to pay additional attention to foreign direct investments (FDI). At first sight, the overall share of services in FDI has not changed so much either. Figures presented with some caveats by UNCTAD a few years ago (Table 3.3) show that their share increased by less than 10 per cent over twenty years, of which trading and finance account for more than half. However, here again it is worth noting an important shift in composition: while developing countries accounted for less than 20 per cent of all FDI inward flows in services in 1990–1992, that figure reaches more than 40 per cent twenty years later, with the share of business services having almost doubled. Consulting, accounting, auditing, data management, customer relation centres, all belong to these new types of business services easily established in developing countries and attracting massive volumes of foreign direct investments. Interestingly, sectors such as health or education, despite often making headlines, remain marginal in comparison, with worldwide inflows of $391 and $814 million, respectively, in 2009–2011. Certainly data on services, let alone data on their international trade, are notoriously complex to come by. The figures are thus likely to remain disputable, as data classification and accessibility remain elusive.

A joint OECD/WTO initiative has attempted to address this issue by producing data disaggregated by the value added in the exchange of goods and services consumed worldwide. According to the methodological note accompanying the publication of the statistics, ‘the break-up of domestic content by direct and indirect sectoral value added reveals that a large chunk of the value originates indirectly from service sectors’ (OECD-WTO, 2013: 11). Well aware that international trade remains dominated by goods even while a large share of global GDP and employment accrues to services, the new metrics produced by the OECD/WTO initiative clearly shed additional light on the significance of services (see Table 3.4). While these estimates are dated, they show that the service sector contributes around 50 per cent of total exports from countries such as the United States, the United Kingdom, and France, with the whole European Union, Japan, and India scoring in the same average, and even close to one-third in the case of China, often seen at the bottom end of low-cost manufacturing value chains. The metrics also provide some estimates of the service content of overall manufactured goods (cf. the aforementioned discussion regarding ‘servitization’). Here too, the figure is significant, with an average of one-third in the reference year of 2009, corresponding to an increase of between 5 and 10 per cent in many countries since 1995.Footnote ⁵

Against this background, it is quite obvious that the offshoring of services is a powerful and significant phenomenon of contemporary capitalism. The shift began in the 1980s with outsourcing contracts in data processing and call centres at the bottom of the value chain. Today, it has moved into much more advanced sectors, with activities such as legal, fiscal, or medical services, financial consulting, and all sorts of business services enabled by information technology, from the entertainment industry to security-related activities. Although the development was still embryonic then, UNCTAD emphasised its significance more than ten years ago: ‘the cutting edge of the global shift in production activity [gives] rise to a new international division of labour in the production of services’ (UNCTAD, 2004: xxv). While firms have embarked on offshoring in many areas, at the same time they have become increasingly aware of the difficulties to be overcome.

The evidence established so far lends us to consider that the jury is still out concerning the rise of service offshoring. In contrast to largely inflated estimates made in the heyday of expanding globalisation, the 25/75 puzzle has not completely disappeared from the picture. The ratio of trade and foreign investment to overall employment and GDP figures in services has remained relatively stable over the last quarter century. There is, however, a significant shift in composition, as developing and emergent countries have more than doubled their share in trade and inward foreign investment – even if the recent development of re-shoring may suggest that this share has reached a tipping point in some domains. Moreover, there is ample evidence of greater diversification, as the offshoring of services has shown a propensity to climb the value chain, especially in sophisticated business services. Finally, compiling a thorough inventory of the internationalisation of services remains plagued with difficulties. At the same time, the integration of services deep into manufacturing instils a significant service-content of exported manufactured goods. How shall we explain, then, the expectations and on-going difficulties with the expansion of services across borders? Moreover, to what extent do these developments prompt new expectations on the role of standards in contemporary capitalism? This is what the following sections examine.

Restrictive versus Extensive Hypotheses

Many services were long considered non-tradable since they required buyers and sellers to be in the same place at the same time. A radiologist was thus supposed to be located in the hospital where the patient had her/his x-ray done. Similarly, an insurer relied on the knowledge of local agents to assess risks of local firms. ICT has removed many of these constraints and made services much more tradable. So-called digitability is the foremost driver of service offshoring. Not only can all kinds of information be digitally stored and directly exchanged almost anywhere in the world, but computer technology also allows knowledge to be digitised, codified, fragmented, and re-organised in such a way that the production process can be spread across the globe in complex, disparate, and far-away locations.

While ICT is both a basic infrastructure and a sophisticated knowledge-intensive tool of service offshoring, the second chief driver is the quest for cheap labour costs. Management textbooks make that point time and again (McIvor, Reference McIvor2010; Oshri et al., Reference Oshri, Kotlarsky and Willcocks2015). We have seen how controversial this can be in terms of jobs losses and flexibility for middle-class workers in industrialised countries, as well as low-paid and exploitative jobs in developing and emerging economies. Conversely, the narrowing of labour cost differentials between industrialised countries and developing and emerging economies accounts for a large part of the ‘reshoring’ of services to the United States (The Economist, 2013). Labour costs thus remain a core dimension in the arbitrage of going or not going offshore. Implications in terms of wage erosion, shifts in bargaining power detrimental to labour, and redeployment difficulties towards higher value-added jobs have not, by a long way, disappeared.Footnote ⁶ The ability of service providers and consumers to move is a third factor to take into account. Some services rely heavily on the mobility of experts or basic workers (what the GATS defines as ‘mode 4 – movement of natural persons’). This occurs in the many instances when foreign nationals provide services abroad either as independent suppliers (for instance, accountants) or as employees of a firm (for instance, a construction company). Similarly, services such as education, tourism, and health treatments rely on the mobility of consumers (‘mode 2’ of an international exchange of services, according to the GATS definition). Moreover, the mobility of service providers depends in many instance on legal and regulatory provisions that set out the conditions for affiliates and subsidiaries to establish a commercial presence (‘mode 3’ of the GATS, relating to foreign direct investment and other forms of foreign-owned and controlled companies). Finally, many studies stress the importance of looking beyond ICT, labour costs, and the mobility of service providers and consumers in order to take due account of language and cognitive skills, cultural understanding, and various kinds of geographical links that, though harder to quantify, have increasingly been recognised as important supports of the tradability of services. As Bryson and Daniels emphasise in reference to Peter Dicken’s bestselling study on globalisation, ‘unlike the first “global shift”, the geography of the second shift is determined by the education and language abilities of services workers located in low-cost location’ (Bryson and Daniels, Reference Bryson, Daniels, Bryson and Daniel2007a: 12). This is why, for instance, a large multinational insurer would have a greater propensity to develop micro-insurance policies in a country with some shared knowledge and interpretation of what a supply or demand for cover really means to protect low-income people. And this is probably what promoters of micro-insurance have in mind when they emphasise that ‘slow and steady incremental improvements are … important for fostering a culture of insurance in low-income markets and creating a firm foundation for future expansion’ (Churchill and McCord, Reference Churchill, McCord, Churchill and Matul2012).

Despite these drivers, many analyses stress the factors that continue to hinder the internationalisation of services. A first explanation considers the sectorial specificity of services, whose intrinsic characteristics are seen as an insurmountable obstacle to internationalisation. A particular instance in this regard is the fact, discussed earlier, that some service activities cannot be stored and require direct co-production between clients and suppliers. Similarly, the more services tend to be immaterial, the argument goes, the harder it is to provide them at a distance. In addition, most firms providing services are SMEs and thus more likely to face additional difficulties in projecting their activity internationally. Often, to complement the sectorial account, another explanation focuses on the institutional specificity of services. In this perspective, as the intangibility of many services industries carries with it the risk of market failures and behaviour taking advantage of market power, governments have a greater tendency to intervene in the regulation of markets for services than for goods. Economists usually describe this as regulation driven by political economy considerations. They also take into account the public policy objectives involved in government interventions targeting services directly or indirectly related to the public interest. Such regulatory practices aim at environmental issues, consumer protection, health and safety guarantees, the provision of basic universal services, securing professional skills such as those of doctors, lawyers, accountants, and many others. The distinct legal framework and the pervasiveness of the regulatory environment surrounding a wide range of services are thus seen as ‘non-tariff measures’ which can, if not properly checked, prompt major hindrances to internationalisation (Copeland and Mattoo, Reference Copeland, Mattoo, Mattoo, Stern and Zanini2008). In a report closely examining such non-tariff measures and their embeddedness in trade policies, the WTO goes straight to the point: ‘Given the pervasiveness of services regulation and its commingling with trade protection a clear identification of which measures are trade restrictions, or a neat separation of the protective component in such measures, is fraught with difficulty’ (World Trade Organization, 2012: 78). Ironically, this has not prevented the OECD from launching in 2014 a project aimed at scoring and weighting barriers to services trade in twenty-two sectors across forty-four countries so as to build a new Services Trade Restrictiveness Index (STRI) and identify what the organisation sees as ‘potential scope to unlock growth through regulatory reform’ (OECD, 2017).

The details of what is driving or hindering services offshoring is not only becoming steadily more sophisticated but also highly contestable. Beyond sectorial explanations focused on the specificity of service industries and institutional explanations examining the range of instruments available to governments to pursue policies that can or cannot be trade restrictive, such non-tariff measures often rely on standards, testing, certification, and labelling in order to claim scientific rather than political justification.Footnote ⁷ Conventional explanations of drivers and barriers to services offshoring often fail to stress the extent to which the expansion of the tertiary sector has prompted new expectations on such non-conventional forms of power and regulation in contemporary capitalism. In the light of this, it is important to understand the influence of mechanisms that go well beyond intergovernmental cooperation and trade transactions. Indeed, greater global integration in the supply of services hinges upon a number of informal, non-state processes challenging national regulatory arrangements. It is in this context that international voluntary standards come into play.

From a number of angles, economic analysis has studied the conditions in which standardisation is possible and the resulting consequences for the pursuit of growth and innovation in the service sector. From a microeconomic and rationalistic point of view, standards are instruments designed to improve the reliability of the market for services, by certifying that providers supply services to users according to the agreed terms (Zeithaml et al., Reference Zeithaml, Parasuraman and Berry1990; Johnson and Nilsson, Reference Johnson and Nilsson2003). According to Blind, it is precisely ‘because of the intangible nature of services and the information asymmetries thus caused between management and service provider, [that] the need to introduce quality standards for each stage of the service production is especially high’ (Blind, Reference Blind2004: 167). Where possible, solutions incorporate enhanced use of technology, in particular ICT (Barras, Reference Barras1986). Where it is not, tasks can still be disaggregated into codified and standardised methodologies. From this managerial perspective, standards describe the ‘extent to which tasks in a process can be executed using a set of consistent and repeatable steps’ (McIvor, Reference McIvor2010: 105). Service standardisation is viewed as completing codification, which in turn provides a complete description of tasks along distinct components. A flurry of analyses has examined the scope of services likely to be standardised according to various taxonomies specifying such sectorial determinants. A number of them link the technological characteristics of services to the nature of the market they serve to define patterns of firms’ choices between standardisation and customisation (Boden and Miles, Reference Boden and Miles2000b; Tether et al., Reference Tether, Hipp and Miles2001; Djellal and Gallouj, Reference Djellal and Gallouj2002; Djellal and Gallouj, Reference Djellal and Gallouj2010). Moreover, while studies generally suggest that the information intensity of a service is correlated to its relational intensity and thus potentially hinders its prospective disaggregation, standardisation, and internationalisation, some scholars see it the other way round: they argue that high information intensity makes an occupation more amenable to disaggregation since each process can be well defined and thus codified and standardised (Mithas and Whitaker, Reference Mithas and Whitaker2007). In either case, while focussing on strategic interactions within a market environment in order to determine innovation and competition patterns, microeconomic studies of service standardisation overlook the political economy content which make these processes fraught with contradictions and power practices beyond and across firms’ behaviours and particular industries.

Scholarship inspired by the French régulation theory systematically includes social and political issues which can promote internationalisation of services in its analysis. Following on from Baumol’s pioneering work (Reference Baumol1967), Petit has analysed how the extension of business services at the interface of economic and political spheres contributes to a new growth regime which is imbalanced, dualistic, and based on a form of competition predominantly influenced by changes in consumer habits and lifestyle. Accordingly, the development of standards is hindered by the lack of shared values to orient and appreciate the qualitative changes required for the expansion of the tertiary sector. For instance, venture capitalists remain relatively weakly internationalised, as they suffer from the absence of a certification system that could help them standardise procedures for pooling investors and entrepreneurs (Petit, Reference Petit, Bryson and Daniel2007: 95–96, Reference Petit, Graz and Niang2013). As seen earlier, Gadrey’s typology also focuses on the variety of demand rationales in services. While his account deals primarily with the industrialisation of services, rather than their internationalisation, it also implicitly assumes strong institutional determinants. According to this analysis, an institutional environment with social and gender inequality is likely to support a commodification of services that seek to maximise productivity, economies of scale, and standardisation processes. The richest social group concentrating a large share of national revenue would have a clear interest in being served with industrialised and certified processes by a large (and mainly female) reserve army of poor workers. In contrast, greater social and gender equality is likely to lead to a more progressive understanding of service rationalisation supporting more reflexivity in working routines, and a rejection of commercial services and technologies that promote labour and gender inequality. The axiom would be ‘serve yourself’ rather than ‘be served’ in a standardised and commodified way, bringing with it important limits to market access on an international scale (Gadrey, Reference Gadrey2003: 105ff). For his part, Du Tertre (Reference Du Tertre, Lyon-Caen and Urban2008, Reference Du Tertre, Graz and Niang2013) examines the institutional outcomes that result from the distinct labour relations in services, in particular immaterial and relational activities (such as training, consultancy, and medical services). Here again, uncertainty as to quality and usefulness is understood as the most direct hindrance to the establishment of standards likely to support their internationalisation. According to Du Tertre, the responses to this uncertainty highlight two opposing types of institutional outcomes. On the one hand, there is the option of neo-Taylorist standardisation, designed to reduce the time needed for establishing the relationship between provider and beneficiary. This development, based on industrial logic, favours the use of machines (for instance, automated teller machines instead of bureaux de change) and information and communication technology (e-banking), as well as the formalisation of stereotypical behaviour (the number of refusals a catalogue retailer should accept before halting communications). The opposite approach is that of professionalisation, including the formulation of ethical principles which commit providers and beneficiaries alike. While this is common practice in the national framework of regulated professions such as in engineering or health services, Du Tertre sees it as largely non-existent when it comes to the international offshoring of services. Irrespective of the favoured institutional outcome, his argument goes, production of a service should always go hand in hand with a ‘social relation of accessibility’, defined as a ‘historic and institutional construct’ characterised by considerations such as geographical proximity, temporal synchronisation, and cultural and social understanding (Du Tertre, Reference Du Tertre, Lyon-Caen and Urban2008: 70–71). In this perspective, the internationalisation of services is rather unlikely, unless the trading logic remains close to that for manufactured goods (Du Tertre, Reference Du Tertre, Graz and Niang2013). In a similar vein, Mouhoud et al. have established a typology of services, based directly on their links with territories (Mouhoud et al., Reference Mouhoud2010).

According to institutionalist scholarship inspired by the French régulation theory, the uncertainty inherent in the intangible and relational nature of many service activities should not be apprehended as a problem of information asymmetry skewing the price mechanism, but as the logical consequence of the actual conditions in which wage relations and forms of competition are implemented in a post-Fordist regime of accumulation. Uncertainty as to quality and usefulness reflects the very heart of a service characterised by a high relational and immaterial component. From this perspective, it is quite normal that uncertainty should hinder the establishment of standards promoting the internationalisation of such activities. Standards would have a role only for services close to manufacturing such as those provided in the huge, remote back offices which Du Tertre calls ‘information factories’ (Du Tertre, Reference Du Tertre, Graz and Niang2013: 116, n. 8). This industrialisation-based reasoning favours the use of machines along with information and communication technologies, and assumes the imposition of stereotyped behaviours. All other types of services, however, will run up against socialisation processes – the social relation of accessibility as Du Tertre would call it – reflecting the disparate nature of institutional and localised dynamics. Here, the service-based logic specific to the immaterial and relational nature of the activities in question acts as a hindrance to the development of standards likely to support internationalisation.

Thus, French régulation scholarship sheds light on the socio-political underpinning of potential standards supporting the internationalisation of services. In doing so, it posits a restrictive hypothesis that paradoxically loses sight of political economy power plays. Such a restrictive hypothesis proposes that standardisation and internationalisation are closely defined by the attributes of the service and domestic institutions. In this perspective, standardisation and internationalisation of services indeed rest, on the one hand, on sector specificity, according to which the more the service provided corresponds to the ideal type of a relational, non-material service oriented towards the end consumer, relying on high-intensity labour, the less likely it is to be standardised and internationalised. On the other hand, it also follows institutional specificity, according to which the closer a service is to this ideal type, the harder it will be to find a substitute in the national territorial framework within which service relations are institutionalised, as is the case for regulated professions in medicine and law. This analysis offers a relevant framework for explaining the structural reasons underlying the difficulty of establishing internationally recognised standards in order to facilitate the offshoring of the most relational and intangible services. However, in my view, its hypothesis regarding the conditions for standardisation and internationalisation of service activities is too restrictive because it makes them dependent on sectorial and institutional specificity that rejects broader power configurations.

In sectorial terms, the specificity of ‘productive configurations’ and forms of competition for activities close to the ideal type of relational, non-material services oriented towards end consumers and relying on high-intensity labour does not in principle exclude rationalisation as a substitute for face-to-face transactions and objectivation of the high cognitive component of such transactions. Like money and law in traditional Marxist analysis, quality standards could represent a new general equivalent in a capitalist economy (Hartmann, Reference Hartmann, Graz and Niang2013). There is no better example than the battery of international quality and security standards for management systems developed by ISO from the 1980s onwards. These are still the best-selling standards in the world, and thus arguably the most widely used.Footnote ⁸ Mention may also be made of management instruments of similar nature marketed by many private players, such as Motorola’s Six Sigma method (which aims to identify and drastically reduce standard deviation from all the given specifications in a production process). Rationalisation of a service relation may thus apply to many highly relational services such as in education, healthcare, and consulting, for which many standards exist and support a market increasingly outsourced on an international scale.Footnote ⁹ As I shall explore in further detail in Chapter 7, it is by following this logic in particular that India has become the world’s office, providing offshore services on an ever-higher rung of the value chains.

On the institutional level, a restrictive hypothesis on standardisation and internationalisation of services arises from a highly local and territorial conception of institutional dynamics. Territorial and local roots of the socio-historical construct of economic activity, and the weight of political action engaged in at this level, certainly have an influence on the expected development of services, even in the field of ICT conventionally viewed as the least territory-based.Footnote ¹⁰ This also applies to national environments and institutional complementarities as highlighted in studies inspired by the French régulation theory (Aoki, Reference Aoki1994; Amable, Reference Amable2000). However, there is no reason why these considerations should be exclusive. A service relation may also be supported by procedures that are formally documented, certified, and accredited beyond the regulatory scope or political action of local, regional, and national authorities. This could include – and thus question – measures relating to social or gender inequalities and the power plays these suppose. For instance, it is precisely with this in mind that a quality standard for customer contact centres in France (NF 345) has been devised, some parts of which have been used at the European level (EN 15838:2009). Analysis of the controversies surrounding the establishment of ISO standard 26000 on social responsibility may be conducted from a similar perspective (Ruwet, Reference Ruwet2009; Capron et al., Reference Capron, Quairel-Lanoizelée and Turcotte2010; Barraud de Lagerie, Reference Barraud de Lagerie and François2011). In both cases, the institutional base of the service relation rests on a polymorphous and partially transnational space. The ‘social relation of accessibility’ of a service may come into play at this scale.

It is against this background that I propose an extensive hypothesis in order to overcome sectorial- and institutional-dependent explanations of services standardisation likely to support internationalisation. Viewing the nature of the service and its embeddedness in national institutions as determining factors does not fully do justice to the great variety of responses that international standardisation is likely to provide to the intangible and relational dimensions of many types of services. Service standards can link national economies to the global marketplace by responding to quality and security uncertainties in many, let alone opposing, ways. Far from stereotypes that deny cultural and labour issues involved in the service economy, the prospects of services offshoring rests on state and non-state regulatory arrangements which are not necessarily isomorphic. My extensive hypothesis emphasises a range of quality and security requirements likely to be standardised notwithstanding sectorial and local institutionalised specificities. Service standards can thus accommodate opposing political economy objectives and power configurations. On the one hand, they can promote a broadening and deepening of minimal market rules; on the other, they can include a number of provisions with the aim of defining labour-, environment-, gender, and any other socially-based specifications likely to be instrumental to more vulnerable actors in the production process and to ensure differentiated usage of services. Such an ambiguous juxtaposition of power instances confers authority to new actors on all sorts of new issues across sovereign spaces. This prompts us to explore the extent to which standards are a key instrument of transnational hybrid authority likely to have strong social and political implications in the quest for regulatory convergence supporting market access to services.

Standards-Defying Services?

In the introductory chapter of this book I defined standards as voluntary technical specifications explicitly documented and published as tools for the organisation of production and exchange of goods and services. Standards thus codify technical specifications regarding measurement, design, performances, as well as side effects of products, industrial processes, and services. I also emphasised the considerable overlap between mandatory standards embedded in regulations set by public authorities and voluntary specifications set by standard-setting bodies not based on sovereign State authority such as the International Organization for Standardization (ISO). ‘Standards hover between state and the market, to quote Schepel once more … standards are very rarely either wholly public or wholly private, and can be both intensely local and irreducibly global … standards can be seen as links between these spheres and institutions’ (Schepel, Reference Schepel2005: 4). The relationship between such ambiguous juxtapositions of standard-setting agencies and society are thus controversial, in particular with regard to services, where deep cultural and societal values and elusive labour issues are very likely to be involved. What is more, deregulation, liberalisation, and privatisation are likely to prompt new service standards to compete with previous rules governing public utilities.Footnote ¹¹

One of the clearest ways in which standards contribute to a non-conventional form of power is in their ability to reinforce path-dependant oligopolistic trajectories in technological innovation (David, Reference David1985; Mock, Reference Mock2005). Such a political economy of innovation inspired by Schumpeter has led Dudouet et al. (Reference Dudouet, Mercier and Vion2006) to conceive standardisation as a process of appropriation with significant incidence on market power and competition. Short of exclusive exploitation rights included in patents, standards codify technical specifications that can de facto exclude alternative technology and contending processes. Innovative technology conforming to such standards often includes patented technology, for instance in large-scale and forward-looking IT-enabled business services relying on extensive databases and complex algorithms. The neo-Schumpeterian analytical lens provides a persuasive explanation why large firms are likely to use standards as artefacts for maintaining their domination on distinct technologies. It is less focused, however, on how the power of such standards redefines the articulation between economic and political spheres. Accordingly, the following question remains unanswered: how do institutions underpin the power of such standards or, in other words, which institutional fora are privileged to exercise such power? Moreover, the approach tends to overcome the extent to which standards are used as levers of power on technological innovation, and reproduction is likely to vary according to distinct objects and processes. This leads to a second unanswered question: how does the issue to be standardised affect the power configurations at stake? For instance, to what extent can the establishment of sustainability standards – including labour, environment, and gender provisions – be clearly distinguished from standards established as instruments to develop technical interfaces to industrialise the provision of services? Finally, neo-Schumpeterian approaches do not explicitly examine how international standards may undermine the constitutive principles of the territorial sovereignty of states. Conversely, the role that states and intergovernmental organisations play in supporting the authority of international standards deserves further attention. This leads me to a third question: on which transnational space can technical specifications be defined, diffused, and recognised among sovereign states and, subsequently, what transfer of authority does this imply for contemporary modern liberal democracies?

These three questions on who standardises what and where bring me back to the analytical framework set out in the previous chapter. The ambiguous juxtaposition of power instances set in motion by what I call a transnational hybrid authority overcomes conventional oppositions. As Busch point out, ‘private standards and public regulations are two similar and sometimes overlapping forms of governance’ (Busch, Reference Busch2011: 27). Following the weight of common belief in the separation between the public and private spheres in modern capitalism and liberal democracies, legal provisions still often contradict this understanding and confine most international standards to the domain of private voluntary market tools provided by private firms.Footnote ¹² It is worth noting in this regard that the WTO analysis of non-tariff measures has largely gone beyond this narrow understanding. The following definition from its 2012 Report is unequivocal: ‘Although cast as “voluntary” in nature (because they are imposed by private entities), private standards (i.e. ISO & other standards) may become de facto a necessary condition for market access even if not imposed by law. … the effect of a particular private standard, if pervasive, could be greater than that of a government regulation of a smaller country’ (World Trade Organization, 2012: 211). Moreover, the ambiguous power of standards rests on the multiplicity of arenas in this domain and their wide range of institutional status (Djelic and den Hond, Reference Djelic and den Hond2014). The multiple standards authorities entail numerous agents who play or claim to play a role as new actors gain power to regulate a wide range of issues recognised as such across borders.

Consequently, the nature and implications of standards shaping the internationalisation of services can be appraised along the three-dimensional framework of analysis exposed in the previous chapter. This prompts us to aggregate the three following categories (Figure 3.1): the actors – i.e. who has the authority to set standards; the objects – i.e. what is standardised; and the space – i.e. where and whence standards are implemented. As we have seen, expectations regarding quality, security, and capacity to deliver services relate both to the private sphere of economic activities governed by market constraints and the public sphere of political action in the general interest of society. At the same time, they are closely intertwined with societal values and face regulatory diversity across countries. These dimensions of the transnational hybrid authority of service standards have some resemblance to what Busch has called the tripartite standards regime.Footnote ¹³ They provide a comprehensive yet differentiated picture of the wide range of actors involved in setting standards, the breadth of issues concerned, and the deterritorialisation of sovereignty conveyed by the multiple systems of certification and accreditation worldwide. We will examine those three dimensions in more detail hereafter.

Figure 3.1 Transnational hybrid authority

The first dimension of the framework is the institutional continuum of actors in the public and private spheres. As we have seen, there is much ambiguity on situating standards along this continuum and that ambiguity plays its part in conferring authority on a range of actors previously denied such capacity. Market mechanisms and policy choices both affect the agents involved in standardisation, but in various ways. Technical specifications belong to the private sphere of economic activities governed by market constraints; they affect social and technological change from that angle. However, they remain related to the public sphere of political action directed toward the general interest of society – for instance, by determining a certain level of risk and setting principles of liability or, a contrario, by allowing rent-seeking behaviour and market power. Hence, even in the circumscribed field of technical specification, standards relate as much to capital accumulation and technical progress as to social improvement or various instruments of the welfare state. When they are mandatory, enforceable, and general, technical specifications become part of public law and enjoy the status of government regulation. While some environmental, health and safety performances are defined under such procedures, they are often established by non-governmental actors on a voluntary basis. In such cases, technical specifications involve standard-setting bodies, whose private or public statutes vary considerably from one country to another. The wide range of actors and standardisation bodies able to set international standards are likely to address the distinctive aspects of the service sector in various ways. This is precisely the issue in understanding what can be standardised.

The second dimension charts the material continuum that outlines what can be standardised along the two poles of the physical and societal worlds. In aggregating the relationship between human beings and nature, technical specifications range from natural and invariable physical measures to constructed and historically bound societal values. Following an approach instigated by the Actor Network Theory in social studies of science and technology, Busch also views standards across the modern divide among nature, science, and society: ‘standards permit us to create complex socio-technical networks. As people are used, people and things are tested, and we shall determine what shall count. Those people and things that pass the tests or make the grade are drawn into various networks’ (Busch, Reference Busch2011: 12). This dimension highlights the extended scope of international standardisation. Long thought of only for physical specification such as those set for nuts and bolts, standards now cover all sorts of issues with clear and present implications for the societal world. One example among many is the European Technical Specification CEN/TS 16880:2015 for service excellence that sets out guidance ‘in order to create outstanding customer experiences, exceed customer expectations and achieve customer delight’.Footnote ¹⁴ As regards services more generally, this indisputably raises further pressing questions about what is a service standard. Do service standards concern the material support enabling the delivery of services (protective equipment used in the leisure sector, IT interface of call centres, etc.)? Do they concern mere procedural and generic aspects of services irrespective of the cultural context of their provision (e.g. billing, complaint redress, information provision, security requirements in the domain of business processes continuity)? Are they able to take into account more substantial expectations related to cultural and societal values, as well as labour processes including implicit skills embedded in the co-production of intangible and relational services (special needs for disabled people, the elderly, and children, or customer-oriented behaviour specifications for employees in tourism)? Clearly, the internationalisation of the service sector depends on standards for the development of technical interfaces to industrialise the provision of services. Yet it also relies on shared cultural values involved in the relationship of co-production between producers and consumers. Service standards can follow various paths in addressing such relationships, whether on a sectorial basis, taking basic cultural and societal values into account, or on a generic and horizontal basis, reflecting stereotyped behaviour that denies cultural and elusive labour issues involved in co-production activities. While those specifications can be defined as voluntary or mandatory on a national plane, if included in international standards they must be recognised beyond state borders.

The third axis of our analytical framework is the spatial continuum where the jurisdictions that support the system of recognition of standards overlap. While international standardisation is driven by attempts to homogenise technical specifications across national jurisdictions in order to reach a higher level of market and regulatory convergence, it faces a plurality of standards or tools for assessing conformity with them. From this point of view, compliance to standards is ambiguous. It rests on the dual nature of sovereignty: the endogenous logic of the territorial State and the exogenous logic of transnational capitalism. The endogenous principle depends on a system of obedience in agreement with the territorial space of state sovereignty. It supports a system of compliance to standards established on the assumption that a development process and an assessment of conformity procedure based on territorial sovereignty is what confers authority to an international standard. This is why, for instance, only one standard-setting body per country is eligible to membership in the ISO, even if many of them are non-governmental bodies identified as the ‘most representative of standardisation in their country’. When it comes to official international standard organisations, we are clearly faced with an ambiguous mixture of private and public bodies. ISO procedures make it even more ambiguous, as the actual work of defining standards is done in working groups where private experts sit in their own right, in contrast to plenary meetings of technical committees, composed only of national delegations. At the other end of the spectrum, when it comes to the exogenous principle guiding the compliance to standards, it is first and foremost their use by market actors across the globe that defines their spatial diffusion. A flurry of standards are set and used away from any defined system of national representation and delegation, and nevertheless entrenched in sovereign contract law. This is particularly the case for so-called consortium standards that define an agreed specification, usually openly accessible, yet developed by a restricted number of market players. They first dealt primarily with specifications required for the development of new products (like CD-ROM or DVD), or with interoperability in the field of ICT (such as the TCP-IP protocol used for connecting computers into a single worldwide network). Recently, however, a range of initiatives appear to promote such standards with a much broader scope. Corporate social responsibility benchmarks and sustainability labels in global value chains are cases in point (such as the Roundtable for Sustainable Palm Oil (RSPO) and the Roundtable for Responsible Soy (RTRS)). A number of initiatives also affect the service economy more directly, such as the Global Reporting Initiative (GRI) guidelines on presentation of information, specifically designed to make the environmental and social impact of major companies more visible. As a final point, it should be noted that, in the absence of harmonised regulation or standards, mutual recognition of standards and regulations reflects a step toward the exogenous end of the continuum. Mutual recognition is in fact designed to ensure that governments recognise each other’s standards and accept results of testing, inspection, certification, or accreditation bodies in specified industries. In Europe, for instance, products not subject to technical harmonisation at the EU level should in principle be exchangeable freely so long as they conform to domestic specifications and regulations of a member state. As we will see in the next chapter, this is also one of the key issues in provisions on regulatory convergence and non-tariff measures in the new generation of mega-trade deals.

* * *

This chapter has shown that the growing significance of services in the world economy involves being able to reach some common understanding on the quality and security of service provision. Far more than a simple question of asymmetry of information, the qualification of service activities is intrinsically political and relies on broader social institutions. It involves new patterns and agents of change through formal and informal regulatory practices of a wide range of non-state actors. In this context, service standards are likely to play a crucial role. Most existing literature shares, however, a restrictive view on the potential for standardisation of services, which would narrowly depend on domestic institutions and intrinsic characteristics of the industry concerned. This book takes a broader view. It proposes an extensive hypothesis that goes beyond a sectorial- and institutional-dependent analysis. The question is not whether services standardisation can or cannot occur across sectors and various domestic institutions. It is that when it does, it rests on the ambiguity of the statutes of actors setting such standards, of the issues eventually standardised, and of the space on which they are recognised. Service standards thus reflect a form of transnational hybrid authority that blurs the distinction between private and public actors, whose scope extends from physical measures to societal values, and that reinforces the deterritorialisation of regulatory practices in contemporary capitalism. That is the analytical framework within which the following chapters provide an overview of developments in the field of service standardisation. I start with the institutions which provide authority to standards as de jure or de facto regulatory instruments supporting the expansion of the tertiary sector across borders.