EUROPEAN MARKET MAKING AND POLICY DUPLICATION

In 1978, the Commission of the European Communities sounded an alarm. The member states had repeatedly interfered with the free movement of goods, it warned, and the common market had been jeopardised (Commission of the European Communities 1979). In the decade that followed, to the surprise of many, the situation was completely turned around; market integration was now a success (Garret and Weingast 1993). And this success is significant as it concerns not only markets but also states: market integration is widely held to be an important political tool in achieving an ‘ever closer union among the peoples of Europe’ (The Treaty of Rome). This success has been widely discussed (Sandholtz and Zysman 1989; Dehousse and Majone 1994; Héritier 1999), and is typically regarded as a successful change from negative to positive integration (Scharpf 1996; Majone 2005: ch. 7). Negative integration is about the removal of national barriers by deregulation, a kind of integration by liberalisation. Positive integration, however, is integration by making common rules. Using this distinction the history of the making of the common market can be summarised as consisting of an initial period when institutions were established and began to function (1958–69), a period when negative integration prevailed (1970–85), and a period when positive integration prevailed (1986 onwards) (Fligstein 2005b: 198–99). In this light the alarm was just an early call for more positive integration.

But the alarm had a broader meaning. The Commission also pointed to national technical standards and norms as the source of especially harmful barriers; they were said to, first, ‘practically isolate’ the markets of the member states from one another and, second, paralyse harmonisation by the Community in the specific area to which they applied (Commission of the European Communities 1979: section 102). The Commission's concern was about private standards, such as those developed by DIN and BSI (Deutsche Institut für Normung and British Standards Institution, respectively). It appears in this case that the process of political market making develops, on the one hand, around public regulation (EU directives, member state legislation and so on) and on the other hand, around private regulation (primarily technical standards). For market making to be successful, the Commission points out, these two parts need to be coordinated. At the time, however, the Commission did not have the means to accomplish this. This was an awkward situation for which the Community later found solutions. And in order to more fully understand the immense success of political market making in Europe since the 1980s, we need to systematically take into account this duplication of policy-making. The success has not been based on positive integration alone; it has required coordination of public and private regulation. What is at stake is not only the distinction between negative and positive integration; it is our ability to more broadly observe the transformations in governments, states and politics in the processes of market making.

In this paper, I further develop the distinction between negative and positive integration; I present a historical analysis of the making of the EU's internal market and show the importance of this duplication of policy-making while also, surveying efforts that were made to achieve coordination. My analysis is based on around 400 documents on the removal of barriers to the free movement of goods issued by political institutions of the EU between 1958 and 2000. In what follows I will first discuss negative and positive integration by drawing on the literature about transnational political governance and related issues (section 2). Then I will present a case study of how private regulation, in the form of technical standardisation, has gradually become part of European market making (section 3) and then I will present my conclusions.

Perhaps the single most important and encompassing project in the history of the EU is that of making a common market. Since the founding of the European Economic Community in 1958, it has been an on-going aim of the EU to make such a market, just as it has been a recurring challenge in the literature on European integration to understand the nature and history of the political changes that have occurred in this process. In many ways, market making has defined the EU (Fligstein 2008a), and in analysing market making in Europe, it is probably no overstatement that the distinction between negative and positive integration (Tinbergen 1965) informs the generally held view. It is a distinction that is commonly used without further explanation and often implicitly. The distinction contrasts ‘measures increasing market integration by eliminating national restraints on trade and distortions of competition’ with ‘common European policies to shape the conditions under which markets operate’ (Scharpf 1996: 15). Not only has this distinction proved handy in analysing market integration in various areas such as labour markets (Lovecy 1999; Beckfield 2006), immigration policies (Vink 2002) and environmental policy (Rehbinder and Stewart 1984), but it has also given ground for more general analyses of European political integration (Scharpf 1996; Majone 2005: ch. 7, 2009) and of European market making (Fligstein and Sweet 2002; Fligstein 2005a).

The negative/positive integration distinction highlights two distinct ways to make a common market as well as two distinct kinds of processes and sets of institutions. Negative integration calls attention to the member state as a government that needs to restrain or roll back some of its regulations. At the European level, this places particular responsibility on the Commission and the Court, on the Commission because the treaties give it the authority to remove certain barriers and to ensure that the member states do not establish new barriers, and on the Court, because it has developed legal principles that affect negative integration, such as the principle of mutual recognition. By contrast, positive integration calls attention to the member state as a government that is taking part in EU decision-making and attempting to bring its regulation into line with EU harmonisation directives and with regulation more generally. At the European level, this places responsibility on decision-making institutions such as the Council of Ministers and the European Parliament, and on the often complex decision-making procedures of the EU. Seen together, the distinction between negative and positive integration reveals a competency gap, since negative integration often develops more easily than positive integration, with the possible effect that national regulation is dismantled without any equivalent EU regulation to take over (Scharpf 1996; Majone 2005: ch. 7), and, of course, this has implications for democracy. It would be reasonable to expect negative integration to be initially dominant and positive integration to gain in importance over time, and this is exactly how Fligstein and Stone Sweet summarise the history of EU market making (2005b: 198–99). He suggests three relatively distinct phases: from 1958 to 1969 the EU institutions were established; from 1970 to 1985 negative integration and positive integration were both pursued, but negative integration was more prevalent than positive. The final period, ‘from 1986 to the present’, is where we find positive integration becoming most active. Over time, a political feedback loop is created:

In short, we have a dynamic that begins predominantly with negative integration, continues with positive integration, and then goes on to generate another cycle of negative and positive integration. The model focuses explicitly on governments, and thus on public authorities; private authorities and private regulations are left out of this model. One implication of the model is the expectation that the EU would increasingly acquire state-like qualities. Consequently, we would see a shift from markets made by national states to a market made by an EU government.

This view neglects private regulation; it would be possible that the demand for regulation would be for private regulation, and in this case, the ‘underlying logic’ would not necessarily be one of increasing stateness. One question that arises is what would happen to the feedback loop if we more systematically took private regulation into consideration, and how this might help us to understand the historical success of the internal market. Private regulations are defined as explicit rules and norms that are issued by institutions that are not public authorities; although they are not legally binding, they are considered binding de facto for a specific community. Private authority is a name for institutions that issue private regulations. The positive/negative integration distinction is a useful one, as it brings together an understanding of politics with an understanding of markets, political science and economic sociology, to allow us to understand complex historical processes in terms of a single distinction. But we have to supplement the positive/negative distinction with the distinction between private/government regulation to develop a model that can capture a higher degree of complexity than provided by an obligatory focus on states and stateness. The literature on transnational governance can help us do this.

In this literature, transnational political relations have been defined as ‘regular interactions across national boundaries when at least one actor is a non-state agent or does not operate on behalf of a national government or an intergovernmental organization’ (Risse-Kappen 1995: 1, italics original). Thus, transnational politics is distinct from national politics, i.e., politics that takes place within the boundary of a territorial state, and distinct also from international politics, i.e., politics that takes place between states, and where national governments have a monopoly on representing their respective national interests. In transnational politics we have an intermediate situation: the monopoly is broken to a certain degree, and private actors – firms, organisations, social movements and so on – take part in political decision making, and sometimes in the making of regulations. The private regulation may be national or it may be international. There are many examples of transnationalisation (Hale and Held 2011), and they often concern the development of a particular non-state legal order (Dezalay and Garth 1996) of private regulation or politics across national boundaries (Bartley 2007a,b; Nölke and Perry 2007), but they may also be coupled with and interact with national policies (Bruszt and Hozhacker 2009; Bruszt and McDermott 2009) or EU regulation (Jacobsson et al. 2003; Menz 2003; Orenstein and Schmitz 2006; Higgins and Hallström 2007). We thus find private regulation e.g., in the fields of forestry, accounting, finance, internet regulation, corporate social responsibility, technical standardisation and labour market regulation, all of which are relevant in an European context.

Private European technical standardisation organisations, such as CEN and CENELEC, which have participated in the creation of the common market, provide a familiar example (Egan 2001; Schepel 2005). These organisations produce technical standards by inviting and organising interested parties coming from industry, interest organisations and so on, in technical committees and work groups in order to discuss and flesh out specific technical solutions to common problems. Technical standards are neither standards emerging from the market nor as such enforced by political authorities, but constitute an intriguing category of its own between what we may term ‘market standards’ and ‘political standards’. Technical standards are, in other words kind of regulation that, from a strictly legal perspective, is mostly voluntary, but is nevertheless often taken to be de facto binding (Frankel and Højbjerg 2012). It is sometimes described as ‘soft law’ (Snyder 1994; Abbott and Snidal 2000). These organisations are thus regarded as being able to regulate markets de facto (Schepel 2004). Indeed, studies of technical standardisation suggest that national governments and EU institutions are not the only de facto regulators of the common (or national) market. Rather, we have situations where national and European private entities (Voelzkow 1996) and private organisations (Reinalda 2000) participate in the making of the market. The extent of their role is still a matter of debate (Börzel 2010), but there is no doubt that they do play a role.

Our interest here is the specific kind of political governance that takes place when regulation is issued by public as well as private authorities at both a national and at an international level. According to the definition earlier this governance is clearly transnational, but of a particular kind. While the definition focuses on who is represented in political processes, we focus more specifically on the kind of regulations that are produced and how these regulations are linked. With this focus I am intentionally leaving out much of what has been studied as political processes, e.g., how lobbyists, social movements, local governments and various other interests interact when seeking to influence the outcome of decision making. While such political factors are important in the creation of the common market, I propose to take a step back and ask the question: what regulators and types of regulations are given responsibility for creating the common market? Answering this question will help highlight changing attitudes toward private authority.

With respect to actual regulations, the creation of a Common European Market is usually analysed with a focus on EU regulations, national public regulations, and how they are connected (illustrated with the line ‘a’ in Figure 1). This is where the distinction between positive and negative integration would usually direct our attention. However, in addition to these two governmental forms of regulation, national private regulations may also prove significant for the creation of a common market. They may involve national standardisation organisations, and meta-organisations such as European standards organisations. In other words, relations between national public regulation and EU regulation are duplicated in the form of relations between private regulators at a national level and at the European level (illustrated with the line ‘b’ in Figure 1).

We may consider that this also represents a doubling of positive and negative integration; national private regulation can either be lifted by (negative integration) or harmonised with European private regulation (positive integration). But if such doubling does take place, one would also expect that these two kinds of regulation – public and private – would be linked and coordinated as a way for the international body to get closer to the goal of a common market. Therefore, we would also expect coupling between private and public regulations.

Of course, the possibility of such duplication has not escaped attention in the literature on political integration and the sociology of markets (Fligstein 2005a; Frankel and Højbjerg 2007). But there is still something to be learned by examining them more systematically. The institutional sociology of markets, in particular, has not felt fully at ease with the notion of private regulation; it raises concern because the core of institutional economic sociology is the idea that states and markets are co-constituted (Polanyi 1944/1985; Block and Evans 2005; Fligstein 2008b). Perhaps this conviction has led to a focus on public regulation and to models that lead us to expect that the EU will increasingly acquire ‘stateness’.

In the historical account that follows, we will consider the relations that have been established among these four types of regulation in the creation of the Common European Market.

In the language of the EU, our focus is the free movement of goods, and our account will be concerned with the removal of barriers to this freedom. The period covered begins in 1958 and reconstructs the history up to 2000. The key sources include the EU treaties, annual reports, green papers, white papers and proposals by the European Commission, hearing responses by the Economic and Social Committee, resolutions and decisions adopted by the Council of Ministers, debates in the European Parliament and, of course, directives and regulations, a total of around 400 documents. The documents have been selected by using literature reviews and search engines, and by following references from one document to the next until all references concerning the free movement of goods had been tracked. The analysis of these texts focuses on the kinds of regulation and relationships between kinds of regulation that are employed, and seen as essential for creating the Common Market. We do not go into all of these documents in detail, but keep focus on the relative importance of negative integration and positive integration in creating free movement for goods, and, of course, on how private regulation became a part of the market-making project. The analysis is of sources from formal political institutions, and this has the obvious limitation that we do not get to know how private interests take part in this duplication of politics, how they pursue their interests and with what results; alternatively, however, it has the strength that it tells us how and why the duplication of politics has been promoted and authorised by recognised political authorities.

At the 1955 Messina Conference, the members of the European Coal and Steel Community agreed to establish a common market. A few years later, in 1958, ratification of the Treaty of Rome established the European Economic Community along with a set of institutions with the goal of making a common market. The Treaty provided a relatively detailed plan for how to realise the free movement of goods. It focused on barriers at the external boundary of the member state (Council 1960). Customs policy is a case in point: customs are typically levied when goods pass the external boundary of the member state. It was assumed that the common market could be established by removing external barriers and other forms of discrimination between nationally produced goods and imported goods (Commission 1958, 1959, 1961, 1962, 1963). The definition of a barrier was based on discrimination. All measures that ‘especially affected goods imported from another Member State but not the equivalent local products’ (Kommission der Europäischen Ökonomischen Gemeinschaft 1966, my own translation from German) were thus seen as equivalent to customs duties.

This approach to market making implied that the common market – although absent – was already present as a potential: the underlying assumption was that by removing existing barriers, the common market would emerge, and as a result, one can essentially achieve a common market simply by negative integration (removal of barriers). This was institutionalised in the form of a particular division of tasks between the Community and the member states. The Community had the task of surveying barriers to the common market and ensuring that these barriers were removed according to a specified 12-year plan, stipulated by the Treaty (Council 1960). The task of the member states was to remove customs and quotas and any of their variants.

However, not many years after the founding of the Community, a change occurred in the market-making project. This change occurred as new barriers came into view. The first indication was a new concern that the notion of discrimination was inadequate. Non-discriminating measures, so the Commission suggested, could also work as barriers. This provoked some remarkable reactions, e.g., from MP Deringer in the European Parliament. He asked what would now be the Commission's ‘fundamental conception’ of trade barriers? (European Parliament 1966). The answer might at first seem tautological, namely that a barrier is something that works as a barrier, but over time, a new, supplemental definition of barriers was institutionalised, technical barriers to trade. These are not barriers as such. They are measures on the part of a member state that work as a barrier because they differ from corresponding measures in other member states (Commission 1967, 1968; European Parliament 1968a). Measures that regulate the packaging of beer can serve as an example. Such regulations are purely internal measures on the part of the single member state in the sense that they do not discriminate between national and imported goods. But packaging measures became regarded as a technical barrier because they differ from packaging measures required by other member states.

Although technical barrier may seem obvious to many today, they are not mentioned in Treaty of Rome and it took years before it became a well-established ‘natural’ category. In 1968, to give an example, Commissioner Colonna di Paliano used the term ‘black market protectionism’ to refer to technical barriers to trade, thus underscoring both the importance of this type of barrier and how little attention they had received (European Parliament 1968b).

The new focus on technical barriers multiplied the number of barriers to be addressed (Council of Ministers 1973a,b,c), and the project to create a market was now essentially transformed into a project of positive integration. Harmonisation of member state legislation came to be seen as essential to removing technical barriers to trade. The seemingly small step from discriminating measure as such to removal of differences profoundly affected the project of creating a new common market. The market loses its idea of naturalness; it is seen as politically constituted by the measures undertaken by the state. This also implies that the market now became concrete and specific, and the EU now definitely was faced with the task of integrating multiple national and product-specific markets. In other words, the Community's task was not simply to open the floodgates, but instead, a step-by-step process to ensure that the common market was politically constituted (European Parliament 1968a; Council of Ministers 1973c). In fact, it was such a vast undertaking, far beyond the 12-year plan for the realisation of the common market. If technical barriers were to be removed, the Community had no other alternative but to adopt a large number of new legislative initiatives.

It is at this point, in 1978, that the Commission sounded the alarm. Part of what the Commission urged members to pay attention to was that private norms and technical standards, although in principle voluntary, could very well function as technical barriers to trade. At first, private norms and technical standards did not constitute a well-defined category, and political institutions such as the European Parliament (1979, 1980) felt that it was somewhat inappropriate to pay so much attention to technical details, but at the same time, they stressed that this was an issue to be taken seriously. The European Parliament agreed that private national regulation can work as a barrier and that this problem needed to be solved. In order to do this, new mechanisms needed to be invented, since at that time, the EU only had the capacity to remove or harmonise barriers stemming from public regulation by member states. Over the following years, solutions to the problem of private technical standards were institutionalised using at least two significantly different approaches. One is what is known as the new approach to technical harmonisation and standards and the other is the information procedure directive. Each of these initiatives linked public and private regulation in a particular way.

The Information Procedure Directive was first proposed in 1980 and adopted in 1983 (Commission 1980; Council 1983), and it has since been revised (European Parliament and the Council of Ministers 1998). This Directive can be described as the institutionalisation of pro-active efforts against barriers to trade. The Directive obliges national regulators to inform the Commission and its information procedure committee about all regulatory initiatives well before national regulators adopt the regulation in question. The Commission then passes this information on to all the member states. The member states thereafter have the option of requesting a moratorium on specific initiatives and also to initiate negotiations as to whether the issue is better dealt with at the European level or can be dealt with nationally in ways that do not impose barriers to trade. In this way barriers to trade are to be dealt with before they ever occur.

In the context of this article, the most remarkable thing about the information procedure Directive was that it systematically included national technical standardisation organisations in the various member states. In an appendix the Directive listed the national standardisation organisations to which the Member states should report their initiatives. They are required to report initiatives regarding new technical standards, just as public authorities and legislators are required to report their initiatives. Also, there is the option of transferring the initiative by a national technical standardisation organisation to a European technical standardisation organisation, such as CEN or CENELEC (European Committee for Standardization and European Committee for Electrotechnical Standardization, respectively). If we examine the information procedure, we see a Directive that seeks proactively to bring together four different ways in which regulation is made. This is illustrated in Figure 2.

Figure 2 shows how the information procedure committee links four regulatory procedures by making member states and European regulators aware of national initiatives, and by the ability to invoke a moratorium and possible transfer of the initiatives to the EU or to European standardisation organisations. Since the 1980s the committee has processed hundreds of notifications every year about member state initiatives and national standards initiatives (Commission 2000). The procedure resembles negative integration as it prevents new barriers from emerging, and it brings about positive integration when new initiatives are moved from the national to the European level.

The information procedure Directive is a political response to the principle of mutual recognition, stated in 1979 by the European Court of Justice in its famous Cassis de Dijon ruling (C-120-78). According to this principle the member states are to recognise that their respective legislation, despite any differences, is intended to serve the same ends. Therefore, products lawfully marketed in one member state should also be legal to market in other member states. But the principle of mutual recognition does not automatically extend to private regulation: how can the public authorities of member states ensure that private parties mutually recognise private regulations of other member states? There is no way to accomplish this because, in the end, ‘no purchaser can be obliged to recognise foreign products’ (Commission 1990: 3). Therefore, it became necessary to work with national standardisation organisations in order to encourage them to work toward common European standards.

The new approach was proposed by the Commission and adopted by the Council in 1985 (Commission 1985; Council of Ministers 1985) but had been in the works for several years before that time. The main idea of the new approach was to couple EU legislation with technical standards. Thus a division of labour was instituted: the EU was to formulate essential requirements for a group of products, such as toys, construction products or machines. These essential requirements were to be formulated in general terms that were essential to protect health, safety and the environment. The European technical standardisation organisations, on the other hand, were mandated to formulate specific technical standards that translated these general requirements into concrete technical notions. And, as a next step, the European technical standards would then supplant the corresponding technical standards issued by the national technical standardisation organisations. Thus, we might say that the new approach regulation relies on double implementation. It is implemented in the laws of the member states. And it is implemented in European technical standards that supplant national technical standards.

The new approach continues to play a vital role and has been dealt with in the literature, but there is often too little attention paid to the fact that the new approach specifically provides a solution to the problem of private, national technical barriers. The national technical standardisation organisations do not become members of the EU, but as members of European standardisation organisations they can nevertheless be addressed and enlisted in the work of harmonisation.

While the information procedure aims at a systematic incorporation of private authority in the politics of market making, it works to ensure that all regulations issued by private and public authorities are produced in a coherent way. This is where the mandate, as institutionalised in the new approach, comes into the picture. The mandate allows the EU to require technical standards in a specific area. This allows the EU to issue general framework directives, but it also makes the EU depend on European technical standardisation organisations to transform the general requirements into concrete rules. The coupling of private and public authority in the new approach is illustrated in Figure 3.

Figure 3 illustrates two harmonisation processes and the resultant doubling of positive integration: on the left we have the EU adopting directives and thereby harmonising regulation of the member states. On the right we have European standardisation organisations adopting European standards and thereby harmonising national technical standards. These two processes are brought together by a mandate (Commission of the European Communities 2000), negotiated by the Commission and the European standardisation organisation in question. The mandate stipulates the areas where standards are to be made, the amount of funding provided by the EU and sets deadlines. The mandating process establishes a division of labour and a mutual dependence between the two main parties, and thus also the conditions for negotiations concerning the intended meaning and match of EU regulations and technical regulations.

The new approach relieved the EU from the responsibility of adopting what could become hundreds of directives, and permitted it to work instead on broader framework directives. Instead of elaborating legislation on highly specific areas, such as tractor seats, new approach directives often cover broad areas such as toys (2009/48/EC), machinery (2006/42/EC), construction products (89/106/EEC) and electro-magnetic compatibility (2004/108/EC). In tandem with the growth of the new approach, European technical standardisation activity has also grown. Until the 1980s it was not very active compared to national technical standardisation (DIN 1995). But European technical standardisation has experienced a considerable growth spurt, and while there were only around 600 European standards in the early 1980s, there were more than 7000 by the end of the 1990s.

Today there is no doubt that cooperation with technical standardisation bodies is essential to the political creation and maintenance of the common market (Commission of the European Communities 2000). Often left out – or perhaps suppressed – is the fact that this transnational concerted negotiation (of e.g., standardisation mandates) also served as a solution to problems stemming from the private soft law that exists in the member states. This omission is convenient, because the division of labour is generally legitimised by the different status ascriptions for the measures: new approach directives are binding, while European technical standards are voluntary. But while technical standards are considered voluntary at the European level, they are certainly seen as de facto regulation at the national level.

The alarm sounded by the Commission in 1978 did, indeed, herald the start of a series of institutional innovations, and since the mid-1980s the duplication of policies has characterised the structure of the common market; barriers to trade have been removed both by EU harmonisation directives and European standards. The historical review has also demonstrated that these two types of integration are systematically linked through institutions such as the new approach and the information procedure. In this light we should now take a closer look at the usefulness of the distinction between negative and positive integration.

One way to do so is to contrast the historical review just presented with the stages suggested by Fligstein and Stone Sweet. The historical approach has shown that negative integration was already central from the beginning, since it was stipulated as part of the 12-year plan in the Treaty of Rome, which is earlier than Fligstein and Stone Sweet seem to suggest. A second contrast is that their account does not give sufficient weight to the changing character of positive integration since 1985. Public–private coordination also developed extensively during this period, and the duplication of policies has in many ways proved to be a precondition for the success of positive integration of markets. Thus, an overall depiction of the integration process as moving regulation step by step toward the European level, and thus in the direction of ‘stateness’ is inadequate; it risks creating a distorted view by ignoring important elements of market making. Instead, the historical account suggests a more complex process where regulation is also shifting concurrently from public to private authorities. Therefore, it remains a question, conceptually as well as empirically, whether this overall change can be characterised as a move towards greater ‘stateness’ on the part of the EU.

At a conceptual level, this shift in perspective implies that the underlying logic of the sociology of markets model can no longer be seen as evolving around firms and other market actors urging governments to regulate (Fligstein 2005a: 197). We need to look at other logics as well, such as firms and other market actors urging technical standardisation organisations and other non-state regulators to regulate, and also urging governments to leave regulation to the technical standardisation organisations, with the possible outcome that environmental organisations and others find it difficult to monitor and influence the decision process (European Environmental Bureau 1994). It would seem that the political struggle, is as much a struggle about who should regulate as it is a struggle concerning which particular regulation should be adopted or not. Moreover, we cannot know the relative practical importance of the different logics suggested except through empirical inquiry. Economic sociology does not provide us with a single logic, but instead invites us to inquire the relative importance of several logics.

The degree of sovereignty exercised by individual states is variable (Fligstein 2008a: 67), but member states are often treated as if they govern and frame private regulations such as those produced by national standardisation organisations. That may have been the case in the past, but there is reason to believe that European market integration has altered this because national standardisation organisations have become directly involved in standards institutions at the European level that have gained in importance vis-à-vis the EU as they have been given monopoly as providers of harmonised standards. Thus the model that markets can be integrated by merely integrating states (that is, public governance by member states) proves to be inadequate.

When we also take private regulation into consideration, the question of integration becomes somewhat more complex. We then must include the integration of public and private authority, and these are not necessarily congruent. They may represent different clubs (Majone 2009) with different members and different interests: although both the German federal government and the German DIN are national in the sense that they represent Germany, their representation of the nation and of interests may vary so greatly that we can no longer recognise their commonality. There is no necessary logic that prescribes that national technical standardisation organisations should come to the same preferences as their respective national governments. Thus, the outcome of integration is not a more-or-less state-like construct formally defined (by treaties and the EU acquis). We also need to include the image of negotiated structures that include both private and public regulation.

The distinction between negative and positive integration is still useful for an inquiry into market integration. But the account given here also points to limitations in the usefulness of this distinction. The information procedure directive can serve as example. This procedure prescribes notification about both public and private regulation, and it may result in a moratorium at the national level or initiate (public or private) regulation at the European level. In a sense, the information procedure addresses negative integration when it aims to avoid national barriers, but in a sense it also addresses positive integration when it provides a mechanism to initiate new regulation at the European level. But even these characterisations fall short. The procedure is not about negative integration in the sense of removing a national regulation; it is concerned with national initiatives, i.e., with the potential of forthcoming regulation. Nor is the procedure about positive integration in the sense that it directly produces harmonisation directives; it is concerned with the potential for initiatives toward common regulation, i.e., with the possibility of forthcoming regulation. The procedure operates primarily in the time dimension. Like the financial instruments, options and futures, it is concerned with possible futures. Moreover, we cannot adequately understand the information procedure directive in terms of the distinction between positive or negative integration. Rather, it is an integration measure that reflects upon these two types of integration, e.g., by turning negative integration into positive, but the directive itself operates as a meta-instrument to coordinate legislative processes and rationales. We can conclude more generally that the negative-positive integration distinction is useful in relation to distinct couplings, namely between regulation at the European and the national level, but is not useful in situations where other kinds of couplings prevail, such as those established through mandates and through the information procedure.

Does the increasing importance of technical standards imply, for example, that crooked cucumbers will be removed from the market to be replaced by standardised products, such as straight cucumbers? In terms of economic sociology, this could be phrased as a question of whether the EU favours standard markets (Aspers 2011: 88ff). But this line of argument is mistaken. The harmonised standards for toys, for example, do not lead to standard toys in standard markets; instead we have various kinds of toys that comply with general standards. Technical standards often specify a function or a performance, and leave open the specific product design to accomplish this. Technical standards are versatile. They may be a central feature for standard markets but this is only one of their many functions (Brunsson and Jacobsson 2000).

This article has focused on the politics of market making, and this leaves open a number questions for further study. For example, are harmonised standards necessary for creating a common market? Comparisons of the EU and the USA show remarkable differences in this regard (Schepel 2005). In the same vein, there are remaining issues to be studied in regard to what makes technical standards de facto binding. Sometimes they are, and sometimes they are not. The analysis undertaken here does not address this question. What it demonstrates is that in a series of documents, legislation and institutions the EU has conceived of national technical standards as de facto binding, as a form of private regulation, and in response to this, has sought to develop solutions to this problem.

The relation between politics and markets is widely debated, and the positive/negative integration distinction has often been used to shortcut this debate and to conceive of states (public authority) as being necessary for markets to emerge, and of states and markets as being co-constituted. Needless-to-say, there are many other notions of markets that do not privilege states to the same degree. The notion of duplication of policies, as suggested in this article, leaves open to empirical study the extent to which states (public authority) and private authorities are necessary for the constitution of markets. This, of course, suggests future studies, not only of technical standardisation and barriers to the free movement of products, but also of other ambiguous processes of transnationalisation of policies and market making, and of potential new relationships between politics and markets.

I want to thank José Ossandón, Signe Vikkelsø, Ursula Plesner and three anonymous reviewers for constructive comments on earlier versions of this article.

Christian Frankel is Associate Professor at the Department of Organization, Copenhagen Business School. He works on politics, products, organization theory and economic sociology. He has received the Danish textbook award for a book on organisational analysis and has published in journals such as Business & Society and Journal of European Public Policy.