The European Automobile Industry

William A. Maloney, Andrew McLaughlin

• Publisher: Routledge
• Year: 2005
• ISBN:
  9781134829255
• Edition: 1
• Topic: Politics & International Relations
• Subtopic: Politics
• Index: Politics & International Relations

This book presents an analysis of some of the changes that have transformed the automobile industry in the last thirty years illustrating some of the most significant consequences of globalization. Focusing on the response of Europe's policy makers, it analyzes government-industry relations at both national and transnational levels, demonstrating how national policy instruments have been eroded by regional, political and economic integration. There has been a significant and irreversible shift in the locus of decision-making power from nation states to the regional level in the automobile sector.

1: Analysing government-industry relations in the automobile sector

In practice the policy-making process of the car industry is affected by the multi-level structure of the EU decision-making system.

Holmes and Smith (1995:125)

This book provides an analysis of government-industry relations in the automobile industry at the national and transnational level. It also contemplates the evolution of governing structures at the global level. The automobile industry is one of the best examples of an industrial sector in which national policy instruments have been eroded by regional political and economic integration, and the globalization of production.¹ This book examines how policy-makers at each level have managed their relations with the industry and how the industry has organized itself to participate in the policy process. By examining events at the national level using the British case as an example and then at the regional level in Europe, the book identifies the governing structures that have evolved and the relationship between these structures and policy outcomes. This approach reflects our initial suspicions that the steering mechanisms used by policy-makers to govern industrial sectors are shaped in large part by the structure of government-industry relations.

There has been a significant and irreversible shift in the locus of decision-making power from nation states to the regional level in the automobile sector. The European Union (EU) had achieved the full harmonization of national technical regulations by 1996, implemented a strict sectoral state aids framework and started to dismantle import barriers against Japanese cars. A new policy process has emerged at the regional level which co-exists with previous national arrangements. Accounting for policy outcomes requires an analysis of the relationship between public authorities and the industry at both levels. To this end the recent literature on EU policy-making has focused on developing and refining theoretical tools and models which can cope with the multi-level nature of EU governance.

This introductory chapter outlines the sectoral characteristics of the automobile industry, reviews the main conceptual debate in the study of government-industry relations, and examines the utility of the concept of policy networks in understanding decision-making at the national and European level.

What is being analysed? Sectoral characteristics

In global terms and within the European region the automobile industry is an oligopoly. The ten largest automobile companies account for almost 80 per cent of world production. In the major markets of North America, Japan and Europe demand has all but matured and the industry has become a high cost/ low margin business. The sector is suffering from significant over-capacity even though the barriers to market entry in the sector are high. This is due to the international growth strategies of producers from developing regions and the considerable efforts made by producers in established regions (including Europe) to improve their productive efficiency. Thus, somewhat ironically, by embracing Japanese production techniques many European companies have become more productive at a time when demand has levelled out. This environment has led to the development of an extensive array of joint ventures and tie-ups between the leading automobile companies as they seek to balance large-scale production with product diversity.

These market conditions have created testing times for many automobile companies in the 1990s and complex problems for policy-makers. The industry may have left behind the rapid growth phase of the 1945–90 period but it remains of fundamental economic importance to national and regional economies. It is a weather vane for the macro economy and the fact that policymakers in all regions have frequently intervened in the sector is a testament to its enduring importance. The symbolic importance of the sector in most nation states partly explains why much-needed mergers and acquisitions have been slow to emerge.

The economic significance of the automobile industry rests on the fact that it is such a large and multifarious sector. The following sub-groups can be identified within the automobile sector: full-line vehicle assemblers (some of whom are also large-scale components producers and sub-assemblers); specialist vehicle producers; tier-one component companies (multi-product companies with production facilities in several countries and serving global markets); tier-two component companies (original equipment suppliers to EU assemblers); tier-three component companies (direct suppliers to vehicle companies or more usually to tier-one suppliers); and large suppliers of materials and intermediate items (for example, steel, aluminium, glass, plastics, synthetic rubber, paint, and ball bearings). Our focus is on the major automobile assemblers, but the wider industry is pivotal to the manufacturing base and provides an end market for so many other sectors.

The importance of the industry within the European region should not be underestimated. The European automobile market accounts for over a third of both world consumption and production. The industry accounts for 17 per cent of total tax revenues in the EU and for 6 per cent of EU manufacturing output. Moreover, it continues to make a positive contribution to the EU’s external trade balance in spite of sluggish demand in Europe and increased competition in export markets. The industry’s products also have a major impact on society. More than 60 per cent of European households own one or more cars, over half of journeys to work are made by car and over 70 per cent of all journeys are by private car. The industry and its products are heavily regulated as a result (see Appendices 1 and 2 for a detailed account of automotive legislation). Thus, the interests of the industry in the policy process extend well beyond manufacturing into areas such as transport infrastructure, transport regulation and environmental policy. The automobile industry has a wide range of policy-relevant interests and is a regular participant in the policy process at the national and transnational level. As a result the range of contacts with national and EU institutions is extensive.

Government-industry relations

It remains an open but important question whether government-industry relationships may vary more significantly or consistently between sectors than between nations. That certain sector-specific characteristics do recur consistently across a range of national settings is clear. Whether such sectoral characteristics are more significant than variations in national characteristics is an empirically unresolved question.

(Wilks and Wright, 1987:290)

This quotation from Wilks and Wright encapsulates the major research questions facing students of government-industry relations since the 1980s. Until then the literature was dominated by ‘top-down’ macropolitical accounts of policy-making—referred to hereafter as the ‘classic’ approach. This approach relates to studies that rely primarily on systemlevel characterizations of a political economy to inform their position on the nature of government-industry relations. The work of Wilks and Wright (1987) represented a major revision of conventional wisdom. They argued that a disaggregated research agenda was likely to reveal a rich diversity of governing structures at the sectoral level, belying the broad characterization of the classic approach. They suggested that a ‘bottomup‘research agenda—termed here the ‘sectoral’ approach—was at least as valid. The two approaches are outlined in turn below.

The classic approach

The classic approach identifies durable national policy-making traditions and cultural norms amongst political elites which are then reflected in policies developed within national political systems. The research agenda is focused on the beliefs, attitudes, political discourse and modus operandi of elite actors. The essence of the classic approach is that ‘politics determines policy’ (Freeman, 1985). Accordingly, macro-political influences are important in shaping specific industry policies and there is an acceptance that certain values and traditions exist autonomously, independent of actual relations between state and societal actors. The state is identified by characteristic features which exhibit themselves across policy sectors and networks.

The leading exemplar of this approach in Britain was Shonfield (1965) whose early work presented vividly contrasting images of the British and French political economies. The ‘proactive’ French state was characterized as having a sophisticated planning machinery to implement industrial steering, whereas the ‘reactive’ British state relied on non-regulatory policy instruments and where possible an ‘arm’s length’ relationship with business. The most enduring contribution to the classic approach—the notion of ‘strong states’ and ‘weak states’—has its modern origins in Katzenstein’s edited collection Between Power and Plenty (Katzenstein, 1978). This study explored the historical evolution of domestic political structures (i.e. the degree of centralization in state and society), and assessed their impact on economic and industrial policy. States were then classified according to the findings. Katzenstein (1978) concluded that Britain and America had pursued a laissez-faire strategy which relied on market forces and non-regulatory measures to steer industrial sectors. These were weak states because they were either unwilling or unable to intervene in the economy at the micro level. When weak states did intervene, it was usually reluctantly at times of crises and with disappointing results. Successful interventions appeared to require well-developed governing structures. Thus, the strong states such as Japan and France had adopted a neomercantilist approach involving the use of extensive policy instruments and selective interventions.

Katzenstein (1978:299) recognized the dangers of characterizing complex national systems with such broad labels, but he argued that the typol ogy was justified in a comparative context. Thus, while Britain and the USA are different in some respects—the British executive has more power to intervene in society than its American counterpart—in comparison with Japan or France both could be justifiably labelled as weak states. Following Shonfield, Katzenstein suggested that successful intervention in the economy depended in large part on the state’s ability to form productive relations with mobilized societal interests.

Another notable contribution to the classic approach is Zysman’s system-level analysis (Zysman, 1983). This approach relies almost exclusively on one explanatory variable, the nature of financial systems, to explain differences between nations. He identified three types of financial system, each of which has specific implications for a government’s ability to coordinate industrial adjustment. For example, company-led adjustment strategies tend to dominate in capital market-based systems such as that in the USA where finance is raised by companies in financial centres. This contrasts with credit-based systems in Japan and France where the state has greater control in financial markets and can use this position to take a lead role in industrial adjustments. Zysman maintained that his analysis ‘supported the proposition that one can make predictions (about adjustment) based on the nature of the financial system. In Britain the capital market-based financial system which promotes arm’s-length relations between business and government interfered with government efforts to direct the process of industrial change’ (Zysman, 1983:286). Whereas other studies focused on the arm’s length relationship between the state and the manufacturing sector, Zysman suggested partnership between these two may achieve little without the cooperation of the financial sector. Nevertheless, his analysis was very much in the strong state/weak state tradition and reinforced these cornerstones of the classic approach.

A final dimension to the classic approach was the transitory attempt to elevate the notion of macro-corporatism as a description of the British and other political systems (Schmitter and Lehmbruch, 1979; Goldthorpe, 1984). In the macro variant of corporatism attention was ultimately focused on the role of peak organizations in the economy. However, this perspective lost ground by the late 1970s as those who examined attempts to develop British corporatism suggested that the more established traditions of the classic approach—weak state and atomistic societal interests—had persisted. Even the most celebrated evidence of British corporatism, the tripartitism of the 1970s, provided inconclusive evidence on closer inspection (see Marsh and Grant, 1977).

In summary, the classic approach seeks to understand and explain differences between national political systems when accounting for the development of industrial policies and patterns of government-industry relations. The assumption is that industrial governance structures are in large part the product of system-level characteristics.

The sectoral approach

The sectoral approach consists of a body of sectoral and sub-sectoral policy studies motivated by a healthy scepticism of system level characterizations. Within this approach, labels such as ‘strong’ or ‘weak’ and ‘interventionist’ or ‘non-interventionist’ applied to states are dismissed as ‘positively misleading’ (Wilks and Wright, 1987:284). For example, Cawson et al.’s study of telecommunications and consumer electronics in France found little to support the traditional ‘strong’ state thesis. They concluded that the ‘rhetoric surrounding the role of the state in France, and a good deal of academic commentary, assumes a level of co-ordination and a capacity for concerted action within the French state which our research suggests is the exception rather than the rule’ (Cawson et al., 1987:33). Vogel’s analysis directly challenged the laissez-faire images used to describe North American industrial policy. He argued that ‘America does have a highly developed set of industrial policies, which on balance, appear to be no more or less coherent, consistent, or successful than those of its major industrial competitors’ (Vogel, 1987:112).² Several studies since the mid-1980s have confirmed the suspicion that policy-making patterns are likely to vary between sectors as well as between nations: this includes evidence of corporatist structures in some sectors (see Cawson, 1986). In contrast, Smith (1990) describes a state-led programme of price support policies in British farming which suggests anything but a weak state in agricultural policy-making.

In the sectoral approach the idea that ‘politics determines policy’ is rejected in favour of Lowi’s famous argument that ‘policy makes polities’ (Lowi, 1964). This suggests a reorientation in the research goals towards identifying recurring policy problems and challenges faced by policy participants, and examining the structure of ‘sub-governments’, ‘policy networks’ or ‘policy communities’ which cluster around particular policy sectors. Within these arenas state actors and groups form clientelistic relations and attempt to insulate and segment the policy agenda. Thus, the political system is not open to broad characterizations but breaks down into a series of sub-systems in which it may be possible to identify a diversity of policy-making styles and traditions.

The sectoral approach also expects that over time each policy sector or industry will evolve a set of governing structures which will mould and shape the politics of the sector (Campbell et al., 1991). Thus, in terms of the wider debate on what shapes industrial policy, the sectoral approach ‘predicts differentiation within individual countries across sectors and con vergence across nations within sectors’ (Freeman, 1985:486). This reorientation in the goals of public policy research (which preceded the specific debates over government-industry relations) has produced a number of neo-pluralist and neo-corporatist ‘meso’ or sectoral-level models of policy-making. Within this, the policy network model has emerged as a dominant paradigm for organizing detailed empirical evidence and exploring the diversity of policy-making patterns within subsystems. For our purposes the policy network approach should be seen as an extension of the sectoral approach to government-industry relations.

Policy network analysis

The policy network label refers to the clusters of actors forming around specific policy areas and programmes within the political system. If one adopts Freeman’s useful concept of policy sector to indicate a broad and demarcated policy area within advanced societies (Freeman, 1985), then the policy network label can be reserved to describe the various interactions between actors within a policy sector. The policy network terminology is not without its problems because the key concepts have different meanings in different hands. The literature is currently divided over the importance or otherwise of interpersonal relationships and whether networks exist at the sectoral or sub-sectoral or even the micro level. These debates are well reported elsewhere.³

Richardson and Jordan (1979), following Heclo and Wildavsky (1974), originally introduced the term ‘policy community’ to convey the image of a sub-government network with a relatively stable membership of actors with broadly shared values on particular policy issues. The actors in the policy community shared various interdependencies and were integrated to the extent that their common ambition was the containment of conflict within network parameters. Thus Jordan contrasts a policy community with an issues network suggesting that a policy community is:

a special type of stable network which has advantages in encouraging bargaining in policy resolution. In this language the policy network is a statement of shared interests in a policy problem: a policy community exists where there are effective shared ‘community’ views on the problem. Where there are no such shared attitudes no policy community exists, [original emphasis]

(Jordan, 1990:327)

Jordan suggests that policy communities develop around specific issues rather than around policy sectors which may be dealing with a number of issues over time. The thrust of his argument is that while conflict exists within policy communities it is constrained by an underlying agreement on the legitimacy of the views of other participants and the need to insulate the policy sector. In contrast, Heclo’s issue network concept (Heclo, 1978) described a much less ordered pattern of actor interaction in the policy process, w...