Co-Ordination in Context

Institutional Choices to Promote Exports

G.P.E. Walzenbach

• Publisher: Routledge
• Year: 2019
• ISBN:
  9780429873218
• Edition: 1
• Topic: Économie
• Subtopic: Exportations et importations

First published in 1998. This book makes an original contribution to our understanding of policy failures at the European and international level. On the basis of a comparative analysis the study shows how the co-ordination mechanisms available in the European Community and OECD have complicated the regulation of national policies on state aid to exporting industries. This failure can be explained in theoretical terms: international and supranational organisations are not neutral arbiters, but have interests of their own, interests which are not necessarily aligned with those of their member states. In detailed case studies of Britain, France and Germany the book examines how the preference structure of governments in the exercise of their promotion programmes contrasts with the policies enacted by international bureaucracies. Walzenbach's interdisciplinary approach specifies the conditions under which policy co-ordination can have detrimental effects and thus, usefully corrects the benign view held by most regime theorists about transaction-cost reducing and efficiency enhancing role of such arrangement.

PART I
THEORETICAL
FOUNDATIONS

1 Welfare and Institutional Design

Whatever the specific context in which they occur, problems of co-ordination are a basic feature of social life, because they reflect the need to sequence several different actions in order to arrive at certain goals. Far from providing definite solutions to these problems, the development of institutions over time will be interpreted here as an attempt to increase the capabilities of actors to structure their own behaviour in a way consistent with their aims. As Dahl and Lindblom (1976, p. 38) have put it:

An action is rational to the extent that it is ‘correctly’ designed to maximize goal achievement, given the goal in question and the real world as it exists. Given more than one goal (the usual human situation), an action is rational to the extent that it is correctly designed to maximize net goal achievement. When several actions are required to attain goals, rationality requires coordination; that is, the actions must be scheduled and dovetailed so that net goal achievement is not diminished by avoidable conflicts among the actions.

Traditionally this task was conceived of as taking place within the boundaries of the nation-state, but has increasingly been reformulated within the context of international institutions. This leads to second order co-ordination problems, since rule-making at this level also has to be compatible with those rules already in place at ‘lower’ levels, i.e. within the jurisdictions of individual states. In addition, the variations in the degree of institutionalisation of international negotiation systems means that the analysis should include the interaction between them as well, thus producing a co-ordination problem of a third type. Before discussing the latter in Chapter 2 we can look first of all at problems of co-ordination occurring on levels one and two.

1.1 First and Second Order Co-ordination

First order, or domestic, co-ordination frequently implies negative co-ordination, a term coined by Scharpf (1991) to describe conflict avoidance rather than consensus formation in national policy-making. In terms of strategy, this involves the minimisation of potential conflicts through a process of sequential decision-making, whereby issues are divided into smaller segments to reach agreement more easily. The process has the advantage of reducing the complexity of policy issues by keeping functionally related policy decisions separate and accommodating only those new initiatives which are compatible with the already existing, fixed positions of the participating government agencies. Established interests are recognised in return for enduring exchange relationships. By focusing on governments, the approach rejects the unitary actor assumption concerning states, which are instead seen to provide formal structures within which interests can be articulated. The nature of the domestic co-ordination process, however, will not always produce coherently fixed state objectives and negotiating capacities; nor will they be without changes over time, across issue areas and among states. This suggests that we first need to investigate empirically exactly how government policies are formulated within a domestic context. Here, we can expect to show that differences in domestic structure lead to political choices that nevertheless display recognisable patterns across issue areas and countries.

Second order co-ordination refers to the increasing number of multiple and functionally overlapping negotiation systems which are usually examined under the heading of regime analysis, but which are better and more comprehensively characterised as belonging to the domain of international institutions. Keohane (1993, pp. 28–9) offers a definition that describes them as:

persistent and connected sets of rules (formal and informal) that prescribe behavioral roles, constrain activity, and shape expectations. International institutions include formal intergovernmental or transnational organizations, international regimes, and conventions. International organizations are purposive entities, with bureaucratic structures and leadership, permitting them to respond to events. International regimes are institutions with explicit rules, agreed upon by governments, that pertain to particular sets of issues in international relations. Conventions are informal institutions, with implicit rules and understandings, that shape the expectations of actors.

Common to nearly all international institutions is a non-hierarchical, network-type organisational structure, which inevitably leads to questions of effectiveness and efficiency relating to the management of policy interdependence. While these multiple and overlapping systems have a welfare potential residing in the reduction of transaction costs, it is by no means clear that use will always be made of it. Again, this ultimately represents an empirical question that necessitates a thorough analysis of the relationship between the national and the international, rather than simply assuming the whole issue is one of the erosion of state power. At the same time, it is possible to put forward some theoretical arguments concerning second order co-ordination in support of a critical analysis of international institutions.

One of the major shortcomings of regime analysis stems from an overly optimistic expectation as to what happens when institutions over time take on a life of their own and only incompletely express the interests and objectives of a hegemonic state. By maintaining an essentially top-down perspective, this kind of analysis suggests that the rules established by a regime would result in a redefinition of state interests, at least in the long run. However, whereas in most cases it will be impossible to clarify unambiguously the extent to which policy was driven by national interests or international norms, a more plausible approach would be to assume the existence of a dialectical relationship, putting more emphasis on the way state actors actually define their interests. In the words of Katzenstein (1990, p. 18), ‘tracing the policy process makes it possible instead to uncover how interests are defined and redefined’. In doing this it makes more sense to start from the bottom up, thus leaving more room for the contextual factors that determine policy-making during crisis situations and over longer periods of time. This approach enables us to investigate more precisely to what extent the interest structures of state actors who do participate in different, but linked, negotiation systems correlate to their institutional choice. In the final analysis, if a relationship of this type can be established, it then has some instrumental value, in so far as it can be used as a prescription for reform. Once changes in the interest definition of states have occurred, it becomes possible to rethink institutional arrangements and probably reshape them in the light of new policy goals, which is undoubtedly relevant for all those policy areas where the interdependence of markets requires some kind of regulation. At the same time, the complexity of international policy issues makes it necessary to concentrate the study on a single policy area.

Using game theory as their major analytical tool Putnam (1988) and Tsebelis (1990) have both examined two or more co-ordination levels in combination. For our purposes the importance of the Putnam model lies in its incorporation of domestic interests, in order to explain the strength and weakness of international institutions. International rules that are supported by stable domestic coalitions are likely to be characterised by a higher degree of compliance, whilst governments are assumed to act sequentially at the domestic and international levels. In the national context, domestic groups pursue their interests by pressurising government agencies to adopt favourable policies, and politicians seek power by constructing coalitions among those groups. At the international level governmental actors attempt to maximise their own ability to satisfy these domestic pressures, while minimising the adverse consequences of foreign influence on their own policy conduct.

By way of contrast, analysing co-ordination using the nested game perspective of Tsebelis allows for the simultaneous combination of multiple arenas. Starting with a game situation in the domestic setting as the principal arena, the other arenas at the international level are considered as contextual factors that influence the payoffs to be gained by each relevant player at the national level. For example, a game at the national level taking on the form of a Prisoner’s Dilemma between firms and governments is at the same time nested inside a bigger game at the international level, such as a Battle of the Sexes between European governments about which co-ordinated solution to choose. The size of the payoffs (not the order) in each of these games is variable, but the final outcome in the primary domestic game is determined by events in the other arena. Consequently, the payoffs (not the nature) of this end game can change, depending on the amount of benefits available in the other setting. Moreover, other contextual factors - i.e. whether or not the actors can communicate effectively with one another, and whether or not the game is repeated over time - can have a similar effect. This line of reasoning has its identical expression in the concept of externalities (Tsebelis, 1990, pp. 58–60): rather than viewing the interaction between two actors in isolation, the consequences to third parties are usually not insignificant either, as they in turn react to the initial game.

International institutions can constitute such a third party. As the Coase theorem suggests, however, it has to be noted that the presence of externalities alone does not necessarily prevent effective co-ordination between independent actors. Welfare enhancing results are still possible, as long as negotiations are allowed for and compensation payments made possible. This assumes that national governments are willing and able to consider the international repercussions of their policy choices, that they have perfect information as to one another’s intentions; and that the costs of organising and monitoring policy co-ordination were negligible. Market failures would then be managed in a co-operative fashion, without the need to delegate powers to an authority at the international level. Although this would at best ascribe a minimal role to institutions, the observable diversity in reality of both hierarchical systems such as the state and also negotiation systems (as embodied in various different international organisations) point to a more balanced analysis of the exact role of the latter.

Institutions can have detrimental effects, a hypothesis which is also to be found in co-operative game theory. Co-ordination problems can exist whenever the pay-off to a single actor is partially determined by the action taken by another. Furthermore, it is assumed that specific actions are not taken simultaneously, but in reaction to each other. The outcome of such interdependent but uncoordinated action will in general be suboptimal, that is to say both parties gain less than in a situation where they try and prearrange some sort of mechanism to bring their individual behaviour closer into line. Sometimes, basic communication can be sufficient to make participants aware of the potential for realising gains (Kydd and Snidal, 1993). More often, however, it will require co-ordination schemes in the form of institutional arrangements to provide a means of enforcement and arbitration. The former allows for either the imposition of sanctions on those who ignore, or the offering of incentives to those who respect, certain agreed rules. The latter involves the interaction of an intermediary between the parties facing a co-ordination problem. Ideally, this neutral arbiter combines the two functions of selecting a co-ordinated solution and of enforcing that particular choice (Calvert, 1992). The notion of an impartial arbiter, however, will only be applicable in the unusual circumstances of some third party having an interest in their dispute being resolved, but no interest in how it is actually settled. More frequently, the intervention of a third party with its own mixture of conflicting and common interests will make matters more complicated (Goodin, 1976, p. 48; Schelling, 1980, p. 14).

In consequence, where the role of an institution is limited to that of an external mediator, the resulting strategy is to rely on the services of (ideally) competent and impartial actors, such as teams of independent experts acting as honest brokers, who provide information on various co-operative options, and evaluate and compare them with regard to a commonly acceptable solution. These improvements in the information base of international negotiations are in themselves nonetheless insufficient, because there is a second aspect to ensuring the co-operation of an institution. The communication mechanism which provides the information is also the precondition for knowing when the policing and the punishment of non-co-operation is required, but although this resort to coercion should in theory benefit the community as a whole, its actual implementation brings with it a potential for misuse. The alternative is a purely voluntary institutional system which, notwithstanding the advantages of lowering the information cost about other parties, can be rejected, because it would entail rapidly increasing transaction costs (North, 1990). Section 1.2 will discuss this argument in more detail, which North presents in a national context for international institutions. Here, I will put forward some further theoretical arguments as to why institutional neutrality cannot be expected.

It should be noted that even if mediators are by definition only in a position to give suggestive rather than authoritative instructions, they already gain an influential role through their control over communication. Or as Schelling puts it (1980, p. 301):

If we give a third player power to send messages to the original two tacit players, he is in a good position to help them; he is even in a good position to help himself if he gets a payoff that depends on the pair of strategies that the original two players choose.

While this still falls short of manipulation, the activities of a potential mediator do not just comprise putting a constraint on communications. There might be an extension to the limitation on the order of offers and counter-offers, for instance, by simply inventing ‘contextual material’ to support certain suggestions. In fact the mediator’s position opens up ample possibilities to ‘influence the other player’s expectations on his own initiative, in a manner that both parties cannot help mutually recognizing’ (Schelling, 1980, p. 144). Nothing other than outright manipulation occurs if part of the information is suppressed. This, however, can be done in very subtle ways such as inconsistently activating the institutional memory. Obviously, it is not necessary to convert mediators into arbitrators by an irrevocable surrender of authority for these effects to be produced; less severe actions can still satisfactorily achieve the desired result. Goodin (1980, p. 214) argues that in circumstances where there is a conflict about which co-ordinated solution should be chosen, the information need not be biased in favour of the desired alternative, nor even be on balance favourable. It is sufficient that ‘everybody be talking about it’ for an alternative to emerge as the obvious solution. Goodin’s emphasis is on established elites, who alone are able to systematically work this politics-of-the-obvious strategy to their advantage.

In theory, the case of third-party enforcement - the second aspect of institutions - likewise involves a neutral party with the ability to measure the attributes of a contract-like arrangement and to enforce its content. Although in principle it makes no difference whether institutions mediate or arbitrate in the latter, as North (1990, p. 58) argues, it is more recognisable that:

the enforcer is an agent and has his or her own utility function, which will dictate his or her perception about the issues and therefore will be affected by his or her own interests.

This issue of interpretation is superseded, and its consequences further aggravated, once sufficient attention is paid to the impact of ambiguity and unanticipated contingencies in the implementation of substantive rules. With these kinds of ‘incomplete contracting problems’ disputes between the parties are not directly covered by the mandate of institutions, but if they nevertheless claim (or actually possess) autonomy to make the final decision, ‘there is nothing to stop them from abusing their positions’ (Weingast and Garrett, 1993, p. 185). Potential correctives in the form of simple non-compliance, legislative action or new constitutional safeguards can be expected to face the resistance of actors pursuing their own institutional self-interest.

Another form of criticism as regards the role of international institutions deals with the third party problem from the viewpoint of the originally participating parties. Wassenberg (1991, p. 261) uses a ‘triad power approach’ according to which actors are not involved in exchange transactions, but in decisions as to whether or not to invest in the (re)-production of mutual power relationships. The availability of a third party, (i.e. the international institution) is just a potential substitute for an existing coalition, in order to maintain the capacity to enlarge or restrict the room for strategic choices, for oneself or for others. Following the argument of the seminal work by Tversky and Kahneman (1986), institutions can be understood to be a deliberate choice by the parties seeking co-ordination to frame a problem in a particular way, including some factors for the assessment of gains and losses while excluding others. The respective institutional setting to which preference is given a priori partly determines how the consequences of a co-ordinated policy are going to be evaluated. Gygi (1991, pp. 114–8) addresses the issue in terms of a principal-agent framework. The principal delegates certain tasks which are of benefit to him or her to someone else - an agent. In the case of international institutions, a threefold principal-agent relationship exists. Citizens delegate to their governments, governments to diplomats or representatives within an international organisation, and finally these in turn delegate to the managers of the international institution. This delegation causes transaction costs because of asymmetrical information on the part of the agents. Agents are assumed to use those comparative advantages for their own purposes, with the consequence that the more the levels of delegation, the greater will be the cumulative negative effects. Although the principal-agent model can still provide a rationale for delegation if information and credibility gains outweigh the costs of political control (Ordeshook, 1992, p. 122), it requires some important qualifications in the context of international institutions.

Firstly, over time governments try to change the original agreements which led to the creation of an international institution to their own advantage. As Section 1.4 discusses in greater detail, regimes for example aim at a long-term binding of governments instead of giving way to short-term preferences. Likewise international organisations frequently try to stabilise contracts between governments against the immediate interests of the contracting parties. Under these circumstances the problem is to distinguish between the flexibility of international institutions that is actually desired and that which serves only particular interests. Secondly, doubts exist as to whether national representatives in international organisations exercise the control function as demanded by their home governments (Vaubel, 1986, p. 51). Due to the particular position they maintain in the decision-making process there will be strong incentives to exploit information asymmetries to their own advantage. Thirdly, given difficulties in evaluating the output of international organisations, governments focus more on their individual contributions rather than on an org...