http://eiop.or.at/eiop/texte/2005-013a.htm
: 14.10.2005| This paper's comments page | Send your comment! to this paper |
| Synergies and Tradeoffs in International Cooperation: Broadening, Widening, and Deepening(*) |
| Jörg Friedrichs, Jordan Mihov and Maria Popova |
| European Integration online Papers (EIoP) Vol. 9
(2005) N° 13; http://eiop.or.at/eiop/texte/2005-013a.htm |
| Date of Publication in
: 14.10.2005 |
| | Abstract |
Back to homepage |
PDF | | This paper's comments page | Send your comment! to this paper | |
1
International institutions are created to meet a variety of challenges. To tackle these challenges, states broaden the scope of their activities, enlarge their membership base, and reform the institutional setup of international regimes and organizations. This is in line with a more general feature of human organizations, from neighborhood associations and self-help groups to industrial companies and international organizations: they are all distinguished by a complex relationship between the scope of their activities, the range of their membership base, and the depth of members’ commitment. In a word, any organization is shaped by a specific relationship among the following three components: substantive scope, membership range, and institutional depth.
In such a perspective, organizational growth means deeper collaboration and wider membership for the achievement of more substantive goals. From an optimistic viewpoint, the three components are mutually reinforcing. A pessimist, however, would object that it is not less plausible to assume tradeoffs rather than synergies between substantive scope, membership range, and institutional depth. Either way, there seems to be a familiar mechanism at work here. Actors have preferences which they cannot realize without collective action; based on these preferences they shape and re-shape social structures; once these social structures are in place, they shape and re-shape the actors themselves and their preferences (Giddens 1984; Wendt 1987).
European integration is a standard example. Over various decades, the European Community has considerably expanded its institutional depth and membership range, while increasing its competencies to policy fields that were previously regarded as the exclusive prerogative of national regulation and control. There has been a constant increase of the substantive scope, geographical range, and institutional depth of European integration. Of course this does not mean that European integration is necessarily and always a good thing. However, European integration is an ideal object to study the complex relationship between substantive scope, membership range, and institutional depth. Both in the theory and in the practice of European integration, substantive broadening, geographical widening and institutional deepening have been, and are still, the nuts and bolts of the permanent reform initiated in the mid-1960s.
2
The crunch question, however, is whether there are synergies or rather tradeoffs between broadening, widening, and deepening. Concerns about inappropriate timing and counterproductive integration mechanisms, as well as the weak impetus for institutional reforms have initiated a broad political discussion. Pessimists treat the prospects for significant progress with skepticism. Nevertheless it is significant that, despite these concerns, up until now European integration has been moving on. After the accession of ten new member states in 2004, however, the intricate interdependence and harmonious evolution of the three dimensions is more than ever crucially important for the proper performance of the EU, and possibly even for its future existence. If the pessimists are right, the European Union must be careful not to shipwreck in the “Bermuda triangle” of substantive broadening, geographical widening, and institutional deepening. If the pessimists are wrong, the prospects look much better.
In the first section of the article we define and explore, with particular regard to European integration, the relationship between substantive broadening, geographical widening, and institutional deepening. We formulate two competing hypotheses, the first assuming a virtuous and the second assuming a vicious circle between the dimensions. In the second section, we narrow down the object of our analysis to state preferences with regard to each of the three dimensions. The third section is dedicated to the selection of case studies. The objective is to conduct a structured, focused, and historical comparison (Mahoney 2003; George and Bennett 2005) of the preferences of the four largest EU member states in a particularly salient policy field: the international fight against drugs. In the fourth section we run a formal analysis of our dataset, which is specially designed to assess the two hypotheses against each other. It turns out that, at least in the policy field examined, there are more synergies than tradeoffs between preferences on substantive scope, geographical range, and institutional depth. We also discuss whether and to what extent our findings can be generalized.
As already mentioned, organizational growth in the EU depends on the triangulation of substantive broadening, geographical widening, and institutional deepening.
For the supporters of European integration, there is (or should be) a virtuous circle of organizational growth. Against this, Euro-skeptics point to possible tradeoffs. In their view, an increase in substantive scope will eventually make it more difficult to reach a common denominator on institutional reform; expanded membership will impede the further broadening of the scope of activities; and the possibility to reform the existing institutions is hampered by enlargement; insufficient institutional reforms, on the reverse, will curtail the substantive output of legal norms and policy decisions.
Insofar as the European Union provides an organizational structure to facilitate political intercourse among its members and thereby helps to overcome collective action problems, all actors concerned are interested in a balanced evolution of the three dimensions. At the same time, however, the member states are also struggling to preserve their autonomy and often have preferences that run counter to the further evolution of a structure that increasingly conditions and shapes their room of maneuver. Depending or the issue at hand, there may be disagreement or whether to privilege substantive broadening over geographical widening, geographical widening over institutional deepening, and institutional deepening over substantive broadening. This creates a series of dilemmas that can affect the organizational evolution of the EU.
It is an empirical question whether there are synergies or rather tradeoffs between the three dimensions. Accordingly, let us formulate two competing hypotheses to be placed under empirical scrutiny.
Hypothesis 1: There are synergies. The three dimensions are related to each other by a virtuous cycle of positive feedback. When substantive scope broadens, geographical range widens; when geographical range widens, the institutional commitment deepens; when the institutional commitment deepens, substantive scope broadens (and vice versa).
3
Hypothesis 2: There are tradeoffs. The three dimensions are related to each other by a vicious circle of negative feedback. Substantive broadening has a negative effect on geographical widening; geographical widening hampers institutional deepening; institutional deepening runs counter to substantive broadening (and vice versa).
In order to assess the plausibility of either hypothesis, let us further examine the influence of the three dimensions upon each other. First we provide a picture of European integration in terms of a virtuous circle. Subsequently, we follow the more skeptical strand of the literature in depicting European integration as trapped in a “Bermuda triangle” of tradeoffs between broadening, widening, and deepening.
To those who believe in European integration as a unidirectional historical process, integration in different fields should be, and is, mutually reinforcing. Organized cooperation in more and more policy fields yields benefits because it wards off the undesired effects of isolated solutions. For example economic cooperation without environmental protection is insufficient, and this puts an incentive on states to expand the scope of European integration. The existing institutional and geographical framework will periodically proof inadequate, but fortunately there is a burgeoning political elite at the national and at the European level promoting forceful institutions and geographical enlargement. As a consequence, the dynamics of European integration will generate an increasing demand for broadening, widening, and deepening.(1)
As a matter of fact, the substantive increase of the European Union’s competences since the Single European Act (1986) has been accompanied by periodic institutional changes. Reforms strengthening the European Parliament are regarded by some authors as an important step to counter the “democratic deficit” of the Union (e.g. Scharpf 1999: 9).(2) In addition, the application of qualified majority voting to new policy areas has been helpful to smoothen the decision-making process (Dehousse and Weiler 1990; Engel and Borrmann 1991; Hayes-Renshaw and Wallace 1997).
Over the last decades, the EU has fully established itself as a gravitational center for political and economic life in Europe. The price for non-membership has become prohibitive, and this accelerates the race for membership. There is a strong stimulus for potential member states to implement the acquis communautaire and join the Union. At the same time, wider membership leads to increasing socio-economic heterogeneity. This heterogeneity needs to be regulated, which leads in turn to a further broadening of the substantive scope of EU activities. At the same time, enlargement does also require institutional reforms to make the Union amenable to increased membership. This process creates a synergetic circle in which the three dimensions of broadening, widening and deepening mutually reinforce each other.
From the optimistic viewpoint, there is a virtuous circle among the three dimensions of integration. Enlargement enhances the readiness of the new member states to share the goals and values of the organization, this in turn reinforces the process of institutional reform and makes it possible to bring new policy fields under the competence of the organization, which further increases the incentives for ever more member states to join the Union, and so on and so forth (Hypothesis 1).
Whereas the optimistic vision of European integration as a virtuous cycle is outlined in roughly one page, it is significant that the pessimistic picture is more complex. In the relevant literature, much more ink has been spent on possible tradeoffs. (3)
The geographical enlargement of international organizations is connected to an increasing heterogeneity of state preferences. It is easier for organizations with a small number of members to endow the organization with additional competencies. Large organizations are composed of many members with a variety of preferences that must be accommodated. Therefore, it is unlikely for these members to agree on cooperation in a large spectrum of activities. Members of smaller organizations, by contrast, are more likely to have a similar willingness and capacity to act. This results in a greater certainty about policy outcomes, and therefore, higher incentives to cooperate on a broad range of issues (Downs et al. 1998). There seems to be a negative relationship between substantive scope and geographical range.
4
Moreover, a game theorists would argue that, in a heterogeneous framework, weak actors tend to be reluctant against the organization being endowed with new fields of competence, since unpredictable coalitions of powerful partners may suppress their interests and further increase their dependence (cf. Walt 1987: 17-32). For the same reason, one may speculate that weak actors will be reluctant to join an organization with an exceedingly large substantive acquis communautaire in the first place.
Another problem is that large organizations face the problem of a diverging capacity to execute basic obligations, reflecting socio-economic and socio-political differences among their member states. This problem seems to be particularly important in the context of eastern enlargement, the largest enlargement that the European Union has ever experienced. There is considerable concern that the obligations of the new member states largely exceed their capacity to act, and that particularly the Central and Eastern European countries may have joined the EU not so much because they were convinced about integration as a normative goal but rather because they wanted to reap material benefits from “joining the club” (Balazs 1998: 69).
Insofar as new member states do not see integration as an end but rather as an instrument, however, and insofar as they are economically less developed, their incentives and capabilities to push the integration process forward are substantially weaker. Accordingly, it may be difficult for a EU of 25 to delegate new competences to the supranational level, or even to reform its institutional setup. Perhaps it would be premature to end on such a pessimistic note. Nevertheless, everything seems to depend on the extent to which the new member states share the values and aspirations of the Union, and are willing to comply with their substantive and legal obligations.
The critical interdependence between the institutional and the geographical dimensions of integration is reflected in the post-Nice debate about the future of the European Union. Geographical widening may affect the structure, functioning, and efficacy of the EU institutions, such as the Council and the Commission. For that reason, the Nice Treaty aimed to reform them in a way suitable for the forthcoming enlargement. But unfortunately the plans to reform qualified majority voting complicated and slowed down the implementation of the treaty (Dinan 2002), while eastern enlargement was already imminent. At the end of 2003, the enlargement process came close to failure because of this tension between widening and deepening.
Even though the impasse was overcome at the beginning of 2004, enlargement may still have dire consequences for the European Council and the Council of Ministers, which occupy a central decision-making position in the EU. The ability of the Union to act depends largely on the proper functioning of these bodies. Unfortunately, however, increased membership in the absence of the necessary institutional reforms poses a dilemma between capacity and control (Kerremans 1998). The interests of the member states become more diverse, and this creates an additional burden for the legislative process. At the same time, especially in the case of qualified majority voting enlargement reduces the control of individual states over the content of decisions. Moreover, if the institutional acquis of an organization demands too much from the member states, this will deter third countries from applying for membership.
The significant structural and decision-making problems that the Union is facing may endanger its efficiency. In the worst case it may turn out that enlargement was premature, as significant institutional problems have not been resolved by the deadline. Eastern enlargement is qualitatively and quantitatively different from the previous rounds, and this makes the future of the European Union look even more uncertain. Accordingly, the latest accession round has become subject to widespread political skepticism (Laurent and Maresceau 1998: 1). Deviations from the traditional gradualist approach may turn the integration process into a particularly vicious circle, in which enlargement takes place at the expense of institutional viability.
Despite recent institutional innovations, many scholars express doubts about the ability of the EU to deal with an ever-broadening political agenda. They object that an increase in the substantive scope of the Union may undermine institutional reforms. A number of studies suggest that the decision-making process in the EU has deteriorated considerably as a result of expanding legislative activities. Scharpf (1988) and Sbragia (1993) emphasize the inability of the Council of Ministers to cope with the legislative output of the Commission. The incongruence between the two organs slows down legislative processes, reduces efficiency, and threatens to dilute the substantive content of EU legislation (Van den Bos 1994, Héritier 1996).
5
Moreover, the Maastricht and Amsterdam Treaties have delegated competences to the community level in policy fields that previously had been strictly regulated by national governments, namely in the second and third pillar. Therefore, the Council often finds difficulties in reaching agreement on these issues. As Scharpf (1997) argues, the policy of the least common denominator tends to prevent the realization of effective institutional reforms. In his opinion, even the introduction of qualified majority voting has failed to make the decision making process more flexible.(4) The fear that individual states may further loose control over the substantive outcome of EU decision making further weakens the impetus for structural reforms.
The other way round, even successful institutional reform can reduce the prospects for substantive broadening. Extensive changes in the voting rules may affect the willingness of European governments to broaden the substantive scope of integration. As Bräuninger et al. (2001) point out, different voting rules increase the uncertainty about future policy outcomes. Moreover, qualified majority voting makes it more difficult for individual actors to control the decision-making process. All this is likely to make them more reluctant to delegate further competencies to the EU. The loss of control connected to extensive institutional deepening may lead to a reluctance of the member states to transfer new competences to the European Union.
The negative interrelation between substantive broadening and institutional deepening is most clearly observed in those fields where national interests predominate. In the sphere of Common Foreign and Security Policy (CFSP), improvements are hampered by the “instinct” of national governments to keep control over this particular area (Ham 1997: 309). In this respect, the Commissioner for External Relations Hans van den Broek acknowledged that “the problem [is] how to go beyond joint declarations and really act together to tackle specific international problems. […] It is partly a question of political will but partly also because our sense of joint European interests is still insufficiently developed and because we have not yet learned how to take decisions together quickly and how to carry them out” (quoted in Ham 1997: 309).
Although in many areas qualified majority voting may facilitate decision-making, the legislative process with respect to the more recently introduced policy fields, such as CFSP or Police and Judicial Cooperation in Criminal Matters (PJCCM), is still cumbersome and makes it difficult to agree on the necessary institutional reforms.
From the pessimistic viewpoint, there is a vicious circle among the three dimensions of integration. Substantive broadening hampers the proper functioning of institutions in the absence of the necessary reforms; institutional deepening affects the actors’ willingness to broaden substantive scope; geographical enlargement complicates the legislative process and leads to stagnation, and so on and so forth (Hypothesis 2).
So far we have dealt with possible synergies and tradeoffs in the outcomes of bargaining on the institutional depth, geographical range, and substantive scope of European integration. The pessimistic theory postulates a variety of tradeoffs. A more optimistic theory would emphasize that the three dimensions are mutually reinforcing.
So far, so good. However, in this paper we are not going to deal with bargaining or with the outcomes of international cooperation. Instead, we are going to deal with state preferences on international cooperation. Let us briefly outline the reasons for this analytical choice.
We understand international cooperation as the outcome of a policy cycle. Such a policy cycle starts from a real-world problem. If unilateral dealing with the problem is perceived as too costly, it will induce states to formulate preferences on the desired terms of international cooperation. Provided that there is sufficient consensus for an agreement to seem at all possible, states can be expected to negotiate with each other. In some cases, although certainly not always, the bargaining process will engender a substantive and institutional outcome, which is implemented more or less faithfully. Due to problems of incomplete contracting, voluntary and involuntary defection, and unintended outcomes, more often than not the result of the exercise will turn out to be unsatisfactory. Compliance is not automatic, nor are the “real” problems always being dealt with effectively. Due to these and similar hardships, states will formulate new preferences on future international cooperation, and the cycle starts anew (cf. Lasswell 1956; Easton 1965: 29-33; Howlett and Ramesh 2003).
6
For a full understanding of the political process one would have to start from the beginning of the cycle, i.e. with state preferences, and work oneself through the bargaining stage to the agreements that are finally reached; moreover, one would have to consider the feedback of experiences with implementation and compliance into new preferences in a new policy cycle. An isolated analysis of bargaining and/or outcomes will not do the trick if one wants to gain a full understanding of what makes international cooperation work. While it is of course not possible to discuss the entire policy cycle in the context of the present paper, we would argue that the most promising starting point is carefully to analyze the initial preferences of the actors involved. Although it is impossible directly to deduce outcomes from preferences, preferences rather than bargaining or outcomes are the linchpin of the policy cycle.
In line with these considerations, preferences can be defined as “the ways actors order the variety of possible outcomes of an interaction” (Frieden 1999: 42). According to Moravcsik’s (1997) liberal theory of international politics, international cooperation is the result of a two-stage process. In the first stage, state preferences take shape in response to pressure from interest groups embedded in society. The second stage constitutes international bargaining on substantive issues and institutional choices.(5) If one adds geographical enlargement to the list, and if one assumes that it is reasonable to draw an analytical distinction between deepening, widening and broadening already at the stage of preference formation, one has a ready-made framework at hand.
In this paper, we focus on the interplay between the three dimensions of international cooperation at the initial stage of the policy cycle, i.e. before the bargaining stage. We call this preference configurations. We argue that there can be synergies and tradeoffs not only at the bargaining stage or with regard to the outcomes of interaction, but already with regard to the very preferences the actors bring to the negotiation table.
Preferences as intended outcomes are related to broadening, widening, and deepening. Accordingly, there are three types of preferences: preferences on substantive scope, preferences on geographical range, and preferences on institutional depth. Preferences on substantive scope regard the gamut of issues a state wants to be subsumed under a regime. Preferences on geographical range are concerned with the number of states with which a state would like to cooperate under the intended agreement. Preferences on institutional depth refer to a state’s willingness to accept binding commitments. Depending on whether there is a positive or negative relationship between these three dimensions, there will be different patterns of preference configuration.
This leads to a modified version of our initial hypotheses. The modified version of Hypothesis 1 suggests that preferences on each of the three dimensions are mutually reinforcing. From this perspective, a state that wants to cooperate with more states will also endorse the necessary institutional reforms and support the inclusion of new items under the remit of an international regime. The modified version of Hypothesis 2, by contrast, would lead us to expect that preferences emphasizing institutional deepening, geographical widening, and substantive broadening take place at the expense of each other. Both hypotheses sound plausible, and it is an open empirical question whether and to what extent we are going to observe synergies or rather tradeoffs between state preferences on deepening, widening and broadening.
To recap: just like inter-state bargaining and its outcomes, state preferences on international cooperation can also be analyzed with respect to substantive scope, geographical range, and institutional depth. An empirical analysis of the interrelationships between these three dimensions is an important precondition for a true understanding of international cooperation (understood as a policy cycle). In the next two sections, we provide an analysis of state preferences for the particular policy field of international cooperation against drugs. Moreover, we discuss whether and to what extent our findings can be applied to other fields of international cooperation.
Our dataset allows the examination of state preferences with regard to international cooperation against drugs. As a matter of fact, international police cooperation in general, and the fight against drugs in particular, are highly topical fields for the examination of state preferences. Insofar as the monopoly of force is the defining characteristic of modern statehood (Weber 1968, 1992), international police cooperation is one of the most crucial fields of international cooperation. With the exception of the military field and maybe taxation, no other sphere is so directly related to national sovereignty. We have selected international cooperation against drugs because, together with terrorism and illegal immigration, it is one of the three most important challenges to international police cooperation (Mitsilegas et al. 2003; cf. Occhipinti 2003).
7
In our analysis, we have chosen to examine the four largest EU member states: Britain, France, Germany, and Italy. These four countries are suitable for comparison because they are similar in all regards other than political system. Not only are they the most relevant players in the European Union, they do also share a similar socio-economic, cultural, and political background. Moreover, they face common challenges and cooperation problems. However, they vary in their willingness to engage in deepening, broadening, and widening the international regime to fight drugs.
The preferences of our four countries are analyzed across two time periods (1960s/1970s and 1990s/2000s). From a European perspective, the fight against drugs has become a central issue only in the 1960s and 1970s. Until then, drugs were considered to be mainly a problem of the United States, with its huge addict population. By the early 1970s, the European states started to be genuinely concerned about drugs and to see the problem with similar eyes as their American partners. Over time, this led to the adoption of three UN conventions on different aspects of drug trafficking.(6)
During the late 1990s, however, European attitudes started to change. Since then, the American hegemony over the international drug prohibition policies has been challenged by the emergence of an alternative European drug regime (Elvins 2003). Whereas the US continues to favor strict law enforcement, most EU member states are now promoting the so-called balanced approach. The idea is to combine the intentional fight against the supply of drugs with appropriate measures of demand reduction at home, which can either take a more repressive or a more preventive form.
The fight against international drug trafficking takes place at three different levels, namely at the level of general principles to convey legitimacy to the struggle; at the level of finding the right methods to pursue this struggle; and at the level of designing adequate frameworks for the authorization of joint international operations against drug criminals. At each of these three levels, we have tried to identify the most important topic in either of the two time periods under consideration (table 1).
On the first level, the UN is traditionally the most important source of legitimacy in the fight against narcotic drugs and psychotropic substances. The UN is also the venue where the international drugs problem is authoritatively defined in the first place (Thomson 1995: 222-223). In the 1960s and 1970s, there was a US-led attempt to boost the existing UN anti-drugs regime. We have analyzed the response of different European states to this American initiative. At least initially, drugs were considered a typically US-American problem and there was only limited interest on the part of European states in the construction of the emergent regime. But as European societies were increasingly faced with the same drug problems as the United States, the UN approach to drug prohibition was gaining sway. More recently, however, several European countries are fairly unhappy with the UN approach. Some of these states, among them Britain and Germany, have started an attempt to emancipate themselves at least in part from the UN regime. As a consequence, an emergent EU drugs regime is challenging some important components of the UN regime, which formally is still the legal umbrella of international drug prohibition (McAllister 2000: 215-246).
On the methods level, the 1970s saw an unprecedented diffusion of the investigative techniques used by American drug enforcers to the rest of the world (Nadelmann 1993, 1997). These techniques included, among other things, intrusive forms of surveillance, extensive undercover operations, controlled delivery of illicit drug consignments, and reduced charges or immunity from prosecution to known drug dealers in order to “flip ” them into becoming informants. In the face of this challenge to their legal systems, different European states reacted in different ways. Some states, such as Germany, tended to embrace the new investigative techniques, whereas other states were more reluctant. Nevertheless, by the 2000s American drug enforcement techniques have been largely accepted in all European states. The cutting edge of international drug enforcement is now the fight against criminal finance, i.e. the idea of depriving drug criminals of their profits by means of anti-money-laundering measures and asset confiscation. There are different preferences on international cooperation against criminal finance, with Britain and France assuming a leading role.
Whereas the UN regime and American investigation techniques were rapidly advancing, in the 1970s operational cooperation against drug trafficking had not yet gone much beyond episodic trans-border operations against criminal gangs. An important exception were US-French and US-German cooperation agreements after the impasse over the so-called “French connection”, a drug trafficking ring based in Marseilles. Over the 1980s and 1990s, however, a dense network of bilateral and multilateral agreements for the authorization of operational police cooperation has been created. The latest achievement in Europe is a multilateral legal umbrella for joint investigation teams.(7) If joint investigation teams become a routine tool for drug enforcement in the EU, the authorization of operational police cooperation will be placed for the first time within a multilateral legal framework. However, different EU member states have different preferences with regard to the practical application of the new legal instrument. Moreover, the usage of the instrument is optional and not all European states are equally keen on recommending it to their own national police forces.
8
We will now investigate the configurations of state preferences with regard to fighting drugs. As we have seen, the dataset consists of 24 case studies (2×4×3), which span across two time periods (1960s/1970s, 1990s/2000s), four countries (Britain, France, Germany, Italy), and three levels of abstraction (legitimacy, methods, operational).
For every single case study, we have conducted a careful analysis of the documentary evidence. This includes public archives and UN debates for the 1970s, and governmental press releases and other official documents for the 2000s. In some cases, we were forced to use other sources such as press agencies, newspaper articles, and personal interviews. Whenever possible, we have given precedence to primary sources. Since it is beyond the scope of this article to trace the way how our data has been gained, the gentle reader is referred to Appendix 1 for further information.
The aim of the data analysis is to place our two hypotheses under empirical scrutiny. We do this by means of a correlation analysis. For the purpose of statistical analysis, the two hypotheses are operationalized the following way:
Hypothesis 1: There are synergies. Preferences on scope co-vary with preferences on range and depth, i.e. there are three positive correlations.
Hypothesis 2: There are tradeoffs, i.e. there are two negative correlations between the three dimensions, whereas the third correlation is positive.(8)
If the statistical analysis does not provide evidence in favor for either hypothesis, then the Null hypothesis (no significant relationship between the indices) must be upheld.(9)
To make our dataset amenable to statistical scrutiny, we use three indices that are especially designed for this purpose. Each index reflects the preferences of a state on one of the three dimensions. The first index measures the preferences with regard to substantive scope; the second index is dedicated to geographical range; and the third index covers institutional depth. The higher the score of an index, the more the country concerned supports the issue the index stands for. The scores are then converted into ranks to make them amenable to rank-order correlation.(10)
Subsequently, we use rank-order correlation to test which of our two hypotheses is valid. Basically, we compare the ranks of different countries on each of the three indices and try to identify regularities across the indices. For example, whenever a particular country has a high rank on one index which correlates with a high rank on another index, then this points to a synergetic relationship between the two categories.
When run over a sufficient number of empirical cases, the statistical method of rank-order correlation can tell us how strong a relationship is, and whether a relationship stands the significance test.(11) In order to accept the first hypothesis, we need to find strong and significant positive relationships among the indices. For the second hypothesis to be confirmed, we would expect two negative and one positive relationship, all of which should be strong and significant.
Table 2 presents the general results. First of all, it is important to note that all correlations are positive. In particular, the correlation between preferences on substantive scope and institutional depth is very strong and highly significant (at the 1 % level). The correlation between preferences on geographical range and institutional depth is also relatively strong and fairly significant (at the 5 % level). Only the relationship between substantive scope and geographical range is the rather weak and hardly significant. All in all, however, Hypothesis 1 is clearly vindicated.
At first glance this looks like an all-out victory for the virtuous circle in the ‘struggle’ between Hypothesis 1 and Hypothesis 2. The supporters of Hypothesis 2 need not despair, though. A careful analysis shows that, at least in one particular sub-set of our sample, there are tradeoffs rather than synergies.(12) When we analyze the data for France, an interesting pattern becomes visible. The correlations between scope and range, and between scope and depth, are strongly negative. The correlation between scope and depth is significant at the 5% level. Thus, France is the “odd one out”: its preference configurations are based on tradeoffs rather than synergies (Table 3).
9
But if we turn back to the overall results (Table 2), there is an interesting implication of our results. As we have already seen, the strongest and most significant correlations in our dataset are those between broadening and deepening (scope depth), and between widening and deepening (range depth). The correlation between broadening and widening (scope range), by contrast, is neither very strong nor significant. This suggests that preferences on institutional depth is the master variable. In line with these considerations, it is possible to provide a ‘consolidated’ version of the virtuous cycle that is the dominant pattern within our dataset (Figure 3).
To what extent do these findings apply to other policy fields? We would argue that the fight against drugs is an unlikely case for a harmonious relationship between broadening, widening, and deepening. The reason for this unlikelihood is the close link between crime fighting and the monopoly of the legitimate use of force, which is the defining characteristic of sovereign statehood. One would expect that, the more a state wants to increase the substantive scope and geographical range of drug enforcement, the more reluctant it becomes to undergo binding institutional commitments. Despite this expectation, however, we find synergies rather than tradeoffs. Pending further research, our findings are therefore likely to be applicable more widely. (13)
Are our findings likely to apply beyond the European Union? Although we have used the EU for the generation of our hypotheses in the first section, a closer look at the case material examined in the present section reveals that the empirical basis of our analysis extends far beyond the European Union. The case sample includes many instances of international cooperation beyond the EU, such as the UN drug prohibition regime, the fight of the G7/G8 and the OECD against money laundering, and a bilateral cooperation agreement between France and the United States in the early 1970s. Although we have examined the preferences of the four largest member states of the European Union, our empirical horizon is not confined to the EU.
Finally, it is important to stress that the empirical analysis has been dealing with preferences, not outcomes. The four states examined are important players, but none of them is in a position to dictate the terms of international police cooperation. There is no direct way to derive preferences from outcomes. While preferences are at the beginning of the policy cycle, outcomes are towards its end. In between are the stages of bargaining, implementation, and compliance. An understanding of preferences is not sufficient to understand outcomes. However, it is an important precondition for the understanding of the subsequent stages in the policy cycle, including outcomes.
In this paper we have been trying to combine the best of three worlds: the empirical examination of qualitative evidence, grounded in sound theoretical concepts and supported by reliable statistical methods. We have identified two competing possibilities of viewing international cooperation in general, and European integration in particular: either as a virtuous circle of synergies, or as a vicious circle of tradeoffs. The former possibility (Hypothesis 1) represents the optimistic view that broadening, widening, and deepening are going hand in hand. The latter possibility (Hypothesis 2) reflects the pessimistic conviction that the three dimensions are sitting uneasy with each other. In the remainder of the article, we have presented an analysis of preference configurations, i.e. issue-specific constellations of state preferences concerning the substantive scope, geographical range, and institutional depth of inter-state cooperation in a particular policy field: the international fight against drugs.
Is there any significant relationship between the three variables? And if so: are there synergies or rather tradeoffs between them? After a careful analysis of a specially designed dataset concerning international cooperation against drugs, it turns out that Hypothesis 1 is vindicated, whereas Hypothesis 2 and the Null hypothesis are rejected (only in the case of France, Hypothesis 2 prevails over Hypothesis 1). Or, in other words: The results that we arrive at are mostly in favor of the optimistic view. More often than not, preference configurations are characterized by a synergetic constellation. Finally, it turns out that preference configurations gravitate around the dimension of institutional depth. This lends support to those who believe that clever institutional design is the key to successful international cooperation.
10
Where do we get from here? As has been argued in the course of the essay, preferences are at the beginning of a policy cycle that may or may not lead to the desired outcomes. With little alterations, it would be possible to analyze the question whether there are synergies or tradeoffs between broadening, widening and deepening at the end-stage of the policy cycle, i.e. with regard to outcomes. However, let us leave this and similar questions to other authors. In the meantime, our results show some interesting insights into the internal structure of preference configurations. It is important to gain a better understanding of state preferences, because they are a necessary, although not a sufficient condition for real international cooperation as an outcome.
We are using a dataset that is borrowed from a research project on the internationalization of the monopoly of force, based at International University Bremen. The dataset uses three indices to measure state preferences on substantial scope (I1), geographical range (I2), and institutional depth (I3). These indices are derived from six parameters, which cover state preferences with regard to scope (P2), range (P4) and depth (P6), and the question whether a state intends a relative advance in comparison to the status quo, i.e. the scope (P3), range (P5) or depth (P7) already achieved. The six parameters are complemented by a forced-choice parameter (P1), which measures whether a country does want the agreement in the first place. The assumption is that if a country is either against international cooperation in a given area or indifferent, then the declared preferences on details are likely to be strategic and therefore irrelevant.
Table
4
(Note: This table includes the criteria adopted for coding.)
Due to empirical and categorial differences between the parameters, there are some minor but unavoidable problems. The parameters, none of which is interval-scaled, combine nominal (P1) and ordinal (P2-P7) scales. Not all of them have the same interval, nor do they all have their arithmetical mean at {0}. The latter problem does also apply to the indices, which are computed from the parameters.
However, there is a justification to all of these problems. In this study we engage in 'within-case analysis'. Mahoney (2003: 360-61) recognizes that it is typical for this kind of analysis to combine different levels of measurement. Nevertheless, since the level of measurement and format of our data is somewhat debatable, for safety reasons we assume that our three indices are ordinal and use them for rank-order correlations only. This should be sufficient to remedy for the problems mentioned.
To measure the strength of association, we use Spearman’s rho (rs). Spearman’s rho is a rank-order correlation coefficient which measures association at the ordinal level. It is a nonparametric alternative to the Pearson correlation coefficient, which requires interval-scaled values. The formula for Spearman’s rho is
, where D is the difference in ranks. The values for Spearman’s rho range from {+1.0} to {-1.0}. {+1.0} means a full positive correlation between the ranks, while {-1.0} means a perfect negative correlation.(14)
In order to use Spearman’s rho, one first has to assign appropriate ranks. Our data set is not continuous, since it is composed by blocks of four case studies each (we examine the preferences our each of our four countries for every individual topic). If we simply ran the analysis over all 24 case studies, our software program (SPSS) would assume that the data is continuous and would rank each case with respect to the other 23, rather than with respect to the other 3 cases in the block. To prevent this from happening, the indices are pre-ranked in each block of four cases, and only then is the analysis run on the already assigned ranks. Because the indices are not whole number, fractional ranking is used to capture the full extent of variation among the indices.
11
This is how we have dealt with the levels-of-measurement problem. Another possible objection is that, since the dataset is composed of different clusters, there may be autocorrelations. This is typical, however, about within-case analysis more in general (Mahoney 2003: 360-367). Within-case analysis, which is typical for historical comparison, disaggregates cases into their component parts and then examines associations within cases, as well as associations with other cases. The aim is to identify general or partial regularities in the internal configuration of the cases. In such a procedure, autocorrelations is not a problem. To the contrary, we are interested precisely in the impact that sampling has over the results, and we try to find this out by disaggregating out dataset into a variety of sub-samples (table 5).
Balázs, Péter (1998) ‘Strategies for the Eastern enlargement of the EU: an integration theory approach’, in Pierre-Henri Laurent and Marc Maresceau (eds) Deepening and Widening, Boulder (Colo.): Lynne Rienner, pp. 67-86.
Baldwin, Richard E. (1995) ‘A domino theory of regionalism’, in Richard E. Baldwin, Pertti Haaparanta, and Jaakko Kiander (eds) Expanding Membership of the European Union, Cambridge: Cambridge University Press, pp. 25-53.
Bräuninger, Thomas, Tanja Cornelius, Thomas König, and Thomas Schuster (2001) ‘The dynamics of European integration: a constitutional analysis of the Amsterdam Treaty’, in Gerald Schneider and Mark D. Aspinwall (eds) The Rules of Integration: Institutionalist Approaches to the Study of Europe, Manchester: Manchester University Press, pp. 46-68.
Dinan, Desmond (2002) ‘Institutions and governance 2001-02: debating the EU’s future’, Journal of Common Market Studies 40 (3-S): 29-43.
Dehousse, Renaud and Weiler, Joseph H. H. (1990) ‘The legal dimension’, in William Wallace (ed.) The Dynamics of European Integration, London: Pinter, pp. 242-260.
Downs, George W., David M. Rocke, and Peter N. Barsoom (1998) ‘Managing the evolution of multilateralism’, International Organization 52 (2): 397–419.
Easton, David (1965) A Systems Analysis of Political Life, New York: John Wiley.
Elvins, Martin (2003) Anti-Drugs Policies of the European Union: Transnational Decision-Making and the Politics of Expertise, Basingstoke: Palgrave Macmillan
Engel, Christian and Borrmann, Christine (1991) Vom Konsens zur Mehrheitsentscheidung: EG-Entscheidungsverfahren und Nationale Interessenpolitik nach der Einheitlichen Europäischen Akte, Bonn: Europa Union Verlag.
Frieden, Jeffry A. (1999) ‘Actors and preferences in international relations’, in David A. Lake and Robert Powell (eds) Strategic Choice and International Relations, Princeton: Princeton University Press, pp. 39-76.
Friedrichs, Jörg (2007) International Police Cooperation against Terrorism and Drugs: The Preferences of Large European States, to be published.
George, Alexander L. and Andrew Bennett (2005) Case Studies and Theory Development in the Social Sciences, Cambridge (Mass): MIT Press.
Giddens, Anthony (1984) The Constitution of Society: Outline of the Theory of Structuration, Cambridge: Polity Press.
Ham, Peter van (1997) ‘The EU and WEU: from cooperation to common defense?’, in Geoffrey Edwards and Alfred Pijpers (eds) The Politics of European Treaty Reform: The 1996 Intergovernmental Conference and Beyond, London: Pinter, pp. 306-325.
Hayes-Renshaw, Fiona and Wallace, Helen (1997) The Council of Ministers, Basingstoke: Macmillan.
Héritier, Adrienne (1996) ‘The accommodation of diversity in European policy making and its outcome: regulatory policy as a patchwork’, Journal of European Public Policy 3 (2): 149-167.
Howlett, Michael and M. Ramesh (2003) Studying Public Policy: Policy Cycles and Policy Subsystems, Oxford: Oxford University Press, 2nd edn.
Kerremans, Bart (1998) ‘The political and institutional consequences of widening: capacity and control in an enlarged Council’, in Pierre-Henri Laurent and Marc Maresceau (eds) Deepening and Widening, Boulder (Colo.): Lynne Rienner, pp. 87-109.
Lasswell, Harold D. (1956) The Decision Process: Seven Categories of Functional Analysis, College Park: University of Maryland Press.
Laurent, Pierre-Henri and Marc Maresceau (1998) ‘Moving to a multispeed Europe’, in id. (ed.) Deepening and Widening, Boulder (Colo.): Lynne Rienner, pp. 1-22.
Mahoney, James (2003) ‘Strategies of causal assessment in comparative historical analysis’, in James Mahoney and Dietrich Rueschemeyer (eds) Comparative Historical Analysis in the Social Sciences, Cambridge: Cambridge University Press, pp. 337-372.
McAllister, William B. (2000) Drug Diplomacy in the Twentieth Century: An International History, London and New York: Routledge.
Mitsilegas, Valsamis and Jörg Monar (2003) The European Union and Internal Security: Guardian of the People, Basingstoke: Palgrave Macmillan.
Moravcsik, Andrew (1997) ‘Taking preferences seriously: a liberal theory of international politics’, International Organization 51 (4): 513-553.
Moravcsik, Andrew (1998) The Choice for Europe: Social Purpose and State Power from Messina to Maastricht, Ithaca (N.Y.): Cornell University Press.
Nadelmann, Ethan A. (1993) Cops Across Borders: The Internationalization of U.S. Criminal Law Enforcement, University Park: Pennsylvania State University Press.
Nadelmann, Ethan A. (1997) ‘The Americanization of global law enforcement: the diffusion of American tactics and personnel’, in William F. McDonald (ed.) Crime and Law Enforcement in the Global Village, Nunn Hall (KY): ACJS/Anderson, pp. 123-138.
Occhipinti, John D. (2003) The Politics of EU Police Cooperation: Toward a European FBI?, Boulder (Colo.): Lynne Rienner.
Ollenburger, Jane and Anthony Walsh (2001) Essential Statistics for the Social and Behavioral Sciences: A Conceptual Approach, Upper Saddle River (NJ): Prentice Hall.
Pahre, Robert (1995) ‘Wider and deeper: the links between expansion and integration in the European Union’, in Gerald Schneider, Patricia A. Weitsman, and Thomas Bernauer (eds) Towards a New Europe: Stops and Starts in Regional Integration, Westport (Conn.): Praeger, pp. 111-136.
Peterson, John and Elizabeth Bomberg (1998) ‘Northern enlargement and EU decisionmaking’, in Pierre-Henri Laurent and Marc Maresceau (eds) Deepening and Widening, Boulder (Colo.): Lynne Rienner, pp. 43-65.
Rosamond, Ben (2000) Theories of European Integration, New York: St. Martin’s Press.
Sbragia, Alberta (1993) ‘Asymmetrical integration in the European Community: the Single European Act and institutional development’, in Dale L. Smith and James Lee Ray (eds) The 1992 Project and the Future of Integration in Europe, Armonk (N.Y.): Sharpe, pp. 92-109.
Schalken, Tom and Maarten Pronk (2002) ‘On joint investigation teams, Europol and supervision of their joint actions’, European Journal of Crime, Criminal Law and Criminal Justice 10 (1): 70-82.
Scharpf, Fritz W. (1988) ‘The joint-decision trap: lessons from German federalism and European integration’, Public Administration 66 (3): 239-278.
Scharpf, Fritz W. (1997) ‘Economic integration, democracy, and the welfare state’, Journal of European Public Policy 4 (1): 18-36.
Scharpf, Fritz W. (1999) Governing in Europe: Effective and Democratic? Oxford: Oxford University Press.
Schulz, Heiner, and Thomas Köning (2000) ‘Institutional reform and decision-making efficiency in the European Union’, American Journal of Political Science 44 (4): 653-666.
Thomson, Janice E. (1995) ‘State sovereignty in international relations: bridging the gap between theory and empirical research’, International Studies Quarterly 39: 213-233.
Van den Bos, Jan M. M. (1994) ‘The European Community, decision making, and integration’, in Bruce Bueno de Masquita and Frans N. Stokman (eds) European Community Decision Making: Models, Applications, and Comparisons, New Haven: Yale University Press, pp. 17-32.
Walt, Stephen M. (1987) The Origins of Alliances, Ithaca (N.Y.): Cornell University Press.
Weber, Max (1968 [ 1922 ] ) Economy and Society: An Outline of Interpretive Sociology, Berkeley: University of California Press (3 volumes).
Weber, Max (1992 [ 1919 ] ) Politik als Beruf, Stuttgart: Reclam.
Wendt, Alexander (1987) ‘The agent-structure problem in international relations theory’, International Organization 41 (3): 335-370.
Wessels, Wolfgang (1994) ‘Rationalizing Maastricht: the search of an optimal strategy of the new Europe’, International Affairs 70 (3): 445-457.
(*) Research on this paper was carried out in a project on the internationalisation of the monopoly of the legitimate use of force, within a collaborative research centre on transformations of the state. The research project is based at International University Bremen, Germany. Funding by the Deutsche Forschungsgemeinschaft is gratefully acknowledged. We would like to thank the project director, Markus Jachtenfuchs. Thanks are also due to Christiane Kasack, Eva Herschinger and Holger Stritzel for their collaboration on empirical research; to Klaus Boehnke, Axel Domeyer and Margit Schreier for their advice; and to Thomas Beer, Magdalena Cholakova, Moritz Weiss and two anonymous reviewers of EIoP for their comments.
(1) Cf. the neofunctionalist strand of integration theory since the late 1950s, as represented by Haas, Lindberg, Schmitter, and others (for a concise overview see Rosamond 2000: 31-42).
(2) Nevertheless Scharpf emphasizes that, despite these reforms, the ‘democratic deficit’ of the EU persists and even increases.
(3) Although we would define ourselves as relatively sympathetic towards European integration, based on the arguments reported in this section we expected that Hypothesis 2 (tradeoffs) should prevail over Hypothesis 1 (synergies). However, things turned out otherwise. We were ourselves surprised by the results of the empirical part of the paper (see below, sections 3 and 4).
(4) There are also studies to support a more optimistic view. By studying the time lag between Commission proposals and Council decisions, Schulz and König (2000) suggest that the EU is often successful in meeting the demands posed by expanding policy competences. Nevertheless, the speed of decision-making varies significantly across policy areas. It is higher with respect to issues constituting the functional core of the EU and lower in the more recently introduced spheres.
(5) In Moravcsik’s book The Choice for Europe (1998: 24-50) this is extended into a three-stage process: preference formation, intergovernmental bargaining, and institutional choice.
(6) Single Convention on Narcotic Drugs, 1961 (amended 1972); Convention on Psychotropic Substances, 1971; Convention against the Illicit Traffic in Narcotic Drugs and Psychotropic Substances, 1988.
(7) In Amsterdam (1997) and Tampere (1999), the EU member states agreed upon a legal framework for these teams, which are created to combat trafficking in drugs and human beings as well as terrorism and other forms of serious crime; cf. Schalken and Pronk 2002.
(8) It is logically impossible to observe three negative correlations between a set of three variables; nor is it possible to observe two positive and one negative correlations.
(9) What we are actually testing is if there is enough evidence to reject the Null hypothesis. If so, then based on the correlations observed we can see which of the two alternative hypothesis is validated.
(10) For a more detailed overview of the dataset and the statistical methods used please see Appendix 1.
(11) The better the significance, the lower the risk that the null hypothesis is true or, in other words, the more reliable the relationship, whether positive or negative, between a pair of indices.
(12) A critical reader might object that there is a problem of multiple testing. However, this is not the case because our detailed analyses (see Appendix 2) serve the heuristic purpose of examining the internal differentiation of our dataset. They should not be taken as premeditated hypothesis tests.
(13) One of the authors is preparing a book-length study about the international fight against terrorism and drugs (Friedrichs 2007).
(14) One advantage with rs is that, when squared, it has a pre-interpretation effect, indicating how much the risk of wrongly predicting the rank of one variable from the rank of another is reduced (see Ollenburger and Walsh 2001: 183).