Administrative styles in the European Commission and the OSCE- Secretariat: Striking similarities despite different organisational settings

Transcription

1 Administrative styles in the European Commission and the OSCE- Secretariat: Striking similarities despite different organisational settings Paper presented at the 2015 ECPR General Conference, , Montreal, Canada Please do not cite or quote without authors permission Author contacts: Prof. Dr. Christoph Knill Ludwig-Maximilians-Universität München Geschwister-Scholl-Institut für Politikwissenschaft Oettingenstraße München (Germany) Phone: +49 (0) Mail: Dr. Steffen Eckhard Ludwig-Maximilians-Universität München Geschwister-Scholl-Institut für Politikwissenschaft Oettingenstraße München (Germany) Phone: +49 (0) Mail: Dr. Stephan Grohs German University for Administrative Sciences Speyer Chair for Political Science Freiherr-vom-Stein-Str Speyer (Germany) Phone: +49 (0) Mail: grohs@uni-speyer.de Biographical notes: Prof. Dr. Christoph Knill holds the Chair of Empirical Theories of Politics at the University of Munich and is the speaker of the research unit International Public Administration. Dr. Steffen Eckhard is scientific coordinator of the research unit International Public Administration at the University of Munich and a non-resident fellow at the Global Public Policy Institute (GPPi). Dr. Stephan Grohs is currently the substitute Chair for Political Science at the German University for Administrative Sciences Speyer. His research interest is in international and comparative public administration. Acknowledgments: Tba Funding: This work was supported by the German Research Foundation under Grant FOR # 1745 TP 02 BA 3658/5-1. 1

2 Abstract: With the growing importance of international institutions for global governance, international bureaucracies gain increasing influence on policy-making. Whereas for national bureaucracies specific administrative styles have been identified, this article explores for the first time administrative styles of two international organizations. The EU Commission and the OSCE represent most different cases in matters of scope, autonomy and resources. The article analyses the specific patterns of policy initiation, policy formulation and policy implementation of both organizations and reveals striking similarities. Both organizations have developed an entrepreneurial style in policy initiation, a strategic approach to policy formulation and a mediating implementation style. This similarities are explained by two different mechanisms: a paradox of weakness in the case of the OSCE and a paradox of strength in the case of the EU. In particular, administrative styles in international bureaucracies seem depend on the extent to which member states take interest in their work. Keywords: Administrative styles, administrative influence, international public administration, global governance, international organizations 2

3 Administrative styles in the European Commission and the OSCE- Secretariat: Striking similarities despite different organisational settings 1. Introduction The development of public policies beyond the nation state is an increasingly important feature of national and international governance activities. International Organisations (IOs) play a crucial role in this regard (Cerny 2010: 111). Yet, these organisations vary considerably. This holds true not only in terms of their structural features (e.g. in terms of size, policy scope, membership, functions; Lenz et al. 2015), but also when it comes to issues of reputation, authority and policy performance (Barnett and Finnemore 1999; Zürn et al. 2012). Over the past decade, IOs bureaucratic interior increasingly has surfaced as a distinct objective in academic research on public administration and international relations (Barnett and Finnemore 1999; Knill and Grohs 2015; Bauer and Trondal 2015). As we know from research on domestic administrations, bureaucrats may exert considerable influence on the contents and the ways of policy-making. This finding has broadly been confirmed for international public administrations (IPAs). Yet, contrary to numerous studies of this phenomenon for the national context, little is known about variations in influence and the factors accounting for it internationally. In addressing these questions, this paper dwells on the concept of administrative styles (Knill 1998, 2001), understood here as the organizational routines that characterize the behaviour and activities of administrative bodies. To study these patterns of organizational behaviour, we differentiate analytically between three stages of the policy cycle, namely the activities of IPAs with regard to policy initiation, policy formulation, and policy implementation. The study of administrative styles of IOs fills a research gap between micro-level studies that have focused on attitudes of bureaucrats (e.g. Hooghe 2001; Kassim et al. 2013) and studies focusing on the macro-level of organizational structures (e.g. Olsen 2010; Egeberg 2006). In so doing, our analytical focus goes beyond merely considering the structural features of IOs, such as their political autonomy, size, organizational differentiation, resources, tasks, or heterogeneity. Rather, we are interested in the organizational routines IPAs 3

4 develop in order to respond to the opportunities and constraints determined by these structural features. In this paper, we compare administrative styles in the European Union (EU)-Commission (with a specific focus on the Commission s Directorate-General for the Environment) and the secretariat of the Organization for Security and Cooperation in Europe s (OSCE). Both IPAs differ strongly with regard to structural factors identified as critical for the development of distinctive administrative styles. In particular, while the EU Commission by its very design can be seen as one of the most autonomous international bureaucracies, the OSCE secretariat, by contrast, can be characterized as highly fragile and weak bureaucracy (see Bauer and Ege, this issue). Yet, contrasting both IPAs reveals striking similarities in their administrative styles a finding which is in sharp contrast to the expected variation in light of their highly different functional and institutional features. Both bureaucracies act as entrepreneurs rather than servants in policy-initiation, they are strategic rather than pragmatic in policy formulation and they generally pursue a mediating rather than interventionist implementation style. Based on our inductive analysis, we argue that administrative styles depend much more on the extent to which member states scrutinize the activities of IPAs than on the factors we considered when selecting our cases. In the case of the OSCE, this results in a paradox of weakness: During the past decade, member states increasingly ignored the work of the organization. A phase of détente between East and West and the growing relevance of the EU in security policy caused the OSCE to drift into irrelevance. Activities of the secretariat could be performed beneath the principals radars (see Abbott et al. 2015) and did not require formal member state consent thus leaving considerable leeway for a structurally weak administration that has been barely impaired by member state oversight. As a result, the OSCE secretariat could venture beyond its structural potential. By contrast, the case of the EU Commission stands for a paradox of strength: Too much use of the considerable implementation powers of the Commission could result in growing member state concerns with regard to the costs caused by EU policy activities; hence potentially increasing the resistance of member states to new policy initiatives of the Commission. The anticipation of the relationship in addition to effectiveness and efficiency concerns associated with an extensive use of its control and enforcement competencies 4

5 favoured a mediating implementation style of the Commission that generally seeks to avoid open conflict with member states and thus remains below its structural potential. In section 2, we present our concept of administrative styles that guides our empirical analysis. The underlying research design, case selection, and methodology is presented in section 3. In sections 4, 5, and 6 we present our empirical findings for both IPAs under study with regard to different stages of the policy cycle. In section 7, we discuss our findings, while section 8 concludes. 2. The concept of administrative styles In the classical analysis of national administrations, administrative styles are distinctive dimensions of the broader concept of administrative culture. Generally speaking, administrative culture is studied at three different levels: first the micro-level, including the values, roles, and behaviours of individual members of the administration, as well as the attitudes of the general public towards administrations, second the macro-level of administrative traditions, and third the meso-level of administrative styles, understood as the standard operating procedures of administrative behaviour and decision-making (Jann 2002). At the national level, one can find a range of similar concepts, such as policy styles (Richardson et al. 1982) and regulatory styles (Vogel 1986) that have been used to explain administrative patterns of action. By contrast, little is known about administrative styles in IOs and how they affect policymaking. To further develop the concept of administrative styles for the context of IOs we differentiate analytically between three dimensions that characterize administrative styles. These dimensions represent different patterns of administrative involvement in the initiation, policy formulation, and implementation of policies. The first dimension of this heuristic, policy initiation, refers to public administrators behaviour in the pre-formulation phase, which consists of the definition of policy problems and the setting of the agenda for political decisions. Both the definition of the problem and agenda-setting are ultimately the result of a selection process which is determined to a great extent by those actors that are able to assert 5

6 their ideas in the political process. Depending on how different actors are affected by the issue, these ideas can vary significantly. This holds not only for the question whether and to what extent a certain situation is perceived as problematic at all, but also for the selection of politically relevant aspects of the problem, which are to be dealt with on the basis of appropriate measures (Knill and Tosun 2012). We conceptualize this behaviour in a dichotomous way, ranging on a continuum where the IPAs act either as an entrepreneur or as a servant towards its principals. As entrepreneurs we can conceptualize administrations that are highly active in detecting new policy problems and in bringing them on the agenda. They resemble the policy entrepreneur as stylized by Kingdon (1984) as advocates of policy proposals, which show a willingness to invest time and resources in the hope of future return. By contrast, administrations conceived as servants are rather passive IPAs interpreting their own role as one of servant of the principals in a rather Weberian sense. The main questions in this phase are: do IO administrations actually initiate policy developments from the inside by accumulating policy-related knowledge, by constant observation of political opportunities, and finally by strategically mobilizing political or societal support to shape the political agenda, or do they merely respond to external requests, thus pursuing a wait-and-see approach? Consequently, three indicators are relevant to assess IOs administrative styles in defining the policy problem to work with. Relevant indicators for assessing which pole they belong to include (1) whether IPAs actively generate own independent expertise, for instance by interacting with external actors or building own research capacities, (2) whether IPAs map the political space available for policy proposals and (3) whether secretariat staff engages in active coalition-building and interest-mediation activities. The second dimension refers to the styles of policy formulation. It is commonly acknowledged that one of the most important sources of administrative influence is the drafting of policy proposals by the bureaucracy. Although such drafts might undergo additional changes in the political decision-making process, they pre-structure the basic content and the instrumental design of a policy. Previous research has shown that international bureaucrats can exert considerable influence in this stage by strategically taking into account the policy positions of the member states and their committees (Mazey and 6

7 Richardson 1995). Such a strategic style may be rooted in the desire to adopt optimized policy options that are based on a holistic triangulation of the problem at hand, the desired end and the resources at hand (Mintzberg 1978). It is opposed to a pragmatic (or satisficing) administrative style that follows a more instrumental and service-oriented role perception and the perpetuation of existing policies. From such a perspective civil servants would do as requested and not question the substance of their tasks, even if they found them to be flawed. Strategic or pragmatic behaviour emerges in matters of developing the policy content as well as in matters of the procedural approach. The first aspect refers to the extent to which the bureaucracy seeks to influence and shape the content of the policies adopted. As bureaucracies are usually broadly involved in the drafting of policies, they might have ample opportunities to pre-structure the basic content and instruments of the policies adopted. Administrations that follow a strategic style of action evaluate a large number of different alternatives and follow the solution it deems most effective; inversely, pragmatic bureaucracies perpetuate existing policy instruments incrementally. The second aspect refers to the extent to which IPAs pursue a strategic or pragmatic approach in order to structure the process of policy formulation. Administrations that follow a strategic logic are characterized by a strong political sensitivity. They anticipate political red lines and make use of political facilitation mechanisms (e.g. depoliticization), in order to prevent later political rejection by member states. Vice versa, pragmatic administrations show low political sensitivity and refrain from time-consuming political facilitation mechanisms. Relevant indicators in this context refer (1) to the extent to which policy proposals are evidence-based and systematically evaluate policy alternatives, (2) whether they are programmatic in the sense that they are linked to a broader policy strategy or framework, and (3) whether IPAs systematically conceive of specific strategies to facilitate political decision-making (de-politicisation). The third dimension refers to the implementation stage. It is a general finding, that shifts in policy objectives and deviations from the original political intentions are frequently observed during the implementation stage of a policy. Secretariats can pursue either of two styles in relation to how to ensure that their policies are being implemented. On the one hand, they can rely on interventionist 7

8 approaches. Although the secretariat s sanctioning powers might be limited, headquarter bureaucrats can attempt to increase their steering capacity by collecting systematic information on policy effects or by setting up close relationships with involved stakeholders, interest groups, national administrations, or external experts. On the other hand, the secretariat can rely on a mediating approach. In such a case, they will refrain from developing procedures to observe and improve compliance that go beyond their legally specified duties and strongly rely on non-hierarchical patterns of self-regulation. The main indicator for wither style is whether IPAs are willing to risk open conflict with member states or whether they pursue alternative (mediating) strategies. Thus far, variation in administrative styles has typically been explained against the backdrop of country-specific factors, in particular the macro-institutional context of the state tradition, the legal system, and the political-administrative system of a country (Knill 2001: 4). It is doubtful, however, that these factors are of equal importance in influencing the styles of IPAs (Barnett and Finnemore 2004; Biermann and Siebenhüner 2009) in which the identification of comparable features is difficult. In response to these peculiarities of IPAs, it has been argued that administrative styles at the level of IOs can be better explained by focusing on more general functionalist and institutionalist approaches (Knill and Grohs 2015). Accordingly, variation of administrative styles is expected to emerge from differences in IPAs functions (the nature and types of problems they deal with) (Ostrom 2007), the institutional opportunities and constraints of decision-making they are confronted with (North 1991), as well as the cognitive frames and the cultural socialization of their staff (DiMaggio and Powell 1991). 3. Research design, case selection, and methodology In order to assess the plausibility of this theoretical reasoning, we pursue an inductive case study approach based on a most different systems design. The EU-Commission and the OSCE-secretariat differ strongly in terms of their institutional background and the conditions in which both administrations operate. They constitute most different systems in many ways, which consequentially leads us to expect variation with regard to their administrative styles. 8

9 In line with the functionalist perspective, the two organizations differ in terms of the problem structure they deal with. The policy scope of the OSCE is narrowly delimited on security policy whereas the EU Commission is a truly multi-purpose organization (Lenz et al. 2015). In addition, compared to the Commission s centralized setting and focus on regulation, the OSCE s main tasks are to implement policy, not to formulate it. In 2007, 71 percent of all OSCE employees worked in the field, which left only around 400 bureaucrats in Vienna (OSCE 2007), whereas most of the employees of the EU Commission are allocated centrally at Brussels and Luxembourg. Second, and quite importantly, the EU Commission and OSCE represent two ends on a continuum between most autonomous and least autonomous organizations, according to structural features (see Bauer and Ege, this issue). In contrast to the EU Commission, the literature states unanimously that the OSCE secretariat is among the most fragile international bureaucracies: its institutional structure has remained weak (Legutke 2005: 188). This has to do with the OSCE s limited resources, but also with its governance structure. Policy decisions in the OSCE are taken unanimously in the Permanent Council (or the superior OSCE summit). In contrast, many task areas of EU policy-making underlie qualified majority voting. And as opposed to the EU Commission, OSCE member states and the Chairmanship hold the exclusive right to set the agenda for OSCE decision-making bodies. With respect to cognitive factors, we can assume that both bureaucracies are different as they represent different constituencies. The EU Commission is quite homogenous in its membership and staffing, representing only European states. By contrast, the OSCE extends beyond Europe alone with its secretariat comprising staff also from (far) Eastern Europe and Central Asia as well as North America. In addition, whereas Commission bureaucrats are employed on a life-long basis, those in the OSCE stay only temporarily for a maximum of ten years (Shkolnikov 2009: 147). Table 1: Different organizational settings at EU Commission and OSCE Secretariat EU Commission OSCE Secretariat 9

10 Policy-scope Broad (multi-purpose) Narrow (security) Focus in policy-cycle Policy formulation Policy implementation Autonomy Most autonomous Least autonomous Political decision-making Mostly qualified majority voting Unanimous Membership Homogenous Heterogeneous (East-West divide) Staff number Duration of employment Lifelong Max. ten years Source: Authors compilation and data provided by the two organizations as quoted in the section above. Overall, we should expect that these differences cause different administrative styles in how both IPAs engage in initiating, drafting and implementing the policies of their organizations. In order to investigate this expectation, we conducted 21 semi-structured interviews with administrative officials in both IPAs (16 in the OSCE, 5 in the EU Commission) between June 2011 and March In addition, we relied on the analysis of secondary literature and organizational documents. In the literature in the EU Commission, it is consistently emphasized that the different Directorates- General (DGs) differ strongly in terms of the organizational cultures. Although the conception of administrative styles developed in this paper is based on analytical categories that cut across this intraorganizational differences, we focus our analysis on the DG Environment. Unless supported by general findings in the relevant literature, the findings presented for the EU Commission hence refer to this DG only. 4. Policy initiation Against the backdrop of our conceptual discussion, we find that both the EU Commission and the OSCE-Secretariat pursue a highly entrepreneurial style with regard to the definition of policy problems and their placement on the political agenda. EU-Commission 10

11 It is well-acknowledged in the literature that the Commission strongly resembles the entrepreneurial pattern during the stage of policy initiation (Hooghe and Nugent 2006; Knill 2005). Usually, this assessment is based on the Commission s formal powers. With its exclusive right of initiative, the Commission has a particularly important resource to influence the orientation and political priorities of EU policies according to its own preferences and priorities. term. Yet, the strong hand of the Commission in shaping political problems in setting the Community s agenda seems to a lesser extent be affected by its formal powers, but by informal administrative routines that reflect its selfunderstanding as a political entrepreneur. First, the Commission is highly active in generating policy-relevant expertise. This occurs mainly through establishing close contacts with scientific experts, national administrations, societal groups and interest associations. In addition to regular formal and informal consultation procedures, the Commission has set up a dense net of working groups in order to stimulate information exchange and to discuss political problems and potential solutions (Gornitzka and Sverdrup 2008: 745). In environmental policy, the Commission can additionally rely on detailed information on the overall development of environmental quality and pollution across the Community that is provided by the European Environment Agency. These data provide an important source of knowledge and expertise that is used by the Commission bureaucracy for setting up new policy proposals (interviews 2, 4). In this context, the Commission is actively engaged in balancing differences in resources of European interest associations to some extent. For instance, it supports associations in the field of environmental or consumer protection, that generally suffer from weaker resources, as is the case for economic associations (Greenwood 2011). Second, the Commission is actively engaged in mapping the political space for new policy initiatives. This basically occurs through closely watching policy developments at the member state level and close interactions with national administrations (interview 1; Larsson and Trondal 2006; Wonka 2015). Hence, often and notwithstanding its formal right of initiative, many Commission proposals can be attributed to initiatives of other actors who approach the Commission and request it to draw up certain policies. Besides the Council of Ministers and the European Parliament (EP), which formally have the possibility of demanding the Commission to draw up a certain proposal, the informal 11

12 activities of the member states play a particularly important role in this regard. It hence appears that the Commission is frequently more preoccupied with mapping the initiatives of individual states than playing an active role on its own (Knill and Liefferink 2007: 80; Cini 2015: 139). This development is driven by the Commission s interest to get informed about policy developments in the member states and to frame policy proposals accordingly (interview 2). Third, this mapping of national developments usually comes along with deliberate attempts of the Commission to build up both member state and societal support for its policy initiatives. Through clever manoeuvring and capitalizing on the informational advantage vis-à-vis other actors gained through extensive consultations (Schrefler 2010: 315), the Commission is able to shape the political agenda according to its own interests. Although it does not have a monopoly on information and expertise, it can drive the policy agenda by manipulating asymmetries between the member states and different private interests (Hix and Høyland 2011: 183f.). This holds in particular when information is dispersed unequally among actors, when the preferences of the involved actors are not sufficiently defined and access to the political process is relatively open (with fluctuations in the number of participating actors) (Kingdon 1984) conditions that certainly characterize many areas of EU policymaking (Knill 2005). OSCE Formally, the room for manoeuvre for OSCE bureaucrats is extremely limited. Although this suggests that secretariat staff acts as servant rather than entrepreneur, the opposite is true as the analysis of the OSCE secretariat s administrative style unveils. First, at a time when the OSCE budget shrank from 211 million Euro to 151 million Euro between 2000 and 2010 (OSCE 2000, 2011c), the OSCE secretariat built up an extensive network to generate policy-relevant external expertise. In 2000, the German government established an OSCE think tank (the Centre for OSCE Research (CORE) at the University of Hamburg) with around 20 staff. Since 2011, CORE experts established the OSCE Network on initiative of OSCE Secretary General Lamberto Zannier. For the secretariat the network allows reaching out to dozens of think tanks and academic institutions that work on topics related to OSCE activities. In single substantial policy areas, such as security sector reform, the OSCE set up 12

13 data bases to flexibly gain access to a wide range of experts. 1 Overall, the organization is well involved in creating and maintaining epistemic expert communities in order to constantly monitor political developments in the region and relevant policy discourses (interviews 7, 19). Second, heads of departments and the OSCE Secretary General are in constant interaction with the Chairmanship and other key member states. In these informal discussions at the floors of the Hofburg in Vienna, they attempt to sound out the potential climate among delegations. This is mainly meant to detect areas for potential policy initiatives. As several officials said, there is a clear understanding that even informal policy development processes can only be launched upon political request (interviews 6, 12, 13). This is why interacting with member state delegations is an important condition for the OSCE secretariat to sense room for policy initiatives and informally influence the agenda setting process even before officially being requested by the Chairmanship or Permanent Council to launch policy development procedures. Third, in line with this mapping of the political room for manoeuvre, the OSCE secretariat influences the Permanent Council s agenda according to its own policy preferences. This includes building up bureaucratic expertise on subject matters and forging outside coalitions supporting the own position. One recent example illustrates this. Since his inauguration in 2011, Secretary General Lamberto Zannier (2012: 5) repeatedly said that it was among his priorities to develop further steps towards strengthening the OSCE in its capacity to prevent conflicts and to help to settle them. While no specific formal request was formulated by the Permanent Council, the secretariat used the OSCE Network to forge an outside opinion on the matter. With the vested authority of twenty-one academic institutes and think tanks, the network published a report in 2014 that outlined options for The Future of OSCE Field Operations (OSCE Network 2014). Although it remains to be seen whether this will eventually appear on the Permanent Council s agenda, the example demonstrates strategic intent and the hand of a policy entrepreneur. 1 One example is the expert data base POLIS, maintained by the Strategic Policy Matters Unit. 13

14 5. Policy formulation For the stage of policy formulation, we distinguish between two ideal types of either strategic or pragmatic styles of IPAs. Indicators used for either style are whether the administration evaluates policy-alternatives, whether secretariat proposals relate to a general strategy and whether IPAs conceive of specific strategies to facilitate MS decision-making (de-politicization). EU Commission The EU Commission generally pursues a highly strategic administrative style when it comes to the formulation of EU policies. This basic finding has been documented in numerous studies as a general feature characterizing the Commission s role across a broad range of different policy areas (e.g. Pollack 1997; Rhinard 2010; Knill and Grohs 2015). It is hence hardly surprising that these features also come to the fore when taking a closer look at respective activities of DG Environment. In terms of policy content, the specific influence of the Commission during the drafting stage primarily emerges from its systematic approach of anticipating policy preferences of the member states. As emphasized by various officials within DG Environment, the Commission systematically analyses policy developments in the member states (interviews 1, 2, 3, 5; Lenschow 2005). In designing its draft proposals, the Commission heavily relies on the input of national and external experts who can in principle exert considerable influence due to the scarce human and time resources of the Commission. A prerequisite for this, though, is that their ideas fit well with the arguments and proposals of the Commission (Knill and Liefferink 2007, 83). This is due to the fact that the Commission can select those arguments out of the multitude of presented arguments which are consistent with its preferences. The discussions within the working groups enable the Commission to collect arguments on the basis of which it can later substantiate its own position. Yet it should be emphasized that depending on the specific interest constellation the anticipation of political conflicts in the Council, the European Parliament, but also within the Commission (i.e. potential conflicts across different Commission DGs) might generally reduce the Commission s room of manoeuvre for adopting measures that significantly depart from the status quo. 14

15 This strategic style is facilitated by institutionalized routines of de-politicizing the process of policy formulation. In particular the Commission thereby attempts to demonstrate that its policy proposals are based on sound scientific evidence and a systematic assessment of policy impacts. Not only in the environmental field, the impact of policy drafts has to be demonstrated not only in terms of potential ecological benefits, but also in terms of its economic and social implications. There is a strong adherence to the principle of sustainable development, implying that Commission proposals need a highly systematic and scientifically sound backing in terms of potential policy effects and assessments of policy alternatives (interviews 1, 3, 5; Lenschow 2005). For the environmental sector, there is clear evidence that as a result of the financial and economic crisis unfolding since 2007, growing emphasis has to be placed on economic effects of environmental measures. In other words, the crisis caused a shift in priorities put on the evaluation of policy effects in favour of the economic sustainability dimension (interviews 1, 3). Finally, in addition to systematic impact assessment and strong evidence base, the Commission seeks to strategically increase the chances of policy adoption through a highly programmatic approach. First, the Commission relies on long-term planning strategies like action programs and strongly makes use of framework laws that serve as general orientation for subsequent policy initiatives. The action programs of the European Commission ultimately have two objectives. They specify the focal areas of EU legislation for the coming years and determine the fresh strategic policy directions in different fields. They constitute a policy framework to be filled in by appropriate directives. In the environmental field, the Commission relies on environmental action programs that specify central issues and topics to be covered by subsequent legislation. As these programs have been adopted through an official decision-making process by both the European Parliament and the member states, they constitute a very strong source of the Commission in order to legitimize policy proposals that link up with the objectives specified the programme (interviews 1, 3). Second, especially in the environmental field the Commission has been relying on so-called framework laws. These define general legal principles and procedures for a specific issue area (e.g. emissions from large combustion plants) and specify the range of items that shall be addressed by future directives and regulations within his general framework (Knill and Tosun 2015: 99). 15

16 OSCE Although the secretariat is primarily engaged in providing technical services for field offices, it still pursues a strategic style in policy formulation. Similar to the EU Commission, routine documents, such as developing the annual working program for field operations and other OSCE implementing bodies, are developed on basis of a rationalistic planning process over patterns of incremental adjustment. Annual planning procedures are formally linked to the budget process by means of the OSCE s Performance Based Programme Budgeting (PBPB). It is a project management framework that sets out the relationship between political objectives, the formulation of a usually country-specific mid-term program strategy and its operationalization in individual projects with a budget figure (OSCE 2011b). Such projects are, for example, the monitoring of human rights standards in Kosovar prisons or a capacity-building workshop at a municipal administration in Kazakhstan. In terms of policy content, OSCE experts are ambitious to come up with the best possible conflict strategy (interviews 10, 11, 12, 19). They know that many OSCE policies do have significant impact on people s lives. Whether OSCE field missions perform or not affects (the resolution of) violent conflicts, such as in Kirgizstan in 2010 or in the Ukraine since Providing effective post-conflict support is sometimes exacerbated by the Permanent Council. Some member states benefit from conflict, others seek to end them. Most of the times, this boils down to two blocks: the Western block and the Commonwealth of Independent States (CIS) states. 2 Within this constellation, the secretariat has significant leeway in developing proposals confirming with one or the other side. This power emanates from its control over the drafting process. According to interviewees, each time policy planning processes are launched, the office of the Secretary General decides on their timeline and which of the organization s units to include. They also decide which outside experts to invite and thus which voices and arguments to hear (interviews 11, 12). Eventually, the secretariat can use this leeway in a strategic fashion, either to favour one side or which seems more common to facilitate compromise and develop feasible and realistic policy options. 2 Commonwealth of Independent States (CIS). This includes Russia and twelve other Eastern European or Central Asian states. See 16

17 Second and related to the process dimension, the secretariat also attempts to facilitate negotiations in the Permanent Council, specifically in cases where member state positions seem irreconcilable. Compromise is always necessary because the Council can only take decisions unanimously. This is a frequent cause of policy failure in the OSCE. However, as soon as such a controversial subject is being split up in smaller, more technical sub-topics compromise becomes possible. This was also the founding idea of the OSCE with its three baskets. According to interviewees, the secretariat oftentimes assumes this intermediate position of the mediator facilitating compromise between political blocks and became quite apt in depoliticizing issues by spinning them technically (interviews 11, 14, 15, 21). 6. Implementation Although there is a general consensus that international organizations have rather weak implementation capacity in the classical sense (Biermann et al. 2009; Joachim et al. 2008), they do have certain instruments at their hands to see their policy goals to be implemented. We can observe differences between an intervening and a mediating style in two different areas: the impact enhancement activities of the IPA, and the IPA s willingness to risk conflicts with member states. EU Commission Although the Commission has considerable formal powers to sanction non-compliance with EU law (especially in comparison with other international bureaucracies), it generally pursues a mediating rather than intervening style of implementation. The Commission is generally reluctant to make full use of its formal powers. Before appealing to the European Court of Justice (ECJ), bilateral negotiations between the Commission and the member state in question take place with the objective of finding a consensual solution at the last minute. Correspondingly, the number of commencements of proceedings before the ECJ is very low in relation to reminder letters and reasoned opinions (Knill 2015). The fact that the Commission tends to avoid open conflict with non-compliant member states can be traced to several factors. First of all, the self-restraint towards a mediating rather than interventionist implementation style is favoured by the fact that the Commission partially anticipates that an extensive use of its formal powers might undermine the willingness of member states to accept 17

18 the expansion of EU policies in general (Lenschow 2007; Knill 2015; interviews 1, 2). Second, initiating formal infringements is highly resource-intensive, implying that the Commission is generally rather restrictive in making use of its formal powers (interviews 2, 3). Third, and related to this aspect, at least within the DG Environment, concerns about general effectiveness of EU environmental policy are given priority over potential problems of non-compliance in individual member states (interviews 1, 3, 4). Finally, the Commission is well aware of the fact that implementation failures in environmental policy often result from lacking administrative capacities in the member states rather than lacking political commitment. From this perspective, implementation is to a lesser extent based on hierarchically defined and controlled requirements, but is understood as a bargaining process between a great number of organisations and administrative agencies participating in the implementation process. OSCE Compared to the EU Commission, the OSCE has only limited formal powers to prosecute noncompliance among member states. Commitments agreed on in the Permanent Council are only politically binding, not legally: they are a political promise to comply with these standards (OSCE 2011a: xviii). Whether member states actually follow-up on human rights or other standards negotiated by the OSCE, or whether a state follows the advice provided by a field operation is merely a matter of political good will. The OSCE secretariat s powers in ensuring compliance are limited. The OSCE still lacks adequate centralised executive capacity which provides guidance and control, ensures coordination, coherence and continuity (Legutke 2005: 188). Similar to many other nongovernmental organisations (Murdie and Davis 2012), the OSCE s most powerful stick is naming and shaming. However, in order to avoid alienating member states, the OSCE secretariat rarely exposes individual states publically. Instead, it uses alternative strategies to increase its influence on national administrations. This includes, as a first tool, mediating measures such as ex-post specification of implementation routines. For example, providing assistance to police reform is an important part of the OSCE s institution-building agenda. In 2006, the OSCE was the first international organization to 18

19 publish a Guidebook on Democratic Policing, specifying the normative aspects of police reform as delivered by OSCE field operations (OSCE 2008: 7). To ensure that these standards are implemented as part of police reform training, the policing unit developed manuals that prescribe the exact curriculum trainers teach on the ground. This and similar activities serve to ensure upfront that OSCE standards find their way into member state policing practice (interviews 6, 7, 10, 12, 19, 20). The second tool illustrating the mediating implementation style is backdoor negotiation. In reviewing whether member states adhere to the OSCE s security or human rights standards, OSCE experts visit member countries on a regular basis. Their reports are never published or even shared among all member states. Instead, OSCE experts meet with member state officials on a confidential basis and discuss options on how to improve (interviews 16, 17). This is a powerful tool to convince member state administrations to adjust according to OSCE standards. However, it must be mentioned that the OSCE secretariat s mediating implementation style fails in the absence of political goodwill on the side of the host country. Interviewees mentioned the example of Kirgizstan, where the OSCE maintained a rule of law mission. After violent riots in 2010, the Permanent Council decided to dispatch police observers throughout the country. Yet, the Bishkek interim government did not support implementing the plan for what reason these officers never reached their destinations. In the absence of concerted pressure by the Permanent Council, which couldn t be rallied by the secretariat either, the Council decision never was enforced (interview 11) Comparison: The paradox of strength and weakness Summarizing the two case studies, we find striking similarities in administrative styles. Although the EU Commission and the OSCE secretariat show considerable differences in their structural leeway and resources for pro-active entrepreneurial policy-making, both IPAs show similar initiation styles: Both can be considered entrepreneurs when it comes to defining new policy problems. In the case of the EU Commission this follows partly from its mandate; but beyond that the Commission has been 3 For details on the Kirgizstan case see: 19

20 perceived as an entrepreneurial actor; in the OSCE it is partly against its mandate, which should make it more a servant of its member state. It has no formal right of policy initiative and is very dependent on member states. Similarly, the structural potential of the EU Commission and the OSCE is different with respect to their chances of influencing the content and process of policy formulation. The Commission is a centralized organization with high resources for policy research and formulation. By contrast, the OSCE secretariat is small and many staff have to deal with coordinative activities related to administrating field level activities. Nevertheless, both IPAs reveal similarly strategic routines in the way they draft policies such as regulations or mission mandates. Finally, in matters of implementation, both IPAs pursue mediating strategies and avoid open political conflict with member states. Yet, while this finding is highly plausible for the OSCE case, it is rather striking for the EU Commission that compared to other IOs disposes of the most far-reaching formal powers to enforce and monitor member state compliance. The finding of similar administrative styles among the OSCE secretariat and EU Commission is surprising. On the one hand, it is counterintuitive that the OSCE secretariat was able to venture beyond its potential in influencing policy initiation and formulation. On the other hand, and by contrast, while entrepreneurial and strategic policy-making was what we expected for the EU Commission, the mediating implementation style is striking. In explaining the paradox of weakness in the case of the OSCE secretariat, we suggest to devote more attention to the interest member states take in the work of IO bureaucrats. Much more than functional aspects of problem complexity, institutional constraints or cognitive factors, the steady drift into irrelevance of the OSCE since 1999 could explain why the secretariat at the time of research was forced to unfold entrepreneurial behavioural patterns and a strategic approach to policy formulation. During the past decade, the OSCE suffered from a huge loss of attractiveness to Western foreign and security policy. As mentioned earlier, the budget was cut by a fourth during the past decade. Overall, as one observer pointed out, the Western countries which set the agenda have deliberately given up the spheres of competence of the Organization to NATO, the EU and the Council of Europe, thus 20

21 marginalizing the OSCE (Milinkovic 2004: 201). Due to this context, as two diplomats said, the secretariat wants to avoid any image of policy failure and mission creep (interviews 13, 15). Controlling the agenda and being strategic to formulate effective policies are necessary conditions for this objective. By contrast, the EU Commission represents a paradox of strength. The Commission uses its available sanctioning powers much less than expected when overseeing how member states implement its regulative acts. Whereas in the OSCE case, a formally weak bureaucracy could develop rather entrepreneurial and strategic styles as long as it remained below the radar of its member states, the implementation style of the Commission reveals an opposite scenario. It is exactly the very strong position of the Commission that leads to intensified attention of its activities on the side of the member states. The anticipation of strong member states control thus leads to the self-restraint of a structurally strong bureaucracy. 8. Conclusion Administrative styles can be useful as a concept to compare administrative configuration across organisations. They look at administrations at the void between micro-level explanations emphasizing bureaucrats individual attitudes and macro-level explanations of organisational structures. Empirically, we were able to show that administrative styles at the international level differ from the national level. The EU and OSCE represent least likely cases for developing similar administrative styles. As opposed to the domestic context, however, differences in functionality, institutional opportunities and cognitive frames in the two organizations did not cause different styles. These findings challenge the proposed assumptions on causal links between structural features and administrative styles. Theoretically, we could conclude that because of their structural peculiarity their institutional weakness compared to national administrations all IPAs share a common type of administrative style. This is not the case. Instead, our analysis suggests that it is necessary to disentangle the causal mechanisms that lead to a certain style and consider explanatory conditions that differ from those 21

22 relevant for the analysis of national administrative styles. In particular, administrative styles in IO secretariats seem to differ depending on the extent to which member states take interest in their work. References: Abbott, K.W., Genschel, P., Snidal, D. and Zangl, B. (eds) (2015) International Organizations as Orchestrators, Cambridge: Cambridge University Press. Barnett, M. and Finnemore, M. (1999) The Politics, Power, and Pathologies of International Organizations, International Organization, 53(4): Barnett, M. and Finnemore, M. (2004) Rules for the World: International Organizations in Global Politics, Ithaca, NY: Cornell University Press. Bauer, M.W. and Trondal, J. (eds) (2015) The Palgrave Handbook of the European Administrative System, London: Palgrave Macmillan. Biermann, F. and Siebenhüner, B. (2009) Managers of Global Change: The Influence of International Environmental Bureaucracies, Cambridge, MA.: MIT Press. Biermann, F., Siebenhüner, B., Bauer, S., Busch, P.-O., Campe, S., Dingwerth, K., et al. (2009). Studying the Influence of International Bureaucracies: A Conceptual Framework, in F. Biermann and B. Siebenhüner (eds), Managers of Global Change: The Influence of International Environmental Bureaucracies, Cambridge, MA: MIT Press, pp Cerny, P.G. (2010) Rethinking World Politics: A Theory of Transnational Neopluralism, Oxford: Oxford University Press. Cini, M. (2015) The European Commission - Politics and Administration, in M. Bauer and J. Trondal (eds), The Palgrave Handbook of the European Administrative System, Basingstoke: Palgrave Macmillan, pp DiMaggio, P. and Powell, W.W. (1991) The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields, in Powell, W.W. and DiMaggio, P. (eds), The 22