When global and European regimes meet - The case of a national civil aviation authority

When global and European regimes meet - The case of a national civil aviation authority Mathias Johannessen, University of Oslo mathias.johannessen@stv.uio.no Abstract This paper is a case study of the Norwegian national civil aviation authority (CAA) and how the involvement in the global and European regimes for regulation of air traffic impacts national officials behavior. Knowing that the International Civil Aviation Organization (ICAO) and the European Union (EU) are organized differently, could one spur this difference in national level officials behavior? Empirical findings suggest that there is a difference, particularly in downstream processes (transposing and applying law). EU mainly issues regulations which leave little room for discretion when national officials transpose and apply EU-law, whereas ICAO provides guidelines that member states are free to adjust or not implement. Simultaneously, empirical findings show similarities in how upstream activities (preparation and/or revision) impact officials. Leaning on an organizational perspective shows that differences could be ascribed organizational factors. Institutionalism highlights similarities in contact patterns and in upstream phases as part of national authorities institutional heritage. Paper to be presented at the ECPR Graduate Student Conference in Bremen 3-6 July 2012. 1

1. Introduction This paper focuses on implications of the agencification processes at the national and European level (Christensen and Lægreid 2006). More precisely, I will here look at how the global regime for the aviation sectors, ICAO, and the EU-level aviation policy impacts officials in a national Civil Aviation Authority, the Norwegian one (hereafter CAA). The precise research question is to what extent the organizational differences between the global level (intergovernmental) and the European level (more supranational) manifests in different impact on national level civil aviation authority officials behavior. This study contributes on two central points. First, it provides an account of consequences working in a complex, multi-leveled policy field may have for the behavior of national level officials. The organizational composition of the field is in many respects organized as most policy fields: a global IGO, European level institutions and national institutions. In this respect, the findings could have relevance beyond the narrow empirical focus. Second, the paper compares the impact of a pre-existing global level regime with the European level. I will argue that seeing the organizational environment in this field as compound could gain some new insights. By including the global level, the paper recognizes that the EU-level policies do not fill an organizational vacuum. The EU-level is rather interpreted as an additional level of governance to the existing institutional framework dealing with international aviation regulation policy. The added value is thereby to test whether or not there is hold to the thesis of an emergent European executive order (Trondal 2010) by comparing the impact of the two qualitatively different regimes. The assumption is that the EU-level represents something new, or something more, than the traditional, intergovernmental, global organizations. As a small third point, Kassim and Stevens (2010: 270) suggest it fruitful to investigate more extensively the impact of international factors on the dynamics of EU-level decision-making in this particular field. This paper, to some extent, represents an attempt to follow up on this research agenda as well, seen from the national perspective. The empirical findings suggest that national officials behavior is impacted both differently and similarly. The differences are most visible when officials transpose and apply European and global law. Most of the EU-regulations are formulated in a very rigid manner and must 2

be implemented as is by member states while ICAO guidelines could be left unimplemented without consequences for its member states which results in a very different impact when officials transpose and apply law from the two systems. When deliberating new regulations/guidelines or revision of existing law, the organizational systems show more similar impact on national officials behavior. Leaning on an organizational perspective, it is shown that differences could be ascribed the organizational factors. Institutionalism suggests that similarities in contact patterns and way of working upstream could be part of national authorities institutional heritage. The paper is structured as follows: First, I will give an outline of previous findings and draw a map of the field. Second, I will give account for the theoretical perspective and outline theoretical expectations. Thereafter, I will give an overview of the methods and the empirical material before proceeding to the findings. Lastly, I move on to discussion and conclusion. 2. Background: What is known? A lot has been said about the dual delegation of regulatory powers to the European and the subnational level, the consequent regulative networks of national and European level institutions (Coen and Thatcher 2008; Eberlein and Newman 2008; Levi-Faur 2011). Coen and Thatcher (2008: 68) concludes that European level networks of national agencies have been given goals to co-ordinate national regulatory agencies and harmonize the implementation of EU law while at the same time not being granted enough powers and resources to fulfill that task. Eberlein and Newman (2008: 45) emphasizes that the regulatory networks are very well situated to cope with international governance needs as they are able to use national agencies to alter EU-law while the supranational level assists in developing rules. This hints at a European level able to make use of national resources in sectors where this is lacked at the supranational level. Others have looked the relationship between the EU-level institutions and national level agencies (see for example Barbieri and Ongaro 2008; Egeberg and Trondal 2009a, b; Martens 2008). These studies are indeed leaving the impression that EU impacts the behavior of national level officials in many ways. The point of view have often been on the phenomenon as seen from the European floor (Barbieri and Ongaro 2008; Coen and Thatcher 2008; Eberlein and Newman 2008) or in some 3

cases with emphasis on national authorities to see how the European level for instance impacts day to day activities in national agencies (Egeberg and Trondal 2009a, b; Martens 2008). However, it has been looked less into the relationship between EU-level institutions, national agencies and global regimes seen as a compound, organizational environment. Gulbrandsen (2011) makes for an exception, addressing the question of how EU is able to influence national bureaucracies way of implementation of international law into national law, e.g. harmonizing the implementation in a highly internationalized field (maritime safety legislation) (Gulbrandsen 2011: 1034). He shows that EU has impact on the behavior of national level inspectors behavior, among others by serving as an interpretative filter for the understanding of global law (Gulbrandsen 2011: 1048-49). 2.1 Background: The case International Civil Aviation Organization (ICAO) has since 1944 played a key role in coordinating safety standards among its 190 current member states globally (ICAO 2012a). At the European level, the key players are the Commission and EASA. One reason to establish EASA was according to Groenleer et al. (2010: 1221) to enhance harmonization regarding implementation of global legislation throughout Europe. This implies that variation in the implementation of ICAO guidelines/appendixes throughout Europe is a key motivation for EU to establish a common European platform for regulating aviation. Looking at the formal competencies of EU, it is interesting to note that EU evaluates the performance in national level aviation authorities through inspections (Groenleer et al. 2010: 1222) and that the Commission is given the right to sanction when member states show non-compliance with EU-law (European Parliament and Council 2008). At the national level, the formal actors are the Ministry of Transport and the CAA. There are also some intergovernmental organizations in the European area, such as Eurocontrol, but these are given little attention in this paper. Kassim and Stevens (2010) provides insights on the difference between the international level, labeled the traditional regime and the European level, making it very clear that the EU appear to encompass a lot more than the counterparts at the global level, for instance the agenda of liberalization as well as passenger rights. Second, some efforts have been given in order to explain the formal and legal structures of EASA and thereby to give an 4

account of the formal underlying potential for impact. Pierre and Peters (2009), has provided a thorough overview of the process leading from the predecessor of EASA, the closed club Joint Aviation Authority (JAA), to the establishment of the bureaucratic EASA. It is highlighted that the shift from JAA to EASA represented a shift from a situation where national regulatory bodies were more or less principals to becoming rather powerless agents for the EU under the current system for aviation regulation at the EU-level (Pierre and Peters 2009: 353). Schout (2011) in a similar vein asked the fundamental question about the added value of EASA compared the (JAA). He emphasizes that JAA was not as much of an underperformer as viewed by the European level and also that the anticipated changes following the establishment of the more formally competent EASA should not be overestimated as national governments still are active through the board (Schout 2011: 380-82). Groenleer et al. (2010: 1226) concludes that EASA is granted significant formal competencies, but that the most add[ed] value lies in the ability to spur informal learning among national regulatory authorities. Lastly, Chiti (2009) noted that EU has shown some interest in uploading its concerns to the global level in this field. In sum, it is reason to expect an EU with more far-ranging competencies than the global level in this particular field. 3. Theoretical departure This paper mainly applies an organizational perspective (Christensen et al. 2007; Egeberg 2004). An organizational perspective implies that a decision taken by an organization ultimately could be regarded an act of individuals. The degree to which an organization is able to affect the behavior, attitudes and loyalty of individual officials is dependent on the way an organization is structured (Egeberg 2004). The underlying assumption of this paper is thereby that structural design matters and in particular so when officials spend time in different regimes such as the global and European one. By decision-making behavior, it is here meant what concerns an official will pay attention to when for instance implementing international law. Since officials are not able to attend every concern simultaneously, they are acting in accordance with the concept of bounded rationality (Simon 1965). This means that how the organization is structured is a way to preselect which concern each individual official should pay attention to (Egeberg 2004: 202). In this paper, I will focus most on how participation in the secondary structures of (ICAO and 5

the European level) impacts national level officials. The way these are structured could be expected to have impact on the behavior of officials in different ways. Hofmann s (2008) analytical categories direct and indirect administration appears to be particularly useful in this case. In brief, indirect administration represents implementation directly by the member states (Hofmann 2008: 667) and thereby presupposes a large extent of flexibility for the constituencies when implementing international law. Since implementation of guidelines and appendixes from ICAO is implemented and enforced as national law (Kassim and Stevens 2010: 24), which is very visible the identified differences in application of ICAO-guidelines throughout EU (Groenleer et al. 2010), ICAO is assumed to fit fairly well as an example of indirect administration. Direct administration could in brief be said to be implementation of a policy through Community institutions (Hofmann 2008: 667). The EU-level is assumed to be infused with supranational elements and therefore expected to be somewhere on the continuum between the two types of administration. It is thereby central to attempt placing EU somewhere on this continuum. To do so, it is here a goal to understand how EU differs from the ICAO system, as seen by the impact on national level officials. The institutional perspective serves a supplementing function in this paper. The main focus is on institutionalization (Selznick 1997) and consequentially on path-dependency (Krasner 1988). By institutionalization, I here refer to the infusion of routines, norms and procedures over time (Selznick 1997) whereas path-dependency refers to how the history of an institution (or the process of institutionalization) binds the institution to make certain decisions in the present and future (Krasner 1988: 66). The basic assumption is that the process of institutionalization creates a certain set of preferences in which an institutional resistance towards change (Pierson 2000) could arise when these are challenged. In this context, the expanding competencies of the European level are viewed as a potential for such institutional resistance when the national level used to certain routines and procedures adapts to changes. 3.1 Theoretical expectations The theoretical expectations are structured along the lines of two ideal types; an indirect (ICAO) and a direct (EU) administration respectively. The paper follows the entire policy cycle (Howlett et al. 2009) seen as upstream processes (preparation and revision) and 6

downstream processes (transposing and application). The expectations for the impact of ICAO and EU on national officials in these processes are based on four behavioral factors; contact pattern, perceived level of conflict, perceived influential actors and role perceptions. All these factors are anticipated to shed light on how the two international orders impact national level officials (differently). First, on the upstream side in ICAO, it could be expected that the large share of contact between the global level and the national level will happen between ministry officials and ICAO rather than between ICAO and national CAA officials due to the territorial principles of organization. Second the expectation for the level of conflict is somewhat equivocal in upstream processes. On the one hand, ICAO basically produces guidelines and leaves it for member states to implement it (Kassim and Stevens 2010: 24) which should result in a low level of conflict upstream. On the other hand, there are 190 member states (ICAO 2012a) and it would in this light be surprising if there were no conflict at all. Since decision making in ICAO will follow the principle of unanimity, it could be expected that certain national ministries are among the most influential actors in upstream processes. ICAO in itself is expected to have little influence. Looking at the downstream side, the expectations are slightly different. First, since transposed and applied international law is de facto national law when incorporated in national legislation, contact is unlikely to be directed towards ICAO from both the Ministry of Transport and the CAA. Second, the legal status of transposed law furthermore leaves very little room for conflict, at least between the levels, over transposed law. Third, the Ministry of Transport and the national CAA are thereby also expected to be the most influential actors in the field when choosing how to transpose and apply the law. Turning to role perceptions, participating in a secondary structure such as ICAO could challenge preferences and expose officials for new concerns (Egeberg 2004: 203) and ICAO could be expected to do so if an official spend much time in the system. At the EU-level, the upstream processes could be expected to be characterized by more direct contact between CAA officials and the EU level because of the extended formal competencies EU has within this field (Groenleer et al. 2010). Second, the Commission and EASA have shown ability and capacity to run its own agenda (Egeberg 2003: 135-136; Pierre and Peters 2009) in contrast to the global level (cf. Trondal et al. 2012). Being organized 7

more along sectoral and functional lines than territorial ones (Egeberg 2006: 21-24), institutions such as the Commission and EASA at the EU-level could be expected to act more independently from the EU member states in upstream processes. This could lead to a potential for more conflict within the EU-context, as member states have to implement all relevant regulations and directives within this field. Thus EU is also expected to be more influential in the upstream processes. Turning to the downstream processes, given the competencies granted the EU level there could be expected some contact between EASA, the Commission and national officials when exercising discretion. Second, there could be expected to be somewhat more conflict between the levels, in particular when transposing regulations and directives. When applied, however, the level of conflict is less likely to be very potent. EU is furthermore expected to have a more direct role and hence influence on downstream processes than its global counterpart. When turning to role perceptions, officials have shown multi-hatted characteristics in previous studies, meaning that they have several concerns to take into account when taking policy decisions (Støle 2006; Martens 2010). EU is therefore expected to have impact on officials role perceptions and more so than ICAO. From an institutional point of view, the expectation is in brief that the degree of institutionalization matters and that one may tend to stick to established norms and procedures when adapting to changes in the institutional landscape. Since ICAO has been in the game since 1944 (ICAO 2012b), it is fairly likely that the ICAO system is more institutionalized than the European level and this is expected to have implication for the national level adaptation to the EU. 4. Methods and empirical material The empirical findings mainly build on five qualitative, semi-structured interviews (Rubin and Rubin 2005) with informants from the Norwegian CAA. The informants were selected through a snowball sampling process (Biernacki and Waldorf 1981). The selection criterion was involvement in international work (either the EU-level, ICAO or both. The empirical material from these interviews is the primary data source for this paper. All interviews were recorded with a digital recorder and transcribed shortly after the interviews took place and all informants were promised anonymity. It further appears that most of the informants 8

have been involved in both upstream and downstream activities. As I conducted the interviews, one informant just came back from Brussels and another was leaving for a meeting in a few days. Being very active at the European level, none of my informants reported to have significant direct experience in upstream activities in the ICAO system. Therefore, I chose to follow up the interviews by mail correspondence with a CAA official who is more directly involved in the ICAO-system. Process documents have also been used frequently, mainly when preparing the interview guide. Instead of following a specific case, or several cases, this study had the intention to take a snap shot of an average day at work. This implies that the questions were formulated in a fairly general manner. However, my interviewees have been encouraged to make use of examples throughout the interviews as is visible in the empirical findings. 5. Empirical findings From a general point of view, it is interesting to note that informants stress that legislation stemming from EU is formulated strictly and in a so detailed way that it leaves very little room for discretion on the downstream side. Most of relevant EU legislation in the aviation sector comes in the form of regulations. Regulations are normally more strictly formulated than for instance directives, and must be transposed into national law as is. Furthermore, the informants also appear to share a general impression that the EU-level has made it less interesting to follow up on ICAO-guidelines. On the one hand, this seems to be because ICAO guidelines are stated to be the basis for most EU-level regulations. If implemented, ICAO guidelines will have to be removed as soon as EU has made a regulation based on relevant ICAO guidelines. If not, the CAA would be in a position where they have a double legislative basis. On the other hand, by being involved in EU lawmaking through the EEA 1 agreement, Norway is legally obliged to implement EU law. In contrast, it is always possible to report non-compliance if they should disagree on parts of or an entire ICAO-guideline. In sum, this leaves the overall impression that the EU-level has more direct relevance than ICAO, and thereby currently given far more attention than ICAO by the CAA officials on a daily basis (Interviews). 1 EEA European Economic Area. Norway is through the participation in EEA equally obliged to implement relevant EU-law in the aviation sector as EU member states (Interview). 9

On the upstream side, and within ICAO, the CAA is normally represented through Nor-ICAO. Nor-ICAO is a constellation of the five Nordic countries, where the countries rotate on having a representative which participates in meetings and diffuses information. The four other are alternates and by that the natural contact point for the national level officials. According to one informant, upstream activities involved a low level of conflict, but diverging interests among the ICAO constituencies were also suggested to be fairly normal. In any case, it usually did not create a lot of tension because constituencies are well aware that they may report non-compliance if a specific regulation is regarded as unfruitful for their case. An informant said that If ICAO should agree upon something that we would find unsuitable for the Norwegian context we could just in principle attend the meeting, and skip implementing it or let parts of the document remain unimplemented when we get home (Interview). From ICAOs side, a fairly frequent contact with the ICAO-secretariat was suggested. However, the role is fairly limited and mainly involves circulating drafts and documents and taking care of practicalities, which is in line with the role IGO-secretariats normally play. The contact with the Ministry of Transport was furthermore reported to be very close and frequent, in particular when issues at stake had a political character. In addition, there were reported to be very much contact with other European states in general, and with the other Nordic CAAs in particular. It was also reported that it was far from unusual to coordinate positions with for instance U.S. and Canada, if one saw concurring interests in a specific case. EU was perceived as a very influential constellation in the upstream phases and someone whose arguments were paid a lot of attention. EU has no formal representation in ICAO, it only holds an observer status, but it was underlined that EU-states have a tradition for coordinating positions before attending ICAO sessions and thereby acting as a bloc (Interviews). An informant mentioned that there was an ongoing case within ICAO concerning the current ban on liquids. The ban on liquids was launched after an attempted terrorist attack at Heathrow. Consequently, EU launched an initiative in ICAO for banning liquids in-flight which they were able to pass through the ICAO system. Now, this ban has become a source of tension between ICAO and EU. Technology able to detect liquids in hand luggage is now available, and EU has therefore launched an initiative within ICAO to relegalize liquids brought on board. The problem, seen from a global perspective, is that a number of countries, in particular in Asia and Africa, are unable to afford the new 10

technology needed to abolish the plastic bag regime. The informant also meant that this was fairly symptomatic for the EU-behavior within the frames of ICAO and feared an increased tension followed by regional block-building over the next couple of years. EU has no formal seat in ICAO but coordinates the activities of member states before attending important ICAO-sessions. This informant felt that this was creating different blocs of countries around the world that tried to challenge EU (and other influential member states) within the frames of ICAO and that such tension between blocks occurred increasingly frequent (Interview). When asked about role perceptions, it was stated that one could perceive oneself as either an expert or a national representative. It was, however, suggested to be very hard to separate between the expert-hat and the national level representative hat at the ICAOlevel. It was furthermore suggested that at the ICAO-level, the hat depended on to what extent the issue at stake was technical or not (Interview). When working upstream in the EU-context, it was stated that CAA tries to participate in as much as possible within the borders of the CAAs organizational capacity to follow up on items in question. The main emphasis is on participation in preparatory committees rather than on comitology committees 2. When a case hits comitology it is more or less already decided upon. Therefore, it was an outspoken strategy to get involved as early as possible in the decision making processes (Interviews). Knowing the origins of a proposal also seems to be of vital importance in this regard. As one informant said, one thing is to be able to exert some influence on regulations as they are under development. For us, as authorities, it is equally important to understand why a particular initiative is launched (Interview). The CAA to some extent felt that they were out of the loop and thus not informed about for instance planned expert groups or policy initiatives that were in development. Therefore they tried to follow up especially closely matters of particular importance for the specific national context such as airport regulation and a regulations concerning offshore helicopter 3 traffic (Interview). Furthermore, it appears that the norm is that officials have a certain extent of contact with the parent ministry before going to a fair share of the European level meetings. 2 This should not be read as if they do not participate in comitology committees, on the contrary. It simply shows that the focus is on those aspects of the decision making process in which they are not automatically invited. 3 Norway has a very peculiar airport structure, and the officials perception was that a one size fits all regulation on airports would not fit Norway very well (Interviews). 11

One informant stated that the participation in the Single Sky Committee 4 almost exclusively happened accompanied with a representative from the Ministry of Transport when attending meetings (Interview). Another informant reported to on a recent occasion having participated on behalf of Norway in a top level meeting alone. In the top-level meeting, the future relation between EASA and Eurocontrol were subject for discussion and later decision. He was sent to the meeting with a clear mandate from the Ministry of Transport on the national position and stated that he basically participated in accordance with that mandate (Interview). It was, however, also indicated that it was equally normal that they attended meetings with a national expert. One such example could be the expert committee on new regulations on offshore helicopters where an expert from the industry was brought in to join the CAA official in an expert group (Interview). These examples leave an impression of a ministry capable of being an influential actor when the issues at stake get political while keeping a distance when matters have a technical character. Furthermore, EU was suggested also to play a very influential role in upstream processes by constantly launching new policy initiatives and expanding their competencies into areas that were previously regulated through ICAO and other European constellations, such as the regulation of airports (Interviews). In addition, the usual suspects such as UK, Germany and France were suggested to be particularly influential member states. If Norway seeks particular influence in a case, it was stated likely that they would form constellations with other states who they anticipated shared their interests. For instance in the case on offshore helicopter regulations a close contact with UK was reported. It is also interesting to note that the national authorities, particularly the Ministry of Transport but also the CAA, appears to be fond of the former (and current) intergovernmental cooperative bodies in Europe such as Eurocontrol and JAA. This was stated to be because Norway enjoyed (and enjoys) full membership in these frameworks (Interview). The level of conflict were overall suggested to be low, and again the informants stressed that other important contact partners were the other Nordic CAAs also in this context (Interviews). When asked about how they perceive their role in EUlevel, all informants have stated to feel like national representatives or experts. It was from some suggested that the role perception was flexible and depending on the type of issue at stake, while others meant that one would always represent national interests when 4 Single Sky Committee has the purpose of assisting the European Commission in the implementation of the Single European Sky. It is composed of representatives from the EU-member states as well as Eurocontrol and observers from third countries. 12

attending different committees (independent of the issue) (Interviews). If technical issues were to be debated, there were suggested natural to feel like experts. When participating in meetings concerning political issues as outlined above, feeling like a national representative appeared to be the norm (Interviews). In sum, the findings draw a picture of an EU with clear impact on officials in upstream activities and more so than ICAO. At the same time the impact of the Ministry of Transport appear to be of no less importance in the EU context than in ICAO. Turning to the downstream side (transposing and application), it appears that there is very much room for discretion from the CAAs side in the ICAO context. First, and in the transposing of international legislation, if one disagrees with a rule from ICAO, one may basically report non-compliance to ICAO and not transpose it into national law. It is also fully possible to make amendments and/or extend or specify the guidelines if it is regarded suitable for national conditions. This of course leaves little room for conflict when transposing ICAO guidelines and the overall conflict level were in this context suggested to be very low. The most influential actors when working downstream are suggested to be internal in CAA and the Ministry of Transport. It was, however, also reported that one from time to time consulted sister agencies, particularly in the Nordic region to consult as to how they transposed the guidelines as well. Turning to the application, an important aspect which most informants underlined was that the large bulk the EU-regulations within the field of aviation are based on existing guidelines from ICAO accompanied by very detailed appendixes specifying the regulation. ICAO-guidelines transposed in national law are by definition national jurisdiction when applied. If discretion is exercised it appears that the contact pattern mainly stays in-house at the CAA, but sometimes they also consult with people in the Ministry of Transport or sister agencies in the Nordic countries. Contact with ICAO was suggested to not happen directly at all on the downstream side. If at all ICAO were to be contacted, this would normally imply a phone call or an e-mail directed to one of the Nordic representatives in Nor-ICAO for clarification. Overall, CAA appear to have a lot of influence on downstream ICAO processes. When looking at downstream processes at the EU-level, there is suggested a very limited influence for national actors and consequently little room for exercising discretion when 13

implementing EU regulations in to national law. The regulations are very specific, and close to all regulations have EEA-relevance. One example of non-compliance from the Norwegian side was suggested to be regulation 216/2008 which extends EU competences, delegating more power to EASA and giving the Commission competence to give national authorities financial sanctions in case of non-compliance (European Parliament and Council 2008). The non-compliance with 216/2008 was suggested to be a challenge for the implementation of new regulations that presupposes compliance with it but it is now well on its way to be transposed (interview). The Ministry of Transport, according to some of the informants, has been heavily involved in delaying the implementation of this regulation. The reason for this is that, again according to my informants, the regulation would give EU too much influence by granting the Commission competence to launch sanctions on Norway (Interviews). It is here worth noting that the Ministry of Transport s influence appears to be limited to the ability to delay the process for a couple of years. But they have apparently been able to exert some influence since the revised version gives the European Surveillance Authority (ESA) rather than the Commission competence to give sanctions in the Norwegian case (Interview). When applying transposed law, the general impression from the EU-level is that most of the discretion is exercised in the CAA to the extent discretion happens at all. Again, the informants also underlined that EU-legislation is so strictly formulated that they rarely have to exercise much discretion when applying the regulations (Interviews). When they on rare occasions needed clarification on a matter, the contact point would be relevant EUinstitutions such as EASA. One informant underlined that contacting EU-level institutions was fairly normal for their inspectors when they practiced the very detailed regulations (Interview). It is also suggested to be very that sister agencies are consulted, most frequently the agencies among the Nordic countries (Interviews). In sum, downstream activities in the EU context appear to be characterized by specific rules and little room for maneuvering when regulations are transposed and applied. In ICAO, the national level appears to have much more room for making adjustments and/or not implement legislation. 6. Discussion In this section, it is discussed how the general findings relate to the theoretical expectations but also how the European and global level differs along the variables in up- and downstream processes. I will start by summing up and discussing the findings related to 14

expectations from the organizational perspective upstream and downstream in ICAO and EU. Thereafter I will discuss finding regarding role perceptions before turning to a brief discussion about some findings that could be further analyzed with an institutional perspective. In the section on theoretical expectations, it was argued that differences in organizational structure could be explored by looking at the different behavior of officials along the impact national officials contact pattern, perceived level of conflict, perceived influence and role perceptions. Starting with ICAO upstream activities, the empirical findings show that most of the contact with ICAO happened through Nor-ICAO, and consequently one of the national CAA officials in Montreal. This implies that the expected contact pattern, that mainly ministry officials would be in contact with ICAO, did not fit very well with the empirical findings at a first glimpse. However, the ministry was reported to be in close contact with the CAA officials in particular when political issues were at stake. This implies that the ministry holds a pivotal role in upstream processes and is after all a clear sign of indirect administration despite the lack of hands on participation. Other contact points of importance included sister agencies, ministry officials, experts and other countries with concurring interests, which also could be seen as in line with the expectations to an indirect administration (main emphasis on other national actors). Furthermore, the level of conflict was also reported to be low. The theoretical section suggested the level of conflict to be low, but that some tension could be expected. An example of tension could be the ongoing proposed re-legalization of liquids on board. EU, and particularly the large member states, was suggested to be among one of the more influential parties at the ICAO level in addition to countries like U.S. and Japan. The activity of EU appear to be a somewhat surprising finding, given the intergovernmental structure but falls well in line with Chiti s (2009) observation that EU tries to upload concerns to the global level. The theoretical expectation was that certain national ministries would be the most influential actors, whereas the empirical section suggests a mixed answer to whether countries or blocks of countries are the most influential actors. It should here be noted that the EU-countries formally meet as individual countries, but they coordinate the activities before meetings. When asked about how they perceive their roles the informant with upstream experience in ICAO stated to feel like expert or national representative, depending on the type of issue 15

at stake. Furthermore, the informant suggested it very hard to separate between the expert-hat and the national level representative hat. It was furthermore suggested that at the ICAO-level, the hat depended on to what extent the issue at stake was technical or not but the borders were blurred. This finding suggests that the role at the ICAO-level is fairly unaffected by the participation in upstream processes. Turning to the upstream processes in EU, the observed level of conflict is low. This harmonizes well with the expectation for an indirect administration. The theoretical expectation for direct administration suggested that the potential for conflict should be larger in EU-level upstream processes than in ICAO. This expectation finds no support in the empirical material. However, it should be stressed the field is fairly technical in its character. In addition, much of the legislation being discussed has its origin in ICAO and is already to a large extent part of national law. This aspect alone could possibly reduce the potential for conflict as the EU-level legislation thereby does not represent something entirely new for national administrations. Turning to contact patterns, the key contact points are suggested to be the ministry, experts, sister agencies and countries with shared interests. This finding is also in line with the expectations to an indirect administration, and fairly similar to the observed contact pattern in ICAO, but with much more time and resources spent. Influential actors are suggested to be the large member states in central parts of Europe, but the EU level also appear to have the ability to launch initiatives and take up an influential role. This finding is more in line with the expectations. EU was expected to be able to run a more independent agenda setting role than ICAO, and appear to do so also at the ICAO level. Most informants mainly identified themselves with being a national representative or sometimes an expert which is similar to ICAO. Sometimes, informants appear to show signs of perceiving European concerns as making equally much sense as concerns from the Ministry of Transport. A good example could be the underlying support from the CAA for the already mentioned 216/2008. I will argue that this finding shows officials who are taking in European concerns. However, the impact on their role perceptions may be more limited than what previous studies have shown (Støle 2006; Martens 2010). In sum, ICAO and the EU-level appear to impact officials behavior fairly similarly on the upstream side. However, that EU launches initiatives and thereby plays a more independent role in upstream 16

processes could indicates a certain tendency of a more direct EU administration on the upstream side. On the downstream side in ICAO, the contact pattern appears to stay mainly in house but they reported frequent contact with sister agencies and the ministry to clarify the understanding of guidelines and appendixes due for transposing. The absence of contact with ICAO when transposing and applying global law harmonizes fairly well with the expectations, but the ministry is apparently less hands on than expected. Turning to conflict, there was reported virtually no conflict at all over ICAO guidelines and appendixes when transposing or applying them. This fits well with the expectation as ICAO guidelines and appendixes are not binding for the constituencies which leave the playing field open for them to transpose and apply the legislation as it suits their context. Furthermore, the influence was perceived as entirely on the national authorities side which also in the same vein fits fairly well with the theoretical expectations. However, it should here be noted that the EU level aims at harmonizing the European application of global law. This could then indirectly reduce the national independence in terms of the flexibility they have had when implementing global law. In EU, it is notable that very little discretion is exercised when dealing with transposing and application of EU-law. Since the field is characterized by mainly regulations, which leave very little room for exercise of discretion, it seems that there is little contact with other relevant organizations as officials normally can apply standard procedures. The main impression is that they rely very much on in-house knowledge and formal procedures. This is a fairly clear sign of a direct administration where a large bulk of the decisions is taken at the EU-level and must be transposed merely word by word into national law. When (and if) exercising discretion while transposing or applying EU-regulations, CAA officials tend to prefer contact with sister agencies and the parent ministry. However, they also state that EU-level institutions, such as EASA, are consulted much more frequently than ministry officials in these (predominantly technical and very detailed) cases; in particular by inspectors. Conflict, in general, appears more frequently on the downstream than on the upstream side although the conflict level despite this appears to be remarkably low, another finding which goes well with the expectations to an indirect administration. The ministry s attempt to delay the 17

process of transposing regulation 216/2008, which resulted in a minor revision, also draws a picture of a very influential EU level downstream. In sum, ICAO to a large extent meets the theoretical expectations to indirect administration on the downstream side. The EU-level rather leans towards the direct administration side in downstream processes. The clearest signs of direct administration lie in the frequent use of regulations and ditto influence as well as in the reported contact pattern. When turning to institutionalism, it is interesting to note that sister agencies appear to play a pivotal role as contact points both upstream and downstream in the global and European context. In some cases, sister agencies are apparently of almost equal importance as the parent ministry. This appears to be somewhat in contrast to the finding of Egeberg and Trondal (2009b) where sister agencies are seen as important players, but ranking well below parent ministries in general. Since the contact pattern matches fairly well with formal, institutionalized channels at the ICAO level, it could be argued that this is an example of path dependency (Krasner 1988) where existing routines, norms and values plays in. Furthermore, the apparently institutionalized contact pattern with sister agencies on a general basis appear to happen either bilaterally (with actors who share a particular interest) or through the Nordic region rather than through networks with the Commission or EASA as a key player. This suggests that previously institutionalized networks between groups of sister agencies are important. The officials did not put emphasis on the importance of European regulative networks at all (Eberlein and Newman 2008; Coen and Thatcher 2008) when asked about important contact points. This could be a sign of possible resistance towards institutional change (Pierson 2000). Furthermore, the ministry appeared to play a more active role at the EU-level than expected and a less active role, at least hands on, in the ICAO-context. This could be a sign of the ministry treating the EU-level in accordance with established routines for how they deal with the ICAO-level while giving the EU level more attention simply because EU is regarded more important. Lastly, the national level authorities have shown preference for intergovernmental over supranational organizations. This could possibly be explained as part of their institutional heritage; they are simply more familiar with the intergovernmental arrangements both at the global and the European level. 18

7. Conclusion This paper has compared the European level with the global level assuming that organizational differences will manifest itself in national officials behavior. First, one may say that the main discrepancy between EU and ICAO, and thereby the clearest example of direct administration, lies in the impact on CAA officials behavior on the downstream side. The clearest example of direct administration seems to be the very frequent use of EU-level regulations which are directly binding for the EU member states. Another clear example is that the EU-level is directly approached when inspectors exercise discretion. However, it appears that the behavior of officials is fairly similarly impacted when dealing with ICAO and EU in upstream phases, with the possible exception that the EU level is capable of launching initiatives and setting the agenda. There could therefore be argued that the turn towards a more direct administration at this point manifests most clearly at the downstream side. As shown, the similar contact pattern and heavy involvement of national ministry officials in EU upstream phases could very well be interpreted as an example of institutionalization in to the ICAO system and its mode of work. The similarity in the contact patterns and the preference for intergovernmental bodies are signs that institutional history matters. 19

Literature Barbieri, D. & E. Ongaro (2008) EU agencies: what is common and what is distinctive compared with national-level public agencies in International Review of Administrative Sciences 74 (3), 395-420. Chiti, E. (2009). An Important Part of the EU's Institutional Machinery: Features, Problems and Perspectives of European Agencies, Common market law review, 46 (5), 1395-1442 Christensen, T. & P. Lægreid (eds.) (2006). Autonomy and Regulation. Coping with Agencies in the Modern State, Cheltenham: Edward Elgar Christensen, T., P. Lægreid, P. Roness & K.A. Røvik (eds.) (2007). Organization Theory and the Public Sector. Instrument, Culture and Myth, New York: Routledge Coen, D & M. Thatcher (2008) Network Governance and Multi-level Delegation: European Networks of Regulatory Agencies? in, Journal of Public Policy 28 (1), 49-71. Curtin, D. & M. Egeberg (2008) Tradition and innovation: Europe s accumulated executive order, West European Politics 31(4): 639 61. Eberlein, B. & A.L. Newman (2008) Escaping the International Governance Dilemma? Incorporated Transgovernmental Networks in the European Union, in Governance 21 (1), 25-52. Egeberg, M. (2003). The European Commission, in M. Cini (ed.), European Union Politics. Oxford University Press Egeberg, M. (2004) An organizational approach to European Integration: Outline of a complementary perspective, in European Journal of Political Research 43 (2), 199-219. Egeberg, M. (2006). The Institutional Architecture of the EU and the Transformation of European Politics, in Egeberg, M. (ed.): Multilevel Union Administration. The Transformation of Executive Politics in Europe. Basingstoke: Palgrave Macmillan. Egeberg, M. & J. Trondal (2009a). Political Leadership and Bureaucratic Autonomy: Effects of Agencification, Governance, 22(4): 673-88. 20

Egeberg, M. & J. Trondal (2009b). National Agencies in the European Administrative Space: Government Driven, Commission Driven or Networked?, Public Administration, 87(4): 779-90. European Parliament and the Council (2008) Regulation (EC) No 216/2008, on common rules in the field of civil aviation and establishing a European Aviation Safety Agency, and repealing Council Directive 91/670/EEC, Regulation (EC) No 1592/2002 and Directive 2004/36/EC, Brussels Groenleer M., Kaeding M. and Versluis E. (2010) Regulatory governance through EU agencies of the European Union? Journal of European Public Policy 17(8): 1212 30. Gulbrandsen, C. (2011). The EU and the implementation of international law: the case of sea-level bureaucrats, in Journal of European Public Policy 18 (7), 1034-51 Hofmann, H.C.H. (2008) Mapping the European administrative space, West European Politics 31(4), 662 76. Howlett, M., M. Ramesh & A. Perl (2009). Studying Public Policy. Oxford: Oxford University Press ICAO (2012a). Contracting States, [URL]: http://www.icao.int/cgi/statesdb4.pl?en [Visited 2 April 2012] ICAO (2012b) ICAO in Brief, [URL]: http://www.icao.int/pages/icao-in-brief.aspx [Visited 2 April 2012] Kassim H. and Stevens H. (2010) Air transport and the European Union. Houndsmill: Palgrave. Krasner, S. (1988). Sovereignty. An Institutional Perspective, in Comparative Political Studies 21(1), 66-94. Levi-Faur, D. (2011) Regulatory networks and regulatory agencification: towards a Single European Regulatory Space, in Journal of European Public Policy 18(6), 810-29. Martens, M. (2008) Runaway Bureaucracy? Exploring the Role of Nordic Regulatory Agencies in the European Union, in Scandinavian Political Studies 31(1), 27-43. Pierre J. and Peters B. G. (2009) From a club to a bureaucracy: JAA, EASA, and European aviation regulation. Journal of European Public Policy 16(3): 337 55. 21