regulation of food safety in the eu: patterns of multi...
TRANSCRIPT
Prof. Dr. Gabriele Abels
University of Tuebingen
Institute of Political Science
gabriele.abels@uni‐tuebingen.de
Alexander Kobusch, M.A.
University of Tuebingen
Institute of Political Science
[email protected]‐tuebingen.de
Regulation of food safety in the EU:
Changing patterns of multi‐level governance
Paper presented at the Conference of the ECPR Standing Group on Regulatory Governance,
June 17‐19, 2010, University College Dublin
Abstract After several food scandals in the past decades, regulatory regimes in food safety have
been reformed on the level of the European Union and in most EU member states. In most countries, independent agencies have been created and mainly perform tasks in risk assess‐ment. In our contribution, we focus on national institutional choices concerning these agen‐cies in all 27 member states and ask whether or not these choices can be explained by path dependency or rather as a phenomenon of Europeanization at the national level. Based on a survey of all national regimes and the EU level we can observe some trends that require fur‐ther scholarly studies. The main feature is that there are two dominant agency models in food safety governance: either a separation of risk assessment from risk management or an integrated model. The first one is a new development triggered by soft Europeanization ef‐fects, which are due to functional requirements and learning effects. Furthermore, these agencies have a more or less central role within the emerging network between the Euro‐pean Food Safety Authority EFSA and the national competent authorities.
2
1. Introduction Topic of the paper Triggered by a number of major food scandals, European and national food safety policies
and regulatory structures were the subject of profound reforms. While on the European lev‐el a new agency, the European Food Safety Authority (EFSA), was established and risk as‐sessment by scientific experts was strictly separated from risk management tasks, which are left to policy‐makers, most EU member states set up specialized agencies with different or‐ganizational features. In several regulatory designs, a strict separation in the risk analysis process can be observed. Based on a survey including all EU member states and the EU level, we aim to establish patterns of risk regulation regimes.1 We then ask whether or not the emergence of certain institutional designs can be explained by Europeanization trends or by national path‐dependencies. Our hypothesis is that there is a mixture Europeanization and domestic factors that explain changes in regulatory regime and regime design. We also as‐sume that there is substantial transfer, emulation and isomorphism involved in the field, especially among those member states that have introduced “separatist” regulatory regimes regarding the separation of risk assessment from risk management. This transfer is triggered, amongst other factors, by functional requirements in regulatory networks as well as by learning. Convergence to a single model of food safety regulation is, however, unlikely.
The reconfiguration of European food safety regulation In the late 1980s and throughout the 1990s, the European Union was struck by several
food scandals. Especially the BSE crisis induced major reforms in several national regulatory arrangements of food safety and also within the Union. Consequently, food policy and politi‐cal‐administrative designs for food safety regulation became politicized. The BSE crisis “cre‐ated a window of opportunity for the development of a more internally integrated food safety policy (and consumer health policy in general)“ (Ugland/Veggeland 2006, 618). The confession of British Secretary for Health Stephen Dorrell on March 20th 1996, that there might be a link between BSE and a new variant of the Creutzfeldt‐Jakob‐Disease (nvCJD), put the regulatory regime in Great Britain and on the European level under pressure. Both the British government and the European institutions were confronted with severe accusations. Among other aspects, a lack of relevant expertise in committees, the systematic exclusion of critical scientists, lack of timely release of information to the public, and the blending of sci‐ence and politics were criticized (Baggot 1998, 64‐72; Bartlett 1998, 238‐244; Buonanno 2006, 262‐263; Chalmers 2003, 534; Chambers 1999, 99; Jasanoff 1997, 226). While the Brit‐ish “BSE Inquiry” of 1997 reached more modest conclusions, the inquiry report by the Euro‐pean Parliament even suggested misinformation and political pressure by the British gov‐ernment in order to prevent further investigations and major drops in consumer demand for British beef.
An increasing mechanisation of food production (e.g. genetically modified food, “func‐tional food”) and a growing interdependence in food trade both in the European internal market and in global trade are additional factors contributing to the need for reforms. All of
1 We are grateful to Jennifer Träsch and also to Nursemin Sönmez for their supporting background research
on national food safety systems.
3
these aspects raise questions about how to guarantee the quality and security of food, bring‐ing the question of new forms of governance in food safety regulation up for debate. Pro‐tecting consumers from possible health consequences which can result from the production or consumption of food is today an important criterion for developed social statehood. The regulation of risks – also of food‐related risks – has become, besides forms of self‐regulation by the industry, an integral part of a state’s tasks. The regulation of food safety has become a key factor in modern consumer policies, both in the member states of the EU and on the European level.
As the member states felt that a more coherent European approach to food safety would be necessary to secure a high level of protection for consumers as well as the functioning of the Common Market, the General European Food Law (GFL) was adopted in early 2002. Ac‐cordingly, the organization of science and expertise was one of the major issues in the re‐form of the European system of food safety:
The Commission issued a Green Paper (1997) and White Paper (2000) on food safety which eventually led to the introduction of the General Food Law (Regulation (EC) 178/2002 2002)
By Reg. (EC) 178/2002, the European Food Safety Agency (EFSA) was established which provides expertise and coordinates European risk assessments.
The responsibility for food safety was transferred from the Directorate‐General Agriculture to the newly established DG Health and Consumers (DG Sanco), thus separating the promotion of industrial interests (DG Agri) and consumer interests (DG Sanco).
Food safety as part of public health became a horizontal issue which has to be considered in all EU policies
On the national level, responsibilities were equally rearranged and several inde‐pendent agencies were founded
In the European case and in many national cases, risk assessment as scientific process was separated from risk management as political process
An extensive debate on participation and transparency in the European political system was stipulated and eventually inspired the White Paper on Governance (2001)
The reconfiguration of national food safety regulation The General Food Law lays down certain principles on food safety regulation and contains
detailed rules for the institutional design of EFSA. As Majone concedes for the development of regulatory policies in Europe in general: “EC laws, policies and judicial decisions are not the only, but certainly the most important, stimulus to current regulatory developments in Europe“ (Majone 1996, 266). However, the GFL does not provide for any guidelines on the design of national authorities, and different political‐administrative designs have therefore been implemented in the member states (Federal Institute for Risk Assessment 2009). Some member states changed the very foundation of their regulatory system while others stuck to
4
their established structures. In most member states, specialized food safety agencies have been set up which are formally independent but accountable to either the ministry of agri‐cultural or health (with some exceptions). Especially the new member states in Central and Eastern Europe, which joined the Union in 2004 and 2007, had to undertake major efforts to adapt their food law and regulatory capacities to meet the demands of the internal market framework – and not all of them have completed this process yet. Within this process, the so called Twinning Projects2 included the reform of administrative structures within food safety regulation. Even within the associated EFTA member states (Iceland, Norway and Switzer‐land) and the Candidate countries (e.g. Croatia and Turkey), such reforms have been made or are ongoing. Consequently, we can still observe astounding institutional dynamics in food safety regulation.
Besides its key role in risk assessment, EFSA is expected to establish a network between the national Competent Authorities (CAs), which connects both the national and European level, and the CAs amongst each other in order to foster cooperation and data exchange between the member states. Thus, we can observe the establishment of a multi‐level trans‐national regulatory network with EFSA playing a major role. Due to its emergence, both a regular exchange of information and harmonized risk assessment methodologies and stan‐dards are expected. Both should contribute to a more coherent scientific expertise through‐out Europe (see also Kobusch 2010).
Outline of the research paper The paper proceeds as follows: We first draw on theoretical arguments about the prob‐
lem of separating risk assessment from risk management and why this matters. In this con‐text, we will briefly review some definitions for regulation and regulatory agencies in order to develop a framework to characterize national agencies. This framework is then aligned with the regulatory process in the member states with regard to whether or not a CA is part of a specific step in the regulatory process (section 2). Following that, we then discuss the theoretical background of our hypothesis by introducing the rival theories of path‐dependency vs. Europeanization (section 3). After the theoretical part, we will turn to the empirical part of our paper. We briefly outline the national systems of selected food safety regulation in the EU member states and at the supranational level using the framework and discuss some exemplary cases of Austria, Hungary and Sweden.3 Based on our findings, we will cluster the national CAs in groups of ideal‐type designs (section 4). Based on the over‐view, we try to identify patterns of national regulatory designs in the empirical mosaic and link these to the structure of the multi‐level regulatory network (section 5).
2 In the Twinning Projects, single Member States advise and assist a Candidate country in reforming its insti‐
tutional structures and in adapting to the requirements of the aquis communautaire. Often Twinning part‐ners share historical ties or closer relations among each other.
3 For a brief description of all 27 national food safety systems see the Annex.
5
2. Major Principles of European Food Safety Regulation Food safety regulation in the EU has undergone tremendous changes since the mid‐
1990s. Until then, food safety regulation had developed “in a piecemeal fashion” based on different legal bases and “in conjunction with its efforts to eliminate trade barriers arising from different domestic legislation in order to establish an internal market” (Alemanno 2006, 237). The strong call for an “agency solution” (Alemanno 2006, 238) and for a cross‐cutting, “integrated policy approach” to food safety (Ugland/Veggeland 2006, 607) was the dominant political response to the crisis. Yet the crisis did not only leave a political footprint but also a scholarly one. Following the politicisation of food policy, food safety issues have increasingly attracted scholarly interest throughout the last decade. Insights have been heavily drawn from an established body of literature dealing with independent agencies and regulatory governance. Meanwhile, EFSA’s institutional structure and its position in the European food safety regime have become major topics both in policy analysis focusing on foodstuff as well as in cross‐sector analyses and comparisons. Based on theoretical discus‐sions on regulation and agencies, in the following section we develop a framework for char‐acterising agencies.
Regulatory Governance Selznick, for example, who was heavily inspired by the American regulatory system, con‐
ceptualises regulation as “sustained and focused control exercised by a public agency over activities that are socially valued” (Selznick 1985, cited in Scott 2006, 653). Similarly, Bald‐win, Scott and Hood define regulation as “the promulgation of an authoritative set of rules, accompanied by some mechanism, typically a public agency, for monitoring and promoting compliance with these rules” (cited in Jordana/Levi‐Faur 2004, 3‐4). Relaxing the “agency‐condition” for the European case makes sense, since some regulatory regimes are not neces‐sarily based on such authorities (Scott 2006, 653). As will be shown in the empirical part of this paper, agencies are not in all cases the cornerstone of food regulation regimes in the EU member states. Consequently, Baldwin and Cave suggest understanding regulation in differ‐ent senses: as “all forms of social control or influence”, as “deliberate state influence” or as ”a specific set of commands”. The latter they specify in the sense that “regulation involves the promulgation of a binding set of rules to be applied by a body devoted to this purpose” (Baldwin/Cave 1999, 2). Hereby regulation often aims to restrict certain behaviour or gives incentives to refrain from undesirable activities (ibid.). Levi‐Faur argues along these lines when he points to the weakness of Selznick’s definition to exclude non‐public forms of (self‐ or business‐to‐business‐)regulation (Levi‐Faur 2010, 8). He suggests a wider definition that includes various aspects:
“[R]egulation is the promulgation of prescriptive rules as well as the monitoring and enforce‐ment of these rules by social, business, and political actors on other social, business, and politi‐cal actors. These rules will be considered as regulation as long as they are not formulated direct‐ly by the legislature (primary law) or the courts (verdict, judgment, ruling and adjudication). In other words, regulation is about bureaucratic and administrative rule making and not about leg‐islative or judicial rule making” (Levi‐Faur 2010, 9).
As such, bureaucratic and administrative rule‐making regulation is a specification of the wider governance‐framework. According to the definition by James Carporaso, governance is
6
“collective problem solving in the public realm”. The attention is directed towards “the prob‐lems to be solved and to the processes associated with solving them, rather than to the rele‐vant agents” (Carporaso 1996, 32‐33). Regulation specifies both certain instruments used (cf. (Baldwin/Cave 1999, 34‐62) and certain actors (in our case bureaucratic‐administrative authorities; cf. Levi‐Faur above; Baldwin/Cave 1999, 63‐75). In this sense regulation only focuses on a certain part of the ideal‐type policy cycle, namely the policy implementation where legislative acts and goals are transposed into rules and enforcements formulated by administrative rather than legislative bodies. Hence, within the European Union’s political system, regulation constitutes one of several governance or “policy modes” (Wallace 2005). Nevertheless, regulation may become a question of governance itself: “Much of the aca‐demic and public discussion of regulation nowadays deals with the governance of regulation itself (or regulating regulation) rather than governance via regulation” (Levi‐Faur 2010, 20). This also holds true for this research paper with its focus on the institutional design and gov‐ernance of food safety regulation in the EU and its member states.
Regulatory Agencies In many definitions of regulation, the concept has been linked to agencies. Levi‐Faur de‐
fines agencies as organizations that conduct a number of tasks. According to him a
“regulatory agency is a non‐departmental public organization mainly involved with rule making, which may also be responsible for fact‐finding, monitoring, adjudication, and enforcement. It is autonomous in the sense that it can shape its own preferences; of course, the extent of the au‐tonomy varies with its administrative capacities, its ability to shape preferences independently, and its ability to enforce its rules. The autonomy of the agency is also constituted by the act of its establishment as a separate organization and the institutionalization of a policy space where the agency's role becomes ‘taken for granted’. Note that rule making, fact finding, monitoring, adjudication, and enforcement capacities are defining characteristics of regulatory agencies, but also that other organizations, both within and outside the state, can acquire and successfully deploy these characteristics” (Levi‐Faur 2010, 15).
Implicitly, this definition also includes different stages of the regulatory process: (1) fact‐finding, (2) rule‐making, (3) monitoring, (4) enforcement and (5) adjudication. This is an im‐portant insight, as it allows characterisation of different agencies according to their scope of functions and their role within the regulatory process. Many studies on regulatory agencies are for methodological reasons based on a comparison of agencies with similar structures and competencies.4 This paper, in contrast, is focused on only a single policy field and in‐cludes all national agencies in this sector regardless of their internal design. Thus, a charac‐terisation of agencies is necessary to understand their varying roles in national policy‐making. The differences in powers delegated to agencies become even more apparent when the European Authority EFSA is included. The General Food Law foresees a clear‐cut separa‐tion within the regulatory process. Accordingly, the process is split into risk assessment, risk management, and risk communication (Regulation (EC) 178/2002 2002, Art. 3 (10)‐(13)). While EFSA in this account is the solely responsible body for risk assessment, all risk man‐agement decisions are reserved for the democratically accountable institutions (Council, Commission and European Parliament). In this sense, the functions (2) rule‐making, (3)
4 Maggetti (2007, 276), for example, based his sample on “consistent comparability within the agencies’ or‐
ganizational models” and “on the most institutionalized agencies and those that benefit from the greatest powers“.
7
monitoring and (4) enforcement remain with these bodies while EFSA is responsible for (1) fact‐finding only.5 In this regard, and compared to some national bodies, EFSA has a very limited mandate, the existence of which can be explained by primarily political reasons:
“In spite of functional pressures to delegate powers upwards, the bulk of the formal powers are still located at the national level. On the one hand, thus far the political resistance of member states has not allowed any far‐reaching transfer of regulatory powers to a supranational ‘regula‐tory state’; while, on the other hand, the EU framework of rules to which member state regula‐tory regimes are subject does not fully match the functional need for uniform EU rules. [...] The European agencies’ modest endowment with formal powers and resources also shows that these agencies are not oriented towards replacing the corresponding national institutions as some sort of ‘European regulatory state’” (Eberlein/Grande 2005, 95).
Thus, it becomes apparent why the Union’s own definition of agencies is very limited with regard to the tasks assigned to the agencies which are “set up by an act of secondary legisla‐tion in order to accomplish a very specific technical, scientific or managerial task” (European Commission 2010; emphasis AK).
A Framework Characterising Agencies Building upon the classical (and ideal‐type) policy cycle, a framework for describing and
characterising agencies emerges. First of all, let us consider the definition of the problem, agenda‐setting, policy formulation, policy implementation and policy evaluation as the main steps of the process (e.g.Birkland 2001, 221‐223). Regulation as understood here is then the transposition of policies into rules and enforcements by administrative bodies distinguished from legislative or judicial institutions and consequently the policy implementation part of the process. As previously explained, we further divide this process into four steps (fact‐finding, rule‐making, monitoring and enforcement) according to the agency definition by Levi‐Faur. This is very much in line with the assertion that regulation is not limited to rule making but also includes rule monitoring and enforcement (Hood et al. 2001, 6). Taking the separation of risk assessment and risk management into account, we can identify fact‐finding as a task of risk assessment while the remaining tasks are classical risk management responsibilities.
5 The responsibility for adjudication (5) lies with the European Court of Justice (ECJ), whose involvement is
not addressed in this paper. It is only relevant if a CA has the mandate to disseminate binding decision within the risk management process. The possibility of legal action against scientific outputs of the risk as‐sessment stage is so far uncertain. For example, scientific opinions by EFSA do not posses any legal status. Thus, so far the Court of first Instance ruled that only the final decisions and directives by the Commission and not EFSA’s scientific opinions as mere preparatory work to the decision‐making process can be chal‐lenged before the ECJ. For a detailed discussion see (Alemanno/Mahieu 2008).
8
Why separate risk assessment from risk management? In order to better understand different risk regulation designs, distinguishing between
risk assessment and risk management is of the outmost importance. When the founding regulation of EFSA was negotiated, the separation of the regulatory process was one of the main issues. We first want to give a short account of the reasons that theoretically lead to the establishment of independent agencies in general and to the specific design of EFSA in particular. We then focus on the institutional set‐up of EFSA that in all its nuances reflects the general idea behind separating risk assessment from risk management.
The establishment of a European food safety agency was already proposed in the early 1990s. At the time, the European Commission judged this to be unnecessary. So why did the food crises change this necessity? This question is linked to the rationale behind establishing independent agencies. Analogously to Ronald H. Coase’s question of why firms exist even though the market is the perfect mechanism for establishing prices and output‐volume (Coase 1937, 388), Majone asks why non‐majoritarian institutions emerge in democratic polities, where all policies are to be decided by elected personnel (Majone 2001a, 59). Ma‐jone makes an argument largely inspired by Douglas North’s transaction‐cost approach. Po‐litical markets are marked by a number of characteristics: the inherent missing information concerning the situation and others’ preferences; high costs in preparing, negotiating, and implementing political contracts; and, finally and most severe, the inability to make credible commitments due to temporal inconsistency and changing coalitions (Gilardi 2002, 875‐876; Majone 2001a, 61; North 1990, 355‐362). Non‐majoritarian institutions can be installed in order to solve these problems. Either they are designed to lower the transactions‐costs by facilitating and professionalizing the collection of information and delivery of expertise or they can solve the cooperation dilemma by enabling credible commitments. As these institu‐tions by definition are not elected they do not follow temporal inconsistencies or election turnouts and thus “restrict the self‐interest of the ruler” (Majone 2001a, 61). Majone calls these rationales “Two Logics of Delegation” (Majone 2001b).
One of the most important lessons learnt from the BSE crisis was that mingling scientific and policy responsibilities might risk the independence and trustworthiness of the underly‐ing expertise. In the British case, the Ministry of Agriculture, Fisheries and Food (MAFF) was
Figure 1: Regulatory Functions within the Policy Cycle
Problem Definition
Agenda‐Setting
Policy Formulation
Policy Implementation
PolicyEvaluation
Fact‐finding Fact‐Finding Rule‐Making Monitoring Enforcement
Risk Management
Risk Assessment
Risk Management
RiskManagement
Risk Management
9
responsible for both consumer protection and promotion of agricultural and industrial inter‐ests. As mentioned in the introduction, MAFF tried as long as possible to de‐emphasize the risks associated with BSE in order to protect the British beef industry. Thus industrial inter‐ests were deemed to be more important than risks to public health. In a more general per‐spective, interests in a functioning market and well‐organized specific producer interests might rally more political support than diffuse and uncertain risks to consumers, thus en‐dangering regulatory capture (van Zwanenberg/Millstone 2003, 72‐78). Bringing Majone’s logics of delegation and the specific case of food safety regulation together, there are many good reasons to think of European food safety policy as reinforcing credible commitment and consumer trust (Borrás et al. 2007, 589; Hellebø 2004b, 125). Within the policy process, science holds a privileged position. Due to its impartial, intersubjective and objective charac‐ter, science is more comprehensively involved in the policy process than other forms of so‐cietal knowledge. The various crises of food safety have also been crises of expertise as sci‐ence was in danger of losing its character as it was intruded by political and industrial inter‐ests.
At the same time, the ever‐growing incorporation of science into the policy‐making and regulatory process puts its legitimacy under pressure. In this context, Peter Weingart coined the notions “scientification of politics” and “politicisation of science” (Weingart 1999, 160). While the first is crucial for the legitimacy and accountability of non‐elected experts involved in the policy process, the second threatens to undermine the credibility of scientific exper‐tise.6 Thus, the separation of risk assessment conducted by independent experts and risk management conducted by elected and accountable officials, who have to take the expertise into account but are not bound to the scientific opinions, is believed to solve these inherent problems.7
Turning to EFSA, one reason for the separation of risk assessment from risk management was already mentioned above. Facing a very sensitive and carefully balanced distribution of competences within the European Union, neither the member states nor the European insti‐tutions themselves were interested in a full‐blown regulatory agency with competencies comparable to US agencies such as the FDA. There are good reasons why the European agency never been planned to be a superior European scientific institution. Firstly, this would be contradictory to the very logic of science that does not accept any mechanisms to establish superiority. In the logic of science, scientific disputes may not be resolved by politi‐cal hierarchy. Secondly, member states were not interested in establishing “an oracle of Del‐
6 Weingart argues that the “scientification of politics” would necessarily lead to the “politicization of science”
as any measure backed by science could only be opposed by “better” counter‐science. Thus, scientific di‐vergence will be stressed and maybe even artificially constructed in order to legitimize opposing political measures. But this endangers the foundations of science, where truth and objectivity are central categories. This is where the separation of risk management from risk assessment is believed to relieve science from political pressure.
7 The idea of the separation was first introduced in the report “Risk Assessment in the Federal Government: Managing the Process” (1983) by the US National Research Council. The report questioned the pure objec‐tivity of science and assumed science being equally influenced by values. Nevertheless the report advised not to separate assessment and management as both are too closely linked (cf. Millstone 2007, 495). Con‐sequently, in the US an integrated approach prevails while the European Union, possibly also in regard of the fragile balance of political competences, prefers a clear‐cut separation. This regularly leads especially in the definition and the use of the precautionary principle to transatlantic tensions and trade disputes.
10
phi spelling out the ‘truth’ in all scientific matters, but rather they wanted to preserve an opportunity for the incorporation of national perspectives on risk assessment” (Alemanno 2009, 2). Consequently, EFSA is neither equipped with regulatory powers nor are its opinions legally binding for the Commission. The opinions issued rank equal with national opinions. It is the founding regulation that introduces mechanisms to cope with diverging scientific opin‐ions. On the one hand, this might appear to be half a step towards a coherent risk assess‐ment (Alemanno 2008, 16‐17). On the other hand, this reflects conflicts between the Com‐mission and the member states and follows the logic of inherent uncertainty in scientific research. From a normative perspective, however, this is the consequent separation of risk assessment and risk management.
3. Explaining the institutional design of national agencies: path dependency vs. Europeanization
According to a European Parliament report issued in 2000 on European food safety policy,
“(s)ince the turning point [of the BSE crisis], there has been a great deal of change underway. Not a single Member State reported that its food safety policy‐making system is experiencing a period of organisational and/or institutional stability. In several cases major institutional changes have been made … In other countries institutional changes have been less conspicuous, but in most Member States there have been some structural and/or procedural changes.” (European Parliament/STOA 2000, 68).
While on the one hand the institutional responses in the member states differ, on the other hand we do see a number of similarities. Agencification, for example, has been a prominent response both at the EU level and in some, but not all, of the member states. This raises the question: Where do we see differences and commonalities, and how do we ac‐count for them? In what follows, we introduce two theoretical explanations: (a) the path‐dependency hypothesis assumes that different institutional designs of food safety regimes prevail in the member states; (b) the second purports that commonalities can be attributed to effects of Europeanization, implying at least a soft trend towards convergence amongst national regimes.8
The diversity of national regulations – but also the commonalities – becomes even more obvious when all EU member state are included, hence avoiding a selection bias. Most stud‐ies on food safety regimes investigate either case studies or they compare a few cases (small‐n studies).9 Even though the “small n problem” is not specific to studies in this sector, it is still a challenge and establishes limits for the generalization of findings. Our sample en‐compasses all EU member states as well as the EU level itself, which, however, implies that we cannot address the complexity of each case.
Path dependency – history matters in organizational design In general, public administrations – just as all institutions – are very “sticky” (Pierson).
They are deeply embedded in national legacies and in bureaucratic cultures which are hard
8 However, in this paper we cannot test these hypotheses in full, and definitely not with regard to all Member
States. 9 Cf., for instance, Surdej (2006); Vos/Wendler (2006); Janning (2008); Elvbakken et al. (2008); Hellebø
(2004a).
11
to change. Agencies have become an increasingly important element of public administra‐tions. Originally stemming from a US‐American tradition of regulatory policy‐making, in the 1960s and 1970s the American model was widely adopted and imitated by European regula‐tors, especially in new fields of social regulation (cf. Majone 1996, 52f.). This, in itself, al‐ready created quite a shift in regulatory policy‐making in Europe. Yet, the design of agencies under conditions of parliamentary systems differs from the US experience. Yesilkagit and Christensen (2010) argue that in parliamentary systems cultural factors are even more im‐portant than political ones in the design of agencies.
In addition, this shift towards specialized agencies has become a prominent approach at the EU level as well, where we can observe a trend towards what has been called “agencifi‐cation” (see Christensen/Lægreid 2005). The term implies a new relationship between politi‐cians and public administration moving away from a clear top‐down command and control pattern towards an outsourcing of regulatory policies to specialized agencies endowed with a certain degree of autonomy and relying on technocratic rule‐making and implementation (cf. Fleischer 2007, 213). Until today, a total of 29 agencies has been set up at supranational level, most of them in the old first pillar (23 agencies) – EFSA being one of the most recent ones10. Despite a number of differences between these agencies regarding their mandate, they share an important feature: “These bodies are not (yet) fully fledged regulatory agen‐cies. They neither make nor implement regulatory policies. For the time being, their func‐tions are essentially limited to the collection, processing and dissemination of policy‐relevant data and information” (Majone 1996, 274). By and large, Majone’s assessment, even if 15 years old, still holds true today. While the 2002 Communication on the framework for Euro‐pean Regulatory Agencies (European Commission 2002) contains some innovative proposals, there is no change regarding the key issue of delegating substantive regulatory powers to agencies: “agencies may not be empowered to make rules, i.e., quasi‐legislative measures of general applicability” (Majone/Surdej 2006, 9). Nevertheless, newer agencies – the so‐called “third generation of agencies” – possess a number of innovative features and, in the long run, may be more likely to develop into fully‐fledged specialized regulatory agencies.
Historical institutionalism provides a powerful explanation for the institutional design of regulatory food safety regimes, for agency change and inertia. The key argument of (ra‐tional) historical institutionalism is that the initial design of institutions is based on actors’ rational preference of maximizing their interests and that these initial decisions have tempo‐ral causal consequences which are difficult to revert. However, these preferences may be based on short‐term considerations whereas, over time, lags between decisions and long‐term consequences occur. Due to lock‐in effects over time and to attempts to immunize new institutions from potential changes by opponents (i.e. changes in government parties), insti‐tutions are not under full supervision from governments but enjoy some autonomy which they may use to develop a life of their own (cf.Pierson 1996, 132‐136). In the evolution of institutions, path dependency becomes a common feature.
In general, institutions are very resistant to change: “Institutions are seen as excessively static and likely to remain on the same path unless some effort is made to divert them.” (Olsen 2009, 3) Simultaneously, organisational learning and adaptation to changes in the
10 For a comprehensive list of all EU agencies cf. Wonka/Rittberger (2009, 15).
12
environment are key survival strategies for institutions. According to historical institutional‐ism, critical junctures are key events that allow institutions to leave the previous path and to adapt to new functional demands. Regarding the policy sector of food safety, the BSE crisis was a seismic event that – followed by a number of several other food scandals – can be conceptualized as a “critical juncture”. It challenged existing organisational routines and shed light on performance failure to protect consumers from food‐related diseases, thus leading to comprehensive institutional change – on the national as well as the supranational level. Before the BSE crisis hit the EU, the EU approach to food safety was marked by incon‐sistencies, many random elements and a lack of structural connectedness in the sense that different and non‐co‐ordinated actors and institutions were responsible for different aspects of food safety. The BSE crisis also changed these institutional barriers (Ugland/Veggeland 2006, 620).
The way the critical juncture of the BSE crisis was politically framed and the design of the new EU food safety regimes simultaneously shows some signs of institutional trajectories. While originally EFSA was supposed to be modelled after the US‐American Food and Drug Administration (FDA), after a long and controversial discussion the mandate was finally lim‐ited to risk assessment and communication, while the regulatory activity itself – risk man‐agement – is left to the EU policy‐makers (cf. in detail section 2). Thus, even though amazing innovations have been introduced, there a clear limits to the revolution: EFSA is an inde‐pendent agency but with a tightly restricted mandate devoted to scientific risk assessment only while real regulatory power was not delegated to EFSA; it clearly remains with the pol‐icy‐makers. In the end, it was rather the EU medicinal agency EMEA that served as the model, and not the powerful US‐American FDA.
The dynamic development at the EU level is reflected in the developments in the member states. Again, Europeanization effects become domestically framed “when Europe hits home” (Börzel/Risse 2000a); this way European and domestic effects intermingle. Given very different national legacies and bureaucratic cultures, we can expect that this is also the case regarding food safety regimes. Yet, we can identify a number of commonalities amongst the national systems, which allows the development of a first classification based on the institu‐tional design of regimes. In order to develop network structures that are most essential in EU regulatory politics, there is a need to create a sense of identification and commonly ac‐cepted ways of doing things. This is particularly the case if institutional change is induced by a performance crisis as was the case in the food safety sector. Olsen (2009, 24) argues that
“feedback from the environment is in particular important when large‐scale failures and per‐formance crises generate demands for more coordination. Then institutional developments are more likely to be influenced by the interaction, collisions, conflicts, meta‐rules, and power struggles between several institutional spheres, adapting to each other”.
Transnational regulatory networks provide for such environmental feedback. In risk regula‐tion this becomes a crucial issue, since “(r)isk assessment has traditionally been an area of policy in which national sentiments are strong” (Ugland/Veggeland 2006, 618). Given that regulatory agencies are highly technical organizations, professional reputation and “mutual recognition” as competent authorities is an important part of organizational norms guiding organizational adaptation. Majone argues that
13
“a regulatory agency which sees itself as part of a transnational network of institutions pursuing similar objectives and facing analogous problems […] is more motivated to resist political pres‐sures. This is because the regulators have an incentive to maintain their reputation in the eyes of fellow regulators in other countries; a politically motivated decision would compromise their credibility and render co‐operation more difficult to achieve in the future.” (Majone 1996, 273)
In sum, we assume that there are rational incentives to incorporate rules that allow agencies to engage in transnational networks and to cooperate with partner organizations at the EU level as well as in other EU member states. This takes us to the issue of Europeanization of national food safety agencies.
Europeanization of national food safety systems Europeanization research is today the major growth area in EU studies. In his overview ar‐
ticle, Bulmer (2007) identifies a total of five variants of what Europeanization means. The core usage is one that is also identified by Radaelli and Exadaktylos as the common ground despite a number of different definitions and conceptualizations: A point “most definitions have in common is the identification of Europeanization with the domestic impact of Euro‐pean integration” (Radaelli/Exadaktylos 2010); it refers to the “central penetration of na‐tional systems of governance” (Bulmer 2007, 47; emphasis in original)(Bulmer 2007, 47; em‐phasis in original).11 A key assumption is that the fact of being a member of the European Union (or the wish to become a member in the case of candidate countries) involves funda‐mental domestic changes affecting policy, politics and polity. Hitherto, policy studies are clearly overrepresented in Europeanization research whereas “Europeanization of the polity has received much less attention” (Bursens 2007, 115; cf. also Lenschow 2005);. Yet there is agreement that Europeanization does not leave national polities unchanged. Europeaniza‐tion may have domestic effects on political and economic institutions, intergovernmental relations, judicial structures, public administration, state traditions and collective identities (seeBörzel/Risse 2000a, 60). While earlier studies in Europeanization treated integration as an independent variable, more recent studies identify explanatory variables at the domestic level (cf. Bulmer 2007, 49f.). Knill et al. (2009) rightfully argue that the major focus in Euro‐peanization studies has so far been on the following two aspects: “(1) the degree of EU‐induced changes in national arrangements and (2) the extent to which these changes lead to the convergence of national structures, institutions or policies”.
When considering mechanisms of Europeanization, there are two broad notions prevail‐ing in the literature: “Europe as pressure” and “Europe as usage”. In general, the pressure perspective is very strong in studies on regulatory policies. Yet, since we are not interested in the policy but in the polity dimension of food safety regulation, we refer to the second notion of “Europe as usage”. It is particularly fruitful to combine this broad Europeanization approach with the concept of policy transfer in areas beyond “positive” and “negative” inte‐gration, including the transfer of institutions and administrative arrangements. This allows studying convergence beyond the usual convergence mechanisms such as harmonization, mutual recognition, regulatory competition or functional necessities (cf. Fischer 2007, 351). While the General Food Law outlines some basic features of the new food safety regime and requires national competent authorities to cooperate with EFSA, it does not entail a tem‐
11 For some recent overviews of Europeanization studies cf. Axt et al.(2007), Bulmer (2007) and Radael‐
li/Exadktylos (2010).
14
plate for a specific model of food safety authorities at the national level. Not even the agency model is an EU requirement. Thus, we are dealing with a field of soft governance where transfer can only be based on alternative mechanisms such as learning.
Much of the literature refers to Europeanization as a top‐down process from the EU to the national level. However, there are also broader conceptualizations which are more use‐ful for our study and which can be linked to the concepts of transfer and diffusion. Europe‐anization can be conceptualized not only as a vertical process, but may also include a hori‐zontal dimension. Bomberg and Peterson, for example, “suggest that mutual influence, transfer, benchmarking and lesson‐drawing do not necessarily presuppose that the EU is there all the time to provide Europeanizing templates. A country can also imitate or import or learn from another bilaterally, although the EU often provides the architectures and the procedures for horizontal transfer” (Radaelli/Exadaktylos 2010, 193). In sum, Europeaniza‐tion in this concept includes also the diffusion of policies across member states. Europeani‐zation then becomes a policy transfer concept.
Another broad, yet helpful conceptualization comes from Ladrech (1994), who refers to an “EU as usage” concept. He argues that “the presence of a clearly defined European policy is neither a sufficient nor a necessary condition for Europeanization to take place” (Radaelli/Exadaktylos 2010, 193). In fact, “domestic constellations of actors can still be influ‐enced by the ideas, discussions, suggested policy templates, etc. discussed in the EU arena. What is necessary for Europeanization to occur […] is the presence at the EU level of a forum of discussion, an arena for negotiation, or a political architecture for interaction and dis‐course” (ibid. 194). In addition to the policy transfer model proposed by Bomberg and Peter‐son, Ladrech introduces a requirement, i.e. that the EU is somehow a facilitator of Europe‐anization.
While such broad concepts have been employed in studies on the effects of the Open Method of Coordination, our argument is that they can also be very fruitful for analyzing regulatory networks because of the very nature of these multi‐level networks. We argue that, in our case, the evolution of transnational food safety networks provides for such an EU‐induced arena, a forum where EFSA works as a facilitator for policy transfer in the new political architecture. A high density of communication, such as in the EU, is most likely a favourable condition for transfer (cf. Fischer 2007, 352f.). This network of food safety au‐thorities provides opportunities for vertical as well as horizontal effects of Europeanization. The core of this argument has also been advanced by Eberlein and Grande (2005). They ar‐gued that transnational regulatory networks have become important forums for European coordination and for an informal, soft harmonization of member states’ regulatory activities. Rules for best practice, procedural rules and the like are examples of what these networks achieve. Given that these regulatory networks build on technical expertise, professional in‐formation/ knowledge is a key resource for soft governance instead of hierarchical modes of “command and control” governing. The power of European agencies is often the power of an information broker. Trust and the willingness to cooperate are preconditions for success‐ful information brokerage. The willingness to cooperate is, according to Eberlein and Grande (2005), especially high if the network participants become conversant with the same profes‐sional standards and share a regulatory philosophy. These arguments are supported by Ma‐jone (cf. section 2). From a Europeanization perspective, especially if a broader concept is
15
applied, such mechanisms of adjustment, learning and transfer are clear Europeanization effects on domestic regulatory actors. This does not imply that we expect full convergence and only one model to prevail, yet in important dimensions there has to be a substantial set of commonalities or at least a trend towards enforcing commonalities. In addition, Fischer (2007, 351) argues that if governments are under high pressure to introduce reforms, it of‐ten becomes easier and is more efficient to adopt policies that have been successfully im‐plemented in other countries.
An epistemological problem of this broad Europeanization concept is the identification of causal relations. There is, as Radaelli/Exadaktylos (2010, 197) put it, “the risk of assuming that if some domestic changes look similar to ‘what Brussels wants’, this must be an instance of Europeanization”. Yet it may be a mixture of globalization, Europeanization and domestic forces that trigger change at the domestic level. The effect of these concurrent social forces and their differential impact on the national level are so far “neglected faces of Europeaniza‐tion“ (Knill et al. 2009). Process‐tracing becomes a crucial research strategy which adheres to the temporal dimension in order to distinguish effects of different social forces.12
Given that most studies on food safety regulation only analyze regimes from the old EU‐15 and exclude Central and Eastern European countries,13 we furthermore need to take ac‐count of the issue of enlargement and Europeanization. The literature suggests that we can distinguish between “Europeanization of member states” vs. “Europeanization of accession countries”. Eight post‐communist Central and Eastern European countries joined the EU in 2004 respectively 2007. Eastern enlargement was very demanding for the would‐be member states and required dramatic changes given EU conditionality; it required a process of in‐tended Europeanization from the candidate countries. Regarding the field of food safety, in many Central and Eastern European accession countries there were no food safety regimes in place, in others the systems in place had to be transformed. In Poland, for example, a functionally diversified system was in place, yet multiple actors brought along the problem of coordination, and the transformation to a market‐based economy required the rethinking of traditional regulatory approaches (Majone/Surdej 2006, 27); the Hungarian regime also has been subject to profound institutional transformation (cf. in detail section 4). Pre‐accession Twinning Projects – which can serve as means for Europeanization by learning, socialization and institutional isomorphism – have been in place in the area of food safety. In her comparative assessment of Europeanization in Central and Eastern Europe, Héritier (Héritier 2005, 203) concludes: “Europeanization East and Europeanization West differ con‐siderably; however, they reveal commonalities as well.“ In both cases, the output is a change at the national level.
To sum up: The path dependency argument emphasizes the conservative influence of ex‐isting domestic institutions that make change less likely unless critical junctures trigger change. Yet, even if change occurs, the old path may still impact the degree and the specific kind of adaption to the new situation. The Europeanization concept – at least the strand employed in this paper – emphasizes the crucial impact of vertical and horizontal coopera‐tion not only on policies, but also on state structures. Learning, policy transfer, change of
12 Obviously, we cannot do this in this paper, yet we want to give some hints for further studies. 13 An exception is the study by Vos and Wendler (2006) which includes Hungary; for Poland see Surdej (no
date) as well as Majone and Surdej (2006).
16
norms etc. is particularly strong in case of cooperation in transnational regulatory networks – an effect that is inherent to the nature of these regulatory networks and the regulators that interact in them. Therefore, there are strong incentives towards convergence and iso‐morphism in regulatory networks – even if differences remain and even if it seems likely that there is not one single model to develop.
4. Systems of food safety regulation in the EU member states and at the supranational level
In the empirical part of our paper, a crucial reference is the European regulatory regime. On the supranational level, a system was established which fully separates risk assessment from risk management (bi‐institutional or separate model). The following section provides a short overview of EFSA’s design. We then turn to the national level and introduce a typology of regulatory regimes within the EU and dwell on some exemplary cases.
EFSA’s Design: Key principles The European Food Safety Agency (EFSA), based in Parma/Italy, was founded in 2002 in
response to a number of food scandals affecting the EU. The key legal document, “Regula‐tion (EC) No 178/2002 of the European Parliament and of the Council of 28 January 2002 laying down the general principles and requirements of food law, establishing the European Food Safety Authority and laying down procedures in matters of food safety” (Regulation (EC) 178/2002 2002), gives a detailed outline of EFSA’s structure and its tasks. The EFSA de‐sign clearly reflects the theoretical and normative aspects mentioned so far and constitutes an important background for characterizing national agencies and bodies.
(1) EFSA lacks regulatory competencies. EFSA’s task is exclusively focused on risk assess‐ments; regulatory competencies have not been assigned. The whole complex of rule‐making is reserved for the Commission and the Regulatory Committees in the comitology process, to a lesser extent also to the Council and Parliament. Monitoring and enforcement tasks lie with the Food and Veterinary Office (FVO) on the European level and with national bodies in the member states. FVO here holds a coordinating and supervising function. The design is driven, firstly, by the motivation not to establish a scientifically supreme body on the Euro‐pean level and, secondly, by the need to take the separation of risk assessment from risk management seriously, which features prominently in Article 3 of the founding Regulation (EC) 178/2002.
(2) The administrative independence of EFSA is one of the main founding principles in order to avoid any bias by industrial or national interests. EFSA’s main steering body is the Management Board.14 It establishes the internal rules of procedure and the budget code, elects the Executive Director, appoints the members of the scientific committees, steers the work of the agency by deciding on the annual work program and the multi annual strategic
14 Differing from other agencies in the European Union, the principle of one seat per Member State is here not
valid. Instead, the Management Board consists of 15 members, of which is one the representative of the Commission. The candidates are selected solely on basis of their experience and professional capacity (Regulation (EC) 178/2002 2002, Art. 25 (1)). Furthermore, four of the members to be appointed must have a background in consumer or industrial organizations. This system minimizes the potential for political bar‐gaining among the Member States (Vos 2003, 125).
17
plan, and influences the operational work by deciding on the budget and staff of the agency (Art. 25‐28; Art. 43). The Advisory Forum fosters the coordination and information exchange between the European and national level (Alemanno 2008, 8). The members are supposed to be based in “competent bodies in the Member States which undertake tasks similar to those of the Authority” (Regulation (EC) 178/2002 2002, Art. 27 (1))). For the networking function of EFSA, the Advisory Forum takes a key role.
(3) Regarding the scientific independence, EFSA does not have its own scientific staff. All scientists directly employed by the EFSA are situated in the secretariat and are only sup‐posed to assist the scientific committees. By public advertisement on the internet, experts can apply for positions in one of the scientific committees. This has a double function of con‐stantly exchanging the scientific experts and appeasing the member states by clarifying that there is no genuine “European” expertise based in an independent agency. Rather, the ex‐perts are regarded as national ones.
(4) Concerning transparency and participation of the public, within its founding regula‐tion EFSA was given strict rules concerning publication of documents (Art. 38). The very comprehensive archive of documents, requests and conducted risk assessments on the web page of EFSA demonstrates the authority to take these rules seriously.15 For organized inter‐ests, the Stakeholder Consultative Platform offers access. This body is composed of various European associations and is regularly re‐elected. For non‐organized interests, there is the chance of being brought forward in the public consultations which are launched on impor‐tant issues. These calls for opinions are regularly used by individuals but also by the member states and other institutions. Hardly any other food safety agency publishes as much of its work and communicates to the public more transparently than EFSA.
Food safety regulation in member states Highly complex food safety regimes are today in place in all 27 EU member states (for de‐
tails cf. annex 1 and 2).16 Despite variety in certain details, there are key similarities that al‐low the development of a classification system. Different classification schemes are dis‐cussed in the literature. For example, a 2000 report to the European Parliament identifies a classical, a modern and an emerging model (European Parliament/STOA 2000, 58‐64). This classification is based on conceptualizations of the relationship between science, politics and the public rooted in current trends in the sociology of science. According to the classical model, risk regulation is solely based on “sound science” concerns and risk communication is a one‐way and top‐down communication strategy. Risk regulation (policy‐making) is based on scientific expertise. The modern model is still strongly based on the ideal of “sound sci‐ence”, yet acknowledges that risk management needs to consider a wider array of issues such as social, economic and cultural concerns. The model still adheres to a “clear division of
15 In addition, all sessions of the Management Board are accessible as live stream on the web. In general the
sessions are also open to the public. 16 Our overview of the national agencies is by and large based on a study conducted for the German Federal
Institute for Risk Assessment (BfR) in 2008/09, which included a total of 23 countries. Additional informa‐tion is considered as well to include all 27 EU Member States. The BfR study is based on the evaluation of the online presentations of national agencies supplemented by a BfR survey of all national agencies to veri‐fy online information. For a more detailed description of the national systems cf. the annexes to this paper; cf. also (Federal Institute for Risk Assessment 2009).
18
labour between the scientific community, which is believed to assess risks in a socially and ethically neutral way, and policy makers who subsequently take legitimate account of the extent and distribution of the social and/or commercial benefits to be offset against the pos‐sible risks” (European Parliament/STOA 2000, 60). Regarding risk communication, it is less “scientistic” and top‐down. Finally, the emerging model argues in favour of a separation of risk assessment from risk management. It is assumed
“that science always operates within a specific social, political, cultural and economic context, and that risk assessments always involve the analysis of selected scientific information within a prior set of non‐scientific considerations … (O)nce expert scientific risk assessors have reached conclusions about the existence of a risk, its probability and severity, and acknowledged the sci‐entific uncertainties with which they have had to grapple, it assumes policy‐makers need to make a further set of specific down‐stream evaluative judgments to decide how the conclusions of the risk assessment will influence policy decisions.” (European Parliament/STOA 2000, 64)
Until the BSE crisis, most national systems were working on the basis of the classical model. In response to the BSE crisis, one can observe a shift to the modern model (ibid., 68f.): “One key feature of the post‐turning point changes to the national system has been the widespread recognition of the necessity and perhaps the inevitability of institutionally separating risk assessment from risk management.” (ibid., 69). So far only few countries show indications of transforming to the emerging model.
Given that this classification implies – based on general concerns for risk governance – different concepts of how risk assessment and risk management should be organized and related to policy‐making, it is compatible with the distinction of models developed in the comparative study of five EU member states and the EU level by Vos and Wendler (2006). This study identifies three different models based on risk governance in the food sector. A crucial criterion employed, which is also a key aspect of our study, is the relation between risk assessment and risk management (cf. in detail Dreyer et al. 2006)(cf. in detail Dreyer et al. 2006): (1) a bi‐institutional or separate model in which responsibilities are separated be‐tween institutions; (2) a structurally more integrated model with nevertheless usually func‐tionally‐divided responsibilities; and (3) a “chaotic” model with fragmented, partly inte‐grated and partly overlapping responsibilities. The separate model resembles the emerging model, whereas the more integrated model resembles the modern model. The fragmented model still falls into the category of either the modern or still the classical model. The ques‐tion is how the food safety regimes in the 27 member states fit into this classification.
Table 1: Classification of risk governance regimes
Bi‐institutional, separate model
Structurally more integrated model
Fragmented (chaotic) model
EU Austria Denmark Finland France Germany Hungary (today) Netherlands
Belgium Greece Ireland Italy Latvia Portugal Spain Sweden
Cyprus** Czech Republic** Estonia* Lithuania* Slovenia** Romania
Bulgaria* Hungary (until 2007) Malta Luxembourg
19
Poland Slovakia
UK
* first signs of separation; ** = countries with no agency
Table 1 illustrates that most national regulatory regimes can be grouped into the model of non‐separated, more or less integrated systems (the modern model). This implies that the same authority – usually, but not always the competent food safety agency – is involved in risk assessment as well as in risk management. However, within this group there is some diversity regarding the exact role of the agency in risk management. Moreover, in some countries such as Cyprus, the Czech Republic and Slovenia, no independent agency has been set up until today at all; in these cases, all competencies are integrated in a ministerial de‐partment. In Malta, a system of institutional separation of risk assessment from risk man‐agement is not in place; there are, in fact, two agencies involved with overlapping responsi‐bilities. Also in Luxembourg there are – to a lesser extent – institutional and functional over‐laps, and there are overlaps and only first signs of separation in Bulgaria. Hungary fit into this category until the reform in 2007 and is today a clear case of a separated system (cf. below). Finally, the Scandinavian countries are, according the European Parliament report, more or less an example of the emerging model (see European Parliament/STOA 2000, 69). Yet, ac‐cording to our classification, they are to be found in the separated (Denmark, Finland) as well as in the institutional model (Sweden).17 In what follows, we take an exemplary case from each classification and discuss this in more detail with regards to our hypothesis of domestic vs. Europeanization factors. The Hungarian is specific, since it is a transitional one where change has occurred from a rather chaotic and fragmented third type to a clear case of a bi‐institutional model.
Austria – an EFSA‐inspired design with a domestic flavour Risk assessment is separated from risk management in Austria. The key legal document is
the Austrian Law on Food Safety and Consumer Protection (Lebensmittelsicherheits‐ und Verbraucherschutzgesetz, LMSVG) that came into force in 2006 and the 2009 amendment introducing a Health and Food Safety Law (Gesundheits‐ und Ernährungssicherheitsgesetzes, GESG). These legal documents transpose EU laws into Austrian law and set up the national agency. The competent authority is the Austrian Agency for Health and Food Safety (Agentur für Gesundheit und Ernährungssicherheit; AGES). AGES was established in 2002 – shortly after EFSA was set up at EU level – as a private limited company (GmbH) owned by the Re‐public of Austria. AGES is a merger of 18 federal agencies and federal offices in the fields of food research, bacteriology and serology, veterinary medicine and agriculture. Due to the merger, AGES today has more than 1,400 employees and is one of the largest food agencies in the EU. It works under the supervision of the Ministry of Health (BMG) as well as the Fed‐eral Ministry of Agriculture, Forestry, Environment and Water Management (BMLFUW), which act as risk managers.
Given the broad AGES responsibilities, Austria is the only member state in the EU where federal competencies affecting the food supply chain “from farm to fork” are concentrated in one public body. This perspective on the food chain in its entirety is an important innova‐
17 These differences in classification may be due to developments over time or different indicators.
20
tion. Until the restructuring that led to the founding of AGES, the different public bodies in‐volved had responsibilities only in their specific sectors. AGES is supposed to have an inte‐grating and coordinating effect that should allow for better protection of consumers and the environment. Yet, the AGES mandate is restricted: it is only involved in risk assessment and in risk communication, whereas enforcement competencies (food inspection) are entrusted to the Federal Office for Food Safety (BAES), which is integrated into the AGES structure. AGES conducts risk research and also identifies new health risks and issues recommenda‐tions for risk reduction. AGES evaluations are based on the standards laid out in the Austrian food law; it also addresses questions of control and prevention of infectious diseases in hu‐mans. As a result, the AGES mandate goes beyond the field of food safety. The agency serves as the EFSA focal point and is the RASFF contact point.
AGES was set up simultaneously to EFSA, and the institutional design was clearly inspired by EFSA. The division of risk regulation tasks is very similar to the division on the suprana‐tional level. The timing of reforms on both levels is a first indicator of a possible European impact. To interpret the Austrian case, it is helpful to look at Germany, since the Austrian case resembles in several aspects the German case of the establishment of the Federal Agency for Risk Assessement (Bundesinstitut für Risikobewertung, BfR).18 Fischer (2007, 361ff.) persuasively illustrates that the German BfR is a clear case of policy transfer from the European to the national level. Indicators are not only the timing (the BfR was established in 2002) and the mandate; process‐tracing shows that even several actors were simultaneously involved at both levels. Furthermore, the final design of the BfR following the EFSA model was legitimized, firstly, with efficiency arguments due to compatibility with EU structures and, secondly, with concerns for greater transparency and independence of agencies in the separate model. As a result, the BfR – an independent assessment body – represents an in‐novation of German public administration traditions. However, more far‐reaching innova‐
18 Given the similarities between the Austrian and the German case it would be worth investigating the ques‐
tion not only of vertical, but also of horizontal transfer between the two countries.
Figure 2: The risk regulation regime in Austria
AGES Focal Point RASFF
(Agency) BMLFUW BMG (Ministries)
BAES
Fact‐finding Fact‐Finding Rule‐Making Monitoring Enforcement
Risk Assessment
Risk Management
RiskManagement
Risk Management
Risk Assessment
Risk Management
RiskManagement
Risk Management
Risk Assessment
Risk Management
RiskManagement
Risk Management
21
tions, including a re‐structuring of enforcement mechanisms, were inhibited. The German Federal Agency for Consumer Protection and Food Safety (Bundesamt für Verbraucherschutz und Lebensmittelsicherheit, BVL) was simultaneously reformed to match the European level Food and Veterinary Office (FVO). However, the German Länder blocked a further extension of BVL’s enforcement competencies, necessitating a transfer of competencies from the Länder to the federal level. Fischer (2007, 367) claims a convergence between the German and the EU agency with regards to the agencies’ objectives, their organizational principles and the result of the reform process. The specific organization of AGES including so many former agencies leads at the same time to a very high representation of AGES employees in EFSA’s scientific panels. At the same time AGES is the only Austrian institution (aprt from universities) that is represented in EFSA’s panels. In the German case the inclusion is much smaller. Apart from BfR 12 other institutions are represented in the panels, which are mostly public research institutes (cf. Annex 3).
Both countries, Austria and Germany, are federal systems with a clear unitary trend or structure of joint‐decision making. Europeanization studies have illustrated that federalism has a deep impact on the extent and the specific ways in which European impulses affect the domestic level (cf., for example, Börzel/Risse 2000b; Bursens 2007; Große Hüttmann/Knodt 2003; Schmidt 1999). This is most obvious in the German case regarding the limitation of changes to the enforcement system, but also in the Austrian case. In Austria the competen‐cies of the Länder are weaker than in Germany and there is a strong unitary trend fostered by principles of consensus democracy. While the Austrian Länder are involved in monitoring and enforcement of food legislation “on the ground”, there is nevertheless a strong federal risk manager, the Federal Office for Food Safety (BAES) which is actually even organization‐ally integrated into the AGES structure. Such integration – involving a transfer of compe‐tences to the national level – was not possible in the German case, but it was in Austria.
In Austria, considerations to integrate the whole food chain into a single public admini‐stration were also based on efficiency and transparency concerns. This integrative concept of food safety as involving everything “from farm to fork” or “from stable to table” takes this European guiding principle seriously at the national level. Yet in other countries such as Germany, there were limitations blocking more radical reforms. In Austria the public pres‐sure for reform opened a window of opportunity to introduce a comprehensive new concept and a new, encompassing public administration in the food safety and consumer protection sector. Thus, the reform process in Austria is not a coincidental, but a deliberate isomor‐phism.19 While the extent and path of reform is, again, strongly shaped by domestic factors, the Austrian case is a strong example of the impact of Europeanization.
Sweden – a separated system with a high degree of stability An independent food safety agency was set up in Sweden as early as 1972 (cf. in detail
Ferreira et al. 2006). Prior to its establishment, there were tensions between public admini‐strations in the field of health and agriculture, yet with the founding of the Swedish National Food Administration (Svenska Livsmedelverkets, SLV) they were resolved. The SLV is respon‐sible for risk assessment, risk communication and risk management. Thus, we have no insti‐
19 Also other explanations for transfer of the EFSA model such as regulatory competition or economic pres‐
sure can be excluded (same in the German case; seeFischer 2007, 369).
22
tutional separation of risk regulation tasks, but some minor reforms in 2000 resulted in a still integrated system yet with a stricter “intra‐organisational division of labour” (Dreyer et al. 2006, 22). The SLV is clearly the key agency; it serves as the EFSA Focal Point and the RASFF Contact Point. The task of the SLV is food monitoring, rule‐making (decrees), and independ‐ent risk assessment as well as risk communication. The SLV also conducts in‐house food re‐search. SLV activities are open to public scrutiny; educational and informational activities for the general public are an integral part of the SLV mandate. It is active not only at the na‐tional level, but also participates in policy implementation at the regional level. Furthermore, the Swedish SLV is an “outlier” among the EU member states in so far as the agency is com‐pletely independent and not subordinate to any ministry, but reports directly to the gov‐ernment. Nevertheless, the Ministry of Agriculture (RK) is also involved in food safety regula‐tion.
Although there have been debates since 1998 about reform in food safety (resulting from 2000 onwards in stricter intra‐organisational division of competencies between departmen‐tal entities), and although a new food law was introduced in 2005 transposing EU law into national Swedish law, the marginal reform “does not mark a turning point in national food safety policy” (Dreyer et al. 2006, 22; cf. also Elvbakken et al. 2008, 137f.). In contrast to most EU member states, the BSE crisis did not really hit Sweden in terms of organization reform. Overall, there is amazing degree of stability in the Swedish system (cf. Ferreira et al. 2006, 333f.).
How can we explain the overall design as well as the stability of the system in Sweden? Institutional as well as cultural factors are crucial, especially the specific tradition of public administration in Sweden. Key principles of Swedish public administration are, firstly, “a dis‐tinction between a dualist system of government that rests on constitutional separation of policy and administration” (Yesilkagit/Christensen 2010, 59); secondly, the public administra‐tion is perceived to be „a professional elite of regulators sharing an ideal of rational planning based on scientific expertise” (Ferreira et al. 2006, 331). Finally, public administration is
Figure 3: The risk regulation regime in Sweden
RK MZ (Ministries)
SLV Focal Point RASSF JV (Agencies)
SVA (Research Institute)
Fact‐finding Fact‐Finding Rule‐Making Monitoring Enforcement
Risk Assessment
Risk Management
RiskManagement
Risk Management
Risk Assessment
Risk Management
RiskManagement
Risk Management
Risk Assessment
Risk Management
RiskManagement
Risk Management
23
guided by the publicity principle according to which “all citizens have the right of access to governmental documents” (ibid.). Central public authorities are a vital component of Swed‐ish government. These authorities are “expert‐based professional organizations exercising their assigned tasks fully independent of political institutions” (ibid., 332). Public authorities not only implement policies, but also enforce regulations and can, in addition, take initiative in their area of responsibility. The Swedish SLV is a typical example of this kind of profes‐sional organization.
While Europeanization researchers often argue that the impact of the EU usually tends to be more disruptive in unitary member states (as opposed to federal systems with a more dispersed power system; cf. Schmidt 1999), this was clearly not the case in Sweden – a plain case of a unitary country with no regionalization trend – and not in the Swedish food safety regime (cf. Bursens 2007, 123). The national Swedish model prevailed. This model clearly deviates from the separated model established with EFSA and taken up in a number of other countries. It also deviates from the specific design of other countries that fit the classifica‐tion of integrated systems. Following the EFSA model was not possible due to deeply‐rooted institutional misfit. The fact that this misfit was not attended to is due to the merely soft Europeanization pressure (no prescribed institutional model) on the one hand, and the strong national tradition (professional and independent public authorities) on the other hand. The need for efficiency and transparency, which in the German and Austrian case led to institutional isomorphism, in the Swedish case was interpreted as calling for a completely independent and solely expert‐based professional agency.
Nevertheless, the Swedish system is also not free from international and particularly European influences. In 1990 there were debates about inducing structural reforms in Swed‐ish public administrations which were a response to the transnational trend towards new public management and flat hierarchies. The upcoming Swedish EU membership (the coun‐try joined in 1995) gave an extra impetus to the debate. In addition, the principles of risk analysis in food safety were influenced by debates in the FAO and WHO, which resulted in some functional separation in the Codex Alimentarius system (cf. Ferreira et al. 2006, 333f.). Following these international trends, which are also reflected at the EU level, not a profound reform but a stronger inter‐departmental functional separation was introduced. And finally, EU food law was transposed into national law and into SLV ordinances. In sum, the Swedish case is an example of strong domestic factors with a limited impact of Europeanization (and globalization).
Hungary: from a fragmented to a bi‐institutional, separated model20 In the comparative study by Vos and Wendler (2006), Hungary is classified as a chaotic
model due to a fragmented structure with partly overlapping activities and with responsibili‐ties which are not clearly separated. However, the Hungarian system has undergone a tre‐mendous change in the last few years. As a result of Europeanization impacts, Hungary is today a clear case of a bi‐institutional, separate system. Europeanization effects can be ac‐counted to accession in general and, therefore, do not come as a surprise. Yet Europeaniza‐tion evolved in different steps and was also influenced by domestic factors.
20 We are most thankful to Jennifer Träsch for the background information on the Hungarian case based on
her interviews.
24
Food safety regulation has a long tradition in Hungary dating back to the Austrian‐Hungarian Monarchy (Ferencz et al. 2006, 383). In communist times, food safety was not a key issue; the most important regulations were the 1976 Food Act and the ministerial decree 12/1977. With democratic transformation in the 1990s and especially with upcoming EU accession, the pressure for change rose (for details cf. Ferencz et al. 2006): “Accession to the EU was the other main reason for changing the regulatory system” (ibid., 389). In addition, democratic transformation brought along a fundamental change of the whole system of public administration, a change oriented towards a system of separation between politics and administration. Legal harmonisation in food safety, lasting from 1993 to 2003, went along with the transposition of numerous EU laws into Hungarian laws. Finally, in 2003 the Hungarian Food Safety Office (Magyar Élelmiszer‐biztonsági Hivatal, Mébih) was founded.
Figure 4: The risk regulation regime in Hungary
FVM EÜM (Ministries)
MÉHIB Focal Point RASFF (Agency)
MGSZHK (Administration; Reports to FVM)
With accession, the acquis communataire on food safety had to be transposed into Hun‐garian law. This is clearly top‐down Europeanization. This European pressure coincided with pressure at the domestic level. The simultaneous debate over establishing EFSA at the su‐pranational level was perceived as a demand to create a domestic partner for a suprana‐tional agency. Hitherto, the Hungarian system was highly fragmented in particular with re‐gards to monitoring and enforcement. At the domestic level, this led to calls for a more co‐ordinated structure. Therefore, the Hungarian Mébih was primarily established for the pur‐pose of domestic and external (EU) coordination. Since its inception, several reforms have been undertaken. The major reforms were crisis‐induced: A key domestic crisis was the so‐called Aflatoxin21 scandal, when aflatoxin‐contaminated ground paprika was detected in Hungarian spice‐producing companies in 2004 (cf. Ferencz et al. 2006, 422). In response to the deficits in crisis management, the supervision of Mébih was completely transferred to the Ministry of Health. Three years later, oversight was again transferred back to the Minis‐try of Agriculture and Rural Development (FVM) where it still rests today.
21 Aflatoxin is a mycotoxine.
Fact‐finding Fact‐Finding Rule‐Making Monitoring Enforcement
Risk Assessment
Risk Management
RiskManagement
Risk Management
Risk Assessment
Risk Management
RiskManagement
Risk Management
Risk Assessment
Risk Management
RiskManagement
Risk Management?
25
Mébih’s key task is the scientific assessment of risks in food and feed as well as risk com‐munication with the public and also with experts. It currently has 18 employees and thus lacks the capacity for in‐house risk research. It refers to risk research conducted in other Hungarian organizations (such as universities) or in other European countries. Its role is then to adapt the risk assessment to the Hungarian situation; it is partly involved in the assess‐ment of risks based on scientific data. Mébih is the EFSA Focal Point as well as the RASFF Contact Point, which is due to its strong coordinating role in the early days of the system.
While the coordinating role was strong in the beginning, it is less important today, be‐cause with the founding of the Central Agricultural Office subordinate to the Ministry for Agriculture a single risk management and enforcement unit was set up. Previously, three organizations were involved in enforcement, supported by several others. Thus, the 2007 reforms have resulted in a major restructuring of the system. While Mébih lost most of its coordinating functions, it can today focus on its key function of risk assessment. Given the size of the agency, this is a major step.
Mébih’s key principles are credibility, reliability, excellence, openness and integrity. These principles resemble the ones adopted by EFSA. There are several indicators (e.g. principles, advisory forum, scientific panels) that EFSA has been a role model for Mébih, triggered by the idea to improve smooth cooperation between the Hungarian and the European agency (Interview with Mébih director). These are soft, yet still powerful effects of Europeanization based on the perceived need for efficient networking.
In sum, risk assessment, communication and management are today separated so that the Hungarian system no longer fits the fragmented, but is instead an example of the bi‐institutional (or emerging) model. The Hungarian system is no longer in the making, but con‐solidated. There is hardly any functional overlap left, but responsibilities are separated. Among the Central and Eastern European Countries, Hungary is considered to be a forerun‐ner in the field of food safety. It has adjusted to the European model and, thus, is an excep‐tion among the 2004/07 accession countries. While Poland and Slovakia also belong to this group, separation of responsibilities is less clear. Besides effects of vertical Europeanization, we also see some signs of horizontal Europeanization. Germany has been the Hungarian partner in a major Twinning Project of food safety organizations prior to accession. There‐fore, the German choice for a separated system may have been an additional factor in the development of food safety in Hungary.
Finally, the reform has evolved in several steps with crisis serving as a trigger. Crisis led to calls for reforms; a homemade scandal affecting the system in its initial stage shed light on the deficits of the system and increased pressure for further change. In general, a concur‐rence of Europeanization factors and domestic factors is a particularly strong trigger for change. “It is expected that given the need for adaptation because of a mismatch between the European policy mandate and the existing national policy practice, and given domestic key actors’ preference for policy reform […] policy change and a concomitant transformation in administrative structures are more likely” (Héritier 2001, 53). In sum, in the Hungarian case, Europeanization impact – vertical as well as horizontal – was very strong.
26
5. Patterns of food safety regulation in the EU – a first analysis There are considerable variations in the kinds of food safety regimes in place in the EU
member states. Despite all variation, there are two developing models regarding risk gov‐ernance: (a) a model in which risk assessment is separated from risk management (bi‐institutional, emerging model) and (b) a more integrated model with all risk governance tasks in the hand of one competent authority. The authors of the report to the European Parliament were optimistic about the emerging model, which they clearly assume to be su‐perior; they claim that “(o)ne key feature of the post‐turning point changes to the national system has been the widespread recognition of the necessity and perhaps the inevitability of institutionally separating risk assessment from risk management” (European Parliament/STOA 2000, 69). Yet, the current reality is more complex. Nevertheless, we can identify some interesting patterns we would like to sketch out.
Origin of reform: Food safety regimes in the EU member states have similar origins. They are also concerned with similar responsibilities and problems. The systems of food safety regulation have been reformed in most member states – often dramatically. Responding to the general crisis was a key reason. Reforms were especially radical if there were concomi‐tant national crises so that Europeanization effects and domestic pressure joined forces. This was the case, for instance, in the UK, France and Germany but also in Belgium or Hungary. Key arguments employed for regime change concerned the issues of transparency and effec‐tiveness in order to improve the output legitimacy of the food safety system. In those mem‐ber states where change worked in the direction of the bi‐institutional model and where reforms were introduced at about the same time as EFSA was set up, often EFSA was adopted as an explicit model. Institutional isomorphism was defended on grounds that an institutional misfit should be avoided. It was hoped that institutional similarities between the domestic and the EU level would improve cooperation of the national agency with EFSA. This was clearly the case in Germany as well as in Austria. This reasoning is perfectly in line with insights from the sociological‐neoinstitutionalist strand in Europeanization research (cf. Börzel/Risse 2000b, 62). Yet even isomorphism does not rule out the continued existence of misfits.
Influential factors: Singling out individual factors, identifying causal relations and intro‐ducing a model of change is not possible based on the data we currently have analysed. Eu‐ropeanization is most certainly a key factor that can be identified, even though it can only be a soft one given that there is no top‐down pressure for a specific model of regulatory design. Diffusion, transfer and emulation can be identified in many cases – especially for the “sepa‐ratist” countries, but may be also for those countries with evolving systems. In addition, globalization factors have to be accounted for as well. In our case, the Codex Alimentarius is a crucial reference point for standard‐setting and procedures in food safety. When the Co‐dex Alimentarius came up with suggestions for a stricter procedural division between risk assessment and risk management, this also had an effect on a number of EU member states. The otherwise most stable Swedish case, which responded with some intra‐departmental separation, illustrates these globalization effects. Finally, domestic factors are prominent in the institutional design of agencies and national regulatory regimes. National legacies and traditions of public administrations are decisive; this has been obvious in all exemplary cases discussed in this paper – yet to a different degree. Yesilkagit and Christensen (2010, 70f.) argue that cultural factors such as traditions, styles and historical trajectories are important
27
and should be integrated into models of agency design ‐ especially in European parliamen‐tary systems. They conclude that “(i)t is striking to what extent solutions with deep roots in the administrative tradition of each country are preferred” (ibid., 71). By integrating the new impulses into their traditions, they aim to reduce uncertainty. Yet there are variations, and deviation from the path is certainly not impossible. Food safety regimes in a number of countries are proof of change, especially those countries that can be grouped as “separa‐tists”.
Basic regulatory models: It is, firstly, striking that the member states following an inte‐grated model outnumber those following the bi‐institutional model. Secondly, the sheer ex‐istence of two models is a strong argument against a strong functional argument which fa‐vours convergence. Even though there are functional requirements rooted in the logic of regulatory networks and how they operate, this is not to say that only one specific model can be successful. Countries that have undergone radical change can be found in both mod‐els (e.g. Austria, UK). There is, thirdly, some convergence to two models, while the third fragmented model seems to be a transitional one (see the Hungarian example). There are also signs of a more separated system evolving in Bulgaria. Fourthly, among those member states following the integrated model, there is a greater wider variety of domestic designs. Most of the countries in this group have independent agencies, yet Cyprus, the Czech Repub‐lik and Slovenia do not. Furthermore, while in Cyprus all functions are located in one minis‐try, in the Czech Republic there is at least some procedural separation. Estonia and Lithuania have an integrated system, but show first signs toward separation of risk assessment from risk management; they may move toward the bi‐institutional model since the system in both member states is currently still in the making.
Relation to government and agency discretion: Almost all agencies are subordinate to one or even more ministries. Sweden is an exception to be explained by the strong domestic tra‐ditions of professional organizations. The struggle for control over food safety between ad‐ministrations of health and agriculture are reflected in the oversight pattern (cf. also Elvbakken et al. 2008). In most member states food agencies work under the aegis of the ministries of agriculture (n=15), whereas only a few are supervised by the ministry of health (n=6). The division of supervision tasks reflects tensions between producer interests more likely to be promoted by agricultural ministries and consumer interests promoted by health ministries.22 In two states other ministries are responsible (cf. annex 1). In some member states (n=5) supervision is divided between two ministries. France is an outlier: AFFSA is su‐pervised by three (!) ministries. However, ministerial supervision does not mean that there is no agency independence and no room for discretion. The relation to government is just one indicator for their degree of independence.
Often, delegation is conceptualized as a principal‐agent relationship (Pollack 1997; 2002; Thatcher/Stone Sweet Alec 2002). In general, such a relationship might suffer from bureau‐cratic drift when the agent starts to pursue its own interests instead of those of the princi‐pal. The principal disposes of two instruments to minimise these agency losses. Either he restricts the discretion the agent has right from the beginning within the mandate (ex‐ante restriction) or he installs different control mechanisms to supervise the agent’s behaviour
22 The restructuring of the European Commission and the establishment of DG Sanco was also heavily inspired
by this argument.
28
(ex‐post restrictions). Usually these means will be combined in the institutional design (Epstein/O'Halloran 1994, 698‐699; Pollack 2003, 61‐62). Depending on the combinations, different amounts of discretion for the agency will result, where discretion means “those actions that no coalition of principals can overturn” (Franchino 2000, 156). In the literature on delegation several approaches are discussed to conceptualize formal independence of agencies (Gilardi 2005; 2008; Verhoest et al. 2004; Wonka/Rittberger 2009). The differences in the criteria applied are only marginal. There is wide agreement that the following factors are important: (1) the independence of the Executive Directors with respect to procedure of appointment and dismissal; (2) appointment and dismissal of members of the administrative boards; (3) the relation to parliament and/or government in terms of reporting and overrul‐ing agency decisions; (4) budgetary and organizational autonomy; and (5) finally the regula‐tory competence to implement agency decisions.23 Delegation theory points out that auton‐omy in terms of regulatory competencies and supervision (or control) are interlinked. The more discretion a non‐elected body is granted, the tighter the control mechanisms are ex‐pected to be. Nevertheless, there are also some hints, for example, that agencies are more independent, if they have been set up in times of policy conflict, because independent agen‐cies will also be free from the influence of opposition parties (cf. Yesilkagit/Christensen 2010, 56). The literature is inconclusive about strong causal relationships.
Relations to EFSA and networking: Establishing a transnational regulatory food safety network is one of the key objectives of reforms in all member states. The multi‐level regula‐tory network are certainly an important transmission belt for horizontal as well as vertical Europeanization. EFSA is supposed to be the central node, the hub in the evolving European network; yet EFSA’s mandate is restricted to risk assessment and risk communication. It fits the “separatist” model just like a number of member states where more radical changes have taken place. Networks require networking partners. A requirement of the EU food law is that the competent authorities in the member states have to cooperate with EFSA. For this purpose, member states have to nominate so‐called EFSA Focal Points. In 19 member states the national food agency serves as the EFSA Focal Point; this still means that in 8 countries other authorities hold this position. This is obvious in those countries where no food agency has yet been set up, but it is also the case in some countries where a food safety agency is in place: the EFSA Focal Point in Bulgaria is the National Centre for Public Health Protection (NCPHP), in Belgium it is the Federal Public Service for Health, Food Chain Safety (a ministry), in Estonia and Slovakia it is the Ministry of Agriculture, and in Lithuania the State Environ‐ment Health Centre (a public administration supervised by the Ministry of Health). In four countries (Belgium, Estonia, Lithuania, Slovakia) food agencies serve only (!) as contact points for the rapid alert system RASSF, which is essentially a risk management and not a risk assessment network. Finally, in a total of 12 member states food agencies are EFSA Focal Points as well as RASSF Contact Points. Most, yet not all, of these countries belong to the group of “integrationists”.
Formal aspects of national agencies: Some patterns relate to factors such as the sheer size of the member state. By and large, the larger member states have larger agencies em‐ploying more staff,24 conducting in‐house risk research. Risk research is often a problem for
23 For a detailed discussion cf. the formal independence index by Gilardi (2005). 24 Staff ranges from only 8 employees in Luxembourg to 1.400 in the Austrian AGES.
29
the very small member states, which often have to rely on studies conducted in other mem‐ber states or which sometimes do not even have reference laboratories (e.g. Luxembourg). Particularly for those countries, shared norms and procedural standards in risk research are a crucial issue. In addition, there is some preliminary evidence that in general the larger member states are more actively involved in EFSA and its dozens of thematic scientific net‐works (see annex 3; cf. also for a more detailed account Kobusch 2010). Germany, France and the UK are particularly active, but also the Netherlands, Sweden and Austria while other member states that are otherwise considered “larger” (e.g. Italy, Spain, Poland) are less ac‐tive.
Territorial dimension: While, in general, federal systems seem to be less prone to Europe‐anization effects, because of the number of veto players and the division of competencies between levels of government, this is not so obvious for food safety. Radical reforms were possible in federal countries such as Germany, Austria and Belgium. In Germany and Austria, even real institutional innovations were possible (new types of public administrations), how‐ever, in the German case there is evidence that a more radical reform of the enforcement regime was vetoed by the Länder. In all member states, regional and/or local authorities are involved in enforcement (e.g. food inspection). Yet policy‐ and rule‐making competences are limited to the national level; this is also the case in the federal and “federalizing” member states such as Spain or the UK. In some countries with stronger regions there may be coordi‐nating administrations at the regional level (e.g. UK).
Regional patterns and East‐West Europeanization: Often different Europeanization im‐pacts can be linked to regional patterns such as Western European, Scandinavian countries, Southern European countries or Central and East European countries. A regional pattern is not so obvious in food safety regimes. The Western European countries can be found in both models; also the Scandinavian countries – often thought of as a more consistent group, fol‐low different models. Two regional patterns seem to be more consistent. Firstly, all Southern member states fit into the model of integrated systems; none has moved to a more sepa‐rated system. Overall, it seems that there have been less radical changes in the Southern European member states. Secondly, the next more obvious regional pattern concerns East‐West differences: All member states accessing the EU in 2004/2007 have systems still in the making. This is not only true for the Central and Eastern European Countries, but also for Malta and Cyprus. Yet the problems in the post‐communist countries are more severe. In some of these countries there are so far no independent agencies and no signs of change. Even though the food acquis had to be transposed into national law with EU accession, this has not resulted in radical changes in the administrative system. Nonetheless, this is not be‐cause of the “intgrated” model as such, even though most of these countries can be grouped into this model, but the problems are rather due to the challenge of accession (e.g. admini‐stration overload and the like). Hungary is an exception that has undergone reform in sev‐eral steps; it may serve as an example for other post‐communist countries. In Bulgaria, the separation of responsibilities is until today not yet clear and insufficient.
6. Conclusions Food safety regimes at European and at national level are highly dynamic. We can moni‐
tor incredible change since the mid‐1990s, most of which is crisis‐induced. The comparative
30
focus reveals that national context and history are important for understanding change. At the same time there is some convergence leading to the establishment of two different basic agency models: the “separatist” and the “integrationist” model. Yet both models are not static. There is evidence of change within the models. In addition, member states may in the future switch from one model to another one. The “separatist” one is the most recent model and this is where we can examine most transfer, emulation and isomorphism. Europeaniza‐tion effects in a vertical as well as in a horizontal perspective are present. Yet the way in which these soft Europeanization effects impact national systems depends very much on domestic factors such as cultural trajectories and traditions in public administrations.
Regional patters that are widespread in European studies are difficult to detect, with one predictable exception: in the Central and Eastern European member states regulatory sys‐tems are on the whole still in the making. Changes have started in the pre‐accession stage and are, thus, Europeanization effects. However, there is still some variety between these member states due to domestic factors. Hungary is an astounding case of change and con‐solidation.
National agencies are most important for the functioning of regulatory policy‐making in the EU. Regulatory policy‐making in the EU has to be designed in a network fashion. In the field of food safety a multi‐level regulatory network is evolving consisting of competent agencies from all EU member states – usually, yet not always independent agencies – and EFSA, which is the hub in the evolving network. Typical for a European agency, EFSA has to rely heavily on input provided by national competent authorities, input of a technical and a scientific nature. Especially in social regulation, and food safety is part of it, trust among and reputation of risk assessors are important assets and are a requirement for efficient regula‐tion. We argue in favour of the sociological‐institutionalist proposition that structural simi‐larities are a factor that helps improving the effectiveness of cooperation between network actors. This is not to say that all agencies have to be institutionally alike, yet methods, pro‐cedures, and substantive rules are an essential part of rather informal, bottom‐up harmoni‐zation in European regulatory networks. Concurrently these networks provide important enabling structures for the diffusion of designs, standards and practices both in the vertical and horizontal dimension.
The data presented and the tentative results discussed are still work in progress. What is needed is more comparative studies that include all 27 systems plus EFSA and testing some of the hypothesis introduced in this paper in more detail. The results, which are heterogene‐ous and not easy to interpret, are a reminder that the selection bias can be avoided in “larg‐er n” comparisons. By focusing on structurally similar agencies the real complexity of agency design might be missed. Therefore, case studies still deserve a lot of attention in order to understand different mechanisms leading nevertheless to similar results and vice versa.
31
Annex 1: Overview of food safety agencies at national and supranational level
Member States Co
mpe
tent
food
safety
agen
cy
Competent ministry
Separation of RA – RM
Agency competencies
Link to EFSA
health
agric
ul.
othe
rs
yes
no
assess‐
men
t man
age‐
men
t commu‐
nicatio
n
Focal point
RASFF
EU level EFSA X X
Austria AGES X X X X X X X
Belgium AFSCA X X X X X X
Bulgaria NCFS X X3 (X) Cyprus ‐‐ X5 X X X X Czech Rep. ‐‐ X X X Denmark DVFA
DTU X X
X X
X X
Estonia VTA X (X) X X X X
Finland EVIRA X X X (X) X X X
France AFFSA X X X2 X X X X Germany BfR X X X X X Greece EFET X X X X X X X
Hungary MEBIH X X X X X X
Ireland FSAI X X X X X X
Italy ISS X X X X Latvia PVD X X X X X X X
Lithuania MVT (X) X X
Luxembourg OSQCA X X X X X X X X
Malta FSC MSA
X1
X2 X
X X X
X X
X
Netherlands VWA X (X) X X X X X
Poland GIS X (X) X X X
Portugal ASAE X2 X X X X X Romania ANSVSA X X X X X X Slovakia ŠVPS X (X) X X
Slovenia ‐‐ X Spain AESAN X X X X X X X
Sweden SLV X X X X X X
UK FSA X X X X X X
Legend: 1 Ministry of Social Policy, 2 Ministry of Economic Affairs; 3 Government, 4 Ministry of Fisheries
32
Annex 2: Food safety regimes in EU member states (alphabetical order) Austria: Risk assessment is separated from risk management. The competent authority for risk
management is the Austrian Agency for Health and Food Safety (AGES), which also indentifies new health risks and issues recommendations for risk reduction. It conducts examinations and expert reviews according to Austrian food safety laws, carries out veterinary examinations, and deals with combating and preventing infectious diseases in humans. The task field of the AGES therefore also extends beyond the area of food safety. In 2002, 19 federal agencies from the areas of food inspec‐tion, bacteriology and serology, veterinary medicine, and agriculture joined together under the um‐brella of the Austrian Agency for Health and Food Safety. With this move, Austria became the only European state in which the federal competencies in various areas dealing with the food supply chain are bundled into one organization. The integrative view of the food cycle is also new in the examina‐tion of Austrian food safety: Up until the reorganization of the AGES, the representatives of the areas agriculture, food, human medicine, and veterinary medicine were responsible for only those tasks falling within their own field. Furthermore, the agency serves as the EFSA Focal Point and is the RASFF Contact Point. The agency is accountable to the Ministry of Health (BMG) as well as to the Federal Ministry of Agriculture, Forestry, Environment and Water Management (BMLFUW), which act as risk managers. Enforcement competencies (food inspection) are entrusted to the Federal Office for Food Safety (BAES), which is integrated into the AGES structure.
Belgium: Risk assessment and management are not separated. The competent authority is the Federal Agency for the Safety of the Food (AFSCA) founded in 2000. In the aftermath of the various food safety crises, in particular the dioxin scandal in 1999, the Federal Public Service for Health, Food Chain Safety and Environment (FPS) was founded in 2000. FPS is the ministry in charge; it is involved in risk assessment but also has risk management functions such as monitoring and also rule‐making. FPS serves as the EFSA Focal Point. AFSCA reports to the FPS. The European reaction in the form of the Commission’s White Book can also be seen as an impetus for reform in this case. For efficiency reasons, the AFCSA also controls animal and plant protection. It fulfills risk assessment as well as risk management tasks; it serves as the RASFF Contact Point.
Bulgaria: The central norm in the area of food safety was established in 2003. Only the first signs of an inter‐institutional separation of risk assessment, risk management, and risk communication can be seen in Bulgaria up to now. The reform of the food safety system is not yet complete; the delega‐tion of competencies is still partially unclear. The main actors are the Ministry of Agriculture (MZH), which acts as the Contact Point for the RASFF, and the Ministry of Health (MH). Both ministries are active in risk management and risk assessment. The National Center for Public Health Protection (NCPHP), which is under the auspices of the MH, is the EFSA Focal Point and only has tasks in the area of risk assessment and communication. It maintains its own laboratories and can therefore es‐tablish on‐site expertise. Unique to the systems studied is the National Committee for Food Safety (NCFS), under the oversight of the government, which is responsible for coordinating food safety policies and controlling mechanisms. However, the NCFS is not an agency in the classical sense. Simi‐lar to agencies in other states, it is staffed with representatives from the competent agencies but at the same time does not have an institutionally‐anchored structure typical for agencies. The NCFS was created within the framework of the Food Law for coordinating state trade in the area of food safety at a federal level. Because of the timing, it can be assumed that the changes were made with an eye on the planned accession to the European Union and reflect an attempt to harmonize national regu‐lations with EU law. A committee of experts, which was also established by the new law in 2003, is responsible for risk assessment, although it is unclear to which part of the administration it is as‐signed. The committee is made up of food safety experts and is to provide independent advice based on scientific findings. According to the FVO report from 2007, however, it appears that the commit‐tee of experts did not yet exist in the timeframe under consideration (that is, shortly before or at the time of the accession to the EU).
33
Cyprus: There is no independent food safety agency; instead, this is part of the duty of the Minis‐try of Health (MOH). It is supported by a Food Safety Council consisting of representatives of compe‐tent agencies and of stakeholders. There is no separation of risk assessment, risk management, and risk communication. The MOH is responsible for all functions, but shares management and commu‐nication functions with the Ministry of Agriculture (MOA). The MOH acts as the EFSA Focal Point and as the Contact Point for the early warning system RASFF. Especially inside the MOH, however, there seems to be a separation of tasks. Here, the state’s general laboratory (SGL) is primarily responsible for risk assessment and the analysis of control samples, but the SGL is a department of the MOH. Considering the limited laboratory capacities in Cyprus, there are contracts with foreign laboratories. The Food Safety Board, which is made up of representatives from all important actors in the area of food safety regulation, assumes advisory competencies for food safety policies, priority‐setting, and the coordination of the responsible institutions. From the FVO reports, it is not apparent that the Cypriot food safety system underwent a significant change. Still, EU regulations are implemented into national regulations and the structural adjustments necessary for this are made.
Czech Republic: There is no institutional separation of risk assessment and risk management, yet there is some procedural separation since risk assessment is conducted outside the ministries. The Czech food safety agency (the Food Authority) began its work in April of 2005 and has approximately 33 employees. It is responsible for risk assessment and communication as well as the development of the Food Safety Strategy. The Czech Republic is a unique case because it has a food agency responsi‐ble for risk assessment, but this is a direct department of the Ministry of Agriculture (MZE), which is the key actor endowed with all risk‐associated functions. The MZE, and not the agency, is the EFSA Focal Point and RASFF Contact Point. In addition, the Ministry of Health (MZ) is involved in food safe‐ty and is endowed with all risk‐associated tags. The National Institute for Public Health is responsible for risk assessment and answers to the MZ, but does so as a subordinate agency. Functionally, the task fields are therefore separated between the two ministries. For the coordination of food safety, an “Inter‐ministerial Food Safety Co‐ordination Unit”, founded in 2002, is responsible, and its role was consolidated in the Food Safety Strategy of 2007. This strategy represents an update of the Eu‐ropean Strategy for Ensuring Food Safety in the Czech Republic after its accession. This document was supposed to help create a unified and coordinated food safety system and set certain deadlines to meet this goal. The Czech Republic followed the French example and founded a food safety agency later than Hungary.
Denmark: There is by and large an institutional separation of risk assessment and risk manage‐ment. On the central level, the Ministry of Food, Agriculture and Fisheries (MFLF) is responsible; in 2007 it was put under the supervision of the Danish Veterinary and Food Administration (DVFA), whose mandate includes risk management and which acts as the RASFF Contact Point. The DVFA is responsible for risk assessment and communication as well as for risk management. These tasks are functionally differentiated within the DVFA departments. The DVFA and the regional agencies are jointly responsible for controlling tasks. Coordination and administrative tasks are done by the DVFA. With nearly 1,870 employees, it is one of the larger European agencies. In addition, the Danish Plant Directorate (DPD) is also involved in risk management. Risk assessment is the task of the National Food Institute (DTU) of the Danish Technical University. The assessments are then used by the MFLF and DVFA as the basis for risk assessment decisions. The DTU is the EFSA Focal Point. Based on the data, nothing can be said about the reform measures in the field of Danish food regulation.
Estonia: The system of food safety regulation is still in the making. There are so far only first signs of an institutional separation of risk assessment and risk management. The Ministry of Agriculture (PM) plays a leading role and is also the EFSA Focal Point. The Ministry of Social Affairs (SM) is also involved in risk management. The Veterinary and Food Board (VTA) is subordinate to the PM and mainly engaged in risk management and communication; it also serves as the RASFF Contact Point. In its accession process to the European Union, Estonia implemented the aquis communautaire in the areas of agriculture and consumer protection as well as the EU legal program. The central regulation
34
for the food safety sector is provided by the 1999 Food Act. In general, the Ministry of Agriculture is responsible for the area of food safety. A functional and institutional differentiation of risk assess‐ment and risk management can be gathered neither from the ministerial sources nor from the EFSA or the expansion reports. It can be assumed that there is an inter‐institutional separation in the Vete‐rinary and Food Department. A clear delegation of tasks for the ministerial departments cannot be gathered from the sources. Possible actors in the area of risk assessment are the Research and De‐velopment Department or, internally, the Food Safety Office in the Food and Veterinary Department (VFB). Evidence for this is the fact that the Veterinary and Food Department is the EFSA Focal Point and Codex‐Almentarius Contact Point.
Finland: There is a clear organizational separation of risk assessment and management in place. The Finnish Food Safety Authority (Evira) is the key actor. The Finnish Food Safety Authority started its activities in May of 2006. Since this time, the agencies responsible for food safety have been unit‐ed under one roof, following the principle “from farm to table.” In order to establish scientific exper‐tise, Evira conducts its own research and is a Reference Laboratory. With its 750 employees, Evira is one of the largest food safety agencies in the European Union. It is also involved in some manage‐ment tasks such as the organization and implementation of inspection plans. Evira is the RASFF Con‐tact Point as well the EFSA Focal Point. The Ministry of Agriculture and Forestry (MMM) and the Min‐istry of Social Affairs and Health (STM) are entirely in charge of risk management and partially for risk communication. A further agency is the National Supervisory Authority for Welfare and Health (VAL‐VIRA), which supervises the safety of tap water and the implementation of the Gene Technology Act.
France: Since the 1998 amendment, France has had a strict institutional separation of risk as‐sessment, risk management and risk communication. The three ministries involved, the Ministry of Food, Agriculture and Fisheries (MAAP), that of Heath and Sports (MSS), and finally that of Economy, Industry and Employment (MEIE) are tasked with risk management and communication. The respon‐sibility for risk assessment lies entirely with the French Food Safety Agency (AFSSA), which is also the national EFSA Focal Point. AFSSA was founded in 1999, even before the European agencies. The insti‐tute has its own laboratory and can therefore conduct its own research as well as develop expertise. The impetus for establishing a better health system, with a focus on prioritizing health, was not just the BSE scandal. As early as the 1980s, France was shaken by a blood transfusion scandal, and in the aftermath, a new, independent agency for assessing transfusion and medication risks was created in 1993. A far‐reaching reform of the field of food safety was then brought about by the BSE crisis. The scandal demonstrated where the deficits in food regulation were which, until then, had been the responsibility of the Ministry of Agriculture and Finance. In order to fight a lack of transparency, in‐adequate scientific discourse amongst experts, the dominating influence of industrial and agricultural interests, and a lack of controlling mechanisms, the Public Health Code was written in 1998. It created several agencies, including the Agence française de sécurité sanitaire des aliments (AFSSA). Its assessments serve as a basis for the risk assessment in the ministries involved. A further task of the AFSSA is risk communication, which it conducts jointly with other actors. An additional important point is the AFSSA’s responsibility for veterinary medicine, which differentiates it fundamentally from many other national food safety agencies in the European Union. Excellency, transparency, and inde‐pendence are its declared principles. One more particularity of the AFSSA is the possibility for public requests. The MEIE bears responsibility for the early warning system RASFF.
Germany: There has been a system of institutional separation between risk assessment and risk management in place since the reform of the system in 2002. Before then, the Ministry of Health as well as the Ministry of Agriculture were responsible for food safety. After harsh criticism was ex‐pressed about the ministry’s procedures in the wake of the first cases of BSE, the then‐Minister of Health, Andrea Fischer, and her colleague, Karl‐Heinz Funke, resigned. In return, the Ministry for Food, Agriculture, and Consumer Protection, (BMVEL) was founded with Renate Künast as the first minister. Not only was the name intended to emphasize the concerns and interests of the consum‐ers. The key ministry is the Ministry of Food, Agriculture and Consumer Protection (BMELV). With the
35
background of the BSE crisis, the consumer protection agency had lost a great amount of credibility, and therefore the Federal Agency for Risk Assessment (BfR) was also created in the course of the reform. The Federal Agency for Risk Assessment (BfR) is a specialist authority with the BMELV portfo‐lio. The BfR is endowed with risk assessment tasks and also with risk communication (also beyond the field of food safety). The BfR is an institution directly under the control of the federal government and responsible for food safety as well as for consumer protection regarding health. A further cha‐racteristic of the BfR in European comparison is its size. With 699 employees (as of the end of 2008), it is one of the largest assessment agencies in the EU. Not least because of its size, it is capable of conducting its own research and generating scientific reports. Besides the independent assessment of risks, it is also responsible for risk communication. Its principles could be described with the words “expertise, quality, independence, and transparency”. It is the national EFSA Focal Point. In addition to other cooperation areas, since 2004 the BfR has participated in the European Research Network for Prevention and Control of Zoonoses (MED VET NET). There is another agency solely for risk man‐agement, the Federal Office of Consumer Protection and Food Safety (BVL), which is also the RASFF Contact Point. The BVL usually draws on BfR risk assessments. The Food and Animal Feed law book of 2005 is not the only legal basis for German food regulation; it also adapts its national laws to those of the EU.
Greece: Risk assessment, management and communication are not institutionally separated. The Hellenic Food Authority (EFET) is an agency is subordinate to the Ministry of Rural Development and Food (YAAT). In the area of food safety, this ministry is typically responsible for controlling food hy‐giene and safety. The EFET is the national RASFF Contact Point and the EFSA Focal Point. At the re‐gional level, 13 regional directorates of the EFET are charged with ensuring food safety. The Greek food safety agency was founded in 1999 as a government organization under the control of the Min‐istry of Agriculture. The priority of the EFET is ensuring food safety, food hygiene, consumer protec‐tion and compliance with legal norms. The tasks of the EFET include risk management, risk assess‐ment and also risk communication. In the case of risk assessment, the EFET is the central actor in Greece. The EFET carries out its own studies on food and nutrition. In the area of risk management, the EFET oversees the implementation of legal norms in food safety. Besides this, the EFET is also responsible for crisis management in Greece. The principal aim of the Hellenic Food Authority is to take all necessary actions to ensure that food produced, distributed or marketed in Greece meets the standards of food safety and hygiene as described by national and European legislation.
Hungary: Risk assessment, risk management and risk communication are institutionally sepa‐rated. The Hungarian food safety system is no longer in the development phase, the competencies of the individual actors are seldom overlap, and in most cases they are now adequately differentiated. The Hungarian Office for Food Safety (Mébih) was founded in 2003 in order to create a contact part‐ner for EFSA at the national level and more closely coordinate the controlling mechanisms. Mébih was therefore primarily created as an agency for internal and external coordination. It is also en‐trusted with risk management and communication, especially in case of crisis. Mébih acts as the EFSA Focal Point as well as the RASFF Contact Point. Because of its size, at this time only 18 employees, and the lack of capacities such as laboratories, it is unable to develop expertise itself and must rely on the data and research of other Hungarian institutes or other European countries. Besides risk assessment, risk communication with experts and the public is another central task. During the 2007 reform, a major restructuring effort took place, after which the Mébih no longer had any important responsibilities regarding the coordination amongst various institutions. The Ministry of Agriculture and Rural Development (FVM) oversees Mébih. Important for the development of the structures was an EU‐initiated Twinning Project between Germany and Hungary.
Ireland: There is no institutional separation of risk assessment from risk management. The key ac‐tor is the Food Safety Authority of Ireland (FSAI), a statutory, independent body established in 1999 by the Department of Health and Children (DoHC). The Irish Food Safety Authority was founded un‐der the 1998 Ireland Act, enacted in 1999. Besides the name, the date it was founded shows parallels
36
to the British FSA. The Authority is characterized by its independence and a scientifically‐based ap‐proach. It is under the aegis of the Minister for Health and Children. The national responsibility for the coordination of the implementation of laws in the field of food safety lies with the FSAI. Its main principle is: “Putting consumer interests first and foremost”. Some risk management tasks in the area of food and feed are carried out by the Department of Agriculture, Fisheries and Food (DAFF), which also carries out risk assessment on feed.
Italy: Risk assessment and risk management are not institutionally separated. The primary legal basis for food safety law in Italy is Directive 190 (Decreto Legislativo n. 190) on the implementation of the European Directive 178/2002. The Ministry of Labour, Health and Social Affairs (MLSPS) has tasks in the areas of risk management and risk assessment. The Ministry of Agriculture, Food and Forestry Policies (MPAAF) is responsible for risk management and risk communication in the area of plant protection and for quality controls on food, animal feed and plant protection products. The Istituto Superiore di Sanità (ISS) is the leading technical‐scientific institution in the Italian National Health Service. On behalf of the MLSPS, it is responsible for risk assessment in the area of food safe‐ty. ISS is the EFSA Focal Point. It deals with research, controlling, training and consultation in the in‐terest of public health protection. In general, there is little that points to an extensive reform of the food safety system.
Latvia: In Latvia, risk assessment and risk management are institutionally and functionally only minimally separated. Risk assessment is the sole task of the Food and Veterinary Service (PVD) and the Plant Protection Service (VAAD), but these agencies also take on tasks in risk management and are subordinate institutions of the Ministry of Agriculture. Other ministries and agencies only play a limited role in Latvian food policy. The Ministry of Agriculture (ZM) and the Ministry of Health (VM) have competencies in the field of risk management and risk communication. The Food and Veteri‐nary Service (PVD) is a specialist authority within the ZM portofolio and is endowed with risk assess‐ment, risk management, and risk communication tasks. Within the PVD, the Food Center (PC) is re‐sponsible for independent risk assessment. The PVD is the EFSA Focal Point as well as the RASFF Con‐tact Point. In the framework of the EU concept “from farm to table”, which is also recognized in Lat‐via as the most effective protection to secure the consumers’ interests, the Food and Veterinary Ser‐vice (FVS/PVD) started its work on January 1, 2002, under the direction of the Ministry of Agriculture. It was created by the fusion of the state veterinary service and the police health unit responsible for border control. Its mandate was expanded to include the entire food supply chain. Its tasks therefore also include controlling food from animals and plants, monitoring tap water, animal health and con‐trol (including zoonoses) and the authorization of sellers as well as the circulation of veterinary medi‐cations include controlling for medication residue. The competency of the agency reaches from pro‐duction up to the marketing of the products. The central task of the institute is controlling and re‐search. Furthermore, the State Plant Protection Service (VAAD), a further ZM agency, is involved in plant protection and GMO and undertakes risk assessment as well as risk management in these fields.
Lithuania: There is not yet an institutional separation of risk assessment and risk management. The system is still in the making, and there are first signs of a separation. The State Food and Veteri‐nary Service (SVFS) is the central authority for food and feed safety. It is not subordinate to any min‐istry. Risk management is the main MVT task; it serves also as the RASFF Contact Point. Since its ac‐cession to the European Union, Lithuania has tried to consolidate its membership several times. Therefore, the most important principle of the State Food and Veterinary Service (SFVS) is to imple‐ment the priority of food safety, the protection of consumer interests and the avoidance of food crises. Employees of the SFVS are, at the same time, delegates in standing committees and other comitology bodies. In addition, Lithuania has a special veterinary and food safety attaché in the EU. The SFVS was founded in 2000, after the restructuring of the State Veterinary Service and its sub‐institutions, the State Hygiene Inspection under the auspices of the Ministry of Health and the State Quality Inspection as part of the State Service for Competition and Protection of Consumer Rights.
37
The SFVS took on the tasks of these agencies and ensures food supply controls “from stable to table”. In total, it employs 1,810 people. Risk assessment is carried out by the National Food and Veterinary Risk Assessment Institute (NMVRVI), and the risk research is then used by MVT in management deci‐sions. A further body involved in risk assessment is the State Environment Health Centre (VASC); it holds the position as EFSA Focal Point. It is also involved in coordination and implementation. The Centre reports to the Ministry of Health (SAM).
Luxembourg: There is no institutional separation of risk assessment and risk management. Food scandals such as BSE, chemicals in animal feed, or the spoiled meat scandal, together with an abun‐dance of food ordinances from Brussels necessitated a restructuring of the food safety agency as well as a revision of Luxembourg’s food safety legislation from 1953. On April 1, 2009, the Service Center for Food Safety (Organisme pour la sécurité et la qualité de la chaîne alimentaire/OSQCA) opened its doors. The “Service de la sécurité alimentaire” is subordinate to Luxembourg’s Ministry of Health. The analysis of food samples is done by the state laboratory as well as by specialized laboratories so that sampling and analysis are separated from one another. The very small institute only has eight employees. The Organisation for the Safety and Quality of the Food Chain (OSQCA) carries out risk assessment, management and communication. It serves as the EFSA Focal Point as well as the RASFF Contact Point. OSQCA is a specialist body within the Ministry of Agriculture, Viticulture and Rural Development (MAVDR) and the Ministry of Health (MS), which are both involved in risk management and communication. A further body is the Veterinary Service Administration (ASV), also belonging to the portfolio of the two ministries and also endowed with risk assessment and management, but only in the field of food of animal origin.
Malta: A system of institutional separation of risk assessment from risk management is not in place. The responsibility for food safety is shared by the Ministry of Resources and Rural Affairs (MRRA) and the Ministry for Social Policy (MSOP). The Food Safety Commission (FSC) is an indepen‐dent body reporting to MSOP. It carries out risk assessment, management and communication tasks and is also the RASFF Focal Point. The FSC was founded on the basis of the 2002 Food Safety Act and in general is charged with the task of introducing all necessary steps to ensure the highest food and hygiene standards through legal instruments and reputable Codes of Practice (cf. FSC, Homepage, 2010). The Malta Standards Authority (MSA) belongs to the Ministry of Finance and Economic Affairs (MFIN); it is responsible for risk assessment and risk management. It is the EFSA Focal Point. MSA was founded in 2000, and, in the last reform of the Act XXIX in 2007, it was made a separate legal entity from the government.
Netherlands: Risk assessment, risk management and risk communication are, to a large extent, institutionally separated. The Ministry of Agriculture, Nature and Food Safety (LNV) and the Ministry of Health, Welfare and Sports (VWS) are the key ministries endowed with risk management. The Food and Consumer Product Safety Authority (VWA) is the key player. It is an authority within the portofolio of LNV and also delivers expertise for the VWS. Within the VWA, an independent unit is responsible for risk assessment (Risk Assessment Division, BuR). The VWS functions as the RASFF Contact Point and as the EFSA Focal Point. In the Netherlands, there were also tensions between health and agricultural aspects which finally led to a far‐reaching reform in 2000. The complete reor‐ganization of the food safety area included the creation of the VWA. In 2003, the agency was first put under the oversight of the Ministry of Agriculture, but a short time later was transferred to the Min‐istry of Health. Today, the VWA is once again under the control of the Ministry of Agriculture, Nature and Food Safety (LNV). The new, independent agency is responsible for its own budgets and can set its own financial priorities. The institution carries the responsibility for risk assessment. For this pur‐pose, it has an “Office for Risk Assessment” which assesses risks and coordinates the research pro‐grams of the agency. An important difference from many other European states is that the reaction to the BSE scandal did not cause the Netherlands to enact any major reforms of the existing legisla‐tion. The primary law regulating food safety is the Warenwet from 1919, but it contains provisions for including EU regulations.
38
Poland: The primary legal basis for Polish food safety is the Act on Safety of Food and Nutrition from 2006. In Poland, many agencies are tasked with regulating food. The institutional structure of food safety regulation is still being implemented. Because of this, which agency is tasked with risk assessment cannot currently be determined, as it is carried out by several actors. In particular, the Ministry of Health and the Chief Sanitary Inspectorate of the Ministry are active in this area. Other actors in the field of risk assessment include various research institutes. Despite the fragmentation, there is a growing institutional differentiation of risk assessment, risk management and risk commu‐nication. In risk management, particularly the Ministry of Health (MZ) and the Chief Sanitary Inspec‐torate (GIS) at the MZ (created by law in 1985) are active. The GIS at MZ is the EFSA Focal Point as well as the RASFF Contact Point. The MH is responsible for the coordination of food safety in Poland. In addition, it is a participant in the research network MED VET NET. Other actors are the Ministry of Agriculture and Rural Development (MRiRW) and the Ministry of Finance (MF) as well as the Office of Competition and Consumer Protection (UOKiK), which are all involved in risk management. Risk as‐sessment is carried out by the National Food and Nutrition Institute (IZZ), the National Institute of Public Health – National Institute of Hygiene (NIZP‐PZH) and the National Veterinary Research Insti‐tute (PIWET). The auditing processes and FVO control missions in the accession process were impor‐tant for the development of the Polish food safety system.
Portugal: There is no institutional separation of risk assessment and communication from risk management. The Economy and Food Safety Standards Authority (ASAE) carries out all three tasks; it is also the EFSA Focal Point. ASAE is specialist authority of the Ministry of Economy and Innovation (MEI), which is a risk manager in consumer protection and food safety. The agency has a long tradi‐tion in the sense that it can be traced back to a long chain of institutions, each of which was trans‐formed into the next (the first institution can be dated back to the end of the 19th century). It was founded in 2005 under the name Autoridade de Segurança Alimentar e Económica (ASAE). ASAE is seen as the primary food safety control agency in Portugal and enjoys administrative autonomy. In addition, it is responsible for the evaluation of risks and transparent risk communication. It employs a total of 500 people. The Ministry of Agriculture, Rural Development and Fisheries (MADRP) carries out risk management and is the RASFF Contact Point; several reference laboratories are attached to MADRP, and the Directorate‐General for Veterinary Issues (DGV) carries out risk assessment and management for MADRP.
Romania: The country is one of the two youngest member states (accession 2007). So far there is no institutional separation of risk assessment and communication from risk management. In 2004, the National Sanitary Veterinary and Food Safety Authority (ANSVSA) was established. It carries out risk assessment, management and communication. It is also the EFSA Focal Point. The agency is ad‐vised by two scientific committees which began their work in 2005. In addition to the aforemen‐tioned activities, it is also entrusted with tasks in the area of veterinary medicine. On its homepage, it publishes public inquiries – totaling 164 in 2010, they can be considered to be at a middle level. ANSVSA works under the supervision of and reports to the Ministry of Agriculture, Forests and De‐velopment of Rural Areas (MAPDR), which is also involved in all three risk‐associated tasks. Further ministries involved in food safety are the Ministry of Health (MS), which is also the RASFF Contact Point, and the Ministry of the Environment (MM), which carries out risk management and communi‐cation in the field of plant protection and GMOs. On the central level, there are a total of six agencies which are entrusted with controlling food.
Slovakia: The primary legal basis for food safety legislation in Slovakia is the Directive (EC) 178/2002, which was implemented into law at the national level. Risk assessment and risk communi‐cation are largely separated. The key risk managers are the Ministry of Agriculture (MP), which is also the EFSA Focal Point, and the State Veterinary and Food Authority (ŠVPS), which is subordinate to the MP and is the RASFF Contact Point. The MP is responsible for agriculture as well as food and animal feed safety in Slovakia. In these areas, the MP is also the primary actor on the national level. Within the ministry, risk management as well as risk assessment and risk communication are conducted.
39
Directly subordinate to the ministry in the areas of food safety and consumer protection are, in par‐ticular, the State Veterinary and Food Agency (Štátna veterinárna a potravinová správa) (ŠVPS), the Central Institute for Controlling and Tests in Agriculture (Ústredný kontrolný a skúšobný ústav poľnohospodársky) (ÚKSÚP), and the Food Research Institute (Výskumný ústav potravinársky) (VUP). The SVPS is responsible for risk assessment in the areas of animal health and protection, food of an‐imal origin, veterinary medications and residue as well as food hygiene (together with the Agency for Public Health UVZ). Risk assessment for feed, plant health and plant protection products is done at the ÚKSÚP. The VUP is responsible for plant protection products as well as for GMOs together with the ÚKSÚP. To fulfill these tasks, multiple National Reference Laboratories (NRLs) are found under the auspices of the Ministry of Agriculture. Further bodies subordinate to the MP are the Central Institute for Inspections and Test in Agriculture (ÚKSÚP), which carries out risk management, and the Research Institute for Food (VÚP), which conducts risk assessment. The Ministry of Health (MZ) is involved in risk assessment and management. Regarding risk management, it is supported by the Public Health Authority (ÚVZ). No information on the founding or role of the various institutions or on reform measures can be found.
Slovenia: Slovenia (EU accession in 2004) orients itself on EU Food Law (Regulation No. 178/2002) in the area of food safety. Up to the present, no separation of risk assessment, risk management and risk communication can be noted. All tasks are currently located at the ministerial level, and no inde‐pendent food safety agency has been developed. The food safety system itself is still being devel‐oped, however. An important actor is the Ministry of Agriculture, Forestry and Food (MKGP), which is also the EFSA Focal Point and is active in the areas of risk management and risk communication. Agencies belonging to the MKGP are the Veterinary Administration of the Republic of Slovenia (VURS), the Inspectorate of the Republic of Slovenia for Agriculture, Forestry and Food (IRSKGH) and the Phytosanitary Administration of the Republic of Slovenia (FURS). The Ministry of Health also car‐ries out risk management and communication tasks; it is the RASFF Contact Point. It is supported by the Health Inspectorate (ZIRS). The government worked out a national plan for food and nutrition issues (timeframe 2005‐2010), which foresees the development of an efficient food safety system.
Spain: There is no institutional separation of risk assessment from risk management. The key ac‐tor is the Spanish Food Safety and Nutrition Agency (AESAN), which was founded in 2002. Its main tasks are to strengthen health aspects and provide guarantees as well as objective information for consumers and the industrial sector. A 2006 law expanded its competencies in the areas of planning, coordination, and management. A further particularity of the Spanish political system is the extreme‐ly autonomous regions which also must be taken into consideration in the area of food safety poli‐cies. Here, the institutional committee of AESAN plays an important role in coordination. It carries out risk assessment, management as well as communication. Internally, these tasks are divided up between different units. AESAN serves as the EFSA Focal Point and as the RASFF Contact Point. AE‐SAN works within the portfolio of the Ministry of Health and Social Policy (MSPS), the prime risk manager in food safety. Risk assessment and management are also carried out by the Ministry of Environment, Rural and Marine Affairs (MARM).
Sweden: There is no institutional separation of risk assessment from risk management. In 1972, Sweden had already established an independent agency in the area of food safety. The National Food Administration (SLV) is the key agency and is also the EFSA Focal Point and the RASFF Contact Point. The SLV is independent and not supervised by any ministry. It is entrusted with controlling food and independent risk assessment and risk communication. The basic structure of the Swedish food safety system was not reformed. There have been smaller changes regarding the names and competency areas, but the tasks in the area of risk analysis and in‐house research at all times remained with the SLV. The Ministry of Agriculture (RK) Ministries is also involved in food safety.
United Kingdom: Particularly in Great Britain, the emergence of BSE and the connection with people in a new form of Creutzfeld‐Jakob Disease marked a turning point in food safety. Not only the
40
meat market collapsed, but also consumers’ trust in the state authorities. The Ministry of Agriculture, Fisheries and Food (MAFF), which was at the time responsible for this area, revealed its deficits in this crisis. The ministry’s actions were strongly affected by internal conflicts due to its double respon‐sibility on the one hand for food safety and on the other hand for the food and agriculture business. In the 1980s, the first attempts were made to better protect consumer interests. The Consumer Committee founded in 1991 was not able to exercise any influence on policy and failed. The first clear institutional change in Great Britain was represented by the foundation of the Food Standards Agency (FSA). Besides the BSE crisis, the change of government in 1997 was also decisive for this development. With the passing of the 1999 Food Standards Act, a far‐reaching reform of the British food sector was introduced, and the FSA was founded in 2000 as a “Non Ministerial Government Department at arm’s length from Government”. There is no institutional separation of risk assess‐ment from risk management. The Food Standards Agency (FSA) is the central authority on all food safety issues. It is an independent body carrying out risk assessment, management as well as com‐munication without any departmental supervision. Despite this, there is still a functional separation of risk analysis: While the scientific committees provide the scientific background, the FSA works together with the relevant governmental institutions in the areas of risk management and communi‐cation. The FSA also conducts its own research. Its self‐made principles are the following: “consumer first”, “be open and accessible” and “be an independent voice”. It is the EFSA Focal Point and the RASFF Contact Point. For food safety issues not covered by the remit of FSA, the Department for En‐vironment, Food and Rural Affairs (Defra), as a government department, is responsible. It carries out implementation and monitoring tasks.
41
Annex 3: Links of national agency to EFSA’s scientific panels The figures presented here are based on a dataset derived from membership in EFSA’s
scientific panels. Every Member of one of EFSA’s scientific panel – experts that applied to an open call – has to fill out a Declaration of Interest indicating inter alia the employer in order to prevent conflicts of interest. These are published on EFSA’s web page. This data yields 533 individual members associated to 266 national institutions. Employing techniques of Social Network Analysis we established a dataset that links institutions to institutions. According to the nature of the dataset, the links are usually between EFSA and a national institution. Links between national institutions here show that a person is affiliated to more than one national institution. These links are visualized below using Ucinet (Borgatti et al. 2002) and NetDraw (Borgatti 2002). In order to reduce the complexity, universities (n=92) and private consult‐ants (n=15) have been grouped together. Figure 5 shows how many national institutions (excluding universities and private consultants) are linked to EFSA. Figure 6 shows those na‐tional institutions (again excluding universities and private consultants) that hold at least 10 links to EFSA and are thus the most strongly involved in EFSA’s scientific work. For further details on the method and the dataset cf. Kobusch (2010).
Figure 5: National institutions represented in EFSA’s scientific panels
43
References Alemanno, Alberto 2008: "The European Food Safety Authority at Five"; In: European Food and Feed Law
Review, 1, 2‐29. Alemanno, Alberto 2009: "Regulating Organic Farming in the European Union‐Balancing Consumer Preferences
and Free Movement Imperatives"; In: Food trade: new challenges and new standards – European, American and international perspectives, 83‐108.
Alemanno, Alberto/Mahieu, Stéphanie 2008: "The European Food Safety Authority before European Courts. Some reflections on the judicial review of EFSA scientific opinions and administrative acts"; In: European Food and Feed Law Review, 5, 320‐333.
Axt, Heinz‐Jürgen/Milososki, Antonio/Schwarz, Oliver 2007: "Europäisierung ‐ ein weites Feld. Literaturbericht und Forschungsfragen"; In: Politische Vierteljahresschrift, 48 (1), 136‐149.
Baggot, Rob 1998: "The BSE crisis. Public health and the ‘risk society‘"; In: Gray, Pat/'t Hart, Paul (eds.): Public policy disasters in Western Europe, London u.a.: Routledge, 61‐78.
Baldwin, Robert/Cave, Martin 1999: Understanding regulation: theory, strategy, and practice, Oxford: Oxford University Press.
Bartlett, David M. C. 1998: "Mad cows and democratic governance: BSE and the construction of a ‘free market‘ in the UK"; In: Crime, Law & Social Change, 30 (3), 237‐257.
Birkland, Thomas A. 2001: An Introduction to the Policy Process. Theories, Concepts and Models of Public Policy Making, Armonk/New York; London: M.E. Sharpe.
Borgatti, Steve .P. 2002: Netdraw Network Visualization, Harvard/MA: Analytic Technologies. Borgatti, Steve .P./Everett, Martin G./Freeman, Lin C. 2002: Ucinet for Windows: Software for Social Network
Analysis, Harvard/MA: Analytic Technologies. Borrás, Susana/Koutalakis, Charalampos/Wendler, Frank 2007: "European Agencies and Input Legitimacy: EFSA,
EMeA and EPO in the Post‐Delegation Phase"; In: Journal of European Integration, 29 (5), 583‐600. Börzel, Tanja A./Risse, Thomas 2000a: "Conceptualizing the Domestic Impact of Europe"; In: Featherstone,
Kevin/Radaelli, Claudio M. (eds.): The Politics of Europeanization, Oxford: Oxford University Press, 57‐80.
Börzel, Tanja A./Risse, Thomas 2000b: "When Europe Hits Home: Europeanization and Domestic Change"; In: European Integration online Papers (EIoP), 4 (15). Available: http://eiop.or.at/eiop/texte/2000‐015a.htm.
Bulmer, Simon J. 2007: "Theorizing Europeanization"; In: Graziano, Paolo/Vink, Maarten P. (eds.): Europeanization: New Research Agendas, New York: Palgrave Macmillan, 46‐58.
Buonanno, Laurie 2006: "The Creation of the European Food Safety Authority"; In: Ansell, Christopher/Vogel, David (eds.): What‘s the Beef? The Contested Governance of European Food Safety, Cambridge u.a.: MIT Press, 259‐278.
Bursens , Peter 2007: "State Structures"; In: Graziano, Paolo/Vink, Maarten P. (eds.): Europeanization: New Research Agendas, New York: Palgrave Macmillan, 115‐127.
Carporaso, James A. 1996: "The European Union and Forms of State: Westphalian, Regulatory or Post‐Modern?"; In: Journal of Common Market Studies, 34 (1), 29‐52.
Chalmers, Damian 2003: "'Food for Thought': Reconciling European Risks and Traditional Ways of Life"; In: The Modern Law Review, 66 (4), 532‐562.
Chambers, Graham R. 1999: "The BSE Crisis and the European Parliament"; In: Joerges, Christian/Vos, Ellen (eds.): EU committees: social regulation, law and politics, Oxford u.a.: Hart, 95‐106.
Christensen, Tom/Lægreid, Per 2005: "Regulatory Reforms and Agencification", Stein Rokkan Center for Social Studies, UNIFOB AS, Working Paper 6.
Coase, Ronald H. 1937: "The Nature of the Firm"; In: Economica, New Series, 4 (16), 386‐405. Dreyer, Marion/Renn, Ortwin/Borkhart, Karin/Julia, Ortleb. 2006: "Institutional re‐arrangements in European
food safety governance: a comparative analysis"; In: Vos, Ellen/Wendler, Frank (eds.): Food Safety Regulation in Europe. A comparative Analysis, Antwerpen, Oxford: Intersentia, 9‐64.
Eberlein, Burkhard/Grande, Edgar 2005: "Beyond delegation: transnational regulatory regimes and the eu regulatory state"; In: Journal of European Public Policy, 12 (1), 89‐112.
Elvbakken, Kari Tove/Lægreid, Per/Hellebø Rykkja, Lise 2008: "Regulation for Safe Food: A Comparison of Five European Countries"; In: Scandinavian Political Studies, 31 (2), 125‐148.
Epstein, David/O'Halloran, Sharyn 1994: "Administrative Procedures, Information and Agency Discretion"; In: American Journal of Political Science, 38 (3), 697‐722.
44
European Commission 2002: The operating framework for the European Regulatory Agencies. Communication from the Commission, COM (2002) 718, 11.12.2002, Brussels.
European Commission 2010: Agencies of the EU. Available: http://europa.eu/agencies/index_en.htm (05.06.2010).
European Parliament/Directorate General for Research; Directorate A; STOA: Scientific and Technological Options Assessment 2000: European Policy on Food Safety, Luxembourg.
Federal Institute for Risk Assessment 2009: EU Food Safety Almanach, Berlin: Federal Institute for Risk Assessment.
Ferencz, Zoltán/Hajdu, Mariann/Vári, Anna 2006: "Food Safety Regulation in Hungary"; In: Vos, Ellen/Wendler, Frank (eds.): Food Safety Regulation in Europe. A comparative Analysis, Antwerpen et al.: Intersentia, 383‐451.
Ferreira, Celio/Boholm, Åsa/Borkhart, Karin 2006: "Food Safety Regulation in Sweden"; In: Vos, Ellen/Wendler, Frank (eds.): Food Safety Regulation in Europe. A comparative Analysis, Antwerpen et al.: Intersentia, 331‐382.
Fischer, Robert 2007: "Unabhängige Bewertungsagenturen im gesundheitlichen Verbraucherschutz: Ein Beispiel für gelungenen Politiktransfer in der Europäischen Union?"; In: Holzinger, Kathatrina/Jörgens, Helge/Knill, Christoph (eds.): Transfer, Diffusion und Konvergenz von Politiken, PVS‐Sonderheft 38, Wiesbaden: VS Verlag für Sozialwissenschaften, 348‐374.
Fleischer, Julia 2007: "Die europäischen Agenturen als Diener vieler Herren? Zur Steuerung und Rolle von EU Agenturen"; In: Jann, Werner/Döhler, Marian (eds.): Agencies in Europe, Wiesbaden: VS Verlag für Sozialwissenschaften, 212‐252.
Franchino, Fabio 2000: "The Commission‘s Executive Discretion, Information and Comitology"; In: Journal of Theoretical Politics, 12 (2), 155‐181.
Gilardi, Fabrizio 2002: "Policy credibility and delegation to independent regulatory agencies: a comparative empirical analysis"; In: Journal of European Public Policy, 9 (6), 873‐893.
Gilardi, Fabrizio 2005: "The Formal Independence of Regulators: A Comparison of 17 Countries and 7 Sectors"; In: Swiss Political Science Review, 11 (4), 139‐147.
Gilardi, Fabrizio 2008: Delegation in the regulatory state. Independent regulatory agencies in Western Europe, Cheltenham u.a.: Edward Elgar.
Große Hüttmann, Martin/Knodt, Michèle 2003: "Gelegentlich die Notbremse ziehen …: Die deutschen Länder als politische Teilhaber und Ideengeber im europäischen Mehrebenensystem"; In: Österreichische Zeitschrift für Politikwissenschaft, 32 (3), 285‐302.
Hellebø, Lise 2004a: "Food Safety at Stake ‐ The Establishment of Food Agencies", Stein Rokkan Centre for Social Studies, Working Paper 14.
Hellebø, Lise 2004b: "Independent Food Agencies ‐ Restoring Confidence"; In: Policy and Society, 24 (4), 125‐148.
Héritier, Adrienne 2001: "Differential Europe: National Administrative Responses to Community Policy"; In: Cowles, Maria Green/Carporaso, James A./Risse, Thomas (eds.): Transforming Europe : Europeanization and domestic change, Ithaca/New York: Cornell University Press, 44‐59.
Héritier, Adrienne 2005: "Europeanization research East and West: A comparative assessment"; In: Schimmelfennig, Frank/Sedelmeier, Ulrich (eds.): The Europeanization of Central and Eastern Europe, Ithaca, London: Cornell University Press, 199–209.
Hood, Christopher/Rothstein, Henry/Baldwin, Robert 2001: The Government of Risk: Understanding Risk Regulation Regimes, Oxford: Oxford University Press.
Janning, Frank 2008: "Risk Regulation without Political Conflicts? Regime Structures in Food Safety Politics in Germany, Great Britain and the Netherlands"; In: German Policy Studies, 4 (1), 67‐106.
Jasanoff, Sheila 1997: "Civilization and madness: The great BSE scare of 1996"; In: Public Understanding of Science, 6 (4), 221‐232.
Jordana, Jacint/Levi‐Faur, David 2004: "The politics of regulation in the age of governance"; In: Jordana, Jacint/Levi‐Faur, David (eds.): The politics of Regulation. Institutions and Regulatory Reforms for the Age of Governance, Cheltenham [u.a.]: Edward Elgar, 1‐28.
Knill, Christoph/Tosun, Jale/Bauer, Michael W. 2009: "Neglected Faces of Europeanization: The Differential Impact of the EU on the Dismantling and Expansion of Domestic Policies"; In: Public Administration, 87 (3), 519‐537.
Kobusch, Alexander 2010: "Linking the levels: independence and influence in the European two tired network of food safety agencies", Paper presented at the Third Biennial Conference “Regulation in the Age of Crises” of the ECPR Standing Group on Regulatory Governance, Dublin, June 17‐19, 2010.
45
Ladrech, Robert 1994: "Europeanization of Domestic Politics and Institutions: The Case of France"; In: Journal of Common Market Studies, 32 (1), 69‐88.
Lenschow, Andrea 2005: "Europeanisation of Public Policy"; In: Richardson, Jeremy (ed.): European Union: Power and Policy‐making, London: Routledge, 55‐71.
Levi‐Faur, David 2010: "Regulation and Regulatory Governance", Jerusalem Papers in Regulation & Governance 1, The Hebrew University.
Maggetti, Martino 2007: "De facto independence after delegation: A fuzzy‐set analysis"; In: Regulation & Governance, 1 (4), 271‐294.
Majone, Giandomenico 1996: Regulating Europe, London u.a.: Routledge. Majone, Giandomenico 2001a: "Nonmajoritarian Institutions and the Limits of Democratic Governance: A
Political Transaction‐Cost Approach"; In: Journal of Institutional and Theoretical Economics, 157 (1), 57‐78.
Majone, Giandomenico 2001b: "Two Logics of Delegation. Agency and Fiduciary Relations in EU Governance"; In: European Union Politics, 2 (1), 103‐122.
Majone, Giandomenico/Surdej, Aleksander 2006: "Regulatory Agencies in Economic Governance. The Polish Case in a Comparative Perspective", KICES Working Paper 5, Koszalin Institute of Comparative European Studies.
Millstone, Erik 2007: "Can food safety policy‐making be both scientifically and democratically legitimated? If so, how?"; In: Journal of Agricultural and Environmental Ethics, 20 (5), 483‐508.
North, Douglas C. 1990: "A Transaction Cost Theory of Politics"; In: Journal of Theoretical Politics, 2 (4), 355‐367.
Olsen, Johan P. 2009: "Change and continuity: an institutional approach to institutions of democratic government"; In: European Political Science Review, 1 (1), 3‐32.
Pierson, P 1996: "The path to European integration: a historical institutionalist analysis"; In: Comparative political studies, 29 (2), 123‐163.
Pollack, Mark A. 1997: "Delegation, agency, and aenda setting in the European Community"; In: International Organization, 51 (1), 99‐134.
Pollack, Mark A. 2002: "Learning from the Americanists (Again): Theory and Method in the Study of Delegation"; In: West European Politics, 25 (1), 200‐219.
Pollack, Mark A. 2003: The engines of European integration. Delegation, agency, and agenda setting in the EU, Oxford u.a.: Oxford University Press.
Radaelli, Claudio M./Exadaktylos, Theofanis 2010: "New Directions in Europeanization Research"; In: Egan, Michelle P./Nugent, Neill/Paterson, William E. (eds.): Research agendas in EU studies : stalking the elephant, New York: Palgrave Macmillan, 189‐215.
Regulation (EC) 178/2002 2002: Regulation (EC) No 178/2002 of the European Parliament and of the Council of 28 January 2002 laying down the general principles and requirements of food law, establishing the European Food SafetyAuthorityand laying down procedures in matters of food safety; In: Official Journal of the European Communities L 31, 1‐24, 01.02.2002, Brussels.
Schmidt, Vivien A. 1999: "The EU and Its Member‐States: Institutional Contrasts and Their Consequences", MPIfG Working Paper 99/7.
Scott, Colin 2006: "Privatization and Regulatory Regimes"; In: Moran, Michael/Rein, Martin/Goodin, Robert E. (eds.): The Oxford Handbook of Public Policy, Oxford: Oxford University Press, 651‐668.
Surdej, Aleksander no date: "Enlarging the EU Food Safety Regime: Adjustments of Polish Food Safety Regulations to the Requirement of the EU Membership". Thatcher, Mark/Stone Sweet Alec 2002: "Theory and Practice of Delegation to Non‐Majoritarian Institutions";
In: West European Politics, 25 (1), 1‐22. Ugland, Trygve/Veggeland, Frode 2006: "Experiments in Food Safety Policy Integration in the European Union";
In: Journal of Common Market Studies, 44 (3), 607‐624. van Zwanenberg, Patrick/Millstone, Erik 2003: "BSE: A Paradigm of Policy Failure"; In: The Political Quarterly, 74
(1), 27‐37. Verhoest, Koen/Peters, B. Guy/Bouckaert, Geert/Verschuere, Bram 2004: "The study of organisational
autonomy: a conceptual review"; In: Public Administration and Development, 24 (2), 101‐118. Vos, Ellen 2003: "Agencies and the European Union"; In: Verhey, Luc/Zwart, Tom (eds.): Agencies in European
and Comparative Perspective, Antwerpen u.a.: Intersentia, 113‐147. Vos, Ellen/Wendler, Frank (eds.) 2006: Food Safety Regulation in Europe. A comparative Analysis, Antwerpen
u.a.: Intersentia.
46
Wallace, Helen 2005: "An Institutional Anatomy and Five Policy Modes"; In: Wallace, Helen/Wallace, William/Pollack, Mark A. (eds.): Policy‐making in the European Union, Oxford u.a.: Oxford University Press, 49‐90.
Weingart, Peter 1999: "Scientific expertise and political accountability: paradoxes of science in politics"; In: Science and Public Policy, 26 (3), 151‐161.
Wonka, Arndt/Rittberger, Berthold 2009: "How independent are EU agencies?", RECON Online Working Paper 12, University of Oslo.
Yesilkagit, Kutsal/Christensen, Jorgen G. 2010: "Institutional Design and Formal Autonomy: Political versus Historical and Cultural Explanations"; In: Journal of Public Administration Research and Theory, 20 (1), 53‐74.