Lise Hellebø
Stein Rokkan Centre for Social Science Research
University of Bergen
The creation of autonomous regulatory agencies has been a trend in public policy, and is discussed and analyzed as a major governance trend in recent literature. In many countries agencies have been a prominent feature of the administrative landscape. Within the field of food policy, a process of agencification across Europe can be observed in the wake of the 1990s BSE scandal, although food agencies also have existed before the BSE incidence. The paper discusses the establishment of two specific agencies, the Food Standards Agency in Great
Britain and the European Food Safety Authority in the EU. Legitimate arguments in the establishment of the agencies were tied to independence, autonomy and credibility – and a desire to separate risk management and risk assessment, as well as responsibility for policy. Official documents published in the process of establishing the two agencies are analyzed in order to examine central discussions, conflicts and normative choices. Specific characteristics of the two contexts influenced how the agencies were formed. At the same time, characteristics of food policy and its formation through history are important explanatories for different arrangements in the food policy field. The question of independence emerges as a central dilemma, and is a discussion that might inform the general literature about regulation and governance through agencies.
Paper presented at the Scancor/SOG Workshop
Autonomization of the state: From integrated administrative models to single purpose organizations
April 1 & 2 2005
Work in progress, please do not cite without permission

Recent series of food scares across Europe, involving BSE
1
, known as the mad cow disease, salmonella in eggs and chicken, dioxins, and E. coli, to mention a few, have put food safety on the policy agenda. These food scares have spurred public awareness and media attention, and have been linked to an alleged loss of confidence in food industry, farming practices, government food policy and arrangements for handling food safety. The BSE scandal also had a major economic impact, especially in Great Britain. Changes in legislation and in government institutions across Europe have followed. One trend in the wake of these events is the establishment of new food agencies, or the reorganization of existing ones.
Food agencies differ in terms structure, remit and powers, but have a common regulatory field of operation: to secure food safety. Food safety regulation has a long history, with specific characteristics, traditions, and values formed over time (Elvbakken 1997). As such, food safety policy can be seen and studied as an organizational field involving certain interacting key actors, professions and organizations. Powell and DiMaggio (1991) have developed theory and elaborated empirical findings on several such organizational fields. Agencies are observed as one type of actors in an organizational field, and are familiar actors in the field of food safety regulation. The origin of modern food safety regulation and control has been placed within the wider field of public health (Rosen 1958/1993). This is related to the central aim to secure health by preventing disease and promoting wholesome food. At the same time, food involves production, distribution and trade, large businesses and major market interests.
Agriculture has proved a powerful actor with a strong interest in dominating the field.
Elvbakken (1997) has shown how the Norwegian food control early on was defined as a public health issue, but then was gradually taken over and dominated by actors from the agriculture sector. This shift may have occurred in other countries and is relevant for understanding what has happened in many countries after BSE.
The creation of autonomous regulatory agencies is seen as a trend in public policy, and has been analyzed in recent literature (Majone 1996; Pollitt and Talbot 2004; Pollitt, Talbot,
Caulfield and Smullen 2004). The establishment of food agencies can be seen as a part of this development. The creation of independent regulatory agencies is also linked to reorganizations of the state, and to theories of New Public Management (Christensen and
Lægreid 2004, Hood 1991). The establishment of agencies can further be related to processes of deregulation and delegation. At the same time, the establishment of agencies can incite a process of re-regulation , which might increase the overall intensity, complexity and specialization of regulation (Hood et. al 1999, Majone 1996a). The idea of creating agencies has become somewhat 'fashionable' within the realms public management, and in central government reform (Talbot 2004). However, the idea is not new. Sweden and the United
States have had agencies for a long time. Neither are food agencies new. The US Food and
Drug Administration (US FDA) dates back to 1906, to take one example. According to Pollitt et. al (2004), most of the 'real' work of government is carried on through agencies. At the same time, few give clear definitions of what constitutes an agency. The works of Pollitt and
Talbot (2004) and Pollitt et. al (2004) are important exceptions.
A central differentiating factor when studying agencies can be related to different fields of regulation or organizational fields in the terms of Powell and DiMaggio (1991). What is
1 BSE: Bovine Spongiform Encephalopathy, a fatal disease of the brain in cattle that can affect humans and create a similar brain disease labelled vCJD – a new variant of Creutzfeld Jacobs Disease.
1

regulated is important for the understanding of how organizations within the field function and operate. This is a central argument in this paper. Food safety regulation involves regulation of individuals as well as the market. This means that the state regulates powerful interests, with potential and inherent conflicts. This is fundamentally different from when the state regulates its own activities, or activities that do not involve economic interests in such a large extent.
A central defining feature of agencies is that they are created at arm's length from the core of central ministries and departments of state. A general idea is that separation of policy and management issues produces accountability, trust and confidence in decision-making structures and government authorities. Independence is a key notion. This is based on an idea of separating management, administration and execution from policy-making (Rhodes 1996).
The general idea is that policy decision should rest with politically accountable institutions, and be executed by separate and neutral institutions. The question of "real" separatedness and actual neutrality is a key question, but a complicated one. The discussion in this paper will reveal that this is indeed true for food agencies.
Paper outline
The paper sets out to examine the establishment of two food agencies placed at different analytical levels – one national and one European. The aim is mainly empirical: To investigate the process of their establishment and different options and choices made in this process. What institutional choices were available, what were the conflicts, and which solutions were chosen? What were the arguments for the particular choice of structure, and what were the restricting factors? Eventually, the answers to these research questions might be theoretically informative, and can inform the general literature about regulation and governance through agencies.
In focus are The Food Standards Agency (FSA) in Great Britain and The European Food
Safety Authority (EFSA) in the European Union (EU). First, the two agencies are selected because they both were created in the wake of several serious European food scares, most importantly BSE. Second, they are new organizations, but differently structured, and to some extent assemble earlier structures and personnel. This means that similar discussion themes and conflicts at the time of their establishment can be identified, though different solutions surface. FSA integrates risk management, risk analysis and risk communication. EFSA separates risk analysis from risk management and risk communication. FSA has an advisory role, but also regulatory and executive functions. EFSA does not have regulatory functions, but advises the Commission on scientific and legal matters. What are the arguments to support these different solutions, and how are they manifested in terms of organization? Lastly, the two agencies operate at different levels – one is national, the other is European. This clearly gives a different institutional background for the two agencies, and therefore differing sets of restrictions and possibilities in the process of their creation.
The empirical findings are mainly based on qualitative analysis of relevant policy documents, document analysis . Four interviews with five informants in FSA, the Department of Health
(DoH), the Ministry of Agriculture, Fisheries and Food (MAFF) and the EU Commission, carried out in 1999 and 2000, have further informed the analysis.
2
The documents can be characterized as official and institutional : Public, government papers and legal texts. Only
2 Data has been collected in a research project funded by the Norwegian Research Council ("Offentlig håndtering av risiko ved mat – en
sammenliknende studie av bruk av virkemidler"). The study compares food control and food policy in Germany, Sweden, the
Netherlands, and the United Kingdom.
2

official documents directly relevant to the establishment of the agencies are analyzed here.
We have looked for expressions of the agencies' central aims, developed in order to support their legitimacy and authority. What are the agencies supposed to achieve, and how? What are the agencies' proposed remit, the tasks they are supposed to take on and resolve? Which sets of problems are defined as relevant for their operation, and which goals are presented as central? These particular questions can say something about how food safety regulation is understood in the two different contexts and how food safety agencies operate.
The theoretical basis: Risk, regulation and institutions
Food safety regulation concerns the regulation of risk, and can be related to theorizing about risk in society as well as regulation. As well as living in a 'risk society', according to Beck
(1992), most of us live in a 'regulatory state' according to Majone (1996a). In Majone's assessment, the rise of the regulatory state means that that government's role as a regulator advances. Economic and social regulation is becoming "the new frontier of public policy and administration in Europe, at both the national and EU levels" (Majone 1996a, introduction).
Whether or not the idea of a risk society is a true assessment of our modern society is much debated. However, attention concerning risk, risk management, analysis and communication is noticeable across several fields. Risk and safety is often held to be major drivers of contemporary regulationary growth (Hood, Rothstein and Baldwin 2001). Risk regulation, hereunder food safety regulation, can be seen to contribute to the re-regulation in public management that Majone observes. This is contrary to views claiming that deregulation is a dominant feature of modern policy and governance.
Food safety regulation involves particularities of risk regulation. Defining and assessing risks is no simple matter (Hood et. al 2001; Baldwin and Cave 1999). There is a tension between actual or factual risks and perceived risks, between technical, rational or expert approaches to risk regulation, and the perceptions of the public and thereby implications for policy. Public perceptions do not always correspond to figures based on products of probability and magnitude of harm. BSE is a relevant example – the actual risk in terms of probability of being affected is much lower than the perceived danger, the widespread loss of confidence, the consequences for beef consumption and media attention speaks of. Differences in public and expert perceptions of risk can be seen as a key element in the alleged diminishing public trust in regulatory institutions – among them structures for handling food safety.
Pollitt (2003) has argued that it is important to look to specific agencies in order to understand agencies in more general terms. By focussing two food agencies, this paper aims to do just that. With the establishment of food agencies, certain traditions, professions and experts, are placed in a particular type of organization, at arms' length from central government. Joerges and several of his co-authors criticize the idea that regulatory policy, in particular risk regulation, can be seen as a mere technocratic concern that can be delegated to experts in independent agencies (Joerges et al. 1997; Joerges and Neyer 1997; Joerges and Everson
2000). A central argument is that judgments on the acceptability of risk must take into consideration normative, political and ethical factors, not just expert assessment. This militates against the creation of separate bodies of experts responsible for judgement about risk, such as some food agencies. Vos follows up on this, and pointedly remarks that "…risk assessment and management are so complex, uncertain, and controversial that they cannot be conducted without reference to normative social values" (Vos 2000a:230). These arguments apply to risk assessment in food safety. Here, scientific uncertainty or controversy repeatedly complicates risk management. Consequently, risk regulation and food safety regulation incorporates important problems, which make food agencies special. This is important to bear
3

in mind when discussing food agencies, and when placing them within the general agency literature. What the agency does, what responsibilities it has, is important to understand in order to assess its role and functions. The subject of regulation is not of inconsequence.
Our work has an institutional outlook. Thus, food safety regulation and food control is seen as an institutionalized policy field, involving certain characteristics, based in distinctive institutions, comprising traditions and values formed over time (Elvbakken 1997). Food control has developed as control of the marketplace by city authorities, to becoming a central state responsibility. Food regulation has changed with changing perceptions of regulation and state involvement, particularly within public health. It has also changed with the development of scientific knowledge. Through time, food control has concentrated on inspection by touch and smell, to focus on adulteration by chemical analysis and infective agents and disease, with the development of bacteriology. Following Elvbakken (Ibid.), the task of securing food safety can be seen as a particular field of regulation. Food safety regulation can therefore be studied within a framework emphasizing general characteristics of public regulation. Majone
(2000) agrees to the notion that regulation requires detailed knowledge of the regulated activity. This has led to the creation of specialized agencies, capable not only of rule making, but also of fact-finding and enforcement. Food agencies are such agencies.
In her study of the institutionalization of the Norwegian food control, Elvbakken (1997:8) argues the usefulness of Selznick's definition of regulation as:
“…sustained and focused control exercised by a public agency over activities that are socially valued”
(Selznick
1985:363-364). This definition is also referred by Majone in his analysis of regulation in
Europe (Majone 1996a:9). Consequently, food safety is seen as an essential and socially valued task, sustained by a public body. Selznick's understanding of regulation focuses the regulating institution or agency, and emphasizes the significance of embedded social values.
With this, the particular values embedded in food safety regulation, manifested in institutional arrangements and organizations, in different contexts and times, become central. The definition further implies, according to Majone (Ibid.), that regulation is not achieved simply by passing a law, but requires detailed knowledge of, and intimate involvement with, the regulated activity. By tracing values in institutions, particular regulated activities such as food safety can be analyzed and examined.
An institutional outlook spurs attention to the organization of political life and its importance, to the structural features of the society or polity (March and Olsen 1989; Peters 1999). An institutional starting point further leads to considerations of institutional change . If institutions are so important in shaping policy, how do they change? Why does the institutional landscape of food safety regulation change, and how do we explain the establishment of agencies?
Crisis , political, institutional and/or economical, is frequently referred as an important explanatory for institutional and political change. Such changing moments have been understood in various terms, as punctuated equilibrium (Krasner 1984) or critical junctures
(Thelen 1999). Thelen and Steinmo (1992:16) draw attention to what they call institutional dynamism : When changes in context produce situations where old institutions change and new institutions might arise. BSE and other food scares can be seen to represent such events or crises, and might have spurred institutional dynamism in the field of food safety regulation.
This is discussed in the next section of the paper.
We have established that food safety regulation concerns crucial state assignments and arrangements, and that it is a valued task upheld by central public bodies. It also involves big
4

business and large economic interests. Public scandals related to food safety are therefore significant, and have potential for changing public arrangements and regulation within the field. The public scandal related to BSE is important in explaining a reorganization of food safety regulation in Europe since the 1990s. Since the BSE scandal erupted in 1996, new food agencies have been created in several European countries, and existing ones have been restructured.
3
BSE is frequently referred to as important in explaining the establishment of both FSA end EFSA.
4 Recurrent references to BSE in relevant public documents, in newspaper reports and academic literature (Grant 1997; Bovens et. al 1998; Baggot 1998;
Joerges and Everson 2000; Majone 2000; Vos 2000a and 2000b; Boin and t'Hart 2003;
Millstone and van Zwanenberg 2002 and 2003, to name some) demonstrate that the BSE scandal is an important point of reference in public policy analysis.
BSE is primarily judged a crisis because of its large impact. It concerned not only public health and safety, but also involved important trade and commercial issues. It had a large economic impact, affecting domestic production and sale as well as the international beef market. Agriculture and production methods became major concerns, beef consumption dropped, and Member State cooperation and progress in the EU was seriously hampered. BSE also involved definition and use of scientific expertise and judgement, and challenged established views of expert knowledge. However, BSE became a significant crisis not just because of its economic significance, or because of the frightening nature of the disease itself and the fact that it could spread to humans.
5
It became a crisis of significant proportion because of the way it was handled by governing institutions.
The disease in cattle was first discovered in Great Britain in 1986. Through the 1980s and
1990s it developed into an epidemic, spreading to more and more herds and farms.
6
The rapid spread was linked to the practice of using rendered cattle in feed. Critics blamed the excessive industrialization of farming and feeding practices. Initially, the British government repeatedly denied a possible link between BSE and the similar human brain disease vCJD. However, in
March 1996, they changed their position and acknowledged that there indeed was a possible link after all. This prompted massive media attention and criticism, both nationally and in
Europe.
7
The notion was that Government had kept secret that BSE could affect people in order to protect the beef industry. An alleged decline in public trust in government followed
(Baggot 1998, Berg 2000, Kjærnes 1999). In the years following 1996, several official reports, papers and legal texts on the organization and principles of food safety and regulation were published. In 1998, a public Inquiry was set up to review the British governmental actions in response to the BSE prior to 1996 (BSE Inquiry Report 2000). The 'lessons to be learned' were apparently plentiful (Ibid: chapter 14, vol. 1). A similar report was published by the European Parliament, after criticism of the European Commission on its handling of the
BSE case (EP BSE Inquiry Report 1997). Grant refers to BSE in Britain as
“…a clear example of a policy disaster”
, partly caused by tensions between the Department of Health
(DoH) and the Ministry of Agriculture, Fisheries and Food (MAFF) (Grant 1997:346). This
3 This has happened in Austria, Belgium, Denmark, France, Germany, Great Britain, Ireland, the Netherlands, Norway and Spain,
4 to name some.
For instance: EFSA website, About EFSA: http://www.efsa.eu.int/about_efsa/catindex_en.html
, and FSA Annual Report
5
(FSA 2002).
The disease affects the brain, ultimately creating a sponge-like texture (hence the name "spongiform"). The victim gradually loses
6 all control over its bodily functions, and dies. BSE is also known as "mad cow disease".
By the time that BSE was identified as a disease as many as 50,000 cattle are likely to have been infected (BSE Inquiry 2000, vol.
7
1, chapter 1).
Evidence can be found in several special media reports on BSE, for example: BBC News: "BSE and CJD Crisis Chronology"
( www.news.bbc.co.uk/hi/english/static/in_depth/health/2000/bse/default.stm
) The Guardian: "Special report BSE Crisis"
( www.guardian.co.uk/bse/0,8250,388290,00.html
) and CNN: "Mad Cow In-depth Special"
( http://www.cnn.com/SPECIALS/2000/madcow/ )
5

illustrates that diverging interests were involved. Tension between health and agriculture is a characteristic feature of food safety regulation. Baggot (1998) has pointed out that an important factor contributing to the crisis was that BSE initially was defined as a problem of agriculture rather than as one of health. This defined the problem in a certain way and structured the solutions sought. Over time, the problem was gradually defined more as a question of public health and safety. As the health departments in different countries and in the EU became more involved, the nature of solutions shifted.
Judging from newspaper and media coverage, other food scares such as dioxin in Belgian poultry (in 1999) and food poisoning through salmonella, listeria and E. coli also have had an important impact on what can be seen as an overall confidence crisis in food control in the
1990s. The evolvement of the BSE scandal can be seen as a parallel case to that of salmonella in eggs documented by Smith (1991). Here, food policy issues and arrangements became politicized after a long period of depolitisation, because of the way central government handled known risks related to salmonella in eggs. Because of its international impact, BSE had further implications for British EU relations, and for co-operation within the European community. Both the British BSE Inquiry report (2000) and the corresponding European
Parliament BSE Inquiry (EP BSE Inquiry 1997) witness this. Several academic analyses agree. Vos (2000a:227) reports that the spread of the disease into Europe resulted in widespread public distrust in how the Commission functioned, which in turn prompted drastic reform (Ibid. 228). Majone argues that the BSE crisis not only revealed a failure to establish a credible community of scientific experts on food safety, but also exposed serious shortcomings in the overall co-ordination of European policies on agriculture, the internal market and human health (Majone 2000:282). In contrast, Forbes (2004) challenges the notion that the 'episode' of BSE can be characterized as a crisis, in terms of policy failure and policy disaster. He questions the belief that BSE actually did reveal fundamental problems with the political and administrative system in the UK and that trust in government was damaged by the failure to communicate fully the potential risks to the public of eating beef.
Whether or not BSE, and the way it was handled, rightfully can be labelled a crisis or merely an 'episode' or a 'scandal' is not important in our view. The point argued here is that, crisis or not, BSE had important implications for policymaking both at national and at EU level. It was in practice treated as a crisis when it came to legitimizing subsequent changes. This is evident in both policy documents and statements from officials since 1996. Still, other food safety issues have also been important. Our conclusion is that the food scares of the 1980s and
1990s, BSE most importantly, have clearly influenced and prompted policy reform, bureaucratic and institutional change across Europe.
According to Pierson (1996), change is a response to unanswered or dysfunctional elements in the original institutional design that generate a subsequent need for change. Change, or incremental adjustment, can be seen as a means of institutional adjustment to changing demands and inadequacies in the original design (Peters 1999:70). Fligstein and Mara Drita
(1996:3) concur that institution-building moments occur when a social, economic or political crisis undermines the current institutional arrangements. In food policy and regulation, the
'dysfunction' in Pierson's terms, can be seen to lie in the dilemma of balancing differing and often conflicting interests, such as consumer interests, public health interests and market or commercial interests. When these interests clash – as in the case of BSE – it may lead to institutional and organizational change.
6

In the following empirical analysis, we have looked at what is stated as the most important goals and aims for the new organizations. What do the documents reveal of the rhetoric and values operating behind the organizational choices that were made? What do the agencies set out to do, and how is this legitimized? These questions will presumably inform the main research questions stated introductorily. In each case, we have analyzed several preparatory documents leading up to the legislation establishing the agencies. Each section concludes with a presentation of the resulting legislation. The empirical sections are followed by a discussion of the findings relevant to the presented literature on agencies.
Table 1: Official documents establishing FSA and EFSA.
FSA
The James Report on the Food Standards Agency
(James 1997)
Government White Paper: "The Food Standards
Agency – a Force for Change" (White Paper
1998)
Food Standards Act 1999
EFSA
Speech to European Parliament by Jaques Santer
(Speech/97/39)
COM (97) 176 Commission Green Paper on Food Law
COM (99) 719 Commission White Paper on Food Safety
Regulation (EC) No 178/2002 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
Labour leader Tony Blair ordered a report on the establishment of a British food agency in
1996. Simultaneously, Conservatives in office decided to establish a Food Safety Agency.
The 'James report' on the Food Standards Agency was published in 1997, after "a broad round of consultation with public and private interests" (James 1997).
8 When Labour won the following elections, the James report became the groundwork for the new agency.
The James Report
According to the James report, British food policy was faced with a serious loss of confidence
(James 1997:10):
There is a serious loss of confidence in British food which needs to be rectified urgently
The report refers to several public surveys that evidently showed concern in virtually all areas of food safety; the microbiological and chemical safety of food, genetically modified organisms, novel foods and processes and the nutritional quality of the diet (Ibid.). The report notifies that it has been formed through a wide consultation process involving consumer organizations, other public interest groups, public health specialists, members of scientific committees and government agencies. A list of people consulted is included in an appendix to the report.
8 Professor Phillip James, at the time head of the Rowett Research Institute in Aberdeen was appointed to lead the work. He is a prominent figure in British food policy, having advised governments on several issues concerning food safety since the early
1980s (see for instance Cannon 1987:18, 35-36).
7

According to the James report, the 'problem' in British food safety regulation was related to a culture of administrative secrecy, conflicts between public and vested interests, and a lack of coordination in existing food control mechanisms (Ibid:5). Before BSE, responsibility for food safety was divided between the Ministry of Fisheries and Food (MAFF) and the
Department of Health (DoH). MAFF largely took care of food standards (quality and composition), chemical safety, labelling, and food technology, including additives, contaminants and the hygiene aspects of dairy and meat production. DoH had responsibility for health aspects of food, including food hygiene, microbiological food safety and nutrition
(Ibid:11-13). Conflicts of interest, in particular within MAFF, were emphasized as a reason behind the loss of confidence (Ibid:14). According to James, it was at the time, almost universally accepted that all aspects of food safety (policy-making, surveillance, control and audit) should be separated from MAFF (Ibid:14):
What is needed, is a new body which separates the role of protecting public health and safety from that of promoting business. An agency must operate openly so that decisionmaking becomes more transparent and the true balance of interests is subject to public scrutiny.
The report tells that there had been a debate about the establishment of an independent food agency in the UK for some time (Ibid:15). This is related to an alleged widespread distrust of government, science and business, and a need for the public to have faith in systems that are intended to protect them. Building public confidence is a key notion. The introduction of
(Ibid.):
…a high quality authority which manifestly takes public health and consumer protection as its first priority and has investigative and executive powers to rectify problems is a fundamental component of the series of developments needed to build public confidence.
The Agency proposed in the James report has a relatively wide remit. It suggests that the
Agency should advise ministers on food safety and regulation, and coordinate research in food safety, nutrition and consumer protection (Ibid:24). The advantage of including nutrition policy is emphasized (Ibid:10). The report also emphasizes the importance of including consumer interests (Ibid.). Democratic accountability and widespread credibility is important.
The James report advises that the new Agency should be governed by a multi-member board with a majority nominated by public interest groups (Ibid:16 and 18). Its legitimacy rests on consumer and public interest representation (Ibid:31):
The time has come to put consumer and public interest representation onto a firmer footing.
Strictly speaking, the Agency proposed would have an advisory role to Government (Ibid:19).
However, the James report proposes that the Agency should be established as an "Executive
Non-Departmental Public Body (...) with clear statutory powers" (Ibid:18). This means that it should be a separate unit with legal personality and its own finances, and have regulatory powers. It is further emphasized that the Agency should have assigned to it "all current government responsibilities for drafting legislation, developing policy and issuing guidance"
(Ibid:19).
9
It would further take over responsibilities previously organized under other government units (Ibid:20-27). It should be responsible for meat control by incorporating the
Meat Hygiene Services (MHS), and incorporate parts of the State Veterinary Service to coordinate surveillance of animal aspects of food safety. It should also have responsibilities for the safety evaluation of pesticides, transferred from the existing Pesticides Safety
9 This would include the microbiological safety of food, genetically modified foods, novel foods and processes, the chemical safety of food, the nutritional quality of diets and effect on public health, food standards and food labelling, aspects of animal health which may affect food, public education and information on food matters (James 1997:19).
8

Directorate, and assume responsibility for evaluating the public safety of veterinary medicines
(at the time upheld by the Veterinary Medicine Directorate). The Agency would also incorporate a series of advisory committees and coordinate research activities. Local food law enforcement tasks were defined as irrelevant for incorporation, because of the strong tradition of locally based food control. The Agency would nevertheless be responsible for coordinating, monitoring and auditing local food law enforcement activities (Ibid:25). In sum, the Agency would not be established with the aim to separate policymaking from executive functions in a strict sense, neither with an aim to separate risk assessment from risk management. The Food Standards Agency proposed by James would incorporate these aspects rather than separate them.
The James report was clear on the mechanisms of accountability envisaged for the new
Agency. It admits that "Which Minister the Agency should be accountable to is an issue of great debate" , but then states that the Agency should report through Health Ministers with the
Secretary of State for Health taking a lead (Ibid:27). The Agency should report through
Health Ministers in terms of funding as well as operational issues, and the funding should come through the DoH budget (Ibid:28). Consequently, food policy and food regulation would be firmly established as a public health concern, separated from agriculture departments and central government. Independence from agricultural and business interests and central government was seen as the key to restoring public confidence.
The White Paper: A force for change
A government White Paper on the establishment of FSA was issued in January 1998. It supported the main conclusions of the James report (White Paper 1998:2), and Prime Minister
Blair stated (Ibid:introduction):
This government took office committed to setting up an independent Food Standards Agency, which would be powerful, open and dedicated to the interests of consumers. Professor Philip James' report, which I received on 8 May, provided an excellent foundation on which to build this long-overdue reform.
The paper has a common introductory statement signed by the UK agriculture and health ministers. It seems that this is to emphasize unity between the different interests, which the
James report had seen as adversary when it reported on the struggle for influence in food safety regulation.
The Agency would have protection of the public as its essential aim (Ibid:1). The importance of independence was emphasized early on. The Agency would act "…at arm’s length from
Government and independent of specific sectoral interests" (Ibid). As to why the Agency is needed, the White Paper says that it would "put an end to the climate of confusion and suspicion" that was related to the way food safety issues had been handled in the past, in particular (Ibid:2):
the potential for conflicts of interest within MAFF arising from its dual responsibility for protecting public health and for sponsoring the agriculture and food industries
fragmentation and lack of coordination between the various government bodies involved in food safety
uneven enforcement of food law
The diagnosis was, in other words, similar to the arguments in the James Report. The White
Paper further recognized that organizational change was not enough; cultural change would have to follow (Ibid:3).
9

The 'Guiding Principles' of the Food Standards Agency demonstrates central motivations behind the establishment of FSA (Ibid:5). The essential aim is "…the protection of public health in relation to food" (Ibid.). Further important aims are scientific excellence (guiding principle no. 2), independence of sectoral interests (guiding principle no 3), transparency and openness (guiding principle no 5), and efficiency (guiding principle no 9). The obvious motivation lying under these principles is an interest in rebuilding public trust in government arrangements for handling food issues. This is also emphasized directly throughout the document (Ibid:4):
The Government's aim in setting up the Agency is to strengthen food safety and standards policies and procedures so as to rebuild the public's trust in the machinery for handling food issues.
It was clear that the proposed Agency would to have an advisory role (Ibid:1). However, the
White Paper did not distinguish clearly between risk assessment and risk management. It established that risk assessment needed to be based on the best available methodology, drawing on expert scientific advice (Ibid:6). The chance of scientific uncertainties and the need for political judgement was also recognized (Ibid):
Where there are uncertainties about the scientific evidence, an element of political judgement is inevitably involved in reaching decisions on the best course of action.
This particular concern can be directly linked to the alleged errors in the way the BSE scandal was handled. The White Paper also recognized a need to balance different interests. In this, independence would be somewhat restricted. The White Paper said, for example, that the
Agency would need to take account of the effects of its actions on those affected – industry, enforcement authorities and other interested parties as well as consumers (Ibid.). It also stated that the interests of the consumer would be served best, if the Agency and the food industry would work closely together to establish methods of achieving the standards the public interest requires. Independence is, in other words, conditional.
Chapter 3 of the White Paper summarizes what the Agency should do. In short, it should take over responsibility from the Agriculture and Health Departments for advising Ministers on food safety and standards, including important aspects of nutrition, it should have responsibilities for public information and education on food matters, and for representing the
UK in international organizations (Ibid:8). Responsibilities regarding nutrition and farming is however limited. Here, the White Paper states that (Ibid):
In some areas such as nutrition policy and food safety issues which relate to farming practices, Health,
Agriculture and Environmental Departments will retain important policy and statutory responsibilities.
This is an important distinction from the James Report.
10
The intention of having the Agency operating at arm's length from Government is also curtailed, whereas the White Paper states
"…it must not operate in isolation from those Departments and other bodies with
10 The White Paper later on remarks that the recommendation in the James Report that the Agency should have a large role in developing nutrition policy, provoked considerable comment and became a subject of vigorous debate (White Paper 1997:32-
33). The White Paper proposed a split of responsibilities as regards nutrition. The agency would be responsible for monitoring, surveillance, scientific advice and information about nutrient content of foods, advice on the diet as a whole. Health
Departments would retain responsibility for wider public health issues including conditions and consideration of vulnerable groups, health education and more general health surveillance.
10

responsibility for protection of the public." The White Paper further states that the Agency should be more than an advisory body (Ibid):
The Government believes that if the Agency is to achieve the objective of strengthening and opening up the Government machinery for handling food-related issues it must be given a wide ranging remit, and it must have executive as well as advisory powers.
So, the Agency proposed was to be a regulatory agency proper, and not merely an advisory body with limited powers.
The proposal further stated that the Agency should take a strategic view of food safety throughout the food chain . This is popularly known as an approach 'from plough to plate' or
'from farm to fork'. However, regulation on the farm is limited (Ibid:17):
The Government considers that to give the Agency operational responsibility for all aspects of farming practices would risk diverting it from its essential aim of protecting public health.
The James Report suggested the creation of an Agency Commission with a majority of public and consumer interests. The White Paper comments that this attracted particular interest
(Ibid:35). Actors from particular sectors wanted to see their own interests represented, while
'a number of respondents' emphasized the importance of appointing members on the basis of skills and experience rather than as representatives of particular groups (Ibid). The White
Paper concludes that the leading Commission of the Agency should consist of (Ibid:35-36):
…a body of individuals who have a proven track record in relevant fields who together provide a reasonable balance of relevant skills and experience, and a majority of whom come from a wider public interest background without any specific affiliation
With this argument, the independence of Commission Members is re-emphasized – contrary to the direct interest representation that the James Report had envisaged.
The White Paper goes on to specify the relationship between the Agency and its sponsor
Department, DoH. The Agency should report to Health Ministers and consult with Health
Departments over the production of its Annual Report and Business Plans. The Agency staff should be civil servants. The proposed Agency should, as proposed in the James Report, be free to make public its advice to Ministers. At the same time, Health Ministers should be given powers of direction over the Agency in cases where it appears to act outside its legal framework (Ibid:37). Therefore, the Agency should act independently, but in special situations, it would be dependent on DoH and central Government. Chapter 8 of the White
Paper proposes a financial framework for the Agency. The James Report recommended that the Agency should be funded from DoH funds. The White paper agrees to this, and comments that this would mean a transfer of resources from MAFF (Ibid:43). The White Paper suggests that in addition, the Agency should be financed from other sources – such as charges to the food industry (Ibid: 43 and 44-45). The argument is that this would shift the burden of cost away from the taxpayer. It would also make the Agency more independent of public finances and directions from the Ministry.
A paper summing up the responses to the White Paper was published on MAFF's website shortly after (FSA 2000). This paper remarked that the proposals in general received a broad welcome, and that the overall aims and core objectives were widely supported. There was also strong support for the Agency's proposed functions. The paper also says that there was
11

widespread opposition to the proposal of financing the Agency through a levy on food industry. In the end, this particular proposal was not followed through, and the Agency is wholly publicly financed.
Summing up, consumer confidence, independence and the responsibility of DoH continued to be central idioms in the White Paper. Food safety was defined as a problem of public health, and it was important to delimit against agricultural and business interests. The White Paper further emphasized an 'integrated approach' to questions of risk management and risk assessment. Scientific risk assessment and risk management were seen as inseparable, whereas the Agency would formulate policy, have regulatory tasks and integrate scientific expertise (Ibid:2). The documents nevertheless diverged on some points. The James report had envisaged strong agency involvement in nutrition policy. This was toned down in the
White Paper. Instead, a split of responsibilities between the Agency and DoH was suggested
(Ibid:32-33). The aspect of significant consumer and public interest involvement was also downplayed.
The Food Standards Act
After a widely publicized hearing, a bill establishing the Food Standards Agency was laid before Parliament and publicly distributed early in 1999. Despite disputes concerning funding and the Agency’s involvement in nutritional issues, the Food Standards Act was passed in
November 1999 without significant differences from the proposals in the bill. The main objective of the Agency is laid down in article 1 (2):
1. (2) The main objective of the Agency in carrying out its functions is to protect public health from risks which may arise in connection with the consumption of food (including risks caused by the way in which it is produced or supplied) and otherwise to protect the interests of consumers in relation to food.
The explanatory notes accompanying the Act specify that food safety is central to the main objective, but that Article 1 (2) also embraces the Agency's role in relation to nutrition and diet, and protecting the wider food-related interests of consumers – sometimes referred to as food standards. Article 1(3) establishes the Agency as a Crown body. This means it will be a non-Ministerial government department, as suggested in the White Paper
The 1999 Act warrants that the Secretary of State for Health shall appoint members of the
Agency's board (article 2).
11
Article 26 specifically directs that the functions of the Minister of Agriculture, Fisheries and Food related to food safety shall cease to be exercisable.
12
The explanatory notes remark that this means that the Minister of Agriculture, Fisheries and Food will cease to have any statutory functions in relation to most matters within the Agency's remit. It is thus clear that it was vital to create distance to MAFF in order to restore confidence.
The Agency's independence is assured through its powers to freely publish advice and other information as it sees fit (article 19). Further regulatory actions are secured through powers to issue general guidance to local authorities or other public authorities on control of food-borne diseases. The actual independence is nevertheless limited by the close relationship with DoH and the powers of the Health Ministers to instruct the Agency in special circumstances. The
11
12
One member shall be appointed by the National Assembly of Wales, two by the Scottish Ministers and one by the Department of Health and Social Services for Northern Ireland.
Specified as functions under Part I of the Food and Environment Protection Act 1985, the Food Safety Act 1990, and the
Radioactive Substances Act 1993.
12

Secretary of State for Health may give the Agency directions if the Agency has seriously failed to comply with the requirements of the act, or performed according to its duties (article
24). This includes powers to remove members of the Agency from office (article 24 (8)).
The Agency's general functions in relation to food, are expressed in several articles: developing policies relating to matters connected with food safety or other interests of consumers in relation to food, (…) providing advice, information of assistance in respect of such matters to any public authority (article 6) providing advice and information to the general public (or any section of the public) in respect of matters connected with food safety or other interests of consumers in relation to food" (article 7) obtaining, compiling and keeping under review information about matters connected with food safety and other interests of consumers in relation to food, including monitoring developments in science, technology and other fields of knowledge relating to food and carrying out, commissioning or coordinating research on those matter (article 8)
The act also states general function for the Agency in relation to animal feedingstuffs. It has powers to carry out observations in food or animal feed production (article 10), and shall monitor the performance of enforcement authorities (article 12, 13 and 14).
FSA was established in April 2000. It was ultimately designed with powers in its own right.
Not only would it deal with matters of risk assessment, it would also possess regulatory powers. The latter allows it to issue general codes of practice and audit local public health authorities responsible for food control. It was also made responsible for MHS, performing meat control. Furthermore, it was to provide information to the public and advise central authorities. Still, important regulatory powers remained with the Ministry (DoH). The responsibilities of MAFF were on the other hand minimized. After FSA was set up, MAFF experienced further break up and was replaced by DEFRA in 2001.
13
According to Flynn,
Marsden and Smith (2003), MAFF disappeared in part because it could not perform its task.
The changes in British food control seem to have its roots in the longstanding conflict between the central departments of health and agriculture.
14
The principle of seeing food control as a specific public health concern has a long tradition (Elvbakken 1997). This principle was accordingly re-emphasized when the British food policy field was restructured, and regulation of food safety was returned to its origins – to the central institutions responsible for managing public health.
The food sector is an area of regulation that dates back to the earliest days of the European community. The form, direction and goals of food regulation within EU have changed significantly over time. They started as provisions necessary to fulfill the harmonization goals of a common market, but over the years came to form a policy goal in its own right, which ultimately aims at ensuring safe food for European consumers (Hellebø 1999). Hellebø (199) and Jukes (1995) have described a parallel shift, from efforts to approximate national legislation (vertical regulation) to a more widespread use of horizontal framework regulation.
Within EU, the BSE scandal had a double impact. It hit the very foundations of the European community – that of trade between member states. Furthermore, it surfaced at the same time
13
14
DEFRA website: www.defra.gov.uk
For a more historical account in agreement with this view, see French and Phillips (2000) and Smith (1991).
13

as an EU summit was held to revise the Maastricht Treaty – the agreement that had formally established the Internal Market. In response to the recognition of the risks relating to the spread of the disease, an EU ban on the export of British beef was issued. The British government argued they had taken the necessary steps to stop the spread of BSE, and reacted by obstructing decision-making on all matters requiring unanimity at the EU summit.
According to reports in the periodical European Dialogue , this seriously hampered progress at the summit, thereby stimulating a debate on the principle of unanimous voting within the community (European Dialogue 1996:5). Thus, BSE had institutional repercussions even beyond agriculture, public health and trade.
Important changes in the EU were triggered by BSE, although in this case it took longer before they were embodied in an agency. As in Britain, EU institutions – in particular the
Commission – were criticized for their handling of BSE.
15
In 1997, Jaques Santer (president of the Commission) held a speech in which he answered some of the criticisms that were raised. He proposed the establishment of an independent agency, based on the '…positive aspects of the United States Food and Drugs Administration' (Santer 1997). The US FDA can be characterized as a regulatory agency, implying that it is responsible for enforcement and monitoring tasks, as well as educational tasks.
16
Regulatory agencies in the USA are generally more powerful than traditional EU agencies, and many national European agencies (Majone
1996a; Shapiro 1997). The realization of Santer's proposal would therefore imply a break with the existing EU agency approach, characterized by weaker, information-based agencies
(described by Majone 1997; Kreher 1997; Dehousse1997; Shapiro 1997). It would, on the other hand be similar in operation to the proposed British FSA.
Santer's proposal was not to be realized. Instead, a number of institutional reforms were agreed upon (COM (97) 183). Most importantly, responsibility for food safety was transferred within the Commission, from the Directorate Generals of Agriculture and Industry to a reformed Directorate General for Consumer Policy and Health (DG Sanco). Consequently, as in Britain, there was an urge to establish food safety matters as a public health concern – separated from commercial and agricultural interests. This illustrates the central and omnipresence of tensions between health and commercial interests within the field of food safety regulation in general.
EU Green Paper
A Commission Green Paper on The General Principles of Food Law in the European Union was laid before the European Parliament and the member states two months after Santer's speech in 1997 (COM (97) 176).
17 This was the first major policy document to review the totality of food regulation and control in the EU, and it anticipated the establishment of a general food law within the EU. Aspects including both regulation and control were discussed, and issues concerning proper risk assessment were essential. However, the establishment of a food agency was not mentioned specifically.
The stated aims of the Green Paper was to review existing food legislation, invite to public debate on these issues and to (Ibid:2-3):
15
16
17
Criticism was made public by a temporary European Parliament committee set up to report on “…the alleged contraventions or
maladministration in the implementation of Community law in relation to BSE” (EP 1997). A debate on BSE and the inquiry committee report was held in the European Parliament in February 1997.
The tasks and responsibilities of the US FDA are described at the agency's website: www.fda.gov
The Green Paper was presented by the European Commission, Directorate General for Industry. Before the reorganisation of
DG responsibilities for food regulation in 1997, DG Industry shared responsibility for food regulation with DG Agriculture
(see page 15).
14

…enable the Commission to propose measures allowing, wherever possible, to improve the protection of public health laid down in its measures for the internal market and the common agricultural policy, improve the coherence of Community food law, consolidate and simplify it, improve the operation of the internal market, and take into account the increasingly, important external dimension, notably the policies followed by our most advanced trading partners and the requirements of the WTO agreements'
The Green Paper specifically referred to BSE as an important background (Ibid:2):
In the light of certain recent events, in particular BSE, others have raised doubts about the capacity of the legislation to entirely fulfil its objectives to ensure a high level of protection of public health and consumer protection.
It further balanced different goals that can be seen as at least partly conflicting: It set out to ensure a high level of consumer protection, while at the same time securing the internal market and the competitiveness of European industry (Ibid:3). These potentially conflicting goals continue to influence discussions on food safety within the EU. The Green Paper also stressed the need for a regulatory approach that covered potential risk at all stages of the food chain (Ibid:4) similar to the proposals in Britain.
Although the Green Paper did not specifically suggest the establishment of a new food agency, it anticipated several changes (Ibid.):
a proposal for a general Directive on food law
consolidation and simplification of certain provisions, or reformulation of existing legislation
suggestions or proposals of a non-legislative nature, including changes in procedures or working methods
In sum, the Green paper was an important framework for the following development and the eventual establishment of EFSA.
The professors' report
In 1999, the Commission invited professors Pascal, Kemper and James to produce an
"independent report on the future of scientific advice in the EU" (James et al. 1999).
18
This relaunched the debate on the establishment of a food agency in the EU. The professors recommended the establishment of an independent EU Food and Health Authority. The proposed structure was, as in Santer's proposal, an independent authority analogous to the US
FDA, albeit "…more independent of political and industrial interests" (Ibid:7). Independence was again a central idiom. Moreover, as in the British case, the restoration of public confidence was a major concern (Ibid:13-15). The diagnosis was that (Ibid:15):
…the drawbacks of the present system are such that public confidence will not grow unless there are structural changes in the way these issues of intense public concern are handled
The failure of the EU was related to what was seen as an artificial separation of risk assessment, risk communication and risk management (Ibid:9 and 14). The advantage of incorporating the three aspects of risk analysis, as had been done in the British case (the
'integrated approach'), was emphasized. The proposed authority was to concentrate on tasks concerning scientific advice, the monitoring of public health, policy analysis, research policy development, communication, and crisis management (EU Commission 1999).
18 The three authors were all members of the Scientific Steering Committee, sorting under the Directorate general for Health and
Consumer Protection of the European Commission. Professor James is the same Professor responsible for the James Report in the British case.
15

As in the James report, the professors envisaged significant involvement of public interest groups. Their report also emphasized direct accountability to elected representatives in the
European Parliament (Ibid:19). The realization of the proposed Food and Health Authority would, as Buonanno et al. (2001) point out, break new ground in the development of the EU, and for the first time remove the control function for a social policy from the Commission and from the domain of the member states. This would mean a definite step towards further
European integration. The Agency would differ from the weaker, more information-based agencies established directly under Commission control. However, the realization of such an
Agency, with 'real' regulatory functions, would go against central clauses in the Treaty of
Rome underlying Community operation, in particular article 4 that enumerates the European institutions.
The EU White Paper
The Green Paper and the professors' report were followed by a White Paper on Food Safety in
January 2000, which proposed changes in the general principles of food law, and the establishment of a European Food Authority (COM (99) 719).
19 The White Paper did not say much about the actual structure of the Agency, but included a range of proposals concerning its precise remit. The Agency would be limited in scope compared to the Food Agency
President Santer had proposed. It would be entrusted with (Ibid:3):
…a number of key tasks embracing independent scientific advice on all aspects relating to food safety, operation of rapid alert systems, communication and dialogue with consumers on food safety and health issues as well as networking with national agencies and scientific bodies.
It would further provide the Commission with necessary analysis, but the Commission would be responsible to decide on the appropriate response to that analysis. More specifically, it would not be a regulatory agency in the sense of the US FDA or FSA. The White Paper proposal was more in tune with the prevailing EU agency model. Furthermore, it specifically rejected the idea that it should be responsible for risk management. This might result in a dilution of democratic accountability, breach with the designated responsibilities of the
Commission, and possibly violation of Treaty provisions regarding the institutional arrangements of the European Union (Ibid:15).
The White Paper states that the food policy of the EU should be built around high food safety standards "…which serve to protect, and promote, the health of the consumer" (Ibid:6). On the other hand, it recognizes that food regulation has other aspects (Ibid:6):
…The production and consumption of food is central to any society, and has economic, social and in many cases, environmental consequences. Although health protection must always take the priority, these issues must also be taken into account in the development of food policy.
The essential role of the Internal Market in providing consumers with a "wide range of safe and high quality products coming from all Member States" is also recognized (Ibid:6).
The White Paper embraces an 'integrated approach' to food regulation, but this has a different meaning compared to what we have called the integrated approach in the British case. In the
White Paper, an integrated approach refers to assessment, monitoring and regulation at all stages of the food chain – a 'stable to table' or 'farm to fork' approach (Ibid:6-8). Contrary to the British integrated approach, connoting integration of risk assessment and risk
19 The European Food Authority (EFA) was later renamed the European Food Safety Authority (EFSA).
16

management, the White Paper continually emphasizes that risk assessment and risk management should be separated. Risk management should reside with the Commission. This differs from the 'integrated approach' adopted in the British case, but agrees with existing EU institutional arrangements.
As in the British case, restoring consumer confidence is a central aim behind the suggestions of the White Paper (Ibid:7, 14). Again, as in the British case, independence, scientific excellence and transparency is vital in order to achieve this (Ibid:16). At the same time, the
White Paper recognizes that the Agency needs to work closely with national scientific agencies and institutions in charge of food safety (Ibid:16). It also states that the Commission and other EU institutions have a vital role to play in supporting the Agency. These elements might become detrimental to the proposed Agency's independence.
Although the White paper consistently emphasizes that the proposed Agency should have a role in risk assessment and communication only, future changes in remit is seen as a real possibility (Ibid:17):
As indicated earlier, the existing Treaty provisions impose constraints in the activities that can be attributed to the Authority, but this should not be taken to mean that a possible future extension of its competencies should be discounted. Such an extension should only be considered in the light of the experience with the functioning of the Authority and the confidence gained in its operation, including the possible need to change the Treaty
General food law – Regulation 178/2002
Regulation 178/2002 was adopted in January 2002. It followed the main lines of the White
Paper, and established a European Food Authority with responsibility limited to risk assessment (scientific advice, monitoring and research) and risk communication. A principle of separating risk assessment (in the Agency) and risk management (in the Commission) was firmly established. The Agency was set up provisionally in Brussels in 2002.
20
The regulation establishing EFSA has its basis in central objectives that also formally guide the overall approach of European food law. These can be found in a number of articles (for instance articles 1, 5, 8 and 22). Article 1 states:
This Regulation provides the basis for the assurance of a high level of protection of human health and consumers' interest in relation to food, taking into account in particular the diversity in the supply of food including traditional products, whilst ensuring the effective functioning of the internal market.
Several diverging intentions and interests can be detected here. First, there is an interest in securing the protection of human health. Second, a central aim is to secure consumers' interests. These two intentions need not be compatible, and in certain cases may even contradict each other. Third, there is an intention to take into account the diversity of food supply and traditional products. This aim is related to the protection of smaller farming units across Europe. Fourth, Article 1 states the goal of securing the internal market. Summed up, the regulation expresses a dual aim corresponding to that of the British case: to protect public health and secure safe food on the one hand, and to ensure the operation of the (internal) market on the other. In addition, the interest of certain farmers and farming practices and the protection of the consumer is central.
20 A dispute concerning the permanent location continued for some time. Finally, a Council decision late 2003 decided on Parma,
Italy (EFSA 2003).
17

The objective of securing the internal market can be seen as serving further European integration. This has traditionally been done through the harmonization of national mechanisms, and can be seen as an important motivation behind the establishment of EFSA and the general food law. The prominence of this aim is evident also in the opening statement of the regulation (introductory remark no 3):
The free movement of food and feed within the Community can be achieved only if food and feed safety requirements do not differ significantly from Member State to Member State.
The regulation prescribes that EFSA shall consist of a scientific committee and permanent scientific panels (article 28). Until this point, food experts participated in the EU mainly through scientific and advisory committees adjoined to the Commission (Joerges et al. 1997;
Joerges and Everson 2000; Vos 2000a). With the establishment of EFSA, scientific experts in the EU will participate in a more permanent structure detached from Commission activity.
This might prove intrinsic to an 'Europeanisation' of scientific expertise, and might as well prove an important counterweight against competing scientific and political judgments in other major trading countries, such as the USA.
Restoring consumer confidence in the aftermath of the BSE crisis is important in the EU – as in the British case. In the introduction on the regulation proposal the phrase 'consumer confidence' is used no less than eleven times (COM (00) 716:5-25). It is also central in the regulation (points 9, 22, 23, 35 and 40). Consumer confidence relates to confidence in food, food production and the food trade. It also implies confidence in European food regulation through EU institutions, and the European Union in general. The centrality of ensuring confidence further emphasizes the value of food safety in governmental activity and society at large, and its centrality in securing the internal market.
The main instruments in EU food regulation mentioned are scientific excellence , independence and transparency (COM (00) 716 final:13-14). The regulation includes a specific article on independence (article 37): The Agency is to operate independently of the
Commission as well as of perceived vested interests. At the same time, members of the
Agency will have specific backgrounds as well as personal and career histories, which are likely to influence their actions. EFSA's management board includes a representative from the
Commission, and members with backgrounds in consumer affairs and industry, as well as from national ministries and food control authorities (Regulation article 25; Press release
IP/01/1789). An Advisory Forum consists of members from competent bodies in each of the member states (article 27). This secures an intergovernmental element. To what degree these can be affirmed as 'independent' members, is relevant. National as well as professional affiliations might influence members, compromising their independence. Another discussion point is related to EFSA's advisory role to the Commission. Previously, scientific and advisory committees now integrated in the Agency, resided within the Commission. Board members and scientific experts with close relations to the Commission might have impact on decisions within the new Agency. Former arrangements might influence the way the new
Agency operates. People inhabiting the new Agency may carry on older administrative cultures and values. How independent of the Commission the new Agency really will be, is something that will develop over the years to come.
The creation of autonomous regulatory agencies has been a trend in public policy, and are discussed and analyzed as a major governance trend in recent literature. Agencies have been a
18

prominent feature of the administrative landscape in several countries. These organizations have certain commonalities, but are on the other hand different in terms of operation, tasks and remit. The agency label is common in relation to national food administrations, but is nevertheless an ambiguous term. Some food agencies are mainly enforcement and inspection bodies, others primarily provide advice to other governing institutions and/or perform scientific research and risk assessment. Different food agencies have different relations to central government, and sanctioning and regulative powers vary. Food agencies are not identical organizations, but they are likely to possess important similarities because they operate in the same policy field and within similar organizational fields.
Rhodes (1996) introduced the term agencification in studying the establishment of independent agencies in British administration in the seventies (the so-called Next Steps agencies). Since then, there has been a development of agency literature in public administration. Overall, the agency literature seeks to answer general questions about agencies, and many look for common features and focus on how to obtain optimal management (Pollitt 2003, Pollitt and Talbot 2004). Agencies are often seen as a distinctive type of public or state organization, although most authors agree that they are difficult to define and categorize. Pollitt and Talbot (2004) bring on five criteria that are useful in the definition of agencies. The agency should be at arm's length from the main departments of state, it should be carrying out public tasks at a national level, be staffed by public servants, financed by the state budget, and subject to public/administrative law procedures (Talbot
2004:5). Most food agencies, although they might have different structures and operational structures, fit this broad characterization. Depending on how the organizational field is structured, food agencies can belong to the public health field, or the agricultural field. Both
FSA and EFSA belong to the public health sector, after the restructurations that took place in the wake of the BSE scandal. The situation in other countries, Norway for instance, is the opposite (Elvbakken and Hellebø 2005).
Despite the many valuable insights in existing agency literature, there is room for more research. In general, there are few explicit empirical studies. There is also a lack of comparative work. Most authors invite to a general debate about what the establishment of agencies means to public administration and management, and to democratic accountability.
However, they do not regard characteristics of the particular field of regulation in question.
Several contributors of the agency literature have asked for this (Pollitt 2003, Pollitt and
Talbot 2004, Pollitt et. al 2004). This paper has been aimed to give some answers to this particular challenge, by looking at the establishment of modern food agencies.
First, food safety regulation is a particular field of regulation, involving powerful (economic) interests, and therefore inherent tensions and conflicts that occasionally burst to the surface and create changes within the field. The food scares of the 1990s exposed these tensions, and conflicts between different actors in the field surfaced. The result was a restructuration of the field, in several European countries as well as within the European Union. In Great Britain, and in the EU, responsibility for food safety regulation was moved away from the agriculture and industry sectors, to the public health field. In a sense, this means that food safety regulation was brought back to its origins.
We have seen that FSA and EFSA in many respects are very different organizations. FSA can be seen as a regulatory agency with powerful instruments for policy-making. It integrates risk management, risk communication, risk analysis and assessment, while at the same time being
'advisory' to accountable governmental institutions. EFSA on the other hand, concentrates on
19

risk analysis, incorporates scientific experts and committees, but has no regulatory powers in a strict sense. Risk management, or regulation, in this case rests with central institutions: the
Commission. However, EFSA has a large potential for policy-making and policy influence; whereas expert and science-based judgements clearly incorporate judgements also based on ethics and established values. Science is not at neutral activity, and there are powers in agenda setting. These powers are readily available also to FSA.
According to Rhodes (1996:7), the key notion behind agencification in Great Britain was that distance from central departments gave freedom to manage. With the establishment of agencies, responsibility for (daily) management is generally delegated to the agency while accountability for policy remains with the respective minister (Rhodes 1996:12). In the case of EU agencies, policy accountability as a rule remains with the responsible EU commissioner(s). Majone (1996a; 1996b; 2000) emphasizes that by separating policy from administration through the establishment of agencies, the day-to-day administration and implementation of policy becomes more stable and predictable. As a result, regulation should become more credible. On the other hand, the establishment of independent agencies with regulatory powers might jeopardize political accountability. The documents analyzed in the process of establishing FSA and EFSA, reveal that a restoration of confidence, trust and faith in arrangements for securing food safety was a central motivation. The choice of the agency structure seems to be related to this ambition. Two different solutions were chosen; one agency incorporates regulatory powers, while the other does not. This demonstrates that an investigation into particular agencies in the same field is worthwhile.
EU agencies are frequently analyzed in relation to the development of the European Union and European integration in general (Dehousse 1997; Kreher 1997; Majone 1997; Shapiro
1997). The development of agencies in the EU is seen as a response to conflicting pressures following the creation of the internal market – and in particular as a response to shortcomings in the existing regulatory approach of the European Commission (Dehousse 1997:246-247).
Several authors point out that EU agencies lack the independence and powers of other regulatory bodies, and are relatively 'weaker' than agencies in the US (Majone 1997:262;
Shapiro 1997:276-282). Nevertheless, these authors are generally optimistic regarding the potential and influence of European agencies. An establishment of (more) EU agencies is seen as an important step towards further European integration. EFSA can be seen to contribute to this development. Vos (2000b) considers the establishment of more (EU) agencies as advantageous, primarily because agencies facilitate the use of scientific and/or technical experts who are not part of the existing bureaucratic structure, and secondly because they reduce the administrative workload on central institutions, allowing the latter to concentrate on strategic policy-making. In this, her arguments can also be applied at a national level, and are relevant to the establishment of both FSA and EFSA. Vos on the other hand warns that extreme agencification could lead to more fragmentation, and erode accountability (Vos
2000b:1120). The question of accountability, agency autonomy and political control is further central to the agency literature in general (Christensen and Lægreid 2004, Pollitt and Talbot
2004; Pollitt et. al 2004). These discussions are, as we have seen, relevant for both FSA and
EFSA.
Majone argues that administrative or technocratic networks within EU agencies serve as important 'bearers of reputation' (Majone 1997:271). The establishment of networks facilitates the development of common practices, which then supposedly create shared expectations. The assumption is that the establishment of an agency in a certain policy field
(like food safety) can create a genuinely European policy in that particular field, and might
20

result in further identification with the values and priorities of European policy. These aspects are valuable in the understanding of the establishment of EFSA. With the creation of a
European agency responsible for risk analysis in relation to food regulation, disparate experts within the same policy field are brought together in a European context. However, this is not a new idea – food experts have earlier on been united in the long existing committee system of the EU.
21
The new element implemented by the establishment of EFSA is that the experts now reside in a separate unit, organized as an independent body separate from the
Commission, responsible for risk management and regulation. EFSA's remit is to provide scientific advice, and not to regulate or control in a direct sense.
The agency term is seen as ambiguous at the EU level, and has been used for several different bodies (Kreher 1997:227). At the same time, EU agencies in general share some similar characteristics. They are described in a Commission paper (EU 2000) and defined as (Ibid:4):
…a public authority set up under European law, separate from the Community institutions (Council,
Parliament, the Commission, Central Bank, etc.), and enjoying legal personality
EFSA is now counted as one of these agencies.
22
EFSA is accordingly set up in the traditional manner. The idea of a powerful regulatory agency such as the US FDA was not realized; the existing EU agency model and institutional arrangements have prevailed.
As shown in the empirical analysis of this paper and in our presentation of the agency literature, independence is a key notion. In relation to the food agencies, independence implies independence from both vested interests and from central authorities perceived to be more directly involved in policy decisions. Independence is a keyword in an effort to ensure a form of neutrality or objectivity in order to secure public confidence. At the same time, this is a problematical goal, not the least because food safety regulation and risk assessment involves conflicting priorities, interests and value-based considerations.
Transparency, openness and consultation are elements used to assure public involvement, in particular in FSA. EFSA also has adopted a policy of openness, and both agencies publish central policy documents, meeting reports and expert statements on their websites. This can be seen as ways to ensure confidence in decision-making. The recognition of public involvement, visible in the broad representation at the management boards (and advisory forum), demonstrates a fundamental acceptance of diverse interests in food regulation. This could conflict with the agency’s desired independence, and result in a rather ambiguous approach. The inclusion of other 'interested parties' on the other hand opens up for an understanding of risk regulation as something more than a mere neutral technocratic or scientific question. Other values and interests are recognized as important to the regulation of food safety as well. This agrees with Joerges et al. (1997) and Vos (2000a), who argue that separating risk assessment and risk management in order to preclude policy-based decision is a difficult – if not impossible – operation. This inclusion of 'other interests' also takes heed of
Selznick's argument; that regulation involves socially valued activities.
An important question is whether complete independence actually is possible – or feasible.
The integrated approach – involving 'other interested parties' in management and decisionmaking – may be characteristic of the British approach to public policy compared to other countries when studying food policy. It certainly differs from a general trend, which
21
22
Other international food expert networks include the Codex Alimentarius Commission and FLEP (Forum for European Food
Law Enforcement Practitioners.
The EU website lists 13 different agencies, among them EFSA. www.europa.eu.int/agencies/index_en.htm
21

emphasizes the separation of risk analysis and risk management. The British approach recognizes a need to coordinate diverse interests in potentially conflicting policy areas in order to make 'sound' and credible decisions. This mingling of interests, scientific judgement, and policy considerations may on the other hand only complicate decision-making further.
With the establishment of EFSA, the EU seeks to achieve a desired split between risk assessment and risk management, insofar as its scope of operation covers risk assessment and risk communication only. This differs from the British 'integrated approach' whereas FSA incorporates regulative tasks as well. However, by involving national representatives and former representatives of consumer groups and industry, EFSA includes other interests. This might bring the overall EU approach closer to the British 'integrated approach' after all, even though the rhetoric of the EU on the importance of separating policy, regulation and scientific risk assessment in the policy documents studied here, is more explicit.
On the one hand, EFSA can be seen to reduce conflicts between national expertise, and thus create a unified European expertise and 'Europeanize' risk regulation. This can be seen to strengthen expertise claims and the public legitimacy of science-based policymaking. On the other hand, expert agencies are very much engaged in politically significant activities and in the long run many will probably realize that are not only engaged in technical or scientific activities. Information and science is power, and policies can turn on analyses of data. The public might thus demand a politically accountability that the agencies do not, or cannot offer.
Changes in food safety regulation can be explained by reference to recent food scares and their impact on public opinion, media and policies. These scandals, and in particular BSE have had the power to change institutions. This has been manifested by the establishment (or restructuring) of food agencies. The agency literature shows that agencies can be difficult to capture analytically. Pollitt (2003) and Pollitt and Talbot (2004) point out that a universal legal classification of agencies is difficult because national systems differ. By restricting analysis to food agencies, we have examined a particular field of regulation. Agencies operating in the same policy field – such as food control – can, despite differences in context, be expected to enjoy certain similarities across countries and institutional boundaries, because of their similar foundations and tasks. Risk regulation, such as food safety regulation, incorporates particular challenges. It integrates different and sometimes conflicting interests, such as conflicts between commercial and public health interests. It involves tension between expert and policy judgement. Food regulation in different countries also share historical similarities, related to the development of food law, advances in scientific knowledge, and the development of techniques for food control and safety regulation (Elvbakken 1997). These particular tensions or conflicts of interest may be, and have been solved in different ways.
This is shown in our empirical analysis of the establishment of FSA and EFSA.
Understanding the subject that is regulated – in our case food safety – is important in order to understand how the field is regulated and organized. This is also important in relation to the interpretation of how different agencies operate and are managed. Following Selznick, understanding the subject of regulation, the tasks valued and upheld by the institution is important in order to assess the potential impact of a crisis – especially a crisis of legitimacy, attacking the foundational values the institution embodies (Selznick 1985).
By placing public tasks outside the authority of central government (or, in the case of the EU; the Commission), it becomes possible to hold the respective agency accountable for new food scandals. The
22

creation of separate institutions therefore illustrates a wish for stability and credibility with regard to decision-making. Whether or not this actually will be attained, is another discussion.
The idea of separating administration, scientific risk assessment and policy issues, is not unproblematic. Administration itself can embody policy considerations. This raises the important question of accountability. The wish to separate policy and administration was an important motivation behind the establishment of both FSA and EFSA. However, agencies still make decisions that have important policy implications. The separation of policy-making and management is therefore not an easy task. This certainly applies to the field of food safety, where consumer interests, agricultural and business interests, ethical judgements and scientific considerations often are at odds.
EFSA surfaced with restricted powers in terms of regulative authority compared to initial proposals. Its main functions are related to scientific risk assessment. Food control is still the responsibility of member states. The existing structure and division of powers in the EU, between EU institutions and member states, and between the Commission responsible for policy issues and other bodies designed for management purposes, proved a strong model for the new agency. The structure of the British administration allowed for a stronger agency including regulatory tasks. FSA was set up as an independent agency, but placed under the
Department of Health. Here, previous arrangements, such as the organization of the scientific committees as well as regional representation have survived within the new organization. In both cases, the recruitment of people from bodies formerly responsible for food safety issues or from other bodies of central administration might indicate transference of existing values and practices from other government bodies. Whether the alleged independence of the agencies creates new practices and cultures, or reproduces old problems remains to be seen.
Baggot, Rob (1998) The BSE crisis: public health and the 'risk society'. In Gray, Pat and Paul t'Hart (eds.) Public
Policy Disasters in Western Europe, London and New York: Routledge.
Baldwin, Robert and Martin Cave (1999) Understanding Regulation. Theory, Strategy and Practice, Oxford:
Oxford University Press.
Berg, Lisbeth (2000) Tillit til mat i kugalskapens tid. En komparativ kartlegging, med fokus på forbrukertillit og matsikkerhet i Norge, England og Belgia, SIFO- report no. 5, Oslo: Statens Institutt for forbruksforskning.
Boin, Arjen and Paul t'Hart (2003): Public Leadership in Times of Crisis: Mission Impossible? Public
Administration Review 63(5):544-553.
Bovens, Mark, Paul t'Hart and B. Guy Peters (1998) Explaining policy disasters in Europe. in Gray, Pat and Paul t'Hart (Eds) Public Policy Disasters in Western Europe, London and New York: Routledge.
BSE Inquiry Report (2000) The Inquiry into BSE and variant CJD in the United Kingdom, available at: http://www.bseinquiry.gov.uk/
Buonanno, Laurie, Sharon Zablotney and Richard Keefer (2001) Politics versus Science in the Making of a New
Regulatory Regime for Food in Europe, European Integration online Papers (EioP) 5(12), http://eiop.or.at/eiop/texte/2001-012a.htm
Cannon, Geoffrey (1987) The Politics of Food, London: Century Hutchinson Ltd.
23

Christensen, Tom and Per Lægreid (2004) Regulatory Agencies – the Challenges of Balancing Agency
Autonomy and Political Control, Stein Rokkan Centre for Social Studies Working Paper 18, Bergen: Stein
Rokkan Centre for Social Studies.
COM (97) 176: Commission Green Paper on Food Law.
COM (97) 183: Commission Communication on Consumer Health and Food Safety.
COM (99) 719: Commission White Paper on Food Safety.
COM (00) 716: Regulation establishing the European Food Authority.
Dehousse (1997) Regulation by networks in the European Community: the role of European agencies, Journal of
European Public Policy 4(2): 246-61.
Elvbakken, Kari Tove (1997) Offentlig kontroll av næringsmidler: institusjonalisering, apparat og tjenestemenn,
Bergen: Institutt for administrasjon og organisasjonsvitenskap and Senter for Samfunnsforskning.
Elvbakken, Kari Tove and Lise Hellebø (2005) Endring og kontinuitet i norsk matkontroll – konflikter mellom kontrollhensyn og verdier, Paper presented at the Norwegian National Conference for Political Science, 5-7
January, Hurdalssjøen.
EFSA (2003) EFSA Press Release 15 December 2003 IP/01/1789, available at: http://www.efsa.eu.int/press_room/press_release/32_en.html
EP (1997): “Report on alleged contraventions or maladministration in the implementation of Community law in relation to BSE”, European Parliament Report: A4-0020/97, available at http://www.europarl.eu.int/committees/bse_home.htm
EU (2000) “Agencies of the European Union”, available at: http://europa.eu.int/agencies/_docs/resume_en.pdf
EU Commission (1999) “Scientists recommend Food and Health Authority to advise the EU”, press release
IP/99/973, available at http://www.europa.eu.int/rapid/start/cgi/guesten.ksh
European Dialogue (1996) European Dialogue. The Magazine for European Integration, Sep-Oct 1996 Issue 5, at http://europa.eu.int/comm/dg10/eur_dial/frameset_current.html
FSA (2000) Food Standards Agency Business Plan 2000-2001.
Fligstein, Neil and Iona Mara-Drita (1996): How to Make a Market: Reflections on the Attempt to Create a
Single Market in the European Union, The American Journal of Sociology, 102(1):1-33.
Flynn, Andrew, Terry Marsden and Everard Smith (2003): Food Regulation and Retailing in a New Institutional
Context, The Political Quarterly 2003:38-46.
Grant, Wyn (1997) BSE and the politics of food, in Dunleavy et al. (eds.): Developments in British Politics 5,
London: Macmillan Press Ltd.
Hellebø, Lise (1999) Food control and food regulation in the European Union, SEFOS-notat 6, Bergen: Senter for samfunnsforskning.
Hood, Christopher (1991): A Public Management for All Seasons? Public Administration, Vol. 69, Spring 1991
(3-19).
Hood, Christopher, Henry Rothstein and Robert Baldwin (2001): The Government of Risk. Understanding Risk
Regulation Regimes, Oxford: Oxford University Press.
James, Philip (1997) The Food Standards Agency. An Interim Proposal by Professor Philip James, 30 th April
1997.
24

James, Philip, Fritz Kemper and Gerard Pascal (1999) A European Food and Public Health Authority. The future of scientific advice in the EU, December 1999, http://europa.eu.int/comm/food/fs/sc/future_food_en.html
Joerges, Christian, Karl-Heinz Ladeur and Ellen Vos (eds.) (1997) Integrating Scientific Expertise into
Regulatory Decision-Making, Baden-Baden: Nomos Verlagsgesellschaft.
Joerges, Christian and Joseph Neyer (1997) Transforming strategic interaction into deliberative problem-solving:
European comitology in the foodstuffs sector, in Journal of European Public Policy 4(4): 609-625.
Joerges, Christian and Michelle Everson (2000) Challenging the bureaucratic challenge, in E.O. Eriksen and J. E.
Fossum (eds.): Democracy in the European Union. Integration through Deliberation? London and New York:
Routledge.
Jukes, David (1995) Food Law Harmonization within Europe – A Learning Opportunity, Food Control 6(6):
283-287.
Kjærnes, Unni (1999) Food Risks and Trust Relations, Sosiologisk tidsskrift 7(4):265-284.
Krasner, Stephen D. (1984) Approaches to the State: Alternative Conceptions and Historical Dynamics,
Comparative Politics 16(2):223-246.
Kreher, Alexander (1997) Agencies in the European Community – a step towards administrative integration in
Europe, Journal of European Public Policy 4(2): 225-45.
Majone, Giandomenico (1996a) Regulating Europe, London and New York: Routledge.
Majone, Giandomenico (1996b) New Agencies in the EC: Regulation by information, in Kreher, Alexander
(ed.): The New European Agencies. Conference Report, EUI Working Paper RSC No. 96/49, Florence:
European University Institute.
Majone, Giandomenico (1997): The new European agencies: regulation by information, Journal of European
Public Policy, 4(2): 262-275.
Majone, Giandomenico (2000) The Credibility Crisis of Community Regulation, Journal of Common Market
Studies 38(2): 273-302.
March, James G. and Johan P. Olsen (1989) Rediscovering Institutions. The Organizational Basis of Politics,
New York: Free Press.
Millstone, E. and P. van Zwanenberg (2002) The Evolution of Food Safety Policy-making Institutions in the UK,
EU and Codex Alimentarius, Social Policy and Administration 36(6):593-609.
Peters, Guy B. (1999) Institutional Theory in Political Science. The ‘New Institutionalism’, London and New
York: Pinter.
Pierson, Paul (2000) Increasing Returns, Path Dependence, and the Study of Politics, The American Political
Science Review 94(2):251-267.
Pollitt, Christopher (2003) Agencies, Apples and Pears: Mapping the Agency Debate, paper presented to the 7 th
International Research Seminar in Public Management (IRSPM7), Hong Kong: 2-4 April 2003.
Pollitt, Christopher and Colin Talbot (eds.) (2004) Unbundled Government. A critical analysis of the global trend to agencies, quangos and contractualisation, London and New York: Routledge.
Pollitt, C, C. Talbot, J Caulfield and A Smullen (2004): Agencies: How Governments do Things Through Semi-
Autonomous Organizations, Basingstoke: Palgrave.
Powell, Walter W. and Paul J. DiMaggio (eds.) (1991) The New Institutionalism in Organizational Analysis,
Chicago and London: The University of Chicago Press.
25

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 Safety Authority and laying down procedures in matters of food safety.
Rhodes, R. A. W. (1996) The New European Agencies. Agencies in British Government: Revolution or
Evolution? EUI Working Paper /Robert Schuman Centre No. 51, Florence: European University Institute.
Rosen, George (1958/1993) A history of Public Health, Baltimore and London: The Johns Hopkins University
Press.
Santer, Jaques (1997) Speech/97/39, Speech by Jaques Santer, President of the European Commission,
Strasbourg, 18 February 1997.
Selznick, Philip (1985) Focusing on organizational research on regulation. In Noll (ed.) Regulatory Policy and the Social Sciences, Berkeley: University of California Press.
Shapiro, Martin (1997) The problem of independent agencies in the United States and the European Union,
Journal of European Public Policy 4(2): 276-91.
Smith, Martin J. (1991) From Policy Community to Issue Network: Salmonella in Eggs and the New Politics of
Food, Public Administration, 69:235-255
Talbot, Colin (2004) The Agency Idea. Sometimes old, sometimes new, sometimes borrowed, sometimes untrue.
In Pollitt, Christopher and Colin Talbot (eds.) Unbundled Government. A critical analysis of the global trend to agencies, quangos and contractualisation, London and New York: Routledge.
Thelen, Kathleen and Svein Steinmo (1992) Historical institutionalism in comparative politics. In Steinmo, S.,
K. Thelen and F. Longstreth (eds.): Structuring Politics. Historical Institutionalism in Comparative Analysis,
Cambridge: Cambridge University Press.
Thelen, Kathleen (1999) Historical Institutionalism in Comparative Politics, Annual Review of Political Science
(2): 369-404.
Vos, Ellen (2000a) EU Food Safety Regulation in the Aftermath of the BSE Crisis, Journal of Consumer Policy
(23): 227-255.
Vos, Ellen (2000b) Reforming the European Commission: What role to play for EU agencies? Common Market
Law Review (37): 1113-1134.
White Paper (1998) The Food Standards Agency. A Force for Change, Her Majesty’s Stationary Office: Dd
5067760 1/98 48003 Job No 319
26