EIoP: Text 2007-004: Full Text
Accountability deficits in European 'Comitology' decision-making(*)
Gijs Jan Brandsma
European Integration online Papers (EIoP) Vol. 11 (2007) N° 4;
http://eiop.or.at/eiop/texte/2007-004a.htm
Date of Publication in European Integration online Papers: 25.5.2007
| Abstract | PDF |
--Begin page 2--

Contents:

1. Introduction to contents list

The academic debate on comitology’s accountability is quite underdeveloped. On the one hand several authors claim that these committees suffer from an accountability deficit, but on the other hand this alleged deficit has not been looked into systematically. Whenever references to possible accountability deficits are made, these are made in the light of the legitimacy of the comitology system, and hence accountability as such is not looked into in depth (e.g. Demmke, 2000: 300-302; Schäfer, 1996: 22; Schäfer, 2000: 22-23; Larsson, 2003b: 169-171). Others, like Radaelli (1999) and Van Schendelen (2004), place the issue of accountability in the broader context of expert input into policy making. They argue that knowledge can be used politically, and that being an expert is part of a certain self-image that makes it hard to call people to account. However, they too do not present a systematic analysis of accountability.

With so many people echoing that there are accountability deficits in comitology, it is about time to look into this matter in depth. But first it should be demarcated what these concepts actually mean. What is accountability, and when is there a deficit? What are comitology committees, and why should accountability be relevant to them in the first place? What has past research found, and how can it help us to investigate this issue? These are the subjects this paper deals with. It will be argued that accountability deficits come under various guises. This is also illustrated by examples from interviews the author held with committee participants in order to test the ground for accountability research(1).

2. Accountability to contents list

2.1. Definition

What exactly is accountability? There seem to be as many definitions of accountability as there are scholars (e.g. Mulgan, 2000: 555-556; Day and Klein, 1987: 26; Bovens, 2006: 8-9; Pollitt, 2003: 89; Behn, 2001: 3-6). This makes it even more unclear what the aforementioned authors actually mean by a lack of accountability in comitology committees. It seems, though, as if in the accountability literature in general two groups of meanings can be distinguished: one focusing on ‘giving account’ and one focusing on ‘holding to account’(2). --Begin page 3--

In the sense of ‘giving account’, accountability is an act of disclosing information and justifying behavior (e.g. Mulgan, 2000: 555). Giving account in this sense does not necessarily entail the possibility of having to face consequences. This definition effectively makes ‘account giving’ a very broad concept, and comes close to the concept of justification. ‘Holding to account’ entails more strict criteria for defining accountability. Here, actors are not only expected to disclose information and justify their behavior, but they should also be in a relationship with another entity which has the authority of imposing sanctions, or giving rewards (e.g. Bovens, 2006; Day and Klein, 1987: 26). This relationship can take the form of a hierarchical, ‘vertical’ relationship such as between employees and their superiors, but in modern times there can also be ‘horizontal’ accountability relationships with third authorities, such as audit offices (Bovens, 2006). Accountability relationships can therefore go beyond the typical ‘principal-agent’-relationships, and hence they are generally defined as ‘actor-forum’-relationships.

The meaning of ‘holding to account’ probably comes closest to what the comitology authors mean by a lack of accountability. Schäfer (2000: 23), for example, states that ‘[t]he awareness on the part of the representatives, both from the Community and the Member State level, that they may be held directly accountable for their decisions would be likely to affect their actions’, while Larsson (2003b: 157 and 169-170) questions if we now have ‘civil servants who are the politician’s masters’ as opposed to ‘civil servants being accountable to elected politicians’.

A clear definition of accountability in the sense of ‘holding to account’ can be found with Mark Bovens, who defines accountability as “a relationship between an actor and a forum, in which the actor has an obligation to explain and to justify his or her conduct, the forum can pose questions and pass judgment, and the actor may face consequences” (Bovens, 2006: 9). This paper will use his definition. It has been adopted because it clearly distinguishes between what is, and what is not accountability. First of all, one would need someone to render account, and someone to render account to. This refers to the ‘actors’ and ‘forums’ Bovens defines (Bovens, 2006: 9-10). As was mentioned before, accountability relationships do not necessarily need to take the form of principal-agent relationships, and hence Bovens seems very cautious not to use these terms instead. Further, there is a relationship between the actor and the forum, in which the actor explains and justifies his or her conduct to the forum. If necessary, the forum can ask further questions, evaluate the actor’s behaviour in a judgment and make the actor face consequences, either in terms of rewards or of sanctions. Another reason for choosing Bovens’ definition is that other, like Day and Klein’s (1987), only stress the ability of the forum to impose sanctions rather than to give rewards.

Still, even though Bovens’ definition clearly defines what accountability is and which various actors and forums can be distinguished, it does not indicate the process by which actors are held to account. A common approach for this is by distinguishing between ‘police patrol’ and ‘fire alarm’ arrangements (see e.g. Huber, 2000: 401; Pollack, 2003; Hopenhayn and Lohmann, 1996). These terms, first used as metaphors by McCubbins and Schwartz (1984), refer to active and passive forms of oversight by the forum over the actor. With police-patrol oversight, active and direct oversight is meant. The activities of the actors are systematically scrutinized, for example by reading documents and observing behaviour. This is the active form of oversight. The advantage of police patrol oversight is that the actor is under continuous, direct control. On the other hand, though, this form of oversight takes a lot of time and effort from the forum.

This is exactly the other way around with fire-alarm oversight. Here, the forum is passive until it receives complaints about the actor’s behaviour. It therefore relies on others to monitor the actor, and waits for complaints to arrive: to ring the ‘fire alarm’. This takes much less effort from the forum. Furthermore the forum can be sure that action is only taken in cases where things go wrong. On the other hand, though, there are other entities – such as stakeholders or colleagues of the actor – to whom the burden of supervision is shifted. The difference between ‘police patrol’ and ‘fire alarm’ arrangements is important to keep in mind when empirically researching accountability deficits. It is not to say that one form is by definition better than the other, it is just to indicate that multiple forms of arrangements exist, and that one should be aware of this when doing empirical research. A focus on the direct interaction between actor and forum would be insufficient, because the forum may use third parties to monitor the actor, rather than doing it all by himself.

Having defined what accountability is, when can one speak of an accountability deficit? When are accountability arrangements insufficient? Before turning to accountability deficits, a look at the aims of accountability is --Begin page 4-- necessary.

2.2. Aims of accountability and accountability deficitsto contents list

Opinions diverge in the academic literature about what merits of accountability are in the first place. One can distinguish at least two different perspectives. First, there is the democratic perspective. Since any democratic political system ultimately rests upon the notion of government by the citizens, delegation from one authority to the other involves a whole series of principal-agent relationships, with ultimately the citizens as principals (Strom, 2000). Accountability mechanisms are meant for keeping a democratic ‘check’ upon the behaviour of public officials enjoying discretion in order to make sure that ultimately the preferences of the people are translated into policy. Therefore, the emphasis of accountability in this perspective is on the content of decision-making in the light of delegating tasks (Bovens, 2006: 25; Behn, 2001). Problems emerge if the agent deliberately does not act in the principal’s interest (sabotage) or simply does not act in the best interest of the principal (shirking), which is likely if the principals are not able to monitor the behaviour of the agents effectively (Strom, 2000: 270-271). For finding the appropriate forum, one would simply need to look only one level up the chain of delegation. For a minister, parliament would be the appropriate forum, for parliament it would be the citizens, and for a civil servant it would be his superior.

The second view on the purpose of accountability focuses on the prevention of the tyranny of government, or rather the prevention of the abuse of the discretionary powers of people and institutions. This could be labelled as the constitutional perspective. Through accountability mechanisms, actors are to be deterred from increasing their powers in a way that distorts institutional balances. Here, the emphasis does not lie with the content of actions, but rather with the processes that have been followed. Accountability regimes are there in order to make sure that the actors do what they are empowered to, and nothing more than that (Bovens, 2006: 25-26). The legal power can be placed in this tradition. Unlike the democratic perspective, this constitutional perspective is most appropriate for assessing the accountability of institutions.

Having said this, it may look as if accountability is something one can never have enough of. However, this is not the case. As Behn, and Bovens and ‘t Hart have argued, there should be a balance between accountability and efficiency (Behn, 2001; Bovens and ‘t Hart, 2005: 245-262). If the balance is tipped in favour of the former, one may easily spend lots of time on explaining one’s conduct without having much time left to do one’s primary tasks. This may be described as accountability overload.

The debate on comitology is about the opposite: an accountability deficit. This may occur if the practices of holding actors to account are insufficient for fulfilling the aims of accountability. Taking the democratic perspective, this would mean that actors are able to act against the wishes of the citizens, parliament, and any chain in between parliament and themselves – and get away with it. In the constitutional perspective, it would mean the absence of institutions counterbalancing the input of the actor if the actor would exceed his powers, so that the behaviour of the actor is unconstrained.

Are these the accountability deficits the comitology authors are pointing to? Before placing accountability deficits in the light of comitology committees, the next paragraph will shed some more light on the phenomenon of comitology itself. What is comitology, and what makes it require accountability?

2.3. Comitology committeesto contents list

There is no consensus about where the word ‘comitology’ actually comes from. Speculations on its etymology range from the French expression ‘kremlilogie’ referring to politicized bureaucracy, to the sixteen-century word ‘comity’, meaning mutual recognition of institutions between states, and to a polemic depiction of the study of the operation of public committees itself (Bergström, 2005: 6-7). In European politics, it is used to depict committees of member state representatives that assist the Commission in implementing policies (e.g. Van Schendelen, 1998; Andenas and Türk, 2000; Egeberg, Schäfer and Trondal, 2003). It should be distinguished from other kinds of committees, such as Council Working Groups, the ‘Grand Committees’ (European Social Committee, Committee of the Regions), and Commission expert groups (see e.g. Larsson, 2003a). The member state representatives --Begin page 5-- attending comitology committees are generally policy experts from the ministries or other public authorities, but can also be from outside government acting on the government’s behalf (Falke, 1996).

Of all forms of European committee governance, the system of Comitology committees is probably most extensively covered by scholars(3). With respect to accountability, the irony is that comitology committees were meant to be accountability devices in themselves. When the first committees were set up by the Council of Ministers in the early 1960s, they were meant to control the Commission. Since they are composed of delegated experts from the member states, an intergovernmental ‘check’ was placed upon the Commission’s implementing powers. This way, comitology committees ‘bridge the gap’ between Council and Commission by keeping member state control over policies that are to be implemented by the Commission (TEC Art. 202/3; Dehousse, 2002; Vos, 1997: 227). Despite – or perhaps complementary to – their official function of controlling the Commission, the committees can also be used as forums of deliberation between the Commission and member state representatives, in which a spirit of expertise prevails (Joerges and Neyer, 1997: 609-625; Bergstrom, 2005: 52; Sannerstedt, 2005: 105). This second feature of comitology prompted Dehousse to observe: “if comitology has indeed evolved into a freewheeling transnational structure, the key question becomes: who controls this structure, and how?” (Dehousse, 2002).

The control over the Commission can currently take up to four forms, and these four forms mainly specify what the meaning of a vote in a committee is. Up to now, four different types of comitology committee procedures exist: advisory, management, regulatory and safeguard procedures (Council Decision 1999/468/EC). This is specified below in Table 1.

Table 1

Working under the advisory procedure, committees advise the Commission on specific policies, and do so by giving an opinion on Commission proposals by a simple majority of member states. As the word ‘advise’ already indicates, the Commission may ignore their opinion although it is supposed to give much consideration to the expressed position (Council Decision 1999/468/EC). Using the management procedure and regulatory procedure, committees have more power: then their approval is required for the Commission to implement policies. The difference between these two may seem only minor, but in practice it is not unimportant. In the management procedure, the Commission will adopt the proposed measures unless the committee explicitly objects to it by a qualified majority. If the committee does not object to the measure, the measure can be implemented with immediate effect. But if the committee does oppose the measure, the Commission may postpone its implementation for a period for up to three months. During that time, the Council may take a different decision. If the Council fails to oppose the proposed measure, the Commission is free to adopt the policy proposal (Council Decision 1999/468/EC). In the regulatory procedure, the role of member state representatives is bigger. Here, proposed measures are only implemented if the committee explicitly approves it by means of a qualified majority. If it does not, the Commission will submit without delay a proposal to the Council relating to the measures to be taken, and will inform the Parliament. Then, the Council is able to oppose the proposal by qualified majority, bypassing the committees. If no opposition is expressed within three months, the Commission is free to adopt its latest proposal (Council Decision 1999/468/EC). In this regulatory procedure, the Council and the member states have most control over the making of implementation measures (Pollack, 2003: 136). The final type of procedure is the safeguard procedure, which is very rare. It entails an add-on procedure for committees working under other procedures as well. The safeguard procedure is used for dealing with individual (mainly trade-related) cases, for example with respect to airport access rights or cabotage, and every member state participating in a committee working under the safeguard procedure has the power to refer the matter to the Council (Council Decision 1999/468/EC).

Theoretically, there are always other resorts for the Commission when it finds a proposed measure to be opposed. It can ignore the committee’s opinion (advisory procedure), or forward the matter to the Council and hope for the absence of a qualified majority against (management procedure) or a vote result that does not conform to a qualified majority (regulatory procedure, see Hofmann and Töller (1997) for a vivid description of a case like this regarding GMO’s). In practice, though, the Commission rarely forwards matters to the Council. In 2005, 2,582 opinions were delivered by the committees, of which only 11 were referred to it (European Commission, 2006b). --Begin page 6-- The Commission is aware that it needs the member states for implementing their policies Europe-wide, and therefore it is keen on unanimity in committees. Matters are postponed for further discussion frequently, so as to make sure every member state is happy with the eventual outcome. Of course there are occasions where time pressure makes the Commission force a vote if necessary, but generally it can be certain of a majority(4).

Comitology committees as such do not appear in any treaty, even though the Treaty of the European Community empowers the Council to impose conditions on the Commission for implementation (TEC, Art. 202/3). However, the position of the committees is formalized in the basic acts. Typically, a basic act contains an article stating that for the implementation of that particular policy ‘the Commission shall be assisted by a committee composed of representatives of the Member States and chaired by the representative of the Commission’ and ‘Where reference is made to this paragraph, the procedure laid down in Article x of Decision 1999/468/EC shall apply, in compliance with Article 7 (3) and Article 8 thereof’, meaning that a comitology committee will be set up according to the advisory, management, regulatory and/or safeguard procedure, that it shall adopt its own rules of procedure, that ‘the principles and conditions on public access to documents applicable to the Commission shall apply to the committees’, and that the European Parliament is to be informed of agendas, draft measures, voting results, summary records of the participants and the list of authorities to which the officials delegated by the member states belong (OJ 1999 L 184). Concerning public access to documents, this means that to the comitology committees the same rules apply as to the Commission expert groups and the Council Working Groups (OJ 2001 L 145/43). Further, a list of comitology committees is published including references to the basic acts by which they were instigated (OJ 2000 C 225/02), the Commission publishes annual reports since 2001, and an on-line register exists containing references to the documents that are sent to the European Parliament (European Commission, 2006a). The meetings themselves, however, are confidential.

The rules of procedure have in some cases very similar, but in most cases even the very same text. The Council provided a model text for these rules, in which blank spaces were left for the committees to fill in their own names and the date of adoption (OJ 2001 C 38/3). This model text states that the agendas, draft directives, voting results, presence lists (not including names, but only the represented organizations) and summary records must be sent to Parliament. The European Parliament, however, hardly has any powers regarding comitology committees. For a limited number of committees, the European Parliament will be enabled to overturn committee opinions in the near future (see Decision 2006/512/EC), but for the majority of committees the European Parliament remains a toothless tiger. Even though informed of the proceedings in the committees, it only has one power: to criticize implementation measures adopted through the committee procedure if they exceed the implementation powers provided for in the basic piece of legislation. If the Council and Commission fail to adapt the measure taken, the European Parliament can appeal to it at the European Court of Justice (Türk, 2000: 217-253).

2.4. Locating accountability deficits in comitology to contents list

Following the above, comitology committees can be very influential policy makers. But why do the comitology scholars feel accountability is lacking? The distinction between ‘democratic’ and ‘constitutional’ forms of accountability addresses the aims of accountability, and indicates to whom account should be rendered. Yet, this distinction does not indicate who should be accountable in the first place. For this, one can distinguish between the institutional position of the committees, and the accountability of individual delegates to the committees. This multi-level accountability debate is displayed below in Figure 1.

Figure 1

The upper part of the picture, part 1, reflects the current institutional arrangements of comitology committees. In basic acts, drawn up by the legislative power, implementation of policies is delegated to the Commission. The legislative power also installs a comitology committee, composed of member state representatives, which needs to give a formal opinion on Commission proposals. The lower part of the picture, part 2, reflects the interplay between delegated experts and domestic actors.

If this two-level scheme is combined with the distinction between the democratic and the constitutional perspective, a classification emerges as presented in Table 2: --Begin page 7--

Table 2

From a constitutional perspective, accountability is about preventing the tyranny of government. Hence, countervailing powers should be built into the system outside the democratic chain of delegation, that are empowered to investigate and sanction power-exceeding behaviour. This perspective addresses the institutional setup rather than individual behaviour. The democratic perspective ultimately links the actor to the citizens: the ultimate principals. Account is given from agents to a principal located one step backwards along the chain of delegation (Bovens, 2006; Strom, 2000), and therefore the terminology common to principal-agent relationships can be used. Unlike the constitutional perspective, the democratic perspective fits both the collective and the individual levels of accountability.

Which accountability do the comitology scholars mean when they speak of an accountability deficit? Do they refer to the institutional setup of the committees, or do they mean the way in which national delegates carry out their position? Türk (2003: 196-197) and Rhinard (2003: 14-15) exclusively focus on the institutional side of it, by commenting upon the position the committees have in the policy process. Others, like Schäfer (2000: 23), Larsson (2003b: 157, 171, 172) and Van Schendelen (2004: 31) refer to individuals who are active both at the Community and Member State levels, but not to the institutional setup. A nicely structured picture emerges when all findings are classified along the two dichotomies presented above.

2.4.1. Constitutional perspective: collective/institutional accountabilityto contents list

Speaking of the collective/institutional accountability deficit from a constitutional perspective, there are several authoritive non-principals surrounding the committees. The Court of Justice has played an important role in shaping the functioning of comitology committees. In doing so, it first had to come to terms with their very nature. From a formalistic point of view, the committees are not very relevant. The responsibility for implementation lies with the Commission, just as the responsibility for legislation and delegation lies with the Council and, under codecision, also Parliament. When Court cases are lodged because of committee behaviour, all legal eyes turn to either the Commission or Council and Parliament.

A compelling example of this is the Rothmans case. A tobacco company requested access to the minutes of the Customs Code Committee. This committee deals with import regulations of, among other things, cigarettes. According to Decision 94/90/EC, applications for access to documents should be sent directly to the author. As with all comitology committees the Commission sets the agenda and takes down the minutes, and hence the company contacted the Commission for getting access. The Commission refused this application, on the grounds that despite the fact that it took down the minutes, the minutes themselves were, in an intellectual sense, ‘made’ by the Customs Code Committee and not by the Commission. In the subsequent court case, the Court of First Instance ruled that the Committee “does not have its own administration, budget, archives or premises, still less an address of its own” and hence “for the purposes of the Community rules on access to documents, comitology committees come under the Commission itself. It is therefore the Commission which is responsible for ruling on applications for access to documents of those committees, such as the minutes here in question.” (Court of First Instance, Case T-188/97).

The question to what extent this judgment made comitology committees as such come under the Commission itself is still a matter open to debate (Dehousse, 2002), but concerning transparency rules they do, according to this ruling. However, ways of addressing the committees themselves have been avoided. Rather, responsibilities have explicitly been shifted towards the Commission because the Customs Code Committee was not considered to be independent enough by the Court of First Instance. Despite the good intentions of the Court in terms of transparency, the Rothmans case could potentially pave the way for jurisprudence making the Commission accountable for the behaviour of the committees, and not the committees themselves. As long as the Court does not regard the committees as being independent enough, this seems the only legal way forward.

2.4.2. Democratic perspective: collective/institutional accountabilityto contents list

From a democratic perspective, the collective/institutional accountability question is about the behaviour of the --Begin page 8-- committees in the light of the preferences of their principals, being the Council and, under codecision, Parliament (see also Figure 1 above). Therefore, according to the democratic perspective, the committees should be accountable to these two institutions. However, if a committee acts in a way the legislative power does not like, there is nothing it can do about it. As mentioned before, the only thing Parliament can do is to adopt a resolution condemning a committee for power abuse, but for that it would need to prove that the adopted proposal was not in line with the mandate of that particular committee as set out in the basic act. In this sense the powers of the Parliament are similar to those of – and add nothing to – the powers of the Court of Justice(5). However, such a resolution should be adopted within one month after a committee voices its opinion, and since resolutions can only be adopted in the monthly plenary sessions in Strasbourg, this arrangement comes close to being a dead letter(6). The Council does not have any powers at all, unless the Commission refers a draft proposal to it. However, if the committees simply give an opinion on a proposal, which they usually do, neither the Council nor Parliament come in the picture if they do not like the content of that opinion. Here, a clear institutional accountability deficit can be exposed from the democratic perspective. The entities to which policies are delegated can take any measure they like as long as they can agree upon it. If implementation measures are not matching the legislator’s preferences, it is unable to hold them to account.

There are, however, substantial efforts under way to stop this part of the institutional accountability gap. Parliament, Council and Commission recently agreed on amending the 1999/468/EC Comitology act, in which a new procedure is introduced for committees which are involved in changing the content of basic legislation. In this new procedure, the ‘regulatory procedure with scrutiny’, both the Council and Parliament would have the power to overturn a favorable opinion by a committee. The new 'regulatory procedure with scrutiny' therefore can stop part of the accountability gap by keeping the Council and Parliament in the game after setting up a committee. This new procedure has been formalized in Decision 2006/512/EC, and the procedures for the according committees will be updated in due course. Promising as this may seem, it must be kept in mind that this new procedure will not apply to all committees. In fact, it is only meant for committees dealing with changing certain aspects of the basic legislation, for example for adapting it to scientific and technological progress. Committees not doing so will retain their original procedures, and therefore they can still be subject to an accountability gap.

2.4.3. Democratic perspective: individual accountabilityto contents list

Moving away from institutional to individual accountability from the democratic perspective, there are different issues to take notice of. Figure 2 shows the chain of delegation for the individual level:

Figure 2

The arrows in Figure 2 represent the chain of delegation: a set of principal-agent relationships. Accountability is to go in the opposite direction: ministers to parliament, parliament to the citizens, et cetera. An accountability deficit would occur if the comitology participant is able to act against the desires of principals up the hierarchy.

A common line of reasoning is that discretion requires accountability. The more autonomy an actor enjoys, the more there is a need to hold him to account in order to avoid ‘agency loss’ (Strom, 2000: 270-271). This means that there is more need of accountability in situations of high autonomy than in situations of low autonomy. At one end of the scale, there is of course the extreme position of ‘no discretion’. Here, accountability would simply have no point: the actor is under complete control and hence there is no discretion to be controlled. Here one can speak of a power relationship but not of accountability (Keohane, 2002). It must therefore be stressed that discretion is a necessary precondition for locating the need for accountability, and hence for locating accountability deficits.

Past research shows substantial autonomy of delegated experts. Egeberg, Schäfer and Trondal (2003) found that 34 percent takes the position he or she thinks is best on the basis of his/her professional expertise, and just over half of their respondents does not have clear instructions about the position they have to take. This last finding is confirmed by Sannerstedt (2005) who found the same figure for Swedish delegates only. --Begin page 9--

Autonomy and co-ordination are closely related subjects. In the conclusion of a cross-national study into this, Kassim identifies two groups of countries with structural differences: “Some […] aim to construct an agreed position on every issue and to ensure coherent presentation by all national representatives at every stage of the EU policy process. Others have more modest ambitions […] filtering out policies that conflict with higher aims or ensuring that more important information is exchanged” (Kassim, 2000: 243). The first group of countries Kassim describes (of which France, the UK and Denmark are exemplary) is likely to give individual experts little room for maneuver because European policies are centrally co-ordinated. In the second group (with Germany, Greece, Spain and Portugal as main examples, but also Belgium and Italy to some extent) there is less structured co-ordination, and hence agents’ behaviour is less well defined ex ante. Following Peters and Wright, the Netherlands – not examined in Kassim’s study – clearly belongs to the second group as well (Peters and Wright, 2001: 163-165). On the other hand, these two authors also suggest that in areas of low political salience, ‘attempts at centralized control, or even information-sharing, may break down in the face of links with policy experts from other countries and from the EU’ (emphasis added, Peters and Wright, 2001: 166). This important expectation, however, is not substantiated. The literature therefore is ambiguous on this issue: on the one hand accountability deficits seem more likely to occur in Germany, Greece, Spain, Portugal and the Netherlands than in the UK, Denmark and France because of their different co-ordination structures. However, it is also likely that these structures do not matter at all for politically less salient topics.

Thus, many delegates to comitology committees enjoy a high degree of autonomy. Also, the work in committees is of quite a technical nature. Sannerstedt (2005) found that committee participants characterize the work in the group as a ‘spirit of expertise’, and many of his respondents thought that the most important thing is to find the technically best solution (Sannerstedt, 2005: 105, emphasis added). Egeberg, Schäfer and Trondal found that 60 per cent of their respondents to a great extent feel allegiance to their own professional background and expertise, and that 26 per cent believes that their colleagues from other member states act as either independent experts, or have a mixed role between government representative and independent expert(7) (Egeberg, Schäfer and Trondal, 2003; italics in the original).

Another line of thought relates to degrees of expertise. As Radaelli notes, European policy-making is dominated by technocratic experts, and the same is true for comitology committees (Radaelli, 1999: 758). Delegated experts bring their specific expertise with them to Comitology committee meetings. It could be hard for their superiors to hold them to account because they themselves lack this specialized knowledge. One may wonder why there is a need for accountability in the first place. If policy making is about specialized knowledge, one could argue that experts should be doing their work autonomously so that their professional judgment is not affected by other considerations. Similarly, politicians are not experts and may not be able to judge important technical questions’ (Demmke, 2000: 301). However, this approach makes too sharp a distinction between the ‘technical’ and the ‘political’ realms. There are plenty of examples where technical issues are ambiguous politically. For example, the decision to put certain chunks of land on the environmentally protected area list may result in prohibitions of port or airport expansion, which has severe economic consequences. Changes in agricultural export refunds affect farming businesses. Therefore, technical issues can very well bear political saliency. This is also true for technical policies which by themselves are politically contagious, such as the Single European Sky. In this policy, efforts are under way to integrate all national airspaces into a single European airspace. This involves very technical discussions, but also hard politically motivated bargaining related to the competitiveness of current national airports, and the military use of airspace – a matter that directly affects national sovereignty.

These examples underline the point made by Radaelli that expert knowledge is not necessarily universally accepted and uncontested. Through ideas, politics takes place and thus, as Radaelli notes, knowledge has become the terrain of politics (Radaelli, 1999: 759)(8). Therefore, policy experts should be held to account (ibid: 770-771). Technically ‘best’ solutions might neither be the most feasible, nor the most politically desirable solutions (Radaelli, 1999: 759). The balancing between technically sound decision-making and political salience, in combination with the autonomy of actors, makes it important to hold actors to account at the individual level. Holding experts to account, though, can be difficult. As Van Schendelen notes: ‘As “being an expert” is part of their self-image, experts tend to play solo, reluctant to accept instructions from and report to the superiors at home’ (Van Schendelen, 2004: 31). From this, it follows that the ‘technicality’ of specific issues affects the accountability relationship between expert and superior. The more expert knowledge an issue requires, the more --Begin page 10-- difficult it is for forums to hold the respective actor to account, and the higher the likelihood of an accountability deficit is. In sum, autonomy and expertise could affect accountability practices, and therefore also the emergence of deficits.

It must be pointed out here that there are various ways to hold individuals to account. An accountability relationship between a principal and an agent can be direct, but can also be influenced by a third party. Bringing the previous discussion about ‘police patrol’ and ‘fire alarm’ arrangements back to mind (McCubbins and Schwartz, 1984), police patrols are structured and active forms of oversight, while fire alarms are less structured and passive forms of oversight. It could therefore very well be that principals do not actively monitor the agents’ behaviour, but rather rely on others to notify them when things go badly wrong.

Even when the structure of an accountability relationship resembles that of a police patrol arrangement, this does not necessarily mean that it also works accordingly in practice. Even though the agents can be obliged to send reports to their principals on every single committee meeting they attend in Brussels, this does not necessarily mean that these principals read the reports, let alone that they have the expert knowledge to understand and assess the situation on the basis of these reports. In situations like these, a simple police patrol setting is not sufficient because it is unable to effectively monitor behaviour. An (additional) fire alarm setting is then required, in which third, more informed, persons or entities are involved. Then, acts of accountability take place after these third entities, such as stakeholders or perhaps colleagues, trigger the principal to take action. Therefore, even in a perspective which only points at principal-agent relationships, the ‘fire alarm’-function of third entities should not be overlooked. Deficits occur if principals are not informed – be it directly or indirectly through stakeholders or colleagues – of the behaviour of the agent, so that ultimately he is able to go against the wish of democratically elected bodies without risking consequences. Of course these principals, in the end, may not have formulated a position at all, in which case one cannot speak of a deficit: there is no wish to go against in the first place. On the other hand, there is an accountability deficit if agents go against a principals will, if the principal is kept uninformed of actual events, or if the principal is not open to information from the agent.

3. Conclusionto contents list

Researchers who are interested in accountability and comitology have important choices to make. On the one hand, there is a clear gap in research. Many claim that accountability is a problem in this context, but fail to substantiate this. On the other hand, the gap includes several quite distinct aspects of accountability. One could say there are two gaps: one is of accountability at the collective level, while the other is of accountability at the individual level. At the collective level, accountability is about checks and balances and the delegation of authority. Deficits can be found with the actual powers of institutions and the according practices. Researching this level can be particularly burdensome because the institutional arrangements are in flux, with new procedures added and redundant ones removed more or less continuously. At the individual level the focus would be on the delegation of tasks, and deficits may be found with the distribution of expertise and the autonomy of agents. It should be stressed, though, that these factors were suggested by the literature, which itself does not provide any systematic analysis of the emergence of accountability deficits. Researchers should therefore keep an open mind for other possible independent variables that have not been mentioned elsewhere.

Future research should be aware of the issues involved in accountability and comitology research, which I have attempted to summarize in this paper. There is a need for empirical research on both the individual and the collective level.


Referencesto contents list

Andenas, Mads, and Alexander Türk (eds.) (2000), Delegated legislation and the role of committees in the EC, London: Kluwer Law International.
--Begin page 11--

Behn, R. (2001), Rethinking democratic accountability, Washington DC: Brookings Institution Press.

Bergström, Carl Frederik (2005), Comitology: delegation of powers in the European Union and the committee system, Oxford: Oxford University Press.

Bovens, Mark (2006), “Analysing and assessing public accountability. A conceptual framework”, European Governance Papers (EUROGOV), No. C-06-01, http://www.connex-network.org/eurogov/pdf/egp-connex-C-06-01.pdf

Bovens, Mark, and Paul ‘t Hart (2005), “Publieke verantwoording: zegen en vloek”, in: Wieger Bakker and Kutsal Yesilkagit (eds.), Publieke verantwoording, Amsterdam: Boom, pp. 245-264.

Christiansen, Thomas, and Emil Kirchner (eds.) (2000), Committee Governance in the European Union, Manchester: Manchester University Press.

Day, P. and R. Klein (1987), Accountabilities: five public sectors, London: Tavistock.

Dehousse, Renaud (2002), Misfits: EU Law and the transformation of European governance, Jean Monnet Working Paper, http://www.jeanmonnetprogram.org/papers/02/020201.html

Demmke, Christoph (2000), “Comitology in the environmental sector”, in: Mads Andenas and Alexander Türk (eds.), Delegated legislation and the role of committees in the EC, London: Kluwer Law International, pp. 279-302.

Egeberg, Morten, Günther Schäfer and Jarle Trondal (2003), “The many faces of EU Committee governance”, in: West European Politics, Vol. 26: 3, pp. 19-40.

European Commission (2006a), “Register of Comitology”,

European Commission (2006b), “Report from the Commission on the working of committees during 2005”, COM (2006) 446, Brussels, http://eur-lex.europa.eu/LexUriServ/site/en/com/2006/com2006_0446en01.pdf

Hofmann, Herwig C.H., and Annette Elisabeth Töller (1997), “Zur Reform der Komitologie: Gen-Mais in der ‘Filet-Falle’, Working Paper, Hamburg: Hamburg University.

Hopenhayn, Hugo, and Susanne Lohmann (1996), “Fire-alarm signals and the political oversight of regulatory agencies”, in: The Journal of Law, Economics and Organization, vol. 12:1, pp. 196-213.

Huber, John (2000), “Delegation to civil servants in parliamentary democracies”, in: European Journal of Political Research, vol. 37, pp. 397-413.

Joerges, Christian and Jürgen Neyer (1997), “Transforming strategic interaction into deliberative problem-solving: European Comitology in the foodstuffs sector”, in: Journal of European Public Policy, Vol. 4: 4, pp. 609-625.

Kassim, Hussein (2000), “The national co-ordination of EU policy: confronting the challenge”, in: H. Kassim, B. Guy Peters and V. Wright, The national co-ordination of EU policy: the domestic level, Oxford: Oxford University Press, pp. 235-264.

Keohane, Robert (2002), “Political accountability”, Paper prepared for conference on delegation to international organizations, Park City, Utah, May 3-4, 2002.

Larsson, Torbjörn (2003a), Precooking in the European Union – the world of expert groups, Stockholm: ESO, http://www.regeringen.se/content/1/c6/03/63/05/17549c2a.pdf
--Begin page 12--

Larsson, Torbjörn (2003b) ’The Different roles played by National civil servants in the European committees’ in C. Demmke and C. Engel (eds.) Continuity and Change in the European Integration Process, Maastricht: EIPA.

McCubbins, Mathew, and Thomas Schwartz (1984), “Congressional oversight overlooked: police patrols versus fire alarms”, American Journal of Political Science, Vol. 28: 1, pp. 165-179.

Mulgan, Richard (2000), “’Accountability’: an ever-expanding concept?”, in: Public Administration, Vol. 78: 3, pp. 555-573.

Peters, B. Guy, and Vincent Wright (2001), “The national co-ordination of European policy-making”, in: Jeremy Richardson (ed.), European Union, 2nd Edition, London: Routledge, pp. 155-178.

Pollack, Mark (2003), The engines of European integration: delegation, agency, and agenda setting in the EU, Oxford: Oxford University Press.

Pollitt, Christopher (2003), The essential public manager, London: Open University Press/McGraw-Hill.

Radaelli, Claudio (1999), “The public policy of the European Union: whither politics of expertise?”, in: Journal of European Public Policy, Vol. 6: 5, pp. 757-774.

Rhinard, Mark (2003), Committees in the European Union: an empirical and normative assessment, Paper presented for the EUSA 8th International Biennial Conference, Nashville, Tennessee, USA, March 27-29 2003.

Sannerstedt, Anders (2005), “Negotiations in EU Committees”, in: Ole Elgström and Christer Jönsson, European Union Negotiations: processes, networks and institutions, London: Routledge, pp. 97-114.

Schäfer, Günther (1996), “Committees in the EC policy process”, in: Robin Pedler and Günther Schäfer (eds.), Shaping European law and policy, Maastricht: European Centre for Public Affairs, pp. 3-23.

Schäfer, Günther (2000), “Linking member state and European administrations”, in: Mads Andenas and Alexander Türk (eds.), Delegated legislation and the role of committees in the EC, London: Kluwer Law International, pp. 3-24.

Strom, Kaare (2000), “Delegation and accountability in parliamentary democracies”, in: European Journal for Political Research, Vol. 37, pp. 261-289.

Türk, Alexander (2000), “The role of the Court of Justice”, in: Mads Andenas and Alexander Türk (eds.), Delegated legislation and the role of committees in the EC, London: Kluwer Law International, pp. 217-253.

Türk, Alexander (2003), ‘Transparency and comitology’, in C. Demmke and C. Engel (eds.) Continuity and Change in the European Integration Process, Maastricht: EIPA.

Van Schendelen, Rinus (ed.) (1998), EU Committees as influential policy makers, Aldershot: Ashgate.

Van Schendelen, Rinus (2004), “The EU as negotiations arena: diplomats, experts, and PAM professionals”, in: Paul W. Meerts and Franz Cede (eds.), Negotiating European Union, Basingstoke: Palgrave Macmillan, pp. 17-34.

Vos, Ellen (1997), “The Rise of Committees”, in: European Law Journal, Vol. 3:3, pp. 210-229.

Vos, Ellen (1999), "EU Commitees: The evolution of unforeseen institutional actors in European product regulation", in: Christian Joerges and Ellen Vos (eds.), EU Committees: Social regulation, law

--Begin page 13--


Endnotesto contents list

(*) This paper forms part of the research ‘Multilevel governance and public accountability in Europe: Which institutions, which practices, which deficit?’, funded by NWO (the Netherlands Organisation for Scientific Research), and led by prof. dr. M. Bovens, prof. dr. D. Curtin and prof. dr. P. ‘t Hart. This project is registered under no. 450-04-319 in NWO’s ‘Shifts in Governance’-programme.

(1) 15 exploratory interviews were conducted with stakeholders (3), member state representatives (8), Commission officials (3) and a member of European Parliament (1). These included the policy areas of Transport (5), Agriculture (6) and Environment (3).

(2) Some argue accountability can be equated with ‘responsiveness’ or ‘responsibility’ (see Mulgan 2000: 566-569 for a clear summary) by indicating that responsiveness towards the general public or organized interests is essentially the same as being accountable to political actors and eventually the general public. However, this covers only one of the entities to which account can be rendered: the general public. As will be shown onwards, there are more ways to render account.

(3) See e.g. Bergström (2005) for a historical account, or Christiansen and Kirchner (2000) and Jörges and Vos (1999) for general overviews.

(4) Interviews with member state representatives and Commission officials dealing with comitology committees. See footnote 2 for more details.

(5) Article 230, TEU, and interview with a Member of European Parliament

(6) See the interinstitutional agreement between Commission and Parliament on procedures for implementing Council Decision 1999/468/EC, OJ 2000 L 256/19.

(7) Interestingly, 74 per cent believes that their colleagues from other member states act mainly as government representatives. In an analytical sense this does not match the 60 per cent who to a great extent feels allegiance to their own professional background and expertise. This would imply that many respondents of this survey feel different about themselves than about their colleagues. This could point either towards reliability problems in the data set, or towards a self-serving bias among the respondents.

(8) See also Demmke (2000: 300) who, in an analysis of environmental committees, concludes that the decisions national executive experts take ‘have an enourmous impact on the national legal, political and economic systems of the Member States’.


©2007 by Gijs Jan Brandsma
formated and tagged by KH, 16.5.2007



European Integration online Papers | ISSN 1027-5193
© 1997-2014 ECSA Austria | All Rights Reserved. Webmaster Editorial Assistant