Introduction

The accountability of independent agencies has recently received considerable academic attention in Europe. On the one hand, this can be explained by the proliferation of European agencies in the last decade; today there are approximately 35 bodies that fall under the umbrella of ‘independent European agencies’. On the other hand, much attention comes from the portrayal of the accountability of independent agencies as an inherent dilemma: ‘independent, hence unaccountable’.See, e.g., P. Magnette, ‘The Politics of Regulation in the European Union’ (p. 12) and E. Vos, ‘Independence, Accountability and Transparency of European Regulatory Agencies’ (p. 126) in: D. Geradin, R. Muňoz and N. Petit, Regulation through Agencies in the EU: A New Paradigm of European Governance, Edward Elgar 2005; M. Everson, ‘Independent Agencies: Hierarchy Beaters?’ 1 European Law Journal 2, 180 (1995), p. 183. Yet, to what extent is that the case?

This article questions the causal relationship between agencies’ accountability problems and their ‘independence’. It argues that the label ‘independent’ creates misleading expectations with regard to agencies’ accountability and shows that the debate on the executive’s accountability should be extended beyond the somewhat attractive case of independent agencies because agencies’ accountability problems are not caused by their ‘independence’.

To this end, the article looks at the concept of ‘independence’ in its first part. It makes an enquiry into the meaning of the term ‘independent’ and tests independent agencies on four independence criteria distinguished in literature, institutional, personnel, financial, and functional. It shows that independent agencies are not that ‘special’ in the sense of being ‘independent’ and are better described using a less misleading term in order to avoid various misperceptions, such as ‘independent, hence unaccountable’. In its second part, this article looks at agencies’ accountability using Mulgan’sR. Mulgan, Holding Power to Account: Accountability in Modern Democracies, Palgrave 2003. and Bovens’M. Bovens, ‘Analysing and Assessing Accountability: A Conceptual Framework’, 13 European Law Journal 4, 447 (2007). definitions and stages of accountability, information, discussion and rectification. It demonstrates that independent agencies can be held to account, and if they are not accountable, it is not a simple result of their independence. This is important because academic attention seems to be misbalanced in the debate on the executive’s accountability in favour of independent agencies upon a false premise. The article uses a comparative legal approach and focuses on independent agencies in the EU and the US to exemplify the discussed issues.The choice of approach and jurisdictions corresponds to the choices made in the course of the ongoing PhD project of the author of this article. The article does not claim to make generalised conclusions; however, certain findings could also be valid, e.g., for parliamentary systems, especially if one considers the establishment of a direct relation between parliaments and independent agencies. Although the approach is limited to legal issues, an attempt is made to overcome, to a certain extent, its constraints by including empirical findings and the literature of other disciplines, such as political science and public administration.

1.Independent Agencies?

Are independent agencies independent? This depends on how one defines the term ‘independent’. Does it imply that an agency is situated in a separate building and enjoys a formal status of legal personality? Or does it mean that an agency can do anything it wants and that it does not have to report back in any form? In the Oxford English dictionary, the first three explanations of the word ‘independent’ are the following: ‘not depending upon the authority of another, not in a position of subordination or subjection; not subject to external control or rule.’http://dictionary.oed.com/cgi/entry/50115165 (last check October 2010). Therefore, an agency that does anything it wants and does not have to report in any form seems to suit this definition best. But to grasp the essence of the term ‘independence’, let us look at an example contrasting the terms ‘independent’ and ‘autonomous’.

International law offers an illustrative difference. Think about an ‘independent’ state and an ‘autonomous’ region.A similar idea was expressed by Groenleer. Also, he chooses to use the term ‘autonomous’ instead of ‘independent’ with regards to European agencies because ‘an autonomous actor is granted a level of autonomy by other actors or will attempt to ascertain a degree of control over his or her own affairs, but this does not mean that he or she is completely free, without restrictions, independent.’ (M. Groenleer, The Autonomy of European Union Agencies: A Comparative Study of Institutional Development, Delft: Eburon 2009). An independent state is in the words of the Oxford dictionary ‘not depending upon the authority of another, not in a position of subordination or subjection and not subject to external control or rule.’ Although today’s realities make it difficult for a state to stay outside international and regional blocs and organisations, an independent state does not have to participate in international cooperation against its will. North Korea is an example of an independent state that does not take an active part in international cooperation. No one can force it to involve itself because it is independent,Here one could also add the term ‘sovereign’, but this example focuses on the contrast difference between the terms ‘independent’ and ‘autonomous’. and if someone tries to do so against its will (e.g. by war), its independence will be broken. Another example is the American War of Independence, which was fought to gain independence and not to become, for instance, an autonomous region within the British Empire. This is because an autonomous region ‘depends upon the authority’ of some sort of central government, and is thus ‘subject to external control or rule’. There are clearly different degrees of autonomy, an autonomous region in Spain may enjoy more autonomous powers than, for example, an autonomous region in Russia. But, if autonomy can be measured by various degrees, the term ‘independent’ seems to be an overarching notion of a complete absence of any dependence.

Another example, which illustrates contrasting nuances between the terms ‘independence’ and ‘autonomous’ comes from their translation and use in other languages. In Dutch, for example, the counterparts of the US independent agencies are the so-called zbo’s: ‘zelfstandige bestuursorganen’, implying autonomous rather than independent (‘onafhankelijke’) agencies. The Russian equivalents for the term ‘independent’ and ‘autonomous’ are ‘независимый’ and ‘самостоятельный/автономный’ accordingly. There is a world of difference between what these two terms imply. Both the Dutch and Russian terms for ‘autonomous’ imply a certain degree of autonomy and a certain degree of supervision. These terms would be used when talking, for example, about a child that goes to a shop alone. While the child does something on his own, this is because he is allowed to do that by his parents. Similarly, independent agencies have their functions to be performed at their discretion, but they are not independent if they must provide an account of these actions to an institution for the purposes of oversight. As long as there is an action that ‘must’ be performed by an agency, it is not independent.

Therefore, from the points of view of definition and linguistics the term ‘independent’ seems to imply a complete absence of any dependence, it is an overarching notion that excludes any oversight authority over an independent body. To determine to what extent independent agencies in the EU and the US meet this definition (1.3.) let us examine the meaning of the term ‘independence’ in the context of these jurisdictions (1.1.) and test independent agencies of both jurisdictions with the help of the four elements of independence established in literature (1.2.).

1.1.Independence in Context: US and EU

What is an independent agency in the US and in the EU?

In the US, the term ‘independent’ implies limited involvement of the US President in agencies’ operation. Independent agencies have been created to ‘bring together individuals of diverse views, expertise, and backgrounds to tackle legally difficult, technically complex, and often politically sensitive issues.’M.J. Breger and G.J. Edles, ‘Established by Practice: The Theory and Operation of Independent Federal Agencies’, 52 Administrative Law Review, 1111 (2000). p. 1112. To this end, they have been created to be insulated from political interference,A. Morrison, ‘How Independent are Independent Regulatory Agencies?’, Duke Law Journal (1988). p. 253. where the restriction of the President’s control has become the means to achieve their independence. While all executive agencies and departments (ministries in the European tradition) are subordinated to the President, the President’s control over independent (regulatory) agencies is somewhat limited by the restriction on the President’s power to remove their board members.‘Congress has employed many different forms of governmental authority in allocating the day-to-day work of government. <…> The diversity is characteristic of our pragmatic ways with government, reflecting the circumstances of the particular regulatory regime, the temper of presidential/congressional relations at the time, or the perceived success or failure of an existing agency performing like functions, more than any grand scheme of government.’ (P.L. Strauss, ‘The Place of Agencies in Government: Separation of Powers and the Fourth Branch’, 84 Columbia Law Journal 3, 573 (1984), p. 583-585). In addition, some of them avoid other contact with the President’s administration, such as sending budgetary drafts to the President’s Office of Management and Budget (OMB) and being represented in courts by the Department of Justice.Breger and Edles (2000), p. 1151-1153, supra fn 8.

In the EU context, the meaning of independence is not as clear-cut as in the US. In its communication proposing the establishment of a common framework for regulatory agencies,There is no clear definition of ‘a European agency’; there are approximately 40 bodies that could be placed under the umbrella of ‘European agencies’. Generally, these bodies can be grouped into five categories: (1) ‘community’ agencies, agencies of the former (2) second and (3) third pillars, (4) executive agencies, and (5) other agencies, e.g., Euratom bodies. This article refers to the agencies of different types in its examples, except the executive type because they are not considered to be ‘independent’. the Commission states: ‘it is particularly important that [agencies] should have genuine autonomy in their internal organisation and functioning if their contribution is to be effective and credible. <…> The main advantage of using the agencies is that their decisions are based on purely technical evaluations of very high quality and are not influenced by political or contingent consideration.’Communication from the Commission on the Operating framework for the European regulatory agencies, COM (2002) 718. p. 5. In the draft interinstitutional agreement on the operation of the regulatory agencies issued some years later the Commission added how independence is to be ensured: ‘granting of legal personality, budgetary autonomy, collective responsibility and own powers of the administrative board, the independence of the director, of the members of the scientific committees and of the boards of appeal, etc.’Draft Interinstitutional Agreement on the Operating framework for the European regulatory agencies, COM (2005) 59, p. 6. The Commission seems to use the terms independence and autonomy interchangeably, but in either case it is unclear whom European agencies are supposed to be independent of.

Barely any founding act talks about whom European agencies are deemed to be independent of. Some founding acts do not even explicitly talk about independence.E.g., the founding act of the European Agency for Safety and Health at Work (EU-OSHA). Moreover, the European Environmental Agency’s (EEA) founding regulation states the following: ‘The Agency should be granted legal autonomy while maintaining close links with the Community institutions and the Member States.’ (Recital 11 of Regulation (EC) No 401/2009). The founding regulation of one of the first European agencies, the European Centre for the Development of Vocational Training (Cedefop), explains that this agency is independent of other departments of the Commission.Regulation (EC) No 337/75. In the case of the European Food Safety Authority (EFSA) ‘the independence of the Authority <…> mean[s] that it should be able to communicate autonomously in the fields falling within its competence, its purpose being to provide objective, reliable and easily understandable information.’Recital 54 of Regulation (EC) 178/2002, as amended. There is only one agency, the European Union Agency for Human Rights (FRA), whose founding act explicitly states that the composition of its board ‘should ensure the Agency’s independence from both Community institutions and Member State governments.’Regulation (EC) No 168/2007.

It seems that the independence of agencies in the context of the EU indeed implies the independence of ‘both the Community institutions and the Member States’ and hence, in a way similarly to the US, the insulation from ‘political’ interference from the EU’s and the Member States’ interests. There are some additional sources that support this idea. In one of its resolutions, the European Parliament proposed concentrating on issues such as agencies’ ‘degree of independence from the Commission as it is often of particular interest to the legislator.’European Parliament Resolution of 21 October 2008 on a strategy for the future settlement of the institutional aspects of Regulatory Agencies (2008/2103(INI)), Point 28. A recent evaluation of decentralised agencies states that ‘agencies are meant to be autonomous from both the Commission and the Member States.’The Ramboll evaluation of the EU decentralised agencies, 2009. Volume I, p. 11 (In the framework of the latest developments on European agencies an interinstitutional group has been set up. The three main institutions, the Commission, the EP and the Council, work together on a common approach towards agencies, so that European agencies of the fourth generation could be set up and function in a clearer (accountability) environment. To start up those discussions the Commission initiated an external evaluation of 26 decentralised agencies, which was exercised by the Ramboll Management-Eureval-Matrix and was published in April 2010. It will be referred to as ‘the Ramboll evaluation’.).And some scholars, for example, Vos, also say that ‘independence is generally considered to be free of both political and industry interests. In the Community context this also refers to national interests.’Vos (2005), p. 123, supra fn 1. So, this article relies on the concept of European agencies’ independence from the Community institutions and from the Member States.

All in all, the explanations of the term ‘independent’ in the context of the two jurisdictions speak in favour of the argument that the agencies are not independent. This is because if agencies were indeed independent, these contextual explanations would have been unnecessary. Agencies would be independent from everyone and everything. However, independent agencies in the US enjoy some distance from the President, but ‘certainly it does not mean independent of Congress.’M. Shapiro, ‘Independent Agencies: US and EU’, Jean Monnet Chair Papers, 34, 1996, p. 8,‘While public discussion of agency “independence” usually focuses on allegations of improper White House influence, the most powerful and persistent “political” influence over <…> [independent agencies] clearly originates with the congressional appropriations and oversight committees and with other important members of the legislature.’ (R.E. Wiley, ‘‘Political’ Influence at the FCC’, Duke Law Journal (1988), p. 282). European agencies are supposed to be independent from the Union institutions and the Member States, but there are mechanisms of control and of accountability that the Commission, the Council and the European Parliament may exercise. And this will be demonstrated next.

1.2.‘Independence test’ for the US and EU Agencies

The existing literature brings the analysis of independence of agencies down to four criteria, institutional, personnel, financial, and functional.See, e.g., B.M.J. van der Meulen and A.T. Ottow, Toezicht op markten, Den Haag: Boom Juridische uitgevers 2003, p. 51-56 and Groenleer (2009), p. 32, supra fn 6; exact formulations vary.Institutional independence means that an agency constitutes a separate institutional unit, so that it is not a part of or subordinated to a ministry or department. In addition, one needs to consider the possibility of abolishing an agency in order to fully understand how independent the agency is from the institutional point of view. The (re)appointment and removal procedures and influences from the respective authorities upon agencies’ heads are important in determining the personnel’s independence. A separate budget and autonomy in financial matters is implied by financial independence. Finally, in absolute terms functional independence means that an agency does whatever it wants. According to some scholars this element basically determines agencies’ independence, ‘agency autonomy is seen in policymaking and implementation.’M. Thatcher, ‘The Third Force? Independent Regulatory Agencies and Elected Politicians in Europe’, 18 Governance 3, 347 (2005), p. 369. It is on the basis of these four criteria that independent agencies in the EU and the US are to be tested.

• Institutional independence
  Does having legal personality and
  putting an agency in a separate building
  qualify as agencies’ independence?

Institutional independence implies that an agency is set up as a separate organisation so it is not part of a ministry. It enjoys its own legal personality, which allows it to function on its own implying that, among other things, it has the possibility of acquiring property, entering into contractual relations with other organisations and private parties, becoming a member of international organisations and representing itself in legal proceedings. ‘Without a formally autonomous position, agencies may easily be terminated or abolished when the political tide changes.’D.E. Lewis, Presidents and the Politics of Agency Design. Political Insulation in the United States Government Bureaucracy, 1946-1997, Stanford University Press 2003. (cited by Groenleer (2009), p. 32, supra fn 6). ‘Once an organisation’s founders have endowed it with legal personality, it is difficult to alter this status.’K. Verhoest, B.G. Peters, G. Bouckaert and B. Verschuere, ‘The Study of Organisational Autonomy: A Conceptual Review’, 24 Public Administration and Development, 101 (2004), p. 106 (cited by Groenleer (2009), p. 32, supra fn 6). From a legal point of view, however, the procedure of how an agency with legal personality can be set up and abolished and what constraints there are matter very much when deciding whether it is difficult to alter an agency’s status or not. Additionally, the agency that is set up by a constitutional or treaty provision enjoys much more institutional autonomy than an agency established by secondary legislation; this is because the Constitution or Treaty is harder to change. Thus, let us analyse independent agencies in the EU and the US taking into account the specified factors, separate institution and legal personality, and procedures and constraints on changing the institutional status.

Independent agencies in both the EU and the US are distinct institutions. Independent US commissions are separated from executive agencies and departments subordinated to the President.In part this is how the first US independent commission, the Interstate Commerce Commission, received its ‘independence’: the authority of the Secretary of the Interior over it, including regulation of salaries and receiving its reports, was abolished. (Breger and Edles (2000), p. 1128-1129, supra fn 8). European agencies are also separate institutions. Although the exact wording may vary, a typical clause of European agencies’ founding regulations is: ‘the Agency shall be a body of the Community. It shall have legal personality.’ Next to that, they are also ‘physically independent’, because they are dispersed geographically across the whole Union,Only executive agencies (pursuant to Council Regulation 96/2003) and few other agencies, like the agencies of the former second pillar, have their seat in Brussels. although this is part of the idea of ‘bringing the Union closer to the citizen’, rather than of promoting agencies’ independence.

The separate institution and legal personality are supported by all kinds of rights that agencies may have. ‘In each of the Member States [European agencies] <…> shall enjoy the most extensive legal capacity accorded to legal persons under their laws. [They] <…> may in particular, acquire or dispose of movable and immovable property and may be a party to legal proceedings.’Regulation (EC) No 1406/2002. US agencies may possess similar rights, though for some agencies certain issues are regulated by the President’s administration,These issues may include agencies’ property, contracts, employment practices, allowances and schemes, and the protection of national secrets. (Strauss (1984), p. 587, supra fn 10). and it is the Department of Justice that represents the federal government in the courts, and this applies to some independent agencies as well.This is provided by law (28 USC § 516), though with a reservation: ‘Except as otherwise authorized by law, the conduct of litigation in which the United States, an agency, or officer thereof is a party, or is interested, and securing evidence therefor, is reserved to officers of the Department of Justice, under the direction of the Attorney General’ (emphasis added). Some agencies however, represent themselves, e.g., the Federal Trade Commission; Strauss provides some more examples. (Strauss (1984), p. 594, supra fn 10). In this sense, European agencies may (at least formally) enjoy more rights with regards to institutional independence, than their American counterparts.

Institutional independence is not only about granting separate legal status and the right to reside in a separate building, but perhaps more importantly, it is also about how easy or difficult it is for an agency to be abolished. What are the procedures and constraints in this respect in the EU and US?

First, regarding the procedure. In both the US and the EU independent agencies are not foreseen in their ‘constitutions’. In both jurisdictions the legislator makes use of the broad existing ‘constitutional’ clauses, roughly speaking ‘necessary and proper’ clauses.For the US it is Article I § 8 of the Constitution (last clause) and for the existing European agencies it is used to be Article 308 of the Treaty establishing the European Community (TEC), but also other ‘sectoral’ provisions that have open formulations (e.g., Article 95 TEC). So, agencies are established when it is necessary to implement a policy and when the legislator needs to deal with a critical situation.The European Food Safety Authority (EFSA), for instance, was created after a series of food crises in the 1990s. In the course of the current financial crisis the creation of new relevant agencies is being discussed in both the EU (e.g., credit rating agencies) and the US (the Consumer Financial Protection Bureau). Independent agencies established by secondary legislation can also be abolished by it. Procedurally speaking, it is not necessarily difficult if all institutions taking part in the legislative process want it. Clearly, the involvement of different institutions, Congress and the President in the US and the Commission and the European legislator in the EU,Since the Lisbon Treaty the European Parliament’s co-legislating scope has expanded from 44 to 87 areas, and the budgetary power has been equally shared between the European Parliament and the Council. (Building Parliament: 50 Years of European Parliament History, 1958-2008, European Communities, 2009, p. 136-137) So, from the functional and representative perspectives one can talk about the appearance of the European or Union Legislator, which is by the way also mentioned in Articles 4 and 7 §3 of Protocol II to the Treaty on the Functioning of the European Union (TFEU). and politics within and between them may make the actual situation difficult. Yet, if there is obvious misbehaviour or inefficiency and a common desire to abolish an agency, it can be abolished as ‘easily’ as it was created. In comparison, the establishment of the European Central Bank (ECB) is regulated by the Treaty. The procedure for abolishing the ECB is hence more complex (especially recalling the adoption of the Lisbon Treaty), than passing a piece of secondary legislation.

Second, regarding the possible constraints. Groenleer states that European agencies are ‘not easily abolished. <…>. Their constituent acts typically do not contain sunset clauses,’Groenleer (2009), p. 19, supra fn 6. meaning an agencies’ length of life. It is true that such clauses are not foreseen by founding regulation,The Euratom Fusion of Energy agency is the only example of a European agency with the sunset clause. It is created for 35 years. (Article 1 (1) of Council decision 2007/198/Euratom). evaluation and review clauses are however more frequently provided in acts establishing European agencies. The European Network and Information Security Agency (ENISA) and the European Chemicals Agency (ECHA) each have a clause in their founding documents which is somewhere between a review and a sunset clause. These agencies’ regulations contain the date when their existence is to be reconsidered, but leaves the opportunity for an extension of their lives open if they prove to be necessary. In the case of ENISA its mission was prolonged in 2007, whereas ECHA will face this question in 2012. The US Congress also uses sunsets in some cases, such as to test the effectiveness of a programme, the appropriateness of allocated funds or the compliance of an agency with a particular delegated function.W.T. Lifland, ‘Sunset Review – Effective Oversight Tool or New Political Football?’, 32 Administrative Law Review, 209 (1980). But, legally speaking, the absence of such clauses in founding acts is not an obstacle to reconsidering the agency and passing ‘abolishing’ legislation.

Agencies could also be considered temporary bodies because they are created with a specific purpose. Despite the fact that their founding acts may lack an explicit sunset clause, a European agency ‘is simply an instrument of policy implementation.’G. Majone, ‘Managing Europeanization: The European Agencies’, in: J. Peterson and M. Shackleton, The Institutions of the European Union, Oxford University Press 2006 (2nd edition), p. 197. If the instrument is not necessary anymore or does not work, it can be modified, replaced or abolished. ‘An agency should not be an agency forever. If it is not needed anymore, it does not live up to the tasks, then we might consider changing the structure.’From my conversation with Mrs Jensen, Member of the European Parliament’s Committee on Budgets, substitute in the Committee on Transport and Tourism, member of interinstitutional group on agencies, August 26, 2010.

Reconsideration of an agency’s existence is a potential removal mechanism. Congress does not have to use this instrument against agencies too often, the fact that it has done it a few times in the past and that it can do so at any moment already makes it a good incentive for agencies to work hard and be more attentive to Congress,The President is of course also involved in the legislative procedure by signing the bill, but in case of independent agencies he may have fewer incentives (if any) to protect an agency he does not control. this clearly undermines an agencies’ institutional independence. Congress closed the first independent agency, the Interstate Commerce Commission, in 1995, most likely due to the absence of functional necessity. And although the abolition of the Civil Aviation Board (CAB) occurred at a time of de-regulation and was most likely due to economic reasons, maladministration also played a role.M. Derthick and P.J. Quirk, The Politics of Deregulation, Washington: The Brookings Institution 1985. p. 43-44. Finally, Congress can (temporarily) abolish an agency even without repealing its founding statute. The Administrative Conference of the United States (ACUS), an advisory independent agency, did not operate from 1995 until 2010 because it was defunded. At the time of writing, only one European agency, the European Agency for Reconstruction (EAR), has been dissolved on the grounds of ‘needs satisfied.’The Ramboll evaluation, Volume III, p. 222, supra fn 20. Although the Ramboll evaluation states that ‘the existence of established agencies is almost never reconsidered,’ it also concludes that many tasks that agencies perform are still relevant, though a merger of some agencies may be considered.For example, the two ‘police agencies’, the European Police College (Cepol) and the European Police Office (Europol), could have been merged, especially considering the problems that Cepol experiences at the moment; financial mismanagement (see the decisions postponing and refusing its discharge (C7-0198/2009 – 2009/2127(DEC) and (C7-0198/2009 – 2009/2127(DEC) accordingly) and inefficient administration, because it has 27 members on its management board and employs only 24 staff. The only reason explaining the mentioned cases is that each member state tries to get a ‘piece of the EU cake’; however, there should certainly be more efficient methods of ‘dividing the EU cake’. But again, the possibility of abolishing agencies via a ‘simple’ legislative procedure is there, and this has a great impact on agencies’ institutional independence.

To sum up, independent agencies of both jurisdictions are separate institutions enjoying legal personality, which implies certain rights that help them to function on their own, such as the acquisition of property. However, their creation and abolition are not foreseen in the ‘constitutions’, they are set up by ordinary secondary legislation to serve a particular purpose. If the purpose is achieved, if the agency is no longer necessary or performs poorly, it can be abolished by secondary legislation, including budgetary acts. This puts pressure on agencies’ behaviour as they have to prove their necessity and good work, which impacts their institutional independence. In any case, does legal personality status and a seat in another building qualify as agencies’ independence? It seems that other ‘independence factors’ should also be present.

• Personnel independence
  ‘Does the removal “for cause”
  ensure agencies’ independence?’

Personnel independence mainly concerns management boards and agencies’ directors that take final decisions on agencies’ working programmes and budgets. The appointing authority matters, because it is likely to choose loyal people to be able to exercise certain control and influence over the agency’s action. However, ‘once an officer is appointed, it is only the authority that can remove him, and not the authority that appointed him, that he must fear and, in the performance of his functions, obey.’Bowsher v. Synar 478 U.S. 714, 726 (1986). The removing authority is probably the main actor whose influence is difficult to withstand. Let us analyse the influence of the appointing and removing authorities on independent agencies in the EU and the US.

The very meaning of independence of agencies in the US is directly linked to the personnel’s aspects of independence, which ensure their distant place from the President. The President appoints all heads of independent and non-independent agencies according to the same constitutionally prescribed procedure (Article II § 2). He may choose a candidate and propose him to the Senate (the upper chamber of the bicameral Congress) whose consent is required. What makes the difference between executive and independent agencies is that the former is headed by one individual and the latter has multi-member boards. In addition, the multi-member boards have to be bipartisan; normally not more than the majority (e.g., three out of five members of the Federal Trade Commission) can be of the same political colour. The multi-membership and different political affiliations clearly make it more difficult for the President to influence independent agencies’ boards, than a single-headed executive agency. Furthermore, members of independent agencies’ boards normally have a specified term of tenure and appointment procedures, each commissioner ‘serves a term so calculated that the term of only one member expires each year,’P.L. Strauss, Administrative Justice in the United States, Carolina Academic Press 2002, p. 98. so ‘no President in one four year term can appoint a majority of the commissioners.’Shapiro (1996), p. 9, supra fn 22.
However, one should keep in mind that the President may ‘prolong’ favorable to him majority on a multi-member board by not proposing a candidate to a vacant position. Also, the Senate may refrain from confirming a proposed candidate and keep the status-quo on a multi-member board. With time, an agency’s board may thus have more than one vacancy, which can lead to appointments in package deals. Meanwhile, the President can use recess (temporary) appointments in order to ensure the agency remains operational. The National Labor Relations Board (NLRB) is an extremely interesting example in this respect. In the last decade it has seen various compositions of its multi-member board. It had one member for about one month and two members (a democrat and a republican) for 36 months. Interestingly, the Supreme Court has overturned the decisions made by a two-member board on the grounds of the lack of quorum (New Process Steel v. NLRB 8 U.S. 1457 (2010)). This also shows how the power of appointment can undermine the functional ability (and functional independence) of an independent agency. It may thus take time for the President to have a majority loyal to him to exercise a certain influence over an independent agency. So, the appointing authority’s influence on the members of independent agencies is limited. But isn’t it the removing authority and its influence that one ‘must fear’ more?

The President’s ability to remove the heads of independent agencies is restricted. The US Constitution addresses the removal of various officers only via the impeachment procedure conducted by Congress (Articles I §2 and 3 and II §4) whereas the President can remove heads of executive agencies at his will.In Myers v. U.S. (272 US 52 (1926)) Chief Justice Taft, interestingly a former US President, ruled that the President as the Chief of the Executive should be able to remove all executive officers at his will. Using the constitutional lacuna and with the help of the Supreme Court‘The language of the act, the legislative reports, and the general purposes of the legislation as reflected by the debates, all combine to demonstrate the congressional intent to create a body of experts who shall gain experience by length of service; a body which shall be independent of executive authority, except in its selection, and free to exercise its judgement without the leave or hindrance of any other official or any department of the government.’ (Humphrey’s Executor v. U.S. 295 U.S. 602, 625-626 (1935)). Congress has restricted the removal power of the President over independent agencies by imposing different statutory conditions; removal for ‘inefficiency’, ‘neglect of duty’, and ‘malfeasance in office’, which have evolved into an omnibus term: the removal ‘for cause’.

Although this indeed restricts the President’s ability to remove, with the exception of impeachment, nothing in the US Constitution is said about Congressional removal power; it is neither prescribed, nor prohibited. This has led Congress to stretch the boundaries of its legislative power in order to include removal aspects when necessary. In 1930 it simply legislated some commissioners out of office. This is how it ‘reorganised’ the Federal Trade Commission, it terminated their terms by law, which led to the appointment of a whole new board. Cushman stated in this respect that in case of a ‘for cause’ removal provision this remains the only legal way to change the board completely.Cushman R.E. The Independent Regulatory Commissions, New York: Oxford University Press, 1941, p. 450 (referring to Act of June 17, 1930, 46 Stat. at L. 590, 696). Alternatively, Congress may use appropriation riders (to be discussed) as a tool for the removal of unsatisfactory officials. Thus, in 1940, in order to get rid of D.J. Saposs, the chief economist of the National Labor Relations Board, Congress attached a rider forbidding the board to maintain the office, without money there is no post.Ibid, p. 675. Anyhow, the presence of various possibilities for removal ‘for cause’ by the President and ‘without cause’ by CongressAlthough the Supreme Court prohibited explicit statutory removal clauses stating that ‘Congress cannot reserve for itself the power of removal of an officer charged with the execution of the laws except by impeachment’ (Bowsher v. Synar 478 U.S. 714, 715 (1986)), it seems that Congress could use its legislative and budgetary competences as an indirect removal power by restructuring an agency’s board or giving no money for a position. affect the independence of agencies’ personnel on the agencies’ boards.

The situation is slightly different with European agencies. The language of founding acts shows that the representatives appointed by the Commission and the Member States, which is a good picture for a great majority of European agencies’ boards,This is a general picture; the management boards of existing agencies are very different. They may, e.g., comprise various stakeholders (normally without voting rights), representatives of additional non-EU states (normally without voting rights) and representatives designated by the European Parliament. are not supposed to be independent. The founding regulation of the European Monitoring Centre for Drugs and Drug Addiction (EMCDDA) states that its management board consists of ‘one representative from each Member State, two representatives from the Commission, two independent experts <…> designated by the European Parliament’Article 9 of Regulation (EC) No 1920/2006. (emphasis added). Thus, on the surface agencies are not independent of the Member States and of the Commission on the basis of the fact that they must have those representatives.

Digging deeper, one sees that the proportions of their representation differ. The Commission and the Member States appoint and removeThe tenure and removal clauses are not always prescribed by agencies’ founding acts, but presumably each appointing authority removes its representatives. their representatives, so there are potentially 27 representatives of the Member States, which form some sort of ‘mini-Councils’, and one or, as it is usually the case, several representatives of the Commission. Formally speaking, the power-struggle is not in favour of the Commission; having one or two soldiers against a whole army (of 27 people) is not a battle the Commission can win, unless those two soldiers have a decisive advantage. And it seems that they may have one. Some research shows that ‘by virtue of being generally better informed than Member States representatives and having a strong knowledge of the EU legal system and its intricacies’M. Busuioc, The Accountability of European Agencies: Legal Provisions and Ongoing Practices, Delft: Eburon 2010, p. 86. See, also: D. Curtin, Executive Power of the European Union: Law, Practices and the Living Constitution, Oxford University Press 2009, p. 156-157. The Ramboll evaluation also states: ‘the EU interest is under-weighted in the governance arrangements of a few agencies, but that it is at risk of being voiced in a contradictory way in many instances.’ (Volume I, p. 15, supra fn 20). the underrepresented Commission (and at times individual Member States’ representatives) can actually rule the boards and hence agencies. So, neither formally, nor in practice do the management boards of European agencies seem to be independent of the Member States and of the Commission.

European agencies’ founding regulations often talk about the independence of agencies’ directors it is therefore worth considering their independence here. The appointing authority is normally a combination of the management boards and the Commission where the Commission has a stronger selection position.The appointment and removal of agencies’ directors vary greatly from agency-to-agency, so this paragraph makes a very rough generalisation. The term is normally fixed and often can be renewed by the board that can also remove directors. In addition, some regulations prescribe the exercise of ‘disciplinary authority’ over directors for the management boards (or in a few cases the Council), it is not clear what this could mean.‘I have never been the subject of the Council’s disciplinary authority. What does this mean? No one actually knows.’ (Busuioc (2010), p. 121, supra fn 55 (the words of an agency director)). Considering the fact that all the action and outputs (budgetary plans, working programs, and annual reports) are always sent to the board for the approval and the board is his appointing and removing authority, the director is not independent of his management boards. Furthermore, directors are often accountable to the European Parliament for the implementation of the budgets whose negative evaluation may indirectly lead to the director’s removal. There are other possibilities, the Parliament may initiate a hearing within the discharge procedure (to be discussed) and ask for additional information and compliance with a proposed plan attached to its postponing decision for granting the discharge. It could also insert potential sanctions or threats in its discharge decisions, as it has done not so long ago with the European Police College (Cepol) whose discharge decision was postponed in May 2010. The Parliament questioned ‘whether the College’s new Director will be able to address’European Parliament Decision of 5 May 2010 on discharge in respect of the implementation of the budget of the European Police College for the financial year 2008 (C7-0198/2009 – 2009/2127(DEC)) Point 3, observations 4 and 5. specified problems, which is in a way the threat of sanction. The Cepol’s Director responsible for implementation of the budget for 2008 was removed and the postponing discharge decision became the first refusal in history to give to a European agency.European Parliament Decision of 7 October 2010 on discharge in respect of the implementation of the budget of the European Police College for the financial year 2008 (C7-0198/2009 – 2009/2127 (DEC)).

To sum up, the influence appointing and removing authorities have in the US and the EU varies, but they certainly impact the personnel’s independence. The members of the European agencies’ boards represent their corresponding appointing authorities. Thus, the Commission and the Member States and their interests are represented and directly influence agencies’ boards, also they remain the removing authorities for their representatives. In the US, the appointing and removing powers of the President are restricted by law, which may make it difficult for the President to influence independent agencies, there are however, possibilities for Congress to remove them. So, does the restriction ‘for cause’ ensure agencies’ independence?

• Financial independence
  Does the formal requirement of
  having an ‘autonomous budget’
  ensure agencies’ independence?

‘Financial autonomy concerns the extent to which organisations are financially autonomous from external actors. <…> Agencies that generate their own resources are less dependent on their political principals than agencies that rely on them for funding. <…> Organisations that can decide how they spend their financial resources have a high level of financial autonomy. If organisations are restrained or restricted in regard to expenditures, for instance by limiting the possibility of transferring money from one budget item to the other, they have a low level of financial autonomy.’Groenleer (2009), p. 32, supra fn 6. Hence, three criteria are to be considered when discussing European and US independent agencies’ financial independence: autonomous budget, the dependence on political principals for funding, and the freedom of choice of how to spend money.

The great majority of independent agencies of both jurisdictions do not meet these criteria.This article does not discuss few exceptions of US self-financed agencies (Strauss (2002), p. 80, supra fn 45) and few inexplicable cases of self-financed European agencies. Concerning the latter this article refers an interested reader to the study where this and other inconsistencies of application of EU financial rules are analysed: Study of Budgetary Support Unit of the European Parliament ‘Agencies’ Discharge’ (12/12/2006). Formally speaking, European agencies’ independence is often thought to be achieved by providing an autonomous budget. A typical clause of founding regulations reads as follows: ‘in order to guarantee the functional autonomy and independence of the Agency, it should be granted an autonomous budget.’Council Regulation (EC) 768/2005. The cited clause however, normally continues by saying that the agency’s ‘revenue comes from a contribution from the Community as well as from payments for contractual services rendered by the Agency.’ The great majority of European agencies are financed from the EU general budget, which means that they go through the whole procedure of submitting their draft budgets with necessary documentation to the Commission who then sends a general draft to the budgetary authority, i.e., the Council and the European Parliament. The draft budget of agencies can be modified at all those stages by the CommissionTogether with an agency’s budgetary draft the Commission sends normally its recommendation to the budgetary authority; the recommendation may have different estimates than the agency’s draft. and by the budgetary authority.Before the changes brought by the latest treaty it was the European Parliament that had a final say on agencies’ budgets which fell within the non-compulsory part of EU budget. Its specialised committees could decide on the budgets of agencies falling within their jurisdictions and then the Committee on Budgets would listen to the specialised committees’ desires and would make the final choices. Finally, all agencies financed from the general budget are subject to the discharge procedure where they have to prove that they spent appropriated money prudently. Thus, through the annual budgetary and discharge procedures, European agencies are dependent on their ‘political principals’ for funding.

A similar situation applies to independent agencies in the US. They are financed from the federal budget and follow the ‘normal’ annual budgetary procedure. Although the first step of submitting draft budgets to the President’s Office of Management and Budget (OMB) is not followed by all independent agencies,Breger and Edles (2000), p. 1151-1153, supra fn 8. they all send their drafts to the respective congressional committees‘Actually four Congressional committees or subcommittees oversee each agency, a Senate and a House committee responsible for legislation in the agency’s jurisdiction and a Senate and a House committee responsible for its annual appropriations.’ (Shapiro (1996), p. 27, supra fn 22). and then the process begins. Agencies’ representatives are invited to attend numerous hearings. Typically, these hearings are the place where the most important annual mechanism of congressional control is exerted: Congress receives all kinds of information and reports it asks for, there are numerous discussions of programmes, their execution and results, then Congress evaluates agencies’ work, and finally agencies receive ‘appropriate’ amounts of money. In addition, as the independent agencies’ budgets are too small, they are not often the subject of debate outside the relevant congressional committees,Strauss (2002): ‘While debates over general spending levels and major programs such as the defense budget are common in the two chambers, regulatory agency budgets are a small part of overall national expenditures. <...> In 2001, for example, the EPA [Environmental Protection Agency] regulatory budget was about $ 7.3 billion; the Securities and Exchange Commission, about $ 438 million; the Federal Trade Commission, about $ 147 million. In comparison, that year’s defense budget was $ 296.3 billion, the total national budget for the year was $ 1.856 billion.’ (p. 79, supra fn 45) In comparison, the EU-15 budget of 2001 was € 82.5 billion, and the budget of the Netherlands was around € 100 billion. http://ec.europa.eu/budget/library/
publications/fin_reports/fin_report_07_en.pdf (last check February 2010) http://www.cbs.nl/
nl-NL/menu/themas/overheid-politiek/cijfers/extra/belasting-animatie.htm (last check February 2010). thus individual committees and members of Congress may enjoy a great individual power over the budgets of ‘their’ agencies.

Individual members’ or committees’ power can be exercised with the help of, for example, special financial tools, such as appropriation riders and earmarks. Appropriation riders take the form of particular provisions in the appropriation bills that ‘single out a specific regulatory activity and prohibit the expenditure of funds for carrying out that regulatory activity or plan.’J.B. Beermann, ‘Congressional Administration’, 43 San Diego Law Review, 61 (2006), p. 85-86. An example of the appropriation rider is: ‘none of the funds <…> may be used by the Occupational Safety and Health Administration <…> to promulgate or issue any proposed or final standard or guideline regarding ergonomic protection.’ (Omnibus Consolidated Rescissions and Appropriations Act of 1996, Pub. L. No. 104-34, § 102, 110 Stat. 1321 (1996)). The earmarked money is given to agencies to run some parallel programmes that are not directly mandated to agencies as such, but are not explicitly forbidden either. These tools allow (individual) membersThe independence from the President could increase the influence of individual members or committees over agencies’ budget; if the President does not control these agencies, he may have fewer (if any) incentives to argue with the members of congressional committees about agencies’ budgets. of Congress directing agencies’ finances and hence activities by prohibiting certain actions and encouraging others. The agencies’ funding thus depends greatly on, and can be influenced by, political institutions.

Finally, what is financial independence, if agencies may not freely transfer money from one programme to another? In both jurisdictions independent agencies are not allowed to freely transfer without permission. In the EU, pursuant to the principle of specificationChapter 6, Articles 22-24 of Commission Regulation (EC, Euratom) No 2343/2002 on the framework Financial Regulation for the bodies referred to in Article 185 of Council Regulation (EC, Euratom) No 1605/2002 on the Financial Regulation applicable to the general budget of the European Communities. all appropriations are given to specific programmes and all transfers can be made upon notification or permission of the management board. In the US, the money bills may contain lump-sum appropriations, i.e., general headings. They are however, supported by not formally binding but informal compulsory reports on how the money is supposed to be spent, transfers can be made upon an informal authorisation from the relevant congressional committee.S. Streeter, The Congressional Appropriations Process: An Introduction, Congressional Research Service Report prepared for Members and Committees of Congress, December 2008, p. 10.
Find at: http://assets.opencrs.com/rpts/97-684_20081202.pdf (last check October 2010). Informal reports and consent from congressional committees seem to eliminate any freedom to reallocate the money. The absence of the free choice of how to spend money and the dependence on political institutions for funding greatly affects agencies’ financial independence. So, does the formal requirement of having ‘autonomous budget’ ensure agencies’ independence?

• Functional independence
  Do decisions, such as
  ‘determining the number of computers’‘We have reasonable autonomy in day-to-day decision-making, such as determining the number of computers, the choice of free-lances or other suppliers.’ (A. Kreher (ed.), The EC Agencies between Community Institutions and Constituents: Autonomy, Control and Accountability, EUI, The Robert Schuman Centre 1997, p. 102).
  qualify as agencies’ independence?

Functional independence in its absolute term means that an agency can do whatever it wants. According to some scholars, it is in fact the functional element that truly determines agencies’ autonomy; ‘agency autonomy is seen in policymaking and implementation.’Thatcher (2005), p. 369, supra fn 24. Following from that, it can be concluded that although all the criteria for agencies’ independence are in a sense equal, some are more equal than others; without functional independence an agency would lack its ‘real’ independence. To determine whether independent agencies in the EU and the US enjoy functional independence the following questions need to be addressed: what are independent agencies empowered to do and how much discretion do they enjoy in exercising their powers? Can they do whatever they want?

Without keeping the reader waiting, the answer is negative. In neither of the two selected jurisdictions can independent agencies do whatever they want to and even if they enjoy certain functional discretion, there are various mechanisms to check and to influence agencies’ decisions and to hold them to account for the decisions they take.

With respect to powers that agencies enjoy, independent agencies in the EU and the US hardly withstand a comparison. If one could illustrate the difference between their formal powers with that of the strength of two football teams, one would need to compare San Marino and Brazil. It does not mean that San Marino is very bad, it is just that Brazil is too good. US independent agencies are very powerful. They mainly enjoy two functions, rulemaking and adjudication.There are also some US independent agencies with information-gathering and advisory functions, such as the Administrative Conference of the United States (ACUS) that was created ‘for the purpose of developing recommendations to improve the fairness and effectiveness of the rulemaking, adjudication, licensing, and investigative functions of federal agency programs.’ (http://acus.gov/ (last check October 2010)). Rulemaking is a kind of legislative power;This terminology is however very tricky in the US context. (See, e.g., Whitman v. American Trucking Associations 531 U.S. 457 (2001)) Pursuant to the ‘non-delegation’ doctrine the legislative power can be exercised only by Congress. The rulings of the Supreme Court use various terms, including administrative, rulemaking, decision-making and ratemaking, referring to agencies’ functions. agencies make rules concerning the subjects of the policy they regulate. Rules are of general applicability and of future effect. Adjudicatory decisions concern, in contrast, past individual behaviour. As an example, the Security and Exchange Commission’s requirement that companies that sell securities should disclose specified information and register with the SEC could be a rule, and ‘the Nuclear Regulatory Commission’s decision to assess a $ 2 million penalty against an electric utility for rule violations in running a nuclear power plant’Strauss (2002,) p. 199, supra fn 45. is an adjudicatory decision.

Only a few European agencies enjoy similar functions.These are the Office for Harmonization in the Internal Market (OHIM), the European Aviation Safety Agency (EASA), the Community Plant Variety Office (CPVO) and the European Chemicals Agency (ECHA). The Office for Harmonization of Internal Market (OHIM) for example, is ‘the official trade marks and designs registration office of the European Union.’http://oami.europa.eu/ows/rw/pages/OHIM/index.en.do (last check September 2010). The registrations it issues give the applicant the right to protect the commercial origin of a product. By registering trademarks and designs in individual cases the agency regulates the protection of intellectual property rights in the internal market. Furthermore, there is a middle category of agencies that give advice on certain decisions, but the Commission takes the formal decision. This is, for instance, the case with the European Medicines Agency (EMEA) whose positive opinion ‘can result in the granting of an EU-wide marketing authorisation by the European Commission.’‘EU agencies: Whatever you do, we work for you’, the European Commission, 2007, p. 22. Finally, next to these somewhat more powerful decision-making and advisory agencies, there are agencies with information, cooperation, and service types of function; the names of their functions speak for themselves. An ‘information agency’ is, for instance, the European Agency for Safety and Health at Work (EU-OSHA) that ‘collects, analyses and communicates’Ibid, p. 6. information across the EU. The European Union’s Judicial Cooperation Unit (Eurojust) is an illustrative example of an agency with cooperation tasks; its mission is to improve and stimulate the coordination of investigation and prosecution between the Member States. A typical service oriented agency is the Translation Centre for the Bodies of the European Union (CdT) that provides translations of various documents, probably one of the most important administrative services for the bodies of a Union with 23 official languages. Regarding their formal powers, the great majority of European ‘regulatory agencies [are] without regulatory powers’G. Majone, Dilemmas of European Integration: The ambiguities and pitfalls of integration by stealth, Oxford University Press 2005, p. 92-93. and cannot be put on equal footing with their US counterparts.Van Ooik, for example, concludes in this respect that the importance of European agencies should not be exaggerated at the moment, at least before they have been delegated with ‘more intense responsibilities’. (R. van Ooik, ‘The Growing Importance of Agencies in the EU: Shifting Governance and the Institutional Balance’, in: D. Curtin and R. Wessel (eds.) Good Governance and the European Union, Intersentia 2005, p. 152).

The powers that independent agencies in the EU and the US have are very much related to the ‘non-delegation’ doctrines that these jurisdictions have. The US Supreme Court’s standard of delegation (‘intelligible principle’‘If Congress shall lay down by legislative act an intelligible principle to which the person or body authorized to fix such rates is directed to conform, such legislative action is not a forbidden delegation of legislative power.’ (J.W. Hampton, Jr., & Co. v. U.S. 276 U.S. 394, 409 (1928)).) is very broad,The fact that there are two cases in history overruling delegation, both decisions issued in 1935, also tells a lot. (These are Panama Refining Co. v. Ryan 293 U.S. 388 (1935) and Schechter Poultry Corporation v. U.S. 295 U.S. 495 (1935)). but it requires that Congress prescribes procedural and substantive standards for agencies to follow when exercising their delegated powers. Congress has passed numerous statutes with the purpose of regulating agencies’ behaviour. All these statutes could be put under the ‘APA +’ umbrella.The Administrative Procedure Act (APA) passed in 1946 regulates rulemaking and adjudication procedures; the steps that agencies have to follow when taking their decisions, the internal appeal and judicial review issues. These procedures establish the cornerstone of agencies’ functioning however, Congress has also passed additional statutes to regulate more specific issues. As they often supplement and are sometimes codified within the APA, they seem to form a common system of agencies’ operation, which I call the ‘APA+’ body of laws. Some of these laws are mentioned later in Section 2.1. See more details of statutes that regulate the rulemaking procedure, e.g., in: J.A. Lubbers, Guide to Federal Agency Rulemaking, Chicago: American Bar Association 2006. The informal rulemaking procedure, for instance, follows the steps of the notice, the commentary and the final decision. All interested parties and individuals may comment on the proposed rule, those comments have to be taken into account by the respective agency, and the substantial comments have to be reflected in the preamble of the final rule.See, e.g., Independent U.S. Tanker Owners Committee v. Dole 809 F.2d 847 (1987). So, US independent agencies cannot do whatever they want, they have to follow the prescribed substantive and procedural requirements laid down by the ‘APA+’ statutes as well as the case-law.The case law fills in statutory gaps and areas of ambiguity thus supplementing the administrative rules of procedure that agencies have to follow. See, e.g., Lubbers (2006), supra fn 85, for the case-law that governs rulemaking. In the EU, the delegation of powers is governed by the Meroni doctrine,Meroni & Co., Industrie Metallurgiche, S.p.A v High Authority of the European Coal and Steel Community, Case 9-56. which imposes a very strict limit on the scope and nature of delegated powers. In short, it allows the delegation of ‘executive powers’ only; such powers cannot involve policy-making discretion. It basically prohibits ‘real’ regulatory agencies. Similarly to the US ‘intelligible principle’ the delegation in the EU has to be supported by procedural guidance and oversight. So, European agencies cannot do whatever they want to do either, but have to follow the procedure and guidance from their founding acts and possible supplementary legislation.

Apart from the powers agencies can enjoy, the question of how much discretion they exercise is of considerable importance for functional independence. What is discretion? ‘Discretion is commonly understood as the net of the powers delegated to agents and the various oversight mechanisms established by principals (i.e. delegated powers – oversight = discretion).’J. Tallberg, ‘Executive Power and Accountability’, in: S. Gustavsson, C. Karlsson and T. Persson, The Illusion of Accountability in the European Union, Routledge 2009, p. 114. Using this formula, also including procedural and substantive constraints, let us look at how much discretion is left to agencies in the US and the EU.

Next to the procedural and substantive constraints prescribed by the ‘APA+’ statutes its oversight function gives Congress the possibility of influencing agencies. Congress can hold ad hoc hearings and make various ‘suggestions’ for agencies’ rules considering (individual) preferences that members of Congress and their constituencies have during various types of formal and informal contact agencies have with the members of Congress. ‘Keeping Congress happy’ is very important for independent agencies,Shapiro (1996), p. 28, supra fn 22. but keeping Congress happy means, among other things, following what Congress desires. Next to this, functional independence is closely related to financial independence. Without money no functions can be performed, and via the annual budgetary procedure Congress may also direct agencies’ activities. Moreover, if something goes wrong, Congress enjoys extensive investigatory powers to find out what has happened and to undertake appropriate action. These are supported by additional tools, i.e., the powers to subpoena witnessesHaving received a congressional invitation (subpoena) an individual has to testify before the body that requested it and has only limited grounds for objections (see, the Senate Rule XXVI (1) and the House of Representatives’ Rule XI (2) (m) (1), (last check September 2010)). and to hold in contempt.Generally speaking, it means imprisonment for a period of time before compliance with the congressional request or as a punishment for non-compliance. (For more information on various types of contempt, see, e.g., Congressional Oversight Manual, CRS Report for Congress, Updated in May 2007, p. 36-38 and L. Fisher, Constitutional Conflicts between Congress and the President, Kansas: University Press of Kansas 2007, p. 160). Finally, pursuant to the Congressional Review Act (CRA) (5 USC §§ 801-808) ‘major agency decisions’5 USC § 804 (2) defines the term ‘major rule’ which means ‘any rule <…> [that] has resulted in or is likely to result in – (A) an annual effect on the economy of $ 100,000,000 or more; <…>.’. have to be submitted to Congress and can be overruled by a legislative act. The prohibition of legislative vetoes‘Legislative veto’ is a provision that Congress may put in its legislative acts prescribing the following procedure. An agency takes a decision and sends it to the relevant committee in Congress. Congress has a certain period of time to review the decision and to react: either Congress simply does nothing, in which case upon expiration of the prescribed period of time the decision enters into force, or Congress issues a resolution of disapproval, in which case the decision is annulled. The Supreme Court declared this practice unconstitutional in INS v. Chadha (462 U.S. 919 (1983)). has resulted in the creation of the CRA; legislative vetoes remain however, in informal practice.L. Fisher, ‘The Legislative Veto: Invalidated, It survives’, 56 Law and Contemporary Problems, 273 (1993).

It is also important to mention that the President’s power over independent agencies is restricted due to the political ‘power struggle’ with Congress, rather than by constitutional constraints. There is, for instance, the ‘opinion’s clause’The President ‘may require the Opinion, in writing, of the principal Officer in each of the executive Departments, upon any Subject relating to the Duties of their respective Offices.’ (Article II § 2). in the US Constitution. It remains a very delicate issue in US constitutional law whether an independent agency is an ‘executive department’, even after the very recent case Free Enterprise Fund et al. v Public Company Accounting Oversight Board et al.561 US _ (2010) Find at: http://www.supremecourt.gov/opinions/09pdf/08-861.pdf (last check October 2010). In this case the Supreme Court ruled that ‘the Commission is a freestanding component of the executive branch, not subordinate to or contained within any other such component, it constitutes a “Department” for the purposes of the Appointment Clause.’Ibid, p. 31. This sentence was supplemented with the footnote: ‘we express no view on whether the Commission is thus an “executive department” under the Opinions Clause.’Ibid. Although the Court is very hesitant in this respect, a number of Presidents have been advised to include independent agencies in their executive orders that they use to govern their administrations.‘Both President Carter and President Reagan were advised that they had authority to include the independents in their executive orders promoting economic analysis of proposed rules as an element of regulatory reform.’ In 1981 Vice President Bush sent a letter asking independent agencies to voluntarily comply with some sections of an executive order ‘to demonstrate to the American people the willingness of all components of the Federal Government to respond to their concerns…’ Several agencies did comply. (Strauss (1984), p. 592-593, supra fn 10). Clinton’s Executive Order 12,866 included, for the first time, the obligation for independent agencies to comply with its content, though only with its planning and agenda provisions (Section 4).The text of the Executive Order is printed in: Lubbers (2006), p. 617-635, supra fn 85. In addition, the President may exercise certain influence over agencies’ policies and agendas via the chairmen of agencies’ boards who are normally the President’s appointees.Breger and Edles (2000), p. 1177-1178, supra fn 8.

So, putting all components into the formula, the discretion of US independent agencies can be described as follows: ‘discretion = delegated powers (which is a considerable amount) – the “APA+” procedural and substantive requirementsThe more precise the legislation is, the more restrictive for discretion this factor is.(including public participation) - congressional oversight (formal (including during the budgetary process) and informal influence) – legislative vetoes and the CRA – internal appeal and judicial review – the President’s influence’. Thus, although US agencies enjoy a lot of powers, the discretion of how to deal with those extensive powers is also restricted by many mechanisms of control and of accountability.

European agencies’ procedures and substantive discretion is regulated on an agency-by-agency basis.Interestingly, as regards the European agencies’ discretion Shapiro stated: ‘A number of the EU independent agencies <…> are described as essentially information gathering agencies. <…> Information gathering operations are purely managerial not political. They involve no discretion. Facts are facts.’ (p. 14, supra fn 22).
However, further he argued that ‘the “information” functions of agencies have policy dimensions of key importance.’ (Shapiro (1996), p. 22, 24, supra fn 22). Generalising here is quite tricky, because the cases are simply too different. However, pursuant to the Meroni doctrine, agencies’ founding acts have to prescribe how agencies have to deal with the functions they are empowered with, and no policy-making discretionary powers can be delegated to agencies. Thus, each agency receives its ‘executive’ competences with the procedures it has to follow. The whole procedure of granting and refusing registration falls within the remit of decision making agencies, such as the OHIM, as can be seen in their founding acts. Also, these decisions are subject to judicial review (Article 265 TFEU). Agencies that provide advice do not take legally-binding decisions. Although there are known cases of recommendations being followed with the highest degree of precision,Groenleer states in this respect that the Commission ‘rubberstamps’ the EMEA’s recommendations without any discussion. (Groenleer (2009), p. 131, supra fn 6.
From my conversation with MEP Jensen (supra fn 39): ‘We did have some trouble a couple of years ago with an aeroplane that crashed. It was with certain type of aeroplanes. It happened in my country, Denmark, and I think there was some trouble in Austria as well. So, in Denmark and in Austria the authorities grounded these aeroplanes, but the European agency said that this type of aeroplane could go on flying; they did not see any problems. And I asked the question to the Commissioner of Transport and said to him “well, what is happening? Is the European agency being too closely linked to the producer ‘bombardier’ or do you not take the same view or you are not really being made aware of the physical things?” And the Commissioner of Transport could only do one thing, he read up a piece of paper that was written by the agency. And when I tried to follow up the question [laughs], he could not really put anything on to the answer. He is sort of somebody I can devolve the attacks, and he can take all the blows but he has not got many possibilities for action really which I think is really a problem. And I can see on the body language of the Commissioner, he was very irritated at the situation he was brought in, because well he could basically just read a piece of paper somebody had written.’
However, one should also keep in mind that as the Commission can be influenced by the expertise of agencies and take their advice literally, so the agencies’ boards can be influenced by the Commission’s representatives and involvements into the working plan preparations. it is the Commission that formally takes the decision and bears the responsibility for that.The issue of responsibility between agencies and the Commission is likely to be one of the points on the agenda of the interinstitutional working group on agencies. Furthermore, some founding regulations may give a task to an agency without concrete steps to take. In such cases the regulations specify an oversight activity by another institution.E.g., the European Agency for the Management of Operational Cooperation at the External Borders (Frontex) has a task to develop a risk analysis model. The model then needs to be submitted to the Commission and the Council. (Article 4 of Council Regulation 2007/2004). As well as this, the issue of money is an influential factor. Agencies depend on funding from the EU’s general budget and are held to account for the implementation of the appropriated money before the European Parliament via the discharge procedure. In both procedures agencies can be exposed to certain pressures, and the budgets may influence the scope and direction of their tasks. Additionally, there is the possibility of using reports, evaluation and other methods for oversight to provide an additional check on their activities as will be discussed in the part on accountability. Finally, the Commission’s influence affects agencies’ functional independence. This can be exercised by its representatives on the management boards and via formal obligations to prepare working plans with the Commission’s consultation.‘In the case of the Environmental Agency, for example, the Commission expressed reservations regarding 18 of the 93 projects in the agency’s first multi-annual work programme. These projects were subsequently excluded.’ (M. Groenleer, ‘The European Commission and Agencies’, in: Spence D. and Edwards G. (eds.) The European Commission, John Harper Publishing 2006 (3rd edition), p. 169).

So, the ‘discretion formula’ for European agencies can be presented as follows: ‘discretion = delegated powers (which is formally not so much) – procedural and substantive requirements prescribed by founding actsThe more precise the legislation is, the more restrictive for discretion this factor is. – influence during the budgetary and discharge procedures – internal appeal and judicial review (in the case of decision-making agencies) – overruling by the Commission (in case of advisory agencies) – the Commission’s influence (through the boards and approvals of the working papers) – the EP’s and the Council’s scrutiny (hearings, reporting) – in some cases stakeholders’ participation.’ Thus, European agencies enjoy far fewer powers than their US counterparts, and the existing mechanisms and influences reduce the discretion they have in exercising those limited powers.

Interestingly, a correlation can be observed between the scope of delegated powers and the presence of existing mechanisms for control and accountability: the broader the scope of delegated powers is, the more mechanisms seem to be available to counterbalance that. In the case of the US, the delegated powers are stronger and there are more (in quantity and perhaps even influence/quality) mechanisms restricting an agency’s discretion. In the EU, on the contrary, the delegated powers are not as strong, and there are fewer mechanisms available to limit the discretion.

Various mechanisms reduce the functional independence of agencies. This does not mean that each and every decision is subject to all mentioned controls, it is clearly ‘expensive’, for example, for Congress to conduct investigations and hearings on each and every of the 4,000 regulationsIn contrast Congress issues approximately 400 statutes per year. (lecture of Prof. Strauss during the Amsterdam-Leyden-Columbia summer programme in American law, July 2010). that agencies issue on average per year. In their article ‘Congressional Oversight Overlooked’ McCubbins and Schwartz argue that instead of conducting continuous ‘police-alarm’ oversight Congress opts for a more effective ‘fire-alarm’ (ad hoc) oversight which is exercised by establishing ‘a system of rules, procedures, and informal practices that enable individual citizens and organized interest groups to examine administrative decisions (sometimes in prospect), and to charge executive agencies with violating congressional goals, and to seek remedies from agencies, courts, and Congress itself.’M. McCubbins and T. Schwartz, ‘Congressional oversight overlooked: Police Patrols versus Fire Alarms’, 28 American Journal of Political Science 1, 165 (1983), p. 166. This is its choice, but even then congressional oversight can be ‘overbooked’ due to the great number of issues that may need to be overseen. Thus, although not all agencies’ actions are actually being checked, no action of an agency will escape the check by default. The legislative rules and procedures that must be followed, (informal) influences from agencies’ principals and various possibilities for checking agencies’ actions affect agencies’ functional independence.

1.3.Independent Autonomous or Multi-headed Agencies

So, are independent agencies that ‘special’ in the sense being independent? The analysis of four elements of independence in both contexts does not seem to come to the meaning of the term ‘independent’ discussed from the points of view of their definition and linguistics. Agencies of both jurisdictions fulfil some (parts) of the four elements of independence, but the term ‘independent’ is an overarching notion implying a complete absence of any constraints, and there are a number of constraints that, in one way or another, affect each discussed element of agencies’ independence. The terms ‘autonomous’ or ‘multi-headed’ agencies,This article focuses on the argument that independent agencies are not ‘independent’ replacing the misleading formulation by more nuanced terms, such as ‘autonomous’ or ‘multi-headed’. However, it does not claim that ‘autonomous’ and ‘multi-headed’ are the best terms or only possible alternatives. however, allow a more nuanced and less misleading perception of these bodies, because these terms imply the presence of an oversight authority over the granted autonomy. This is important with respect to the misleading causality or dilemma of ‘independent, hence unaccountable’, because a nuanced term for agencies brings more conceptual clarity which, in turn, defeats the misleading causal relationship. It is the misleading causal relationship that seems to misbalance the academic attention in the debate on accountability of the executive in favour of ‘accountability of independent agencies’.The classic dilemma is indeed irresolvable: as long as a body must render an account of its action, it is dependent, because it can always face consequences if it misbehaved, and knowing this fact impacts its behaviour in prospect; and as long as an agency does not have to do anything, it is independent. Although it could be suggested that it is possible to talk about degrees of independence to provide a more nuanced picture, it is more correct to talk about degrees of autonomy, because ‘independence’ is an overarching notion which implies a complete absence of dependence.

2.Non-accountable ‘independent’ Agencies?

Accountability is a problematic issue when talking about independent agencies. ‘Accountability is important to provide a democratic means to monitor and control government conduct, for preventing the development of concentrations of power, and to enhance the learning capacity and effectiveness of public administration.’Bovens (2007), p. 462, supra fn 3. It is ‘a symptom of a growing public anger at individuals and institutions that are supposed to pursue the public’s interests but refuse to answer the public’s questions or accept their directions.’Mulgan (2003), p. 1, supra fn 2. In this respect, the problem with independent agencies seems to be even bigger, because they seem to be independent and hence seem to have no obligation whatsoever to ‘answer the public’s questions and accept’ any directions. Setting aside what has been argued in the previous part on agencies’ independence, let us focus here on the classic dilemma’s logic and address the question of whether independent agencies are in fact unaccountable simply because they are independent. Accountability is a complex notion and therefore needs to be defined first.

According to Mulgan, accountability is ‘the obligation to be called “to account”’, it is ‘a method of keeping the public informed and the powerful in check,’ but also ‘accountability is incomplete without effective rectification.’Ibid, p. 1, 9. Bovens proposes asuccinct definition of accountability which reads as follows: ‘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 judgments, and the actor may face consequences.’Bovens (2007), p. 450, supra fn 3.

Following the theoretical insights and definitions of Mulgan and Bovens, three stages of accountability can be distinguished: (1) the information stage where an actor ‘has an obligation to justify his or her conduct’; (2) the discussion stage where ‘the forum can pose questions’ and the actor ‘has to explain and to justify his or her conduct’; and (3) the rectification stage where the forum passes its judgments and the actor faces consequences. Without information the discussion is futile; having information without a possibility of discussion can prevent the rectification of mistakes; accountability without sanctions is incomplete as the presence of a possibility to sanction ‘makes the difference between non-committal provision of information and being held to account.’Ibid, p. 451. These stages are however only ‘roughly separable,’Mulgan (2003), p. 30, supra fn 2. because, for example, providing information and threatening sanctions can take place during one discussion.

Alongside the three stages of the accountability there are different dimensions in which the accountability can be found. Mulgan and Bovens put them in four questions. (1) Who is accountable? (2) To whom? (3) For what? And (4) how?Mulgan (2003), p. 22-23, supra fn 2; Bovens (2007), p. 454-455, supra fn 3. Let us relate these issues to this article’s specifics. The first dimension, or ‘who is accountable’, is given in this article, these are independent agencies. The question of before whom independent agencies should be accountable requires more explanation. As agencies are delegated with some public authority and are supposedly independent of the government institutions that enjoy public authority, it is implied that they have to be called to account before the people directly or indirectly via their elected representatives. In this sense the traditional delegation-accountability chainK. Strøm, ‘Delegation and accountability in parliamentary democracies’, 37 European Journal of Political Research, 261 (2000), p. 269. flows first from the people to their elected representatives to agencies with regards to delegation and the opposite way with regards to accountability. Following from this idea the article focuses on independent agencies being brought to account before representatives’ of the people (legislators).Although such approach ‘captures part of emerging practice’ only, the limits of an article do not allow considering all possible forums. (D. Curtin, ‘Holding (Quasi-) Autonomous EU Administrative Actors to Public Account’, 13 European Law Journal, 4 523 (2007), p. 523). But even with such a restriction it is possible to demonstrate the point the article wishes to make: independent agencies can be held accountable and if they are not, it is not simply caused by their independence. Furthermore, the subject of accountability (for what?) seems to be quite a clear-cut issue: accountable for the exercise of the public authority delegated to it or in other words for the compliance with the legislative intent and procedures that determine the agency’s remit and govern it. The last element of ‘how’ includes various mechanisms of accountability which can be ‘conveniently classified into three stages’Mulgan (2003), p. 29-30, supra fn 2. discussed above. These three stages serve a structural purpose, the accountability environments in which European and US independent agencies operate are assessed in three parts; information (2.1.), discussion (2.2.) and rectification (2.3.) stages. The final section (2.4.) concludes whether independent agencies can be held to account if they are independent.

2.1.Information Stage

The information stage is ‘an essential pre-requisite enabling actors to be held to account by accountability forums in various ways.’Curtin (2007), p. 532, supra fn 122. If there is no information-giving from one person to another, there is no contact and subsequently no (accountability) relationship. Do independent agencies in the EU and the US have an obligation to inform their parliaments and people?

The answer is positive. There is no single European agency that does not have any obligation to report on what it has done. The amount of institutions-recipients as well as the number and types of reports vary greatly from agency to agency, but every agency issues at least some kind of report and submits it to at least one of the following authorities, the Commission, the Council and the European Parliament. The list of institutions can be extended in some cases with the European Economic and Social Committee and the Court of Auditors. Founding acts and the Financial Regulation (EC) No 1605/2002 (the Financial Regulation further) are two existing legal sources of agencies’ reporting obligations that require different types of reports. The most ‘popular’ (in the sense of applicability to almost all agencies) are annual reports and working plans of founding acts and annual activity reports of the Financial Regulation.

Pursuant to the Regulation on public access (EC) No 1049/2001, European agencies are obliged to give ‘the fullest possible effect to the right of public access’ (recital 4) to all documents in their possession, subject to the principles, conditions and limits of this regulation. The aim of this act is to ‘enshrine the concept of openness’ (recital 1) which ‘enable[s] citizens to participate more closely in the decision-making and guarantees that the administration enjoys greater legitimacy and is more effective and more accountable to the citizen in a democratic way’ (recital 2). Agencies have to make sure that direct access to all documents and rules is available via their websites. If direct access to a document is hindered, a person may file a request, which must be granted within 15 days, unless one of the prescribed exceptions applies. The exceptions include, among others, the protection of the public interest and the privacy of individuals’ issues.

As Pray points out, in the US ‘reporting requirements have a history nearly as long as that of Congress itself.’J.G. Pray, ‘Congressional Reporting Requirements: Testing the Limits of the Oversight Power’, 76 University of Colorado Law Review, 297 (2005), p. 304. Agencies submit all kinds of reports and information required by their enabling acts and the ‘APA+’ statutes mentioned earlier. For instance, concerning rulemaking, the APA’s ‘notice-and-comment procedure’ (5 USC § 553) informs the public in advance about future rules and provides the possibility of participating in their making. In addition, the other statutes ‘+’ include information-giving provisions. These are the Federal Register Act (44 USC §§ 1501-1511) which requires agencies to publish their proposed and final rules; the Freedom of Information Act (5 USC § 552), which obliges agencies to disclose information about themselves and any material they hold upon request; the Information Quality Act (44 USCA § 3516 Note), according to which agencies have to issue various guidelines to facilitate comprehension of the disseminated information; pursuant to the Government in the Sunshine Act (5 USC § 552b) independent agencies’ meetings have to be held in public and under the E-Government Act (44 USC § 41) agencies have to provide all kinds of information online. In addition, to prevent unnecessary paperwork the Paperwork Reduction Act (44 USC §§ 3501-3521)requires agencies to submit information collecting requirements that are imposed on the public to the Office of Information and Regulatory Affairs, a part of OMB.Independent agencies may overrule OMB’s negative decisions by a majority vote (44 USC § 3507 (f)). The Negotiated Rulemaking Act (5 USC §§ 561-570)establishes a procedure whereby interested groups could be involved in the rulemaking proceedings at the ‘embryonic’ stages.For a more detailed inquiry into these and other relevant statutes, see: Lubbers (2006), supra fn 85.

Next to reporting requirements there are also numerous evaluations, reviews, ‘extensive statistics, lists and registers and databases, all of which make a contribution to shedding light on the various activities assumed by administrative actors.’Curtin (2007), p. 533, supra fn 122. Individual agencies may also be obliged to submit special thematic reports in the field of their activities.For instance, the European Centre for Disease Prevention and Control (ECDC) sends an annual evaluation of existing and emerging threats to health to the European Parliament, the Council and the Commission (Article 10(2) of Regulation (EC) No 851/2004). More than 4,000 additional staff working for the three principal congressional support agencies assist US Congress, individual members and (sub) committees, to receive necessary information.These are the Congressional Research Service (around 700 personnel), the Congressional Budget Office (238 personnel) and the Governmental Accountability Office (3129 personnel). Congress employs 17,586 staff. (The data is for 2009 and is taken from: US Office of Personnel Management, Federal Employment Statistics. Find at: http://www.opm.gov/
feddata/html/2009/January/table2.asp (last check October 2010)). Individual members and committees may request special topics with respect to independent agencies to be investigated and reported. In addition, one of these congressional agencies, the General Accounting Office (GAO), conducts regular audits and investigations of various agencies and the effectiveness of various programmes.

Also, the European discharge procedure is a very valuable procedure in terms of getting information from agencies. Almost all agencies have to submit their provisional accounts and reports on the implementation of their budgets and the financial management of the European Parliament, the Council and the Commission, which subsequently sends a consolidated version of the reports from all institutions and agencies to the Court of Auditors. The Court of Auditors enjoys extensive powers regarding the request of additional information. Agencies ‘shall afford the Court of Auditors all the facilities and give it all the information which the Court of Auditors considers necessary for the performance of its task’ (Article 142 (1) Financial Regulation). The Court’s reports together with the agencies’ responses to its observations are published annually in the Official Journal. Finally, before giving the discharge the European Parliament may also request additional information, which, among other things, includes aspects of the economy, effectiveness and efficiency.According to the principle of sound financial management of Commission Decision No 2343/2002 (Chapter 7, Article 25).

Do independent agencies have obligations to provide information? Yes, they do. Another question is whether the recipients of that information are being informed? This depends very much on whether the legislators oblige all agencies to report back and whether they read the information submitted. On both accounts there are some problems in the selected jurisdictions.

In the EU, all agencies submit at least some kind of report however, there are big differences in reporting obligations. The European Parliament and the Council are not always the recipients of the annual reports. There is no clear logic behind the differences in reporting obligations. One could think of some relationship between the agencies’ type and the institutions-recipients. For instance, executive agencies submit their reports to the Commission, because the Commission is responsible for them. The requirement to report of agencies of other types however does not follow this logic. Agency reports of the former second and third pillars are normally submitted to the Council, but in some casesThe European Union Institute for Securities Studies (ISS), the European Union Satellite Centre (EUSC), the European Police College (Cepol) and the European Union’s Judicial Cooperation Unit (Eurojust). the Parliament also receives a copy forwarded to it by the Council. A few ‘Community’ agencies, for instance, the Community Plant Variety Office (CPVO), are required to submit their reports to the Commission and not to the European legislator, which is normally the case with this type of agencies.

As well as the inexplicable differences in the obligation to report in the EU, the quality of the reported information is often unsatisfactory. The Ramboll evaluation of European agencies stated that ‘performance reporting is almost non-existent.’The Ramboll evaluation, Volume I, p. 23, supra fn 20. Roughly speaking, existing reports would list the action that the agencies have taken, rather than assess what the agencies have achieved, whether it was necessary, efficient and actually intended, and if so, why. In addition, ‘the bulk of evaluation efforts apply to periodic agency evaluations, which fall short of concluding on results and impacts, and therefore add little value in terms of accountability.’Ibid, Volume I, p. 26. Similarly in the US, a study of the Congressional Research Service concluded that ‘in a sample of several hundred reports, one-third were “no longer serving a useful purpose”.’Pray (2005), p. 301, supra fn 125 In addition, thanks to the Congressional Review Act (to be discussed) the amount of information sent to Congress, i.e., each ‘major’ rule issued by agencies, has increased enormously.

The poor quality of reporting might be a signal of the absence of interested and knowledgeable readers. The draft report of the European Parliament’s Committee on Employment and Social Affairs stated: ‘relations between the decentralized European agencies and the European Parliament are inadequate. MEPs are often very poorly informed of the functions these agencies fulfil and what they actually do.’Draft Report on the proposal for a regulation of the European Parliament and of the Council establishing a European Training Foundation (recast), 2007/0163 (COD), 07.01.2008, Committee on Employment and Social Affairs, p. 6. However, the situation does not seem to be so in all cases. The empirical findings of Busuioc demonstrate a mixed picture. Some of her respondents indeed share the sentiment of the mentioned parliamentary draft report: ‘it’s very good to be accountable but if you have to explain each time to everyone what you are doing, what is your task, what will be your future, it’s a bit tiring. The Parliament, the Commission, everyone, they want a lot of reports but what do they do with these reports?’Busuioc (2010), p. 114, supra fn 55. However, there are also examples demonstrating the contrary situations. ‘What I have seen a few times, the discussions in the Transport Committee I have been impressed by the expertise some parliamentarians have on very technical issues. <…> what they discussed there was really technical, not at all political.’Ibid, p. 108. Thus, some parliamentarians are very much involved, interested and have knowledge of the subject matter while others might have no idea on what is going on in an agency.

Similarly in the US, the interest members of Congress have may vary. Individual members can be more interested in the issues relevant for their constituencies,Strauss (2002), p. 71, 84, supra fn 45. because this is where they can be re-elected. Moreover, it is difficult to say how many reports Congress receives each year. Several attempts to study the number of reports have been undertaken by the GAO, however, ‘there is no system for tracking reporting requirements or for determining whether reports were actually submitted in accordance with the statutes.’Pray (2005), p. 299, supra fn 125, referring to H.R. Rep. NO. 96-1268, at 1 (1980), A Wall Street Journal estimation of 1991 was approximately 3,000 reports, at an annual cost of $ 350 million. (Ibid, p. 300). Once enacted, reporting requirements remain in place ‘virtually indefinitely’.Ibid, p. 304. Such a system can negatively affect the desire of members of Congress to read all those thousands (if not millions) of pages of ‘questionable usefulness’.

Independent agencies are thus subject to various information-providing requirements. Generally speaking, the information flow is there, however, if formal reporting obligations vary, if the quality of reporting is low, and if the other end of the flow does not simply receive ‘the signal’, because it does want to or does not know what to do about it, these are flaws in the system. These shortcomings are very likely to include not only independent agencies but also other institutions. The mechanisms are or can be there, but if they are not (properly and consistently) used, it is not due to agencies’ independence.

2.2.Discussion Stage

Providing the information is the first and crucial element of establishing a contact. However, a contact is not a relationship. According to Bovens, the accountability is ‘a relationship’ when, after the information stage, discussion plays a crucial role. Justifying information leads to debates which provide additional possibilities to ask for more or better (explanatory) information. Debates, in turn, lead to the identification of the mistakes made and the improvements of an actor’s behaviour or policy. After all, does one want to simply check and punish an actor or also improve his behaviour? It would be ‘costly’ for the legislator to check the behaviour of an actor and simply punish him, including, for example, the abolition of an agency. The legislator may have invested time, money and other resources in the actor’s creation and operation and it also needs it, otherwise, why would the actor (still) be there? So, do independent agencies in the EU and the US have possibilities for discussion? Can the legislators ask questions and do agencies justify their action?

Discussions normally take place when agencies and parliaments meet. Independent agencies of both jurisdictions can have formal hearing obligations and informal meetings. As agencies are often attached to specialised parliamentary committees according to their jurisdictions, they often have direct contacts with ‘their’ committees, but also budgetary committees. In their founding acts a number of European agencies have obligations to come to the hearings. There are two types of ‘hearings obligations’: (1) before his/her appointment the candidate for the position of the agency director has to come to the European Parliament (and to the Council) to make a statement and answer questions, and (2) the director can be invited by the European Parliament and the Council at any time to report on his activities. Furthermore, the discussion stage may also take a written form. There is, for example, an interesting practice in the Council of giving conclusions on the annual reports of the European Union’s Judicial Cooperation Unit (Eurojust). These conclusions are to be commented on by the agency and followed up in the next year’s report.Busuioc (2010), p. 124, supra fn 55. Finally, the discussion between agencies and the European Parliament also takes place within the annual budgetary and discharge procedures. For example, the European Parliament has established informal contacts and meetings in the course of the budgetary procedure where members of various specialised, budget and budgetary control committees come together with agencies for discussion.Annual meetings organised by the Committee on Budgets together with the Committee on Budgetary Control responsible for the annual budgetary and discharge procedures respectively (see, e.g., Working Document on a meeting with the decentralized agencies on the PDB for 2008, Committee on Budgets, 23.05.2007. DT\666715EN.doc.). The discharge procedure gives additional power to the Parliament to request information before granting the discharge. The discharge competence may in a way compensate for Parliament’s weak powers of scrutiny. This is because although the European Parliament can set up various special temporary committees to investigate specific events, e.g., the BSE committee in response to the ‘mad cow’ crisis, it has quite limited formal competences for requesting information and demanding a hearing, especially in comparison to Congressional oversight.

Congressional oversight is ‘the review, monitoring, and supervision of the executive [including independent agencies] and the implementation of public policy.’Organization of the Congress, Final Report of the Joint Committee on the Organization of Congress, December, 1993. Find at: http://democrats.rules.house.gov/archives/jcoc2.htm (last check April 2010). The scope of congressional oversight ‘is as penetrating and far-reaching as the potential power to enact and appropriate under the Constitution.’Barenblatt v. US 360 US 109, 111 (1959) ‘The power of the Congress to conduct investigations is inherent in the legislative process. That power is broad. It encompasses inquiries concerning the administration of existing laws as well as proposed or possibly needed statutes. It includes surveys of defects in our social, economic or political system for the purpose of enabling the Congress to remedy them. It comprehends probes into departments of the Federal Government to expose corruption, inefficiency or waste.’ (Watkins v. U.S. 354 U.S. 178, 187 (1957)). It would not be a gross exaggeration to state that congressional committees may question agencies at any time on any matter within their jurisdiction, formally and informally, upon a statutory requirement or without any. US independent agencies have in a way a ‘more difficult’ situation, than the executive agencies and departments, because they have no political protection from the President. A commissioner, ‘who had previously been an important official in the White House, remarked that being in an independent regulatory commission meant “having to appear naked in front of Congress,” without the political protection a connection to the White House could bring.’Strauss (2002), p. 82, supra fn 45. Furthermore, Congress enjoys special tools to help ‘persuade’ agencies’ heads to accept its invitation, the earlier mentioned powers to subpoena witnesses and to hold them in contempt.

As the possibility of a hearing is not always formally prescribed, various informal meetings have been established. At times the relationship between agencies and parliamentary committees can be very intensive; a Former Chairman of the Securities and Exchange Commission, W.L. Cary, wrote in his book: ‘Congress has been more than willing to exercise its budgetary authority [and oversight within this process] over <…> regulatory commissions during the Kennedy and Johnson administrations. Indeed, that is one reason why any chairman of an agency must spend a very substantial part of his time on Capitol Hill. In my own case, our sub-committee chairman, the late Albert Thomas, made the remark, “I recall the first year Mr. Cary was here I thought he was in the office with us over here all the time…”’W.L. Cary, Politics and the Regulatory Agencies, McGraw-Hill Book Company 1967, p. 36. According to MEP Jensen ‘informally a lot of things are happening’ between the European Parliament and European agencies and ‘agencies themselves are quite interested in having a good relationship with the Parliament. That’s my experience anyway. But I think that it should be more formalised to make sure that the Parliament has a stronger say.’MEP Jensen, supra fn 39. This is probably because informal contact relies on the ‘good will’ of agencies and can be difficult to enforce. Formal requirements for hearings have a stronger effect for the compliance of both agencies’ officials and parliamentarians, but they are rarely prescribed by European agencies’ founding acts, which leads to the shortcomings in accountability systems.

Generally speaking, the European legislator may be involved in hearings and the questioning of agencies’ directors. However, speaking specifically, the relevant hearings’ provisions are only prescribed for 13 (out of almost 34) agencies’ directors, excluding the executive agencies created by and accountable before the Commission. And this pattern applies to all kinds of accountability-related practices. Diversity in agencies’ creation and practices is not necessarily bad, unless there is good reason for it to be so. The reason for flexibility in the establishment and operation of agencies, which is behind the Council’s retaining position in the creation of a common framework for agencies’ operation, does not seem to be a good reason with regards to ‘should-be-universal’ accountability mechanisms.

Furthermore, the absence of interest from and knowledge of parliamentarians could also have negative implications for the discussion stage. Limited scope of hearings precludes a general performance assessment. Members of Congress focus on the issues relevant to their constituencies. ‘EP’s hearings and “grillings” are impressive but tend to focus on a limited number of sensitive issues, rather than on the overall performance of an agency or its directors.’Busuioc (2010), p. 113, supra fn 55. Irregular involvement in discussion and giving no feedback when assessing agencies’ work‘I never get any feedback. This is also a bit frustrating. No feedback whatsoever. I would also be pleased to have sometimes criticism why not, or congratulations on the work done, because it was not so easy.’ (the words of an agency director, Ibid, p. 114). may reduce the incentive to work hard and lead to a questionable quality of implementing policies in addition to maladministration.

Both jurisdictions experience certain problematic issues at this stage, including randomly inserted legal obligations to come to hearings and different interest of parliamentary committees in the oversight of ‘their’ agencies. However, as a whole, there are various possibilities for legislators to engage in a debate with independent agencies about agencies’ programmes, behaviour and submitted reports; perhaps more in the US, than in the EU. If these mechanisms are not (properly and consistently) used, it is due to the flaws of the whole system applicable to all kinds of institutions, and not only to independent agencies, let alone on the grounds of their independence.

2.3.Rectification Stage

‘Accountability is incomplete without effective rectification.’Mulgan (2003), p. 1, 9, supra fn 2. ‘Some would argue that a judgement by the forum, or even only the stages of reporting, justifying and debating, would be enough to qualify a relation as an accountability relation.’Bovens (2007), p. 451, supra fn 3. However, it is ‘the possibility of sanctions – not the actual imposition of sanctions – [that] makes the difference between a non-committal provision of information and being held to account.’Ibid. Therefore, the mere presence of sanctions without using them very often gives a crucial incentive to the actor to comply with the previous two stages of accountability. What kind of (possibilities of) consequences can independent agencies face in the US and the EU?

Generally speaking, all consequences agencies may face could be divided into two groups, individual and institutional. The former implies consequences that agencies’ heads may face, while the latter includes various changes that can happen to an agency as an institution, such as modifications of agency’s structure and of the scope of its authority.

With respect to the individual type of consequence, there are several possibilities for removing commissioners in the US. In terms of the US Constitution, independent agencies’ heads are ‘officers of the United States’ (Article II § 2), and ‘officers’ can be impeached for the specified ‘high crimes’ (Article II § 4). In addition, according to enabling acts commissioners can be removed by the President ‘for cause’. Finally, several historical examples cited above show that Congress could legislate commissioners out of office or attach a rider to the office to remove officials in its disfavour.

In the EU the situation is somewhat complicated. The management boards normally consist of representatives of all Member States and the Commission, and can also be in some cases supported by the representatives from the European Parliament and other stakeholders. Founding regulations do not always stipulate their tenure, so as the removal clauses are not there. Presumably, the appointing authorities replace designated members according to their own rules. However, many founding regulations stipulate the agency director as representing the agency, accountable before the European Parliament for the implementation of the budget and before its management board for the agency’s activities. Following this logic, the director of the agency is likely to face the consequences, if something goes wrong. It is also more practical to hold a single director accountable and remove him if necessary, rather than, for instance, the board of 78 members of the European Foundation for the Improvement of Living and Working Conditions (Eurofound). The Directors’ removal clauses are prescribed in agencies’ founding acts, and they vary from agency-to-agency.

There can also be ‘softer’ sanctions in a way of public ‘shaming and blaming’. The ‘shaming and blaming’ hearings that are conducted by the US Congress and by the European Parliament can be not a pleasant place to be. MEP Jensen states in this respect that ‘there has been harsh criticism. And we do follow up on this criticism. I can tell you that managers of the agencies do not find it nice to come and being exposed to this public criticism. So, our experience is that they actually do deal with it…’MEP Jensen, supra fn 39. Congressional hearings can be even more intimidating. A summoned person ‘stands in the witness box and must answer questions asked by a small number of Congressmen who sit above him like judges.’N. Jabco, The politics of central banking in the United States and in the European Union, NEWGOV, Ref. No 19b/D04, 2005.2005), p. 10 (describing the hearings of the Federal Reserve Chairman in Congress).

Institutional changes in both systems imply the statutory change of an agency’s powers, its discretion, programmes, and even an agency’s very existence. An agency’s founding act can be changed. Additional statutes and regulations can increase or decrease the authority and staff of an agency. Congress and the European legislator may reorganise the programme in question, making it more efficient and responsive to the public need, if necessary. They may restructure the agency making it more efficient or abolish the agency completely. Some legislation explicitly includes review and sunset clauses that could be used as an incentive for an agency to work hard otherwise it will be abolished.

The financial consequences also form part of the institutional type of consequences. The agency’s appropriation bills may see changes, for example, the increase of funds in case of rewards or a decrease in case of punishments. The financial consequences can be used as a result of a financial evaluation and as a punishment for another action, such as poor performance or failure to meet a deadline, which is not necessarily directly related to a financial failure. As the presence of money determines whether the agency may exercise its powers, the money sanction is probably the most intimidating sanction for agencies, with the exception of abolition.

Speaking more specifically, there is a ‘menu with US specialties’ of sanctions which includes the Congressional Review Act (CRA), legislative vetoes, deadline hammers, appropriation riders and earmarks. Pursuant to the CRA, each and every rule has to be submitted to Congress for review. Thus, Congress enjoys the authority to annul any rule, though via the constitutionally prescribed legislative process involving the President.‘Critics have questioned the efficacy of the review scheme as a vehicle to control agency rulemaking through the exercise of legislative oversight.’ (Congressional Research Service’s Report for Congress ‘Congressional Review of Agency Rulemaking: An Update and Assessment of The Congressional Review Act after a Decade’, May 2008, p. 1. Find at:
http://www.fas.org/sgp/crs/misc/RL30116.pdf (last check October 2010)). This could be because a bill overruling an agency’s rule has to follow a standard lengthy legislative procedure. With regards to independent agencies, however, the President may have fewer incentives not to sign a ‘CRA-based’ bill. Declared unconstitutional, legislative vetoes did not disappear, but rather went ‘underground’Fisher (1993), supra fn 95., i.e., they have become informal consents of agencies’ actions or decisions by congressional committees in particular cases. Moreover, Congress may impose deadlines for agencies’ actions and also attach sanctions, the so-called ‘hammers’, in case of agency’s failure to meet a deadline. An example of such a deadline hammer is when Congress gave the Department of Health and Human Services and the Food and Drug Administration 18 months to issue regulations for food imports registration; the hammer in this case was that if the deadline was not met, importers could have registered in the manner they wished, for which these agencies were responsible.Lubbers (2006), p. 16, supra fn 85. Finally, individual members of Congress may also use appropriation riders as a sanction targeting specific programmes that they dislike, and the earmarked appropriations could be used to put an agency on the ‘right’ track.

In the EU, the discharge procedure is a unique opportunity for the European Parliament and Council (via its recommendation) to express dissatisfaction with the behaviour of an agency and to improve the situation. In the course of this procedure agencies’ financial accounts are checked by the specialised institution, the Court of Auditors, and then by the European Parliament. Not only financial indicators but also the ‘soundness of financial management’ are examined. Thus, if an agency performs poorly, the discharge decision may at first be postponed and then denied. At this stage ‘what is being asked for is more information, some clarification, <…> and the promise that you would want to solve the problems mentioned, that promise from the responsible people that they would do some things to remedy problems that have been revealed.’MEP Jensen, supra fn 39. The postponement discharge decision is normally supported with a plan of actions how to improve the situation.

Postponing the discharge is clearly the first sign of disapproval of agencies’ performance. Article 147 (1) of the Financial Regulation states that agencies ‘shall take appropriate steps to act on the observations accompanying the European Parliament’s discharge decision and on the comments accompanying the recommendation for discharge adopted by the Council.’ Non-compliance with the recommendations made when the decision was taken to postpone will eventually lead to the refusal to grant the discharge,As was mentioned before, Cepol has become the first European agency whose discharge was refused. Its director was removed, and it is currently under a direct supervision of the European Parliament that in its negative discharge decision prescribed the obligation for the new director to report to it each six months about the progress on the implementation of the action plan. (European Parliament Decision of 7 October 2010 on discharge in respect of the implementation of the budget of the European Police College for the financial year 2008 (C7-0198/2009 – 2009/2127(DEC), Point 11). and the President of the European Parliament may bring the case to the Court of Justice.Article 265 TFEU; Article 6 (3), Annex VI European Parliament’s Rules of Procedure (last check October 2010).

So, independent agencies face a number of (potential) consequences of individual and institutional character. Clearly, there are practical limits to the overseeing capacity of Congress and of the European legislator, because the administrations are simply huge. However, this relates to the whole apparatus, rather than to just independent agencies. Whether these sanctions have been used and how often are the questions for empirical studies to address. But the sanctioning possibilities do exist, and this is what matters with regards to holding independent agencies to account.

2.4.Accountable Independent Agencies

So, are independent agencies unaccountable because they are independent? From a legal point of view, if mechanisms to ensure information, discussion and rectification stages are in place, independent agencies can be held to account. In both jurisdictions studied there are various mechanisms that oblige independent agencies to give information and to hold discussions, and they can face various sanctions, if necessary. Another question is however, whether the mechanisms in place work properly and serve the needs of accountability. Here, both systems experience some difficulties. The EU accountability system shows inconsistencies, which are not supported by good reasons for having such differences. Why are few agencies formally obliged to come to hearings and the great majority of them are not? This is probably because ‘it has not always been prudence that has guided the hands of the legislators of Europe when agencies have been decided.’MEP Jensen, supra fn 39. Millions of pages of questionable quality sent to Congress and the discussion of few subjects interesting for individual members’ constituencies are some examples of the shortcomings in the US accountability system. As a whole, however, the revealed problems relate to independent agencies as to non-independent agencies and institutions alike. Independent agencies can be held to account, and if the accountability systems fall short on certain accounts, the reason for that does not necessarily lie with agencies’ independence.

Conclusion

‘Independent, hence unaccountable’? This article has demonstrated that agencies’ independence need not impair the possibility of holding them to account and that it would be better to replace the somewhat attractive, yet misleading, term independent. Refuting the misleading causal relationship between accountability and independence does not mean that there are no problems concerning agencies’ accountability. This article has pointed out on several occasions that multi-headed agencies can be held accountable. But there is a world of difference between ‘can be’ and ‘are’. The jurisdictions studied have demonstrated certain shortcomings in holding agencies to account. However, as these shortcomings are not caused by agencies’ ‘independence’, why focus so specifically on the accountability of independent agencies?

This article therefore offers to extend the debate on accountability of the executive beyond the somewhat narrow focus on multi-member boards. Without doubt, comparative analysis on accountability problems of various types of institutions of the executive branch could be insightful because such studies could shed more light on the scope of existing accountability problems and hopefully increase the range of possible solutions. It might also be interesting to see whether these studies could show to what extent multi-headed agencies actually differ from non-independent agencies. In this light, what is the essential element for agencies’ ‘independence’? Is it being placed in a separate building? Is it being protected from the removal by a ‘for cause’ clause even though the appointment process is highly politicised? Is it having a statutory clause of enjoying an autonomous budget which is negotiated by political actors every year and which defines what an agency is going to do because if there is no money, there is no business? Or is it all those functions that are delegated to an agency to be performed by politically appointed persons and in accordance with the given budget? A comparative study could put the term ‘independent’ under even more pressure. Finally, as a number of discussed accountability shortcomings come from the lack of knowledge and at times desire on the side of the overseeing institutions, such as parliaments, the question for further investigation becomes whether it is possible to have a successful accountability system at all. And what is successful accountability?