The handle http://hdl.handle.net/1887/46112 holds various files of this Leiden University dissertation
Author: Mickler, T.A.
Title: Parliamentary committees in a party-centred context : structure, composition, functioning
Issue Date: 2017-02-22
Chapter 4
Analysing Structural Features of Committee Systems
Committee Structures in Thirty-Three Parliaments: An Extended Overview
T
H E focus of the first analysis of this dissertation lies on variation in structural features of committee systems. Although “it is customary to refer to legislative committees as if they were a well-defined phenomenon, in reality they come in almost endless varieties” (Mattson & Strøm, 1995, p. 257). Rarely mandated or regulated in constitutions, large sections of standing orders are dedicated to the instalment and functioning of committees, making them exist in a great variety of size, duration and powers. The 33 legislatures with strong parliamentary party groups in the case selection have developed distinctive committee systems which can be described by a variety of formal criteria. The subsequent analysis uses the hypotheses deduced earlier to test the implications of the congressional theories and the ‘keeping tabs’ perspective on this. Due to the unavailability of data which capture committee systems along multiple dimensions in all cases included in the study, new data were collected.The list of items builds to a great extent on the framework proposed by Mattson and Strøm (1995)52, but is extended with some items suggested by S. Martin (2011), and with other items. The exhaustive list of items allows for a good depiction of committees with regard to their structure, procedures and powers.
Data were gathered for all cases in the ‘complete’ case selection of 33
52 More recent publications considering committee structure (S. Martin & Depauw, 2009; S. Martin, 2011) mainly converge with this framework.
parliaments.53
The primary sources for data acquisition were standing orders and rules of procedures of the respective parliaments. The latest versions of these documents were obtained via the parliamentary websites. For several indicators which are not based on standing orders (i.e. the number of staff ), the parliamentary websites were consulted for information. After completing the data acquisition, the data were sent to country experts via an online survey in late 2013 with single reminders in early 2014. The experts were asked to review the data, propose improvements and provide missing information. This form of data validation was used to check the initial results and to minimise any possible biases and misinterpretations of the rules.54 In case the experts contradicted existing information expert judgement prevailed. If country experts of the same country contradicted each other, the information was checked again and a final decision was made by the author. The data is listed in Appendix 2 (p. 254). Data were also gathered for the legislatures of Australia, New Zealand and Croatia but country experts were unable to check the data yet.55These three countries, therefore, had to be excluded from the analysis.
The General Features: Types, Number, and Jurisdiction of Committees Across countries there is great variation in types of committees.56 Almost all legislatures delegate tasks in their daily operations regarding legislation and oversight to permanent, specialised committees. The number of specialised committees varies from 6 (Malta) to 28 in the Austrian Nationalrat. Although the number of the Maltese House of Representatives seems very small it should be noted that it is also one of the smallest legislatures in the list with 69 members.
The UK House of Commons is the only parliament in the case selection which still relies on a system of ad hoc committees to deal with legislation. In
53 The data reflects the situation in Winter 2013 and refers (if not indicated otherwise) to specialised committees, thus excluding inquiry and housekeeping committees.
54 For checking the individual country data I am deeply indebted to Wolfgang C. Müller, Sam Depauw, Christopher D. Raymond, Zdenka Mansfeldová, Martin Ejnar Hansen, Vello Andres Pettai, Mihkel Solvak, Matti Wiberg, Olivier Costa, Theodore Chatzipantelis, Gabriella Ilonszki, Shane Martin, Reuven Hazan, Marco Giuliani, Naofumi Fujimura, Janis Ikstens, Patrick Dumont, Astrid Spreitzer, Alvidas Lukošaitis, Cynthia van Vonno, Knut Heidar, Agnieszka Dudzi ´nska, Ana Belchior, Marek Rybáˇr, Miro Haˇcek, Magnus Hagevi, Juan Rodríguez Teruel, Adrian Vatter, and Phil Larkin (order of names based on alphabetical order of countries).
55 Several country experts were approached by email but were not able to validate the data before the finalisation of this project.
56 Following (Mattson & Strøm, 1995) the overview distinguishes the following types. Ad hoc committees, denoting those committees who are dissolved after their task is completed and permanent committees, which are further classified into those (1) law-making by function (thus preparing legislation), but which are not specialised by policy area (2) specialised, referring to committees which are divided by policy areas, and (3) non-legislative committees, e.g. those which do not correspond to the area of competence of a ministry (privileges, immunities, procedures and impeachment). Additionally, the overview lists the number of joint committees in bicameral parliaments. The discussion of this chapter mostly focuses on specialised committees, as these are in the primary focus of the analysis.
this executive-centred legislature, the committee system is not supposed to exert detailed scrutiny of bills. In the UK House of Commons, proposed Public Bills57 which are not considered by a Committee of the Whole House are referred to a Public Bill Committee (General Committee).58 Public Bill Committees consider a particular bill (named after the respective title of the bill, e.g. Public Bill Committee on the Energy Bill 2009-10) by focusing on individual clauses. During the committee stage, amendments are proposed and a report containing conclusions including any amendments on the bill is reported back to the Commons. Although the committee system was restructured, the basic nature of the usage of committees was not altered. Committees still come into action after bills are examined for compliance with the House’s rules (Second Reading) and are dissolved after their task is completed, meaning they have kept their ad hoc status. Within Public Bill Committees the parliamentary party group lines are “usually drawn fairly firmly with the government generally able to get its legislation through the committee stage” (Larkin, 2008, p. 97). Only the select departmental committees which are involved in oversight of governmental departments are permanent.
Another peculiar committee system is found in the Irish Dáil, the lower chamber of the Irish Parliament (House of the Oireachtas). Dáil committees were established in their present form in the mid-1990s. Before this reform, draft legislation was debated by a Committee of the Whole House. These committee stages were “often very long, were attended only by a small number of legislators interested in that particular topic and tied down parliamentary time disproportionately” (Oireachtas Brief, 2015b). The reforms aimed to handle the committee stage of draft legislation more efficiently. The post-reform Dáil relies on several types of committees: Standing Committees, Select Committees, Select Sub-Committees, Special Committees59 and Joint Committees. The term ‘standing’ committees relates to those committees that have to be established after every general election. They are called ‘Standing’
Committees, as the usual phrasing of the Standing Orders is that they ‘shall stand’ established.60 Apart from these stipulated committees, additional
57 Public Bills are the most common type of bill to enter parliament. Public Bills change the law as it applies to the general population. Most Public Bills are initiated by Government ministers but MPs or Lords may also initiate them (in this case they are called Private Members‘ Bills).
Their treatment in committee is different from Private Bills which are usually promoted by organisations, like local authorities or private companies. These aim to change law which applies to specific individuals or organisations and not to the general public.
58 Until 2006, these were referred to as ’Standing Committees‘ (labelled A, B, C). The change followed a report by the Parliament’s Modernisation Committee in 2006 which criticised the inefficiency of the system and the terminological confusion. The report recommended a number of changes. The 2006 report not only strengthened the legislative committees (allowing them to take written or oral evidence from officials and experts outside of Parliament), but also renaming them into General Committees, of which several types exist, depending on the type and scope of the bill.
59 As an example, in 2016 the Dáil established a Special Committee on Housing and Homelessness to review the implications of the problems of housing and homelessness and to make recommendations.
60 “There shall stand established, following the reassembly of the Dáil subsequent to a General
committees can be established via motions which define the powers and determine the number of members to serve on them. These are referred to as Select Committees because each house ‘selects’ a committee from its own members only. These committees divide up jurisdictional policy areas and mirror one or several ministerial departments (i.e. Select Committee on Agriculture, Food and the Marine). Similar to the lower house, Select Committees are also appointed by the upper chamber, Seanad Éireann. To avoid duplication of oversight, as the respective committees in both houses mirror the same ministerial departments, the relevant Select Committees of each house join together to form a Joint Committee. The work of this committee is limited as only the (Dáil) Select Committee can consider proposed legislation, proposed estimates for expenditure and international agreements that involve a charge on the public purse. The Joint Committee is limited in this regard. It may only consider and report on the need for legislation or expenditure and international agreements that do not involve a charge on the public purse.
Subcommittees are subdivisions of committees themselves. According to Mattson and Strøm (1995, p. 272) subcommittees “represent a trade-off between the benefits of a greater division of labour among members and the costs of an additional step in the legislative process”. Due to their small size they
“can narrow the range of political interests represented at the committee stage”
(Mattson & Strøm, 1995, p. 271). Subcommittees can, theoretically, increase the capacity to consider many issues simultaneously.61In general it can be stated that subcommittees are not extensively regulated. Most Standing Orders are relatively ambiguous about their establishment. The Latvian Saeima is one of the few legislatures which regulates the number of subcommittees per committee. If not regulated, the number of subcommittees can rise rapidly.
However, there are hardly any legislatures which make extensive use of them.
An exception is the Polish Sejm. This legislature is unusual because of the high number of established extraordinary ad hoc subcommittees to consider bills (275 as of June 2014 since the beginning of the legislative period in October 2011). Some legislatures have specific resources attached to subcommittees or to chairmanships. As an example, the chairs of subcommittees in the Finnish Eduskunta receive a monthly supplement of 474 - 717 Euro.62
A central issue with regard to the ability of committees to exert government oversight or be efficient in its legislative scrutiny is whether the jurisdiction of committees corresponds closely to the ministerial departments of the country.
Election, a Standing Committee, to be known as the Committee on Procedure and Privileges,[...]”.
(99. (1) SO Dáil). By common parlance, the word standing committee is also used for permanent committees. In this dissertation I refer to these committees as permanent committees.
61 The U.S. Congress is ‘exceptional’ in this regard as well. It has established an elaborate system of sub-committees. Having their own budget and staff due to reforms in the mid-1970s (Rohde, 1974) their importance and power have been growing, leaving some authors to conclude that
“the modern Congress has been transformed to a state of ”subcommittee government." (Hall &
Grofman, 1990; Deering, 1982; Deering & Smith, 1985).
62 This supplement is not paid if a subcommittee chair is entitled to a committee chair’s supplement.
Correspondence to the ministerial departments “facilitates influence through expert knowledge and enables individual committee members to build personal networks” (Mattson & Strøm, 1995, p. 270) with government agencies as well as interest groups. Most parliaments have more or less crude correspondence between their committees and the existing governmental departments, although some committees cross- or double-cover multiple departments (for example in the Hungarian Országház or the Australian House of Commons). In the British House of Commons only those committees that exert government oversight (permanent select departmental committees) have a direct correspondence to the governmental departments.
For most legislatures, the number and jurisdiction of committees stayed relatively stable over the last two legislative periods. Seven legislatures have changed their number of committees and their jurisdictions in more than two instances.63These changes of the jurisdiction of committees are sometimes a direct effect of restructuring at the ministerial level. In some countries, the Standing Orders specifically regulate a committee for each ministry. When the ministerial portfolios change, parliamentary committees do as well. In others, these correspondences are more informally regulated. Eleven parliaments have formal rules with regard to multiple membership restrictions, some of which are only applicable to independent MPs (Denmark). In the Norwegian Storting, the number of available committee seats is equal to the number of legislators.
This allows for the assignment of each legislator to exactly one committee.
Committees‘ sizes vary considerably. Most parliaments, however, have established committees with relatively small sizes based on the number of legislators. Mattson and Strøm (1995) argue that small committee sizes create incentives to specialize. This is based on the assumption that the “possibility of monopolizing expertise in parliament increases as the size of the committees decreases” (Mattson & Strøm, 1995, p. 268). If this is true, then committees in the French Assemblée Nationale provide the lowest incentives. Although the legislature is of considerable size, it is constitutionally restricted to eight permanent committees (the overview also counts the European Affairs Committee as a ninth specialised committee), giving it by far the highest number of members per committees (up to almost 80 MPs if all MPs are a member of a committee). Committees with relatively high numbers of members can also be found in the Japanese Sh ¯ugiin (30-45 members), the Italian Camera dei Deputati (43-47) and the Spanish Congreso de los Diputados (45).
63 The German case is an example of how a new legislative period can change data. In the 15th to the 17th legislative period (elections in 2002, 2005 and 2009) the same committees have been established, respectively only names have changed. However, after the election of 2013 several portfolio shifts took place. Committees and their jurisdictions have changed considerably.
All in all four committees have changed their jurisdiction which means a reorientation of a considerable amount of MPs.
The Procedures of the Committees: Different Patterns of Majority Rule and Minority Rights
In addition to the description of basic committee structures (number and size), information was collected on the procedures of the committees. These refer to different patterns of majority rule and minority rights within parliaments. Four different aspects are distinguished: formal committee assignments, chair selection and allocation, openness of committee meetings, possibility to draft minority reports.
Most parliaments leave the responsibility to assign members to committee to the parliamentary party groups or on a vote in the House formally.64 The rules regarding chair selection and allocation follow the same reasoning as those on committee assignment. Although one should be careful not to overestimate the power of a chair, the prestige and possible remuneration65 attached to the position make them highly sought after positions. Almost all legislatures leave the formal approval of the choice of the chairpersons in the hands of the committee.66 Only few Standing Orders actually regulate a proportional distribution of chair positions (the Polish Sejm, the Portuguese Assembleia da República, the Slovenian Državni zbor). In other cases, this process is regulated by informal rules. Committees chairs in the House of Commons (Canada), the Estonian Riigikogu, the Althingi (Iceland), the Camera dei Deputati (Italy), the Sh ¯ugiin (Japan), the Chambre des Députés (Luxembourg), the House of Representatives (Malta) and the Congreso de los Diputados (Spain) are mostly held by the majority parliamentary party group or the government coalition. The other parliaments have either a perfect or crude proportional distribution of chairs across all parliamentary party groups in parliament.
Committee openness refers to the question whether committee meetings are private or public. Mattson and Strøm (1995, p. 282) argue that “public meetings dissipate some of the informational advantages committee members acquire”
and also run the risk to turn committee meetings into potential advertising fora for committee members. Most Standing Orders merely allow for open and closed meetings. Several legislatures (the Austrian Nationalrat, the Danish Folketing, the Finnish Eduskunta, the French Assemblée Nationale, the German Bundestag, the Italian Camera dei Deputati, the Luxembourgian Chambre des Députées, the Norwegian Storting and the Riksdag in Sweden) have in principle closed sessions which can be public under certain conditions.
The extent to which minority reports are allowed is seen by Mattson and
64 The information focuses purely on the formal regulations beyond the influence of parliamentary party groups. The factors which play a role in the assignment process are part of the second analysis of the book.
65 Remunerations for chairpersons are not uncommon. Chairpersons in the Finnish Eduskunta receive between 717 and 1183 Euro per month extra, depending on the committee. In Dáil Éireann Chairpersons receive a remuneration of 8740 Euro per year extra.
66 It is very likely that these are usually votes which confirm decisions taken in a different venue, i.e. through informal discussions between or within parliamentary party groups.
Strøm (1995, p. 283) as an additional incentive for committee members to
“specialise and to take their task in committee seriously”. Minority reports are a possible resource for the plenum to gain further insight into hitherto unconsidered policy options or inform the chamber about a missing consensus in committee. When minority reports are not allowed, the governing parliamentary party group(s) has/have an important instrument to determine the agenda and legislative initiatives. Half of the legislatures allow for minority reports, i.e. to issue a statement against the position of the committee’s majority.
There is no apparent pattern in the way that legislatures assign this right.
Committees in the Danish Folketing are not able to table these reports (despite frequently having minority governments) while other Scandinavian legislatures (the Swedish Riksdag and the Norwegian Storting) allow them.
Committees’ Powers in the Legislative Process
The third dimension differentiates committees‘ powers in the legislative process. The powers refer to “the committees’ ability to influence legislation, independent of such external actors as parliamentary party group leadership, chamber majorities, and the government” (Mattson & Strøm, 1995, p. 285). The authors employ a distinction between negative and positive committee power.
Positive committee power “is the ability to move policies away from those that the floor prefers and towards those that a committee prefers. Negative power is the ability to prevent a change in policy despite the fact that policy is not at the floor’s ideal point.” (Epstein, 1997, p. 273). The debate basically refers to the question whether committees have effective gate-keeping power, i.e. “power to enforce the status quo against the parent body’s wishes” (Epstein, 1997, p. 271).
A crucial aspect of determining the role and impact of a legislature’s committee system is the scheduling of the committee stage during the decision-making process. In sequenced legislative deliberations, committees come into action at a certain moment in the treatment of a bill. Standardised deliberation chains can either consist of several stages or debates. These debates are referred to as ’readings‘.67The timing when committees come into action on a bill is crucial to the power the committee has. Generally speaking, when committees come into action after a plenary debate on the content of the bill, their impact is smaller than when they consider a bill before a major plenary debate on the bill has taken place. The committee stage in most parliaments is immediately after the first ’reading‘/stage which is often a formality and does not include a debate on the content of the bill. In some
67 The introduction of a bill into parliament is usually called the ’first reading‘ of a bill. The practice of giving every bill three separate readings originates from an ancient parliamentary practice in the United Kingdom in which the clerk literally read a bill out loud three times. Generally, two reasons are put forward which have influenced this practice. First, due to the inability to reproduce large numbers of copies at low cost at that time, merely a single copy of a bill was introduced in handwritten form. Second, three readings were also necessary in order for MPs to discern the content of the bill due to widespread illiteracy (Parliament of Canada, 2009).
legislatures, like the Hungarian Országház or the Norwegian Storting, the committee stage is even before the first reading. In some legislatures, e.g. the Dáil, committee stage follows the general debate. This limits the influence of the committees as the draft legislation is already debated in general terms.
An important source of power is the right to initiate legislation and therefore to set the legislative agenda. Thirteen legislatures allow committees to initiate bills, while the majority of the cases do not provide such a right or put some restrictions on it. A related power is the ability to split or merge bills. Two-thirds of the legislatures do not have a formal provision which allows committees to split or merge bills. Only the Austrian Nationalrat, the Kamer van Volksvertegenwoordigers (Belgium), the Knesset (Israel), the Camera dei Deputati (Italy), the Saeima (Latvia), the Chambre des Députés (Luxembourg), the Riksdag (Sweden) and the Nationalrat in Switzerland allow for both.
One of the most striking features of committees in the U.S. Congress is the ability to kill a bill (death-by-committee). Congressional committees make extensive usage of this power. W. Wilson (1885, p. 58) called committees “dim dungeons of silence whence[a bill] will never return”. Between 1967 and 1968 93.6 per cent of bills introduced in the House and 66.6 per cent of those introduced in the Senate died in committee (Polsby, 1971, p. 90). A look at the legislative statistics shows that a vast number of bills referred to a committee is not reported after it has been referred to a committee (see Figure 4.1). Most bills do not reach the floor after they have been referred to a committee but are pigeon-holed (i.e. put aside and ignored), or simply never discussed. On average, almost 95 per cent of bills do not make it through committee stage. The combination of this power and the committees’ extent to modify legislation and determining the future of the bill has led to the observation that these committees “monopolise the deliberative activities of the legislature and are, in effect, exercising an autonomous influence on the policy-making process”
(Mezey, 1979, p. 69). In the analysed legislatures, the right to kill a bill is only available to a small number of cases: the Austrian Nationalrat, the Kamer van Volksvertegenwoordigers (Belgium), the Althingi (Iceland), the Camera dei Deputati (Italy68), and the Japanese Sh ¯ugiin. The vast majority of committees has to send the bill back to the plenum after it has worked on it. This leaves the question, what committees can actually do with draft legislation, e.g. whether committees have the right to redraft the original text, propose amendments to the original text or formulate a recommendation for a decision, and which document is finally considered by the plenum. Most committees possess extensive powers to rewrite bills.
If committees are not allowed to rewrite a bill and report a new ‘clean bill’ back to the plenum, they are usually restricted to propose amendments which are then voted on in the plenum. In around 1/3 of the legislatures, the committees are not allowed to rewrite the bill at all, while the majority of committees have the ability to redraft the bills by proposing and voting on amendments. The committees of
68 This only applies when the committee is in legislative capacity.
Figure 4.1: Number referred and reported bills. 93rd - 113th US Congress
Own depiction. Data collected from https://www.govtrack.us/congress/bills/statistics. Access date: October 30, 2013 (Data for 113th Congress reflects date on access date as legislature was still in session)
the French Assemblée Nationale, for example, have a weak institutional position.
Although committees were relatively powerful in the French parliament during the Fourth Republic, changes introduced in the Constitution of the Fifth Republic in 1958 cut back their decision-making power. Each committee works under tight schedules, cannot revise the purpose of legislation and its redrafting rights are restricted to technical modifications.
A special case in this regard is the proceeding in the Dutch Tweede Kamer.
During committee stage, the committees are not allowed to rewrite a bill and may not include amendments either. The most common procedure is as following: A draft bill, after it has been submitted to the chamber, is referred to a committee. During the committee stage, committees provide their views on the bill only in written form. After calling for input from the parliamentary party groups on the bill the committee drafts a report. Each paragraph depicts the parliamentary party group’s position in narrative form (“The members of parliamentary party group A have taken notice of the draft bill regarding[...].
followed by suggestions and the assessment of the parliamentary party group.”).
This report (Dutch: verslag) is send back to the government which drafts an answer in form of a report (Dutch: nota naar aanleiding van het verslag). This submission of written comments is mostly done in two rounds (Bovend’Eert &
Kummeling, 2010, p. 225), but remains in written form. At the end of these exchanges, a final report is drafted which concludes that the bill is sufficiently prepared for consideration in the plenary session. If members in committee want to propose an amendment with regard to revision of bills outside of these written consultations, then these need to be introduced and voted on in the plenum. Only there an oral debate takes place in which motions and amendments can be tabled.
General consultations (algemeen overleg) contain an (oral) debate in committees. These do not concern a specific piece of legislation but serve as an exchange of views between the government and the committee. The basis for
these consultations are letters from the government which introduce plans for future policies. Based on the consultation a report is drafted. If, following the discussion during the general consultation, a legislator or parliamentary party group wants to propose a motion, the same restrictions as with bills are applicable. Votes on motions cannot be done in committee but need to be held in plenary session. In such cases, the report of the general consultation (Dutch:
algemeen overleg) is placed on the agenda of the plenum (this is referred to as verslag van een algemeen overleg in Dutch, usually abbreviated to as VAO). Even during this debate, there is no extensive discussion. Each parliamentary party group gets two minutes of speaking time (also referred to as ‘two minutes debate’ - tweeminutendebat) after which motions can be tabled. The Rules of Procedure offer additional, though limited possibilities for committees. These were introduced in 1994 in order to break through the “inertia and overuse” of written exchanges (roughly translated from Bovend’Eert and Kummeling (2010, p. 225: “traagheid en verschriftelijking”). A ‘debate on broad outlines’ (Dutch:
hoofdlijnendebat) is a plenary debate during which the necessity of a bill is debated before the committee starts its proceedings. These debates are very rare (Bovend’Eert & Kummeling, 2010, p. 225). Since 1994 it is also possible for committees to hold oral debates on a draft bill. These are referred to as legislative consultations (Dutch: wetgevingsoverleg). They were introduced in order to relieve the plenary session of some of its workload. These debates can take place at every stage, including after the plenary debate has already taken place. They are mostly used for “dealing with draft bills which contain many technical/specialist elements” (Bovend’Eert & Kummeling, 2010, p. 225). In any case, votes still can only be taken in a plenary session.
The control of a committees‘ own timetable, whether sessions of the plenary and a committee can be scheduled simultaneously, grants greater committee autonomy. Most legislatures do not allow for such meetings or at least put some restrictions on it (e.g. only in case no votes are scheduled). In the Estonian Riigikogu, the Finnish Eduskunta, and the Portuguese Assembleia da República the meetings are regulated in the parliamentary working schedule with the possibility to have extraordinary meetings. In most instances, the committee’s timetable is decided by the committee. Only a few legislatures do not allow for any right to recall and leave the decision on when to hold a session completely to committees: the Folketing (Denmark), the Assemblée Nationale (France), the Vouli ton Ellinon (Greece), Althing (Iceland), Chambre des Députés (Luxembourg), and the Sejm (Poland). However, most legislatures allow for a right of recall, meaning that the decision can be overruled, either by the House or a deliberative body of the legislature or the Speaker.
The Information Gathering Powers of Committees
The capability to gain the best possible knowledge about an issue is crucial for committees. The power of a committee to acquire information, specifically the
right of summoning witnesses and documents, gives MPs on the respective committees a tactical advantage over their colleagues who are not on the committee. Most legislatures grant extensive rights to committees to demand documents from outside. Committees in the Danish Folketing do not have any formal power to demand documents. This is also the case in the French National Assembly, the Knesset (Israel), the Maltese House of Representatives, and the Slovakian Národná rada (National Council).
Additional information on the extent and usage of committee staff was collected. No legislature even remotely reaches the level of professionalism of committees in the U.S. Congress69, but several countries can be classified as being relatively generous with regard to own staff. In the German Bundestag committee staff varies between 2-15 staff members, the Knesset (4-14) and the Seimas (Lithuania, up to 11 per committee) also have relatively high number of staff per committee. Related to the size is the task of these staff members. In most legislatures they are involved in administrative work, carrying out research and preparation of documents. Several legislatures restrict their task to only administrative work or preparation of documents and leave the research to other entities (Poslanecká snˇemovna (Czech Republic), Vouli ton Ellinon (Greece), the Camera dei Deputati (Italy), the Sh ¯ugiin (Japan), the Chambre des Députés (Luxembourg), the House of Representatives (Malta), the Tweede Kamer (Netherlands), the Storting (Norway), the Sejm Rzeczypospolitej Polskiej (Poland), the Assembleia da República (Portugal) as well as the Slovakian Národná rada). In other legislatures, the staff is tasked with research as well. The UK House of Commons steps out of line in this regard as well. Procedural support and expert advice come from central bureaus. Similarly, the Japanese Sh ¯ugiin does not have committee staff but relies on a legislative bureau.
This overview indicates great variation in committee systems in the analysed legislatures (the detailed overview is listed in Appendix 2). These structures determine the ‘playing field’ for committee members. The first analytical part of this dissertation focuses on the variation of the structural variation by testing the predictions of both the congressional theories and the
‘keeping tabs’ perspective. The following section clarifies the concepts used in the analysis and gives more information how the theoretical framework was operationalised.
Analysing Structural Variation: The Concept of Committee Autonomy
In order to compare committee systems across multiple countries, a meaningful indicator of committee structure is needed. Following the example
69 Congressional committees are particularly well-equipped. Shaw (1979, p. 387-388) noted more than 250 subcommittees and the high number of congressional staff working for committees (770 in the House and 620 in the Senate). This number has increased in recent years to 1300 in the House and over 900 in the Senate in 2009 (Ornstein et al., 2013, p. 104)
of Martorano (2006) the committee systems of the countries are analysed using the concept of committee autonomy. Autonomy, in general, refers to the differentiation of an organisation from its environment (Atkinson & Nossal, 1980, p. 290). Fenno (1966) uses the concept of committee autonomy to describe the degree to which decision-making processes in committees are influenced by members or outsiders (see also Smith, 1986). Since then committee autonomy as a concept has been used in studies of the U.S.
Congress and state legislatures (Deering, 1982; Martorano, 2006) and parliaments outside the U.S. (see for the Canadian House of Commons Atkinson & Nossal, 1980). It is assumed that the strength of committees depends directly on the degree to which they have the authority to conduct their own affairs (Atkinson & Nossal, 1980, p. 290).
Martorano (2006) operationalises the concept by drawing on the first four of five dimensions of committee autonomy proposed by Rosenthal (1973).
According to Rosenthal, autonomous committees are able to 1) receive legislation, 2) screen legislation, 3) shape the nature of legislation 4) affect the passage of legislation and 5) study problems and initiate legislation.
Subsequently, her measure of committee autonomy is based on an additive index of eleven procedures. This reasoning is also taken into consideration in the study at hand. Committees have multiple functions. Therefore, any measure of committee structure which merely focuses on a single issue (i.e. only on the right of committees with regard to the passage of legislation) does not give an accurate picture of the structural features.
The index of ‘committee autonomy’ used for this analysis was constructed using the overview on committee structures compiled in the course of the project. This was done by rank-ordering all answers in each item according to the lowest amount of autonomy granted to a committee to the highest level. As an example, the item measuring whether simultaneous meetings with the plenum are allowed has three levels (0= not possible at all, 1 = possible under certain circumstances with special permission (from the plenum, Speaker, etc), 2 = always possible; with 0 being the lowest level of autonomy). Another example would be whether committees are able to recruit their chairpersons from within the committee (0= no, 1 = yes). More complex items have more levels, i.e. the control of a committee’s timetable varies between the government setting out the deadline via a motion (ranked as ‘0’, indicating the least amount of autonomy) to parliamentary actors which are not on the committee (Speaker, etc.= 1) to the Chairperson of the committee (2) to the committee members themselves with no right of recall by the plenum (highest rank of autonomy= 3).
In the same manner, more committees mean higher autonomy.70Subsequently,
70 Mattson and Strøm argue that the number is “positively correlated with committee strength”
(Mattson & Strøm, 1995, p. 259). This reasoning is built on a claim by Smith (1980, quoted in (Mattson & Strøm, 1995, p. 260)), who argued that “[...] the greater the number of small groups, the less amenable to governmental control they are than a single, large one”. All else being equal,
“the more committees, the more bills can be dealt with at the same time” (Mattson & Strøm, 1995, p. 260).
Table 4.1: List of items measuring the formal structures and operationalisation of committee autonomy
Question Less autonomy↔ More autonomy
Number permanent committees (Legislative per function /
specialised/ non-legislative)? Not used for index
Number ad hoc com’s (no legislation/ legislation)? Not used for index How many joint committees exist? Not used for index What is the min. and max. size of ad hoc committees Not used for index What is the min. and max. size of permanent committees Not used for index
Are committees jurisdictional to ministerial portfolios Subject based, crude coverage ↔ Complete correspondence
Do multiple membership restrictions exist (limit)? Yes↔ No restrictions Are substantive (non-procedural/administrative) sub-
committees used? How many?
Not possible↔ Possible Which body is formally involved in the assignments of members
to committees?
External group (e.g. whips)↔ House How are committee chairs formally selected? Speaker↔ Vote in committee How are chairs allocated to the PPGs? Only cabinet PPGs↔ Proportional Same committees established in the last 2 leg. periods? Substantive changes↔ Yes Are committee meetings public or private? Public↔ Private Are rights granted to publish minority reports? No↔ Yes, automatically
How are bills allocated to committees? Minister, Speaker↔ Board of Chairmen When is the committee stage? After plenary debate↔ Before debate Do committees have the right to initiate legislation? Not possible↔ Right exists Can committee split or merge bills? Not possible↔ Right exists Do committees have the right to redraft legislation or propose
amendments?
No power to amend / redraft ↔ Extensive amendment power Do committees have the right to kill a bill? No↔ Yes
Which document is considered by the plenum at the end result of committee negotiation?
Original bill with proposed amendments ↔ Redrafted “clean”
bill; resolution recommendation Who controls the committee’s timetable? Meetings regulated↔ Committee (no
right to recall) Can committee sessions be scheduled simultaneously to plenary
sessions?
No↔ Yes
Do committees have own staff? No↔ Yes
Number of supporting staff for each committee? No support staff↔ Highest number Secretariat’s functions: Administrative/ Research / Document
Preparation?
None↔ Combination of all three Do committees have the right to invite/ compel witnesses? If so,
who are they allowed to compel?
Invite but not compel ↔ Compel unrestricted.
Are committee hearings public or private? Public↔ Private Can committees demand documents from outside parliament?
If so, from whom?
Not possible↔ Yes, unrestricted Source: Own depiction.
each item was scaled to an interval between 0 and 1. The overall score of autonomy is the mean of the ranked score of all included items. These values serve as the dependent variable. Table 4.1 lists the items and how these indicators measured the autonomy of the committees.
Initially, the data were analysed in order to extract multiple dimensions by means of a principal component analysis. However, the results did not indicate any meaningful unobserved factors and presented no interpretable results. It was therefore concluded that not much would be gained by classifying the data into several dimensions. The aggregate autonomy scores of each legislature are
shown in Figure 4.2 which indicates variations from the mean autonomy score (0.58). Committees with low autonomy scores are found in the British House of Commons, the Maltese House of Representatives, the Dutch Tweede Kamer, the Vouli ton Ellinon in Greece, and the Irish Dáil. The strongest committees are found in the Nationalrat (Switzerland), the Storting (Norway), the Kamer van Volksvertegenwoordigers (Belgium), the Camera dei Deputati (Italy), the Bundestag (Germany) and the Riksdag (Sweden). No correlations to other data sets can be provided due to different case selection and focus on different items, but the placement of countries here resembles prior comparative studies (e.g.
S. Martin, 2011, p. 353). Small differences can most likely be accounted for by the different number of items considered in this study and by the calculation method of the index. There are some notable exceptions. The low score of the Dutch Tweede Kamer might be surprising as it contradicts the findings of L. W. Martin and Vanberg (2011, p. 50). Based on the formal characteristics the low score is, however, not surprising as committees in this legislature are very restricted with regard to a number of features. During the revision of laws, committee meetings consist of written questions and answers, while no amendments can be proposed or included by the committee in a draft.
Additionally, these committees possess no rights to initiate, split or merge bills.
The proceeding in the Dutch Tweede Kamer was depicted in greater detail on p.
76.
Figure 4.2: Committee autonomy scores in 30 legislatures
Source: Own calculation. The plot depicts variation from the mean autonomy score (0.58).
Analysing Variation in Committee Autonomy: Operationalising the Independent Variables
To explain the variation in committee autonomy several hypotheses were deduced from the congressional theories and the “keeping tabs”-perspective. A partisan organisational logic is supported when committee autonomy depends on the seat share of the government parliamentary party group or coalition (hypothesis 1). Using the ParlGov database (Döring & Manow, 2012) the average seat share of government parliamentary party group(s) was calculated by aggregating the seat share of the governing parties (excluding caretakers) of a particular country across all legislative periods since 1990. Data for Israel were obtained from Carr’s Election Archive (Carr, 2014) and cross-checked with data from the Election Guide database (Election Guide Database, 2014).71 Across countries, the seat share of the cabinet parties is on average 53.35 per cent (SD
= 8.066) of the votes.
The distributive theory argues that committees ultimately serve the interest of individual legislators. This distributive rationale is tested using two hypotheses. First, the extent to which the electoral system allows for a personal vote (hypothesis 2) and second, the connection of individual MPs to electoral districts (hypothesis 3). For the extent of personal vote incentives, rankings for all countries were obtained from the data set compiled by Johnson and Wallack (2012) which contains information on most of the world’s electoral institutions by electoral tier. Based on the four indicators72proposed by Carey and Shugart the data set ranks various institutional combinations according to the relative value to legislators of personal reputations versus party reputations.73
Apart from the incentives to cultivate a personal vote it was hypothesised that more autonomous committees are established in those countries in which individual MPs have a clear connection to single districts due to their electoral systems. Whether this predictor can account for the variation in committee autonomy is based on the effect of a mean district magnitude score. This score is calculated by dividing the number of seats in parliament by the number of electoral districts. As an example, all 308 MPs in the Canadian House of Commons are elected via simple plurality in single-member districts, giving each legislator a clear connection to a single district. At the other extreme, MPs in the Dutch Tweede Kamer, the Israeli Knesset or the Slovakian National Council are elected in a single district which coincides with the country borders.
71 The status as coalition-PPG was only given in case parliamentary party groups joined the coalition right after the election.
72 (1) The degree of control parliamentary party group leaders exercise over ballot rank in list systems, (2) whether votes cast in the general election are pooled across entire parties, (3) the number and types of votes cast and (4) district magnitude.
73 The data set contains two rankings. The results presented in this paper use the rank of the tier with the greatest incentives to cultivate a personal vote (denoted PERS_ RANK). Models with the other ranking which ranks countries according to their most populous tier (DOM_ RANK) presented similar results.
In order to calculate the score information on the number of districts per electoral system was collected, mostly directly from parliamentary websites. No distinction was made with regard to variation across districts. As an example, the 101 seats in the Estonian Riigikogu are distributed among 12 multi-mandate electoral districts (6-13 mandates each), meaning the score was calculated as 101/12 = 8.42. In the case of mixed-member proportional systems, the divisor is the sum of the number of single-member districts (e.g. 299 in Germany) and the number of districts in which members are distributed via party lists, i.e. 16 in Germany as the lists are entered in the 16 federal states (German:
Bundesländer). The divided score is skewed and therefore was logged (Mean(logged)= 2.23, SD(logged)=1.33).
Hypothesis 4 connects the existence of strong committees to the level of corporatism in a country. Corporatism has been “one of the most heavily studied concepts in comparative political economy over the past two decades”
(Kenworthy, 2000, p. 2). Operationalising the concept is complicated by the fact that it is a “multi-purpose concept” (Molina & Rhodes, 2002) and has been
“embedded in contradictory meanings and generated apparently inconsistent usage” (Winkler, 1976). The well-known ideal-typical definition by Schmitter (1974, p. 93-94) of corporatism is:
[A] system of interest intermediation in which the constituent units are organised into a limited number of singular, compulsory, non-competitive, hierarchically ordered and functionally differentiated categories, recognised or licensed (if not created) by the state and granted a deliberate representational monopoly within their respective categories in exchange for observing certain controls on their selection of leaders and articulation of demands and support.74
Many authors have attempted to establish quantitative indicators of corporatism (for an extended overview, see Kenworthy, 2000). Arguably the most widely used measures of corporatism are the rankings by Lijphart and Crepaz (1991) and Siaroff (1999). Both measures are based on experts’
evaluations and are highly correlated (Bloodgood et al., 2014). They differ in the countries they cover: the Lijphart ranking covers eighteen countries, the Siaroff ranking includes twenty-four industrial democracies. Unfortunately, several countries of Central and Eastern Europe of my analysis are not included in either of these rankings.75
Fortunately, a more recent index by Jahn (2014) covers all countries in my data set with the exception of Iceland. His corporatism score is the result of a factor analysis of eight items to measure the level of corporatism in a country (Jahn, 2014, p. 14). The correlation between his index and the Siaroff index is
74 See also Siaroff (1999, p. 178 - 179) for an ‘ideal type’ corporatist political economy.
75 More specifically, data on the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Slovakia, and Slovenia are missing.
high. When all cases are included the correlation is 0.76. When excluding Greece, Japan, Spain and Switzerland, all cases which deviate and are usually seen as ‘problematic’ in the study of corporatism (Siaroff, 1999, p. 183), the correlation increases to 0.90. The cases of Central and Eastern Europe are mostly ranked in the lower half of the ranking. According to Ost (2000, p. 503), post-communist countries “invested heavily in tripartism” in the years after 1989. This led some authors to the conclusion that these countries are in a phase of “transformative corporatism” (Iankova, 1998). However, several scholars (Bohle & Greskovits, 2012; Contrepois & Jefferys, 2010; Crowley, 2004) argue that they did not succeed in consolidating similar levels of corporatism as other countries, e.g. with regard to union weakness. Ost (2000, p. 503) refers to corporatism in these countries as “illusionary corporatism” and argues:
“While the facade of tripartism is present throughout the region, with duly constituted commissions holding regular meetings bringing together formal representatives of the state, trade unions, and employers, this is in no position to bring about the politically stabilizing and economically inclusionary class compromise that was West European neocorporatism’s great achievement” (Ost, 2000, p. 503).
There are, however, exceptions to this general pattern. Slovenia, for example, has been one of the countries in which corporatism has been an influential doctrine (see Lukšiˇc, 2003). This is also reflected in its position behind Finland and Germany. Due to its coverage and the match with the ranking by Siaroff, the Jahn ranking was deemed to be an appropriate source. For the analysis, the average scores of corporatism for each country were used.76Data for the only missing country (Iceland) in the Jahn index was compensated with its score of the Siaroff ranking.77
The informational theory highlights the ability of committees to increase the efficiency of the chamber as a whole. It is predicted that more autonomous committees are established in larger legislatures ((hypothesis 5) and when a legislature faces a structurally greater workload over an extended period (hypothesis 6). Information on the size of the legislatures was collected from the websites of each parliament. The legislative workload is measured via the following formula:
Legislative workload= average number of bills per year average number of plenary sessions per year
76 The indices of Jahn are z-standardised. They were rescaled to give the lowest country the value of 0.
77 In the Siaroff ranking, Iceland had a score of 3.00 and was placed above Israel. In the Jahn ranking, Israel has a z-score of 0.09 (which corresponds closely to the 3.00 score of the Siaroff ranking).
For the analysis, Iceland received the same score as Israel. Although it would have been preferred to receive an original score by Iceland, this placement hopefully does justice to Iceland’s level of corporatism and will not distort the analysis.
This measure is not as detailed as earlier measurements by Martorano (2006) who also includes the length of bills.78Nevertheless, it still gives a good impression of the ‘demand’ of a legislature to establish committees in order to guarantee an efficient legislative process. The average number of bills per year was obtained for all cases by consulting overview lists of bills or search engines on parliamentary websites (on average over a period of around 15 years). All bills which were introduced by legislators or groups of legislators (Private Member Bills), the government, and committees were counted. The number of plenary sessions per year was also gathered based on the information in parliamentary yearbooks or databases. The data spans fifteen years. Using these two indicators the overall ratio of the average number of bills was calculated.
The average number of bills per year varies considerably, from around 29 per year in Malta to over 3500 in Sweden. The number for Sweden seems extraordinarily high, but it is in line with earlier research (see e.g. Bergman &
Bolin, 2011, p. 268). In this particular case, the number is inflated by the high number of private member bills (Swedish: motioner). These are not only used for the introduction of independent bills but also used regularly as amendments to government bills. The analysis will control for any bias caused by the high number of bills in Sweden.
As was argued above, it could be the case that legislative workload is endogenous to the level of autonomy of committees. It may be that more autonomous committees increase the number of laws a parliament, especially if committees have the right to propose legislation. However, there is only a small positive correlation between the variables measuring the rate of bills and initiation rights (0.123). Although those legislatures which allow for committees to initiate bills indeed seem to have slightly higher numbers of bills, there is no strong evidence that committees which have this right make massive use of this.
One might still raise an objection on a more general level and argue that the number of bills is higher because the specific committees are more autonomous. This argument is not, however, entirely convincing. The decision whether the government or legislators submit draft legislation to parliament is unlikely to be based on whether there is a committee system available to deal with it. Rather it rests on the perception of a particular issue that needs to be addressed which is independent of the existence of a particular committee structure. It is much more plausible to assume that institutional rules are designed because the actors in a legislature have to cope with a growing workload. Looking ahead, legislators purposefully try to design the rules in such a way as to be able to work on these issues. In this sense, the effect of the average number of bills per plenary session on the institutional design is an appropriate test of the underlying rationale of the informational theory.
Following the predictions of the ‘keeping-tabs’ perspective, committees
78 Data on the length of bills is readily available in the U.S. (see for data on pages in the Federal Register, 1936-2012 e.g. Ornstein et al., 2013). This is not the case for all 30 countries included in this study. Gathering data on the length of bills exceeds the time frame of this study.
serve as policing institutions of the coalition partners. Based on this premise several hypotheses were formulated to test this policing demand of various actors. The first hypothesis tests whether legislatures in countries which frequently establish coalition governments have more autonomous committee systems (hypothesis 7). To test the influence of coalition formation versus single-party governments, the number of government parties for each cabinet since 1990 (excluding supporting parties) was determined using the ParlGov database (Döring & Manow, 2012). Each cabinet was then classified as single-party or coalition government. By aggregating the ratio of single-party and coalition cabinets per country the final score reflects the percentage of governments that were formed via a coalition of parties.79
Additionally, it is tested whether a larger ideological distance between coalition partners has a negative effect on committee autonomy (hypothesis 8).
Based on the ParlGov database (Döring & Manow, 2012) the ideological distances between coalition partners were calculated (again excluding those parties which were indicated as caretaker). The overall score is the mean between the difference of the highest and the lowest scores in the left-right, state-market, liberty-authority and the pro-anti EU dimensions across all governments since 1990. These four dimensions in the ParlGov database are aggregated party positions based on data from various expert surveys (time-invariant unweighted mean values on a 0 to 10 scale).80 Data for Israel were calculated based on the scores in Benoit and Laver (2006).
Hypothesis 9 refers to the possibility for parliamentary party groups during minority governments to influence policies via a strong committee system (see Strøm, 1990). To test whether the occurrence of minority governments can account for structural variation in committee autonomy, data on the frequency of minority governments were collected. The ParlGov database (Döring &
Manow, 2012) contains information on the composition of all governments.
Data for governments formed after 2012 is missing from this overview but was added by the author. Data on the governments of Israel was obtained from the website of the Knesset (Knesset, 2015). The time frame of the analysis is
79 An analysis whether countries with a higher average number of parties in coalitions establish more autonomous committees presented similar results.
80 The left/right dimension is based on data from Castles and Mair (1984) (variable used:
‘left/right’), Huber and Inglehart (1995) (variable used: ‘left/right’), Benoit and Laver (2006) (left/right) and the Chapel Hill Expert Survey (CHES) (Bakker, de Vries, et al., 2015; Bakker, Edwards, et al., 2015) (variable used: ‘lrgen’ from the years 1999 and 2002 and 2006); the state/market dimension is based on data from Benoit and Laver (2006) (variable used:
‘taxes/spending’) and the Chapel Hill Expert Survey (CHES) (Bakker, de Vries, et al., 2015; Bakker, Edwards, et al., 2015) (variable used: ‘lrecon’ from the years 1999 and 2002 and 2006). The liberty/authority dimension is based on data from Benoit and Laver (2006) (variable used:
‘social’) and the Chapel Hill Expert Survey (CHES) (Bakker, de Vries, et al., 2015; Bakker, Edwards, et al., 2015) (variable used: ‘galtan’ from the years 1999 and 2002 and 2006). The EU anti/pro dimension is based on data from Ray (1999) (variable used: ‘pos96’), Benoit and Laver (2006) (variable used: ‘euauthority’, ‘eulargerstronger’, ‘eujoining’) and the Chapel Hill Expert Survey (CHES) (Bakker, de Vries, et al., 2015; Bakker, Edwards, et al., 2015) (variable used: ‘position’ from the years 1999 and 2002 and 2006).
twenty-five years (January 1990 - January 2015). For each government formed within this period it was coded whether it was supported by a minority of members in parliament or not. Countries with high numbers of minority governments include Denmark, Spain, Sweden, Norway (more than 3/4 of governments formed were minority governments), but almost all other countries have experienced at least one minority government in the analysed period. Only Germany, Luxembourg, Malta, Switzerland, and the United Kingdom have not experienced a minority government in the last twenty-five years.
The autonomy of committee systems may, apart from the hypotheses deduced from the theories, also be influenced by other factors which are controlled for. The institutionalisation literature views autonomous committees as a sign of a fully institutionalised legislature. In their study on committees in post-communist parliaments D. M. Olson and Crowther (2002, p. 3) argue that committees in legislatures of developed democracies are “the prime organisational means by which parliaments act independently of the executive”.
When autonomy is a critical factor in the capability of a legislature and associated with a process of institutionalisation, then the level of autonomy should be higher the more institutionalised a legislature is. To test this, ideally, data is needed which indicates the exact year of transition to the current parliamentary system. As a proxy, the analysis tests the effect of the number of years the country is classified as a parliamentary democracy. This is based on information provided in the Polity IV Country Reports (Marshall & Gurr, 2010), one of the most widely used measures of democracy. Although the number of years is just a proxy for institutionalisation, the individual country reports offer detailed information on the transition to democratic rule. An additional control variable was included to test the effect of bicameralism on the committee system. It is expected that a bicameral legislature establishes committees with less autonomy than a unicameral legislature. When the upper chamber is an additional controlling body there is no need for the committee system to exercise this task. Table 4.2 contains descriptive statistics on each of the independent variables.
By operationalising the concepts this way the analysis relies on a ‘snapshot’
of the dependent variable (the situation of committee systems in 2013). Several shortcomings of this approach need to be addressed. First of all, one might argue that committee systems evolve gradually and by analysing the specific configuration in one year it is possible to overlook the gradual changes that have led to this setup. It would, of course, be interesting to analyse how specific configurations have evolved. However, it is very difficult to keep track of small changes in committee rules with regard to all of the indicators which are used.
Therefore, a snapshot of the configurations by the time of this study without taking into consideration the gradual changes is a pragmatic choice. The aim of this analysis is to provide a first insight into the rationale of committee structures on a general basis.
Table 4.2: Descriptive statistics of the independent variables for the analysis of committee structures
Variable n Min. Max. Mean SD
Committee autonomy score 30 0.31 0.75 0.58 0.11
Majority seat share 30 38.99 78.05 53.35 8.07
Reliance on ‘personal votes’ 30 1.00 10.33 4.14 3.16 (log)Connection to districts 30 0.00 5.01 2.23 1.33
Corporatism 30 0 3.61 1.55 0.94
(log)Size 30 4.09 6.48 5.34 0.70
Average number of bills per plenary session
30 0.20 27.10 5.25 6.08
Ratio of coalition governments 30 0.00 1.00 0.74 0.35 Ideology distance gov. parties 30 0.00 6.84 2.87 1.63
Frequency minority
government
30 0.00 1.00 0.30 0.27
(log)Years parl. democracy 30 2.64 5.13 4.02 0.72
Bicameral 30 0 1 0.50 0.51
Source: Own data set.
A second objection which might be brought forward refers to the independent variables. When analysing the snapshot of committee structures, which are the result of an ongoing process, with several static variables it is possible to ‘miss’ the actual relationship and analyse the situation at the time of this study with variables which have not led to the establishment of the particular rules which regulate the committee system. This argument is taken into account by relying on averages over a longer period of time as much as possible. This prevents that outliers in a period bias the result. By using averages over several years, the independent variables capture a dynamic process in a single static value. Using these to account for the committee autonomy score is appropriate as rules are only slowly being reshaped and do not immediately follow changing patterns in the independent variables. Even though rules may stem from a time before the analysed period, the variables still explain the implicit decision to maintain the institutional rules. The committee system is seen as an institutional package, not as a number of independent rules. It is true that committee organisation is not stable over time and these rules may be changed at different points in time. Nevertheless, it is assumed that actors purposefully design each rule which leads to the overall committee system.
