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 3
Theoretical Framework:
Congressional Theories in a
Parliamentary Party Group-Centred Context
At least in a European context, “no one would seriously consider any alternative to political parties as the most important political coordination mechanism” (Müller, 2000, p. 316)
Using Congressional Theories Outside of the U.S. Congress: A Defence
T
H E main research question guiding the analysis is formulated as ‘how do specialised committees work in parliaments with strong parliamentary party groups?’. Within this broader question, the research focuses on structural features of committee systems of legislatures with strong parliamentary party groups, the criteria which play a role in the committee assignment process and, once assigned, the relationship between committee members and the parliamentary party groups. In order to tackle these three themes, a coherent theoretical model needs to be set up which allows for the formulation of testable hypotheses. As has been argued above, the main theoretical framework which deals with the issue of committees and legislative organisation is the congressional framework of distributive, informational and partisan theories.These theories are therefore the first perspectives I turn to. Similar to other
studies which have analysed committee workings in parliamentary systems, this study builds on this congressional framework, but adds the only alternative to the congressional bias, the ‘keeping tabs’ perspective by L. W. Martin and Vanberg (2011). It contains enough distinctive elements and is a valuable addition to the framework because it accounts for the usage of committees in coalition government situations.
As the congressional theories were originally developed against the backdrop of a particular legislature, one might object that their assumptions are specific to the United States and thus do not ‘travel’ well to other legislatures.
Some scholars have argued against drawing too much on the congressional theories (Hansen, 2011; Yordanova, 2011). However, the congressional theories are the main perspectives on how committees work. Although committees in legislatures outside the U.S. have since recently increasingly been in the scholarly focus, they lack their ‘own’ theoretical framework comparable to the U.S. theories. The knowledge and insight gained from a long-standing research tradition on the U.S. Congress and U.S. state legislatures can hardly be ignored and needs to be taken into consideration. After being tested on a variety of legislatures (state legislatures and Congress) across multiple points in times, research in the United States could not actually point to the dominance of one theory and became increasingly gridlocked in methodological and theoretical disputes.
Legislatures of parliamentary system of government which possess strong parliamentary party groups provide additional cases for thoroughly testing their predictions as they vary with regard to their internal organisation. Combining the rich literature on committees in the United States to the study of committees in other legislatures is, therefore, sensible. The predictions of the theories are not simply transferred but are adapted to the new institutional setting. Several structural features need to be addressed when looking at differences between the system of the U.S. and the legislatures used for the analysis. These, I argue, do not fundamentally prevent the transfer of the congressional theories, but need to be accounted for. In the following section I will first discuss the limitations of the congressional theories and then present a new framework which utilizes the congressional theories, but makes them applicable to the new institutional setting which comprises strong parliamentary party groups. I do this by using the general concepts of each theory as heuristic devices for the analysis.
Structural Differences Between the ‘New’ Institutional Setting and the U.S.
Congress
Generally speaking, two structural features prevent a direct translation of the predictions of the congressional theories to the legislatures used for the analysis.
A first structural feature which hinders a direct application to the ‘new’
institutional setting is the different electoral connection of legislators. In the United States, clear connections between legislators and individual districts are
established due to the reliance on single-member districts and the simple majority electoral system (usually labelled as ‘first-past-the-post’). A few countries outside of the U.S. rely on the same electoral system as the U.S.
Congress, usually those of the Westminster tradition (e.g. the House of Commons in the United Kingdom and Canada). In those chambers, all legislators possess a clear constituency link. This makes the prediction of the distributive theory, in principle, directly applicable. However, even in those legislatures, the level of connectedness to individual districts is probably weaker compared to the U.S. Congress. Legislators in the U.S. have an exceptionally strong connection with their districts and develop a personal ‘home style’
(Fenno, 1977, 1978) to cater to the needs of their constituents (Owens & Loomis, 2006). This is also underlined in the widespread practice of ‘providing pork’ to bolster the re-election chances of members of Congress, “stoutly defended[...]
as an essential form of representation of their constituents‘ needs” (Evans, 2011, p. 336). This practice is applicable to some countries with a parliamentary system of government, but not applicable to all legislatures. In most countries outside of the U.S., the electoral connection of MPs is much weaker. The reliance on party lists in some electoral system limits the potential for an electoral pay-off to a geographically defined constituency. This lack of strong ties to constituencies “removes many of the incentives that traditionally are seen to produce distributive gains” (McElroy, 2006, p. 10). Although McElroy referred to elections in the European Parliament, equal constraints apply for some countries with a parliamentary system of government. This is especially evident for MPs in the Knesset (Israel), the Národná rada (Slovakia) and the Tweede Kamer (Netherlands) where seats are distributed nationwide and which therefore lack clear constituency links.
Second, a direct translation would underestimate the influence of (by comparison) strong parliamentary party groups in the internal organisation of the analysed legislatures. These are all characterised by the existence of strong parliamentary party groups, defined as “powerful floor coalitions, capable of disciplining their members and passing their programs, [...] effectively dominating the legislative agenda and taking responsibility for the final legislative product” (Cox & McCubbins, 1993, p. 5). This is, for example, visible in the high level of voting cohesion.35In a comparative study of PPG unity in the 1990s, Sieberer (2006) finds average PPG unity of 97 per cent in his cases with Finland being a comparatively ‘extreme’ outlier with only 88.63 per cent voting cohesion. In the other countries, PPG unity in roll call votes varies between 94.75 per cent (New Zealand) and 99.83 per cent in Denmark.
In the congressional literature the question how much central parliamentary party group organisation restricts the behaviour of individual legislators is still unsettled. Although there is some consensus that parliamentary party group
35 Political science research has established several definitions of PPG unity (‘PPG cohesion’,
‘PPG discipline’ and ‘PPG loyalty’) and its determinants. Conceptual clarification between these concepts was provided by Andeweg and Thomassen (2011) who distinguish four primary mechanisms to achieve PPG unity.
organisation is not absent in the U.S., it is certainly weaker. The United States Congress is still characterised as the “terrain par excellence for the individual legislator” (Owens & Loomis, 2006, p. 263, italics in original). Prospective candidates nominate themselves, direct their own campaigns by raising large amounts of money and building up their own publicity to eventually offer themselves to a party. Subsequently, they “win (re-)election as individuals, not because of their party connections” (Owens & Loomis, 2006, p. 263). To a certain extent, the scope of internal cohesion of political groups in the European Parliament is also still debated. In contrast, this discussion is superfluous in the parliamentary party group-centred context of the legislatures in which parliamentary party groups have a major stake in the organisation of the legislature. Müller (2000, p. 316) noted that in a European context, “no one would seriously consider any alternative to political parties as the most important political coordination mechanism”. This statement is expandable to all national legislatures of the countries with a parliamentary systems of government analysed in this study. In these legislatures, there is no question that parliamentary party groups are central actors in the inner workings of the analysed legislatures and have an important role in the assignment procedure and in committee proceedings. Despite great variation in the organisational nature of parliamentary party groups across countries (Heidar & Koole, 2000) their involvement in the internal organisation of the legislature is undisputed.
Parliamentary party groups and committees are the two crucial organisational structures in parliaments, but the former clearly dominate the latter (Damgaard, 1995). Voting instability and majority cycling in parliamentary systems is prevented by strong parliamentary party groups. The presence of strong parliamentary party groups reduces uncertainty to avoid cycling in voting situations and therefore effectively takes out a core reason for the establishment of strong committees in the informational and distributive theories. The collective action problem that was argued to be solved by the committee system is here taken care of by parliamentary party groups as central authorities.
Distributive theory and informational theories view parliamentary party groups as procedurally weak and mostly non-restrictive with regard to the behaviour of individual legislators. These two perspectives have been labelled by Krehbiel et al. (2013) as principally non-partisan. Although they do not lack parliamentary party group organisation per se, they assume that parliamentary party groups are reactive and follow the preferences of individual legislators by being largely non-restrictive.36 This assumption fits with the traditional scholarly consensus which saw congressional parliamentary party groups as weak but allows the theories to explain institutional stability despite such weak and non-restrictive parliamentary party groups. Not parliamentary party
36 Weingast and Marshall (1988, p. 137), for instance, assume in their model that “parties place no constraints on the behaviour of individual representatives”. Krehbiel (1993) finds no convincing evidence for significant parliamentary party group behaviour, i.e. “behaviour that is consistent with known party policy objective but that is contrary to personal preferences” (Krehbiel, 1993, p. 240) and concludes it to be marginal.
groups, but committees bring stability, either as mechanisms for a structure-induced equilibrium or as means to reduce uncertainty on the floor.
This has led to the proclamation of the ‘self-selection’ model of committee allocation. This would hint at the supremacy of the partisan theory, as it already places parliamentary party groups as central actors to explain why legislatures are organised in the way they are. But such conclusions are premature.
A careful reading of the partisan theory reveals some general shortcomings which need to be addressed. First, it was developed with a bipartisan legislature in mind with clear majority and minority situations, letting the majority parliamentary party group acts as a procedural cartel using negative or positive agenda setting. The theory is very explicit about the majority parliamentary party group, but it is “deafeningly silent about the rights, resources, and influence of the minority[parliamentary party group]” (Krehbiel et al., 2013, p.
1). Second, outside the U.S. context, clear majority situations are rare. Looking at the number of governments in Western and Central Eastern Europe between 1989 and 2010 out of 273 formed governments 40 had a majority. Of the 233 governments where no party had a majority, 199 were coalition governments (Bergman et al., 2015). The way coalition situations change the dynamics in committee needs to be addressed in the analysis. A less “U.S.-driven” view on the establishment of committees should also take into consideration this occurrence of multi-party governments and highlight the significance of legislative institutions such as committees in coalition situations. The ‘keeping tabs on coalition partners’ theory by L. W. Martin and Vanberg (2011) is, therefore, a promising perspective for the relationship between parliamentary party groups and committees.
Adapting the Theories: A New Framework with Parliamentary Party Groups as Main Actors
These structural differences need to be taken into consideration when applying the predictions to the new institutional setting. Several strategies have been employed. A first strategy is to broaden the focus of the congressional theories.
As an example, a distributive rationale has in prior studies also been tested using connections to interest groups rather than constituencies (see Yordanova, 2009) to analyse whether an external interest drives the assignment. A second strategy takes each theory as an organisational rationale which can be pursued by parliamentary party groups. This means that the original focal point of the informational theory (the majority of the floor) shifts to the majority of the parliamentary party group. This overarching positioning of parliamentary party groups is e.g. used in an earlier study on committee assignments by Mickler (2013) and is pursued in this study as well.
In order to apply the congressional theories to the new institutional setting the most fundamental adaptation is a redefinition of the role of parliamentary party groups. Generally speaking, parliamentary party groups are the main
gatekeepers within the analysed legislatures and their involvement in the internal organisation is undisputed. In this sense, the partisan theory by Cox and McCubbins (2007, p. 17) is ‘correct’ in its premise that committees are not autonomous. There will be no ‘committee government’, no self-selection with legislators’ preferences being the sole decisive determinant of assignments.
However, this does not make the further empirical implications of the partisan theory correct by default. When premising the supremacy of parliamentary party groups over committees, it can be tested whether parliamentary party groups organise their work in committees according to a distributive, informational or partisan rationale.
It has been argued that, although the congressional theories need to be considered simultaneously, they cannot be merged because they “rest on different assumptions about the dimensionality of the policy space and the actors’ incentive structures” (Yordanova, 2009, p. 261). However, the central concepts of the congressional theories can be united in a common framework in a parliamentary party group-centred context. All three perspectives indicate valid, rational strategies to be pursued by parliamentary party groups. In order to clarify these different rationales, the basic concepts and predictions of each perspective have to be extracted. This has also been suggested by Krehbiel (1992, p. 21) who stated that “by highlighting differences we are able to learn more clearly which exogenous features of the legislative environment are and are not capable of explaining variation in legislative institutions and their use”.
The theories can be combined because, despite having different theoretical and empirical implications, they all share certain assumptions. All assume rational behaviour of actors. In all of these perspectives, legislators are motivated by clear goals, their own re-election and interested in making good policies (but putting varying degrees of emphasis on each one). They are all based around the question how legislators solve the collective choice and collective action problems that they face to generate an equilibrium solution to the problem of legislative instability. Yet the theories differ on how they achieve it. Distributive and informational theory simply work under the assumption that parliamentary party groups are existent but non-restrictive. This assumption of non-restrictiveness of parliamentary party groups advocated by the informational and distributive theory is relaxed for the analysis. In doing so and focusing on cases with strong parliamentary party groups, we can use the predictions to disentangle clear strategies and better understand how committee work is organised.
Extracting the main concepts and empirical implication of the new framework with regard to structure, composition and functioning
A major difference in the congressional theories is their placement within the
‘outlier versus non-outlier’ debate. Broadly speaking, all congressional theories (regardless of their perception of the role of parliamentary party groups) relate
their arguments back to this issue. In the original ‘congressional’ form, the distributive theory predicts the establishment of autonomous and outlying committees which present means to cultivate personal votes and profit from gains from trade. The informational theory argues that committees will not be outlying, but present an incentive structure for MPs to reach specialised decisions with the median floor preference being decisive. The partisan theory makes the outlying composition of committees dependent on the jurisdiction of committees (considering who is affected by its decisions). It is assumed that especially in those that have an effect on the national perception of the party (i.e. those offering uniform externalities37) the preferences of the committee members “will tend to have contingents that are microcosms of their party caucus” (Cox & McCubbins, 2007, p. 184). There might, however, be a bias towards the most loyal MPs. Those committees which offer targeted externalities may be outlying. Committees are utilised to maximise the number of seats the party will win at the next election and to induce loyalty. These represent the core of the perspectives and offer enough room to disentangle broad rationales on how strong parliamentary party groups organise their work in committees in legislatures.
A ‘parliamentary party group centred distributive rationale’ (hereafter referred to as distributive rationale) views committees as means to offer particularistic benefits to groups outside parliament, disproportionally composed of outliers and consisting of high-demanders across the board.
These outlying committees which cluster legislators who have a ‘stake’ in the committees’ jurisdiction are at the heart of the distributive theory. A distributive rationale is further strengthened, if all committee members are granted autonomy and individual legislators have extensive room for manoeuvre in committees. The committee system as a whole is highly autonomous and isolated from the influence of the parliamentary party group.
On the other ‘extreme’, according to the ‘parliamentary party group centred informational rationale’ (informational rationale) committees provide means to deal with the workload of the parliamentary party group. With regard to their composition and functioning, they are not outlying but are (more or less) representative microcosms of the parliamentary party group. Parliamentary party groups which follow an informational rationale with regard to legislative organisation do not allow for outlying committees. Although considerable room for manoeuvre is given to the committee members in order to minimise agency loss, the parliamentary party group remains the main anchor point. The basic assumptions of this perspective are directly applicable. Parliamentary party groups in all legislatures have to take decisions with regard to complex issues under uncertainty. Using the specialised committee system to facilitate the legislative process by delegating authority to legislators who can specialise is certainly a feasible strategy which parliamentary party groups can pursue. The
37 These committees “have the potential to affect the electoral prospects of virtually every member”
(Cox & McCubbins, 2007, p. 180)
policy choices of committee members will reflect the preferences of the majority of members of the parliamentary party group. Outlying committees are prevented. Legislators who have a major stake in the jurisdiction of the committee will not be assigned to it due to the fear of individual opportunistic behaviour. In the policy-making process, individual MPs are ‘closely watched’
agents of the parliamentary party group with clear reporting duties.
A ‘parliamentary party group centred partisan rationale’ (partisan rationale) predicts committees to be part of the reward system of parliamentary party groups to induce loyalty and highlights the conditional nature of the influence of the parliamentary party group leadership with regard to the composition.
This partisan rationale is used in this study for a ‘middle ground’. It highlights the conditional nature of outlying committees. Committees become part of the reward system to induce loyalty. The existence of outlying committees depends on the jurisdiction of the committee. If a committees‘ jurisdiction is important to the parliamentary party groups’ prospective electoral success (similar to the argument made by Cox and McCubbins (1993, p. 199) regarding the national perception of the party), the influence of the parliamentary party group leadership will be visible. Only the central leadership has the possibility to structure the composition of committees. Party loyalty is an important factor in the assignment procedure (Cox & McCubbins, 2007, p. 175: “Our results (...) indicate that loyalty to the party leadership is a statistically and substantively important determinant of who gets what assignment.”) reserved for certain MPs and the committee system is used for negative agenda control. Committees which have ‘uniform externalities’, i.e. those whose decisions affect all non-committee members equally, are closely monitored while the others are given more autonomy. When self-selection is granted, parliamentary party groups are assumed to watch this carefully and decide whether the outcome is contradictory to their seat-maximizing strategy.
The ‘keeping tabs’ rationale predicts committees to be used to monitor the coalition partners in coalition situations. It is a valuable addition to this theoretical framework as it highlights the processes during coalition governments. The structuring factors for all committee-related issues is the government situation. Committees are a central instrument to control other government legislators. This is also reflected in the selection criteria of committee members. Earlier analyses have usually focused on ‘high-value’
seats such as committee chairs and vice chairs. With regard to the working procedures, the ‘rank-and-file’ legislators will be primarily focused on the positions that legislators from the coalition partner propose. They are an early warning system and aim to prevent an eventual outcome that would be disadvantageous to their own parliamentary party group. Table 3.1 summarises the main concepts of each theory and how they relate to structure, composition and functioning of committees.
By amending the congressional theories in this way, the theories are united
Table 3.1: Translation of congressional theories into the new framework
Prediction with regard to:
Parliamentary party group centred...
Structure Composition Functioning
distributive rationale
Provision particularistic benefits
High-demanders, outliers
Autonomy
of individual legislators, all issues
informational
rationale Efficiency Specialisation, non- outliers
Median parl. party group, all issues
partisan rationale
Agenda-control by majority parl. party group
Active influence
in some
committees, loyalty
Intensity of
control depends
on electoral
importance of an issue
‘Keeping tabs’ ‘Watchdog’
see partisan theory, additional chair selection
Enforcing coalition agreement
under a common framework.38 I argue that the theories and their predictions are not broadened beyond reason and that the new framework still takes into consideration the different purposes of committees highlighted in each theory.
Rather, the ‘outlier versus non-outlier’ discussion of the informational and distributive perspectives are well-suited to be incorporated into a parliamentary party group-centred context. The theories indicate valid strategies that can be pursued by parliamentary party groups. For example, parliamentary party groups can use the internal organisation of legislatures primarily to facilitate the re-election chances of their MPs by offering high-demanding legislators a chance to cluster (distributive rationale) in committees. This framework has clear predictions for the institutional design and can be used to identify the different conditions which cause the prevalence of a theory in a given context.
38 Similar different rationales have been proposed by other authors. Yordanova interprets their predictions as to “(1) serve special interests outside the European Parliament, (2) bring informational benefits to the plenary or (3) promote partisan interests”(Yordanova, 2009, p. 255).
Mattson and Strøm (1995, p. 255) state that committees are seen as either “(a) arenas of high demanders; (b) an efficient mode to manage information or (c) extensions of majority parties”.
Hypotheses and Expectations Using the ‘New’ Framework
Hypotheses on Committee Structures: What Explains Structural Variation?
The predictions of the congressional theories of legislative organisation and the
‘keeping tabs’ perspective are first tested on variation in committee structures.
This cross-country analysis includes a large-n of legislatures from countries with a parliamentary system of government (the case selection is discussed further below). The analysis focuses on the concept of committee system autonomy. Fenno (1966) uses the concept of committee autonomy to describe the degree to which decision making in committees is influenced by members and 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. (Atkinson &
Nossal, 1980).
One general notion needs to be made with regard to the nature of the argument of this first part. The analysis of formal committee structures attempts to shed some light on why variation of committee autonomy occurs.
The variation is mostly accounted for by institutional variables which are slow-moving and may not lead to rapid changes. This makes it difficult to establish a causal connection. The argument that is made in this part is therefore not of a causal nature, but rather seeks to establish correlation. It is assumed that, over time, a match will be established between the political situation and the organisation of a parliament. Using the theoretical framework we can deduce testable hypotheses on factors which may go together with particular structures.
The partisan theory highlights the role of the majority parliamentary party group to structure committees. With regard to committee organisation, it follows that “the extent to which committees are autonomous actors in the legislative process depends on the extent of control the majority parliamentary party group wields in the chamber” (Martorano, 2006, p. 209). The smaller the difference between majority and minority parliamentary party group, the more likely it is that the majority parliamentary party group exerts control “over the legislative process to ensure that the ‘agreed upon’ deals come to fruition” (Martorano, 2006, p. 213). However, outside of the special U.S. case, the relationship between the size of governing party and committee autonomy might be reversed. Sieberer and Müller (2015), for example, show in their study that governing parties have an incentive to control the legislative structures to prevent the opposition from using them for obstruction. A governing party with a higher vote share is better able to seek legislative reforms to control committee deliberations. Because of the different institutional setting from the U.S., this expectation is followed.
As has been noted above, while in the United States usually clear single party majorities occur, coalitions between parliamentary party groups are more common in the countries analysed. This necessity to rely on multiple parliamentary party groups to form a government has to be accounted for. It is
expected that the smaller the seat share of the governing parliamentary party group or coalition is, the less autonomy will be granted, or the other way around. A general caveat needs to be made with regard to the usage of ‘fixed’
scores of several independent variables. Committee structures are less volatile than election results. To account for this, averages over a longer period of time are used.
Hypothesis 1: A structurally larger relative seat share of the governing parliamentary party group or coalition (across multiple legislative periods) is positively related to the autonomy of committee system.
The distributive theory argues that committees ultimately serve the interest of individual legislators to secure their own re-election. It follows that the electoral connection of individual MPs is an important organisational rationale.
Carey and Shugart (1995) focus on varying intra-parliamentary party group consequences of electoral laws. They argue that certain configurations induce legislators to rely on their personal reputation, instead of a parliamentary party group reputation, for electoral success. Autonomous committees offer individual legislators the chance to engage in advertising, credit claiming, and position taking (Mayhew, 1974). Therefore, more autonomous committees are expected in legislatures whose MPs can rely more strongly on ‘personal votes’.
In other electoral systems, the necessity for highly autonomous committees is not present.
Hypothesis 2: Electoral systems which allow individual legislators to rely on a
"personal vote" are positively related to the autonomy of the committee system.
An additional hypothesis concerning this perspective takes the district link of individual MPs into consideration. The distributive rationale places the link of individual MPs with constituencies central. This concept can be applied to the study of structural features as well. It follows that committees are expected to present such incentive structures in those countries in which individual MPs have a relatively closer connection to a district. Although reminiscent of Hypothesis 2, the extent to rely on ‘personal votes’ is related, but not the same as the relationship with individual ‘districts’. Open-list proportional representation systems allow for a certain degree of a personal vote, even though they might lack a connection to districts. One example is the Dutch Tweede Kamer, the legislators of which are elected in a nationwide district.
However, the Dutch electoral system still allows for a personalised vote. It is, therefore, important to distinguish both hypotheses.
Hypothesis 3: A stronger connection between individual MPs and districts is positively related to the autonomy of the committee system.
The adapted reading of the distributive theory highlights the outside interest which committees serve. Committees are central venues for an
in-depth investigation of a particular matter. Due to their involvement in the policy-making process, they are also prime organisations to present a communication channel between members of parliament and external organisations such as interest groups and their participation in the policy making process. The formal structures of the committee systems can be organised to facilitate these networks and discussions between interest groups and legislators. Such “institutionalised pattern of policy-formation” Lehmbruch (1979, p. 150) are usually described under the term corporatism. Under corporatism, policy-making involves negotiations (or at least consultations) between the state and interest organisations. Higher levels of corporatism are expected to be related to the establishment of more autonomous committees.
In less corporatist countries, the necessity to establish strong committees is less prevailing.
Hypothesis 4: Higher levels of corporatism of a country are positively related to the autonomy of the committee system.
According to the informational theory, committees are established to improve the efficiency of the legislative process as a whole by providing means to specialise and deal with the workload of the legislature. A straightforward test for this rationale is whether the size of the parliament explains committee autonomy. Larger legislatures offer more possibilities to relocate decisions from the plenary to committees. In these situations, more autonomy is granted to the committee system for efficiency purposes.
Hypothesis 5: A larger size of parliaments is positively related to the autonomy of the committee system.
The test regarding the structural features of a legislature (size) of a legislature is a relatively rough test to account for efficiency maximisation via the committee system. Several studies (Polsby, 1968; Cooper, 1970; Froman, 1968; Hedlund, 1984) have shown that an “increased demand for public policy also resulted in an overall strengthening of committee systems in the legislative process” (Martorano, 2006, p. 211). This argument which utilizes the efficiency maximisation of committees can be directly transferred to the study of parliamentary systems of government. It is hypothesised that more an increasing workload is related to the establishment of more autonomous committees. In order to prevent an overflow of bills and to cope with this workload, a legislature might establish more powerful committees to solve inefficiencies in the legislative process. One may argue that legislative workload is endogenous to the level of autonomy of committees. Instead of high workload leading to more autonomous committees, it may be that more autonomous committees increase the number of laws a parliament has to work on, especially if committees have the right to propose legislation. The analysis will check whether there is bias in the data and the number of bills is indeed much higher in those countries which allow for committees to introduce bills.
Hypothesis 6: A higher workload of a parliament (over an extended period) is positively related to the autonomy of the committee system.
The “keeping tabs”-perspective argues that committees serve as watchdogs to ministers in coalition governments and to prevent ministerial drift. Following this logic, a decisive argument for the establishment of autonomous committees is the occurrence of coalition governments. If this is the case, then the committee system will be granted the necessary tools to fulfil this task and thus be more autonomous.
Hypothesis 7: Frequent coalition government is positively related to the autonomy of the committee system.
Earlier studies have shown that a minister is more likely to be ‘shadowed’ by a committee chair of a coalition partner when the disagreement between the coalition parliamentary party groups increases. This argument can be extended to committee structures: The establishment of a strong committee system is predicted be related to the ideological distance between coalition partners. When this difference is structurally greater (e.g. due to a frequent necessity to establish government coalitions between parliamentary party groups which differ in their ideological views), then the institutional rules guiding the committee system should be structured in such a way as to offer committees to acquire information and amend legislation. The analysis tests whether a larger difference between coalition parliamentary party groups increases committee autonomy. Similar to hypothesis 1 an average over an extended period is used in order to account for the fact that ideological distances between coalition partners shift after the formation of a new government.
Hypothesis 8: The formation of coalition governments with greater ideological distances between the coalition partners on a regular basis (across multiple legislative periods) is positively related to the autonomy of the committee system.
Another argument relating to the efficiency of monitoring the government links the structural variation in committee autonomy to the frequency of minority governments. Minority governments occur when the governing parliamentary party group or governing coalition do not hold a majority of seats in the chamber. They are a common occurrence amongst the legislatures analysed. Amongst European countries the highest frequency of minority governments are Spain, Denmark, Sweden and Norway, but they have also occurred in Westminster-style legislatures of New Zealand and Canada. The connection between the committee system and the occurrence of minority governments was highlighted by Strøm (1990). Strong committees grant the possibility for opposition parliamentary party groups to influence policies. It is likely to assume that parliamentary party groups, when confronted with frequent minority governments, will establish committee systems with high autonomy. The tested hypothesis does not follow the direction of causality of
the argument of Strøm who suggested that strong committees facilitate minority governments. In this analysis, correlation is sought. It is tested whether a frequent occurrence of minority governments is positively related to the establishment of strong committees. The analysis is guided by the assumption that the rules are dependent on the background situation in which legislators organise their work.
Hypothesis 9: A higher frequency of minority governments across an extended period is positively related to the autonomy of the committee system.
Hypotheses Committee Assignments: Who Gets What and Why?
The second analysis of this dissertation focuses on the way parliamentary party groups structure their work within parliamentary committees. In a first step, the question of ‘who gets what?’ is addressed by means of a statistical analysis of actual assignments. Parliamentary party groups are assumed to be major gatekeepers in the assignment of committee members. A completely autonomous ’self-selection’ of legislators to committees is unlikely. There are, however, no fixed, written rules about this process. To test which underlying assignment rationale prevails, the assumptions of the theories can be used to draw conclusions on informal rules of the processes within parliamentary party groups. The different hypotheses allow for several possibilities on how seats in these jurisdiction-specific sub-units are assigned.
The distributive theory highlights the autonomy of committees and the outlying preferences of committee members. Committees are viewed to serve an interest of individual MPs. These individual legislators rely on the committee systems to serve their electorate and secure their own re-election. The electoral connection of legislators is central and needs to be tested. A re-election principle in the assignment process means that those members of a parliamentary party group join those committees which correspond to an
‘electoral demand’ of a legislator. The first choice for testing this rationale is whether a ‘district link’ is present and whether high-demanders join committees in which they can serve their constituency.
Hypothesis 10: Legislators are more likely to serve on committees that correspond to their electoral connection with constituencies compared to legislators who do not have such an electoral connection.
In earlier studies (see Yordanova, 2009) the ‘stake’ in a committee was interpreted more liberally as serving ‘external interests’. One proposed alternative is using ties to interest groups as drives to be assigned to a certain committee. This is used as a proxy of whether an MP is a possible
‘high-demander’ that can shift legislation into a direction that is not representative of the parliamentary party group. These external interests are not involved in the election of an MP, but the general idea is nevertheless applicable:
MPs can be high-demanders in a different way by serving interests outside
parliament. The question arises whether this is a realistic scenario. Why would parliamentary party groups allow for MPs to cater to those interests? Two points need to be made in order to address this. First, it is assumed that parliamentary party groups are aware of the possible negative consequences of such an assignment logic and carefully consider this. Instead of viewing this assignment logic as something ‘shady’ it should be highlighted that the connection to interest groups also brings expertise into the decision-making process.
Parliamentary party groups might value this higher than the risk of possible negative outcomes. This is connected to a second issue. The test so far is merely an analysis on what explains committee assignments, not on what legislators can actually do. Parliamentary party groups still maintain other possibilities to affect the behaviour of legislators in committees. This is addressed in the last section of the dissertation. Understanding that legislators are assigned based on this rationale does not allow for conclusions on their behaviour. An important side-note needs to be made with regard to this hypothesis. Yordanova (2009) argues that being assigned to committees due to external interests does not necessarily imply a distributive rationale - it is the combination of external interests that is important. The distributive rationale predicts the staffing of committees with homogeneous preference outliers, for example, legislators with ties to labour unions in a committee dealing with employment and labour affairs (assuming that this does not represent the ideological position of the other members not on the committee). However, parliamentary party groups might assign legislators with external interests to extract expertise but then match those members with legislators with interests biased in the opposite direction (Yordanova, 2009). To stick to the above-mentioned example, a committee dealing with employment and labour affairs could also contain legislators with ties to employers’ association in order to achieve a balance of heterogeneous preference outliers. The analysis will account for this possibility.
Taking this into account, the formulation of the hypothesis is:
Hypothesis 11: Legislators are more likely to serve on committees whose jurisdiction corresponds to their external interests compared to legislators who do not have such a connection.
The informational theory highlights the importance of specialised, efficiency improving committees. Contrary to the distributive theory, outlying committees are a major concern due to the feared agency loss. Rather than seeing the legislature as decentralised, “(a) rational legislature is one that efficiently taps the special talents of its legislators” (Krehbiel, 1992, p. 136). The informational theory predicts that committees are not outlying, but that strong parliamentary party groups prefer to have policy experts in committees. This prevents the likelihood of outcomes which are not desired by the parliamentary party group. This is a valid strategy not only for the legislature in general but especially for strong parliamentary party groups and their members.
Parliamentary party groups consist of a number of legislators with various
backgrounds. It is expected that, in order to ‘tap the talents’ of their members, a parliamentary party group appoints those members who are especially interested in an area and can specialise at low cost due to advantages in knowledge related to a policy area. This might, for example, be due to their prior experience with the policies that lie within a committee’s jurisdiction. It should be noted that experience and external interests may sometimes be linked. As an example, medical doctors are more likely to join medical associations. The same goes for farmers and farmer’s associations. However, if an informational rationale prevails then their interest group connection is not the reason to be appointed and might even decrease their chances.39
Hypothesis 12: Legislators are more likely to serve on committees in which they can specialise at low cost compared to legislators who do not have similar advantages in knowledge related to a policy area.
The assignment of members can also be analysed in terms of the continuity of membership across terms. Consecutively assigning members to the same committee supports the claim of the informational theory that the committee system enables legislative specialisation among members. One might object that this could also be the case for a distributive or a partisan rationale.
However, the underlying principle of re-assigning legislators is best understood in terms of an informational rationale. This has also been the interpretation in studies of the U.S. Congress. Re-assigning legislators creates an incentive toward greater legislative specialisation as members settle into the same committee slots to “cultivate expertise in a distinct policy field, and spend their time managing legislation and conducting oversight in that field” (Davidson &
Oleszek, 1996, p. 135).
In the congressional context, continuity of membership across terms is sometimes viewed as part of the distributive theory (with reference to ‘property rights’). However, I do not view a continuation of membership as something distinctive distributive. The existence of the ‘property rights norm’ in the U.S.
Congress precedes the formulation of the distributive theory but was, when formulating the theory, added into the theoretical framework. I rather argue that the most important difference between informational and distributive theory is between ‘preference outliers versus non-outliers’. Nevertheless, it is true that solely drawing conclusions based on this variable would not be enough to conclude the informational mechanism of specialisation. To be able to view it as evidence for an informational rationale is only valid under certain conditions: First, the underlying mechanism should be based on expertise. This is why it is important to view the results of this variable together with other variables in the model and with the additional evidence from the interviews. If the variable measuring advantages in knowledge is not related to the assignment of committee members and/or the interviewees do not highlight
39 Such concerns are addressed by controlling for the other rationale in the statistical model and later on in the interviews with MPs.
expertise as an added value (but rather highlight distributive factors), then a continuation of membership cannot have specialisation incentives (see quote above by Davidson & Oleszek, 1996, p. 135). Second, this hypothesis assumes that a continuation of membership to a committee is not merely done due to path dependency in order to avoid transaction costs of a new legislator who needs to familiarise him- or herself with a subject matter. It needs to be verified whether the assignment process at the beginning of a legislative period indeed offers a chance to remove a sitting legislator from a committee.
Hypothesis 13: Legislators are more likely to serve on committees when they have been a member on the committee in a prior legislative period compared to legislators who do have not served on the committee.
The partisan theory highlights the role of parliamentary party groups in controlling committees. With regard to selection criteria, the theory highlights the proactive role of the parliamentary party group leadership. Prior research in the U.S. tested a partisan rationale by analysing partisan stacking, i.e. whether the majority parliamentary party groups seeks to control legislative committees by numerical over-representing its own members on standing committees (Hedlund et al., 2009). This, however, is not a viable hypothesis as proportional assignment is the rule and parliamentary party groups‘ ratios in the plenum are usually mirrored in committee. Rather, the analysis focuses on whether the parliamentary party group leadership has an active role in the assignment process and ’structures‘ the composition of committees. We know from earlier research in the U.S. that committees differ with regard to their level of desirability. An important asset in this regard is the assignment of desirable committee seats and granting transfers to ’more desirable‘ committees. The party leadership controls committee seats as valuable assets and ’reserves‘ seats on some committees for particular legislators.40One of the factors which might be used in this process is the number of legislative periods (parliamentary seniority). Parliamentary party group leaders are expected to prefer experienced legislators to join particular committees and withhold other from serving on them.
In terms of which committees are structured this way we can turn to the prediction of the partisan theory. The partisan theory links the predisposition of committees to be non-outlying to the significance of its jurisdiction for the electoral success of a parliamentary party group. Cox and McCubbins (2007) build their model on the assumption of seat-maximizing leaders. Appointments are influenced based on considerations on the effect of the committees’ policy decisions. Committees which “have the potential to affect the electoral prospects of virtually every member” (Cox & McCubbins, 2007, p. 180) will tend
40 There are also other possibilities to offer legislative benefits. In Congress this is routinely used by leaders on the rank-and-file members by scheduling a bill for vote under favourable conditions, but also to punish disloyalty by withholding such benefits (for a discussion of benefits see Pearson, 2005).
to have contingents that are microcosms of their party caucus (allowing for a bias towards the most loyal MPs). Those committees which only affect smaller subgroups of members may be outlying. This means that the parliamentary party group leadership most likely ’reserves’ seats on those committees whose policy-area concerns an important issue-domain of the party. This argument resembles the seat-maximizing strategy of the partisan theory.
Hypothesis 14: Legislators who have served for more legislative periods are more likely to serve on committees whose jurisdiction concerns an important issue-domain of the party compared to legislators who have served fewer legislative periods.
Parliamentary seniority can be used to test whether long-standing MPs have a higher chance of being assigned to more important committees. The partisan theory views committees as part of the reward system of parliamentary party groups. Research on the U.S. Congress shows that voting in line with the parliamentary party group leaders had a significant impact on the assignment of MPs to prestigious committees (Cox & McCubbins, 1993, see especially ch. 7).
Committees are, as authors have claimed, loyalty-generating institutions in which the parliamentary party group leadership has a significant role (Coker &
Crain, 1994). A more thorough test of this needs to include more fine-grained data on individual legislators and their loyalty to the parliamentary party group.
Loyalty as a concept refers to a firm adherence to the parliamentary party group and its goals. The partisan theory is explicit on which committees are ideologically outlying. When a committee’s jurisdiction is important to the electoral success of a parliamentary party group, then it will be less likely for high-demanders to be assigned to this committee. This might be detrimental to the party as a whole. Rather than being non-outlying and composed of high-demanders, the strategic hand of the parliamentary party group leadership is visible. Instead, legislators who are ideologically closer to the parliamentary party group are expected to be preferred by the parliamentary party group leadership. Those MPs are disproportionally assigned to committees whose jurisdiction concerns an important issue-domain of the parliamentary party group.
Hypothesis 15: Legislators who are closer to the ideological mean of the parliamentary party group are more likely to serve on committees whose jurisdiction concerns an important issue-domain of the party compared to legislators who are further away from the ideological mean of the parliamentary party group.
A central prediction of the ‘keeping tabs’ perspective is that legislative institutions are used to control coalition partners. A test for this rationale with regard to assignment patterns is done by analysing the number of watchdog chairs. An analysis by Kim and Loewenberg (2005) of the German Bundestag showed that between 1961 and 1998 the CDU and the SPD used such watchdog
chairs only about half as often as they could have. In a coalition of CSU and FDP they were put in place nearly all of the possible time (Kim & Loewenberg, 2005, p. 1113). This strategy is an important rationale to consider.
Hypothesis 16: Committee (vice) chairs are assigned to ‘shadow’ ministers from the coalition partner.
The statistical analysis is intended to find patterns in the assignment procedure. The evidence is used to get a first idea of the assignment procedure and serves as a basis for the interviews that were carried out in each case. The analysis of assignment criteria is a first step in understanding the dynamics and is further investigated in the qualitative part of the study. Only when these are also taken into account, a complete view of committees can be given as a particular assignment rationale does not yet give any indication whether considerable autonomy is granted to individual legislators. This might still be the case which is why a sole analysis of who gets what could be misleading.
Expectations Committee Mode of Operation: How Autonomous are Committee Members?
Sartori viewed committees as an ideal arena for conflict resolution and consensus seeking, stating that “what is peculiar to committees is that their members engage in exchanges over time and having especially in view a future time” (Sartori, 1987, p. 229). Therefore, committee work is not fragile but rather
“the optimal decision-forming unit” (Sartori, 1987, p. 237). Relying on committees for the formulation of decisions thus “reflects the very essence of democratic decision-making” (Damgaard & Mattson, 2004, p. 113). The focus of the third part of the analysis lies on how the individual legislators in committees are related to other actors in the parliament. Committees are privileged institutions which, at least theoretically, offer property rights in a given policy area. However, legislators in committees usually are also members of a parliamentary party group. The ‘basic’ relationship of the relationship between individual legislators and his/her parliamentary party groups is depicted in Figure 3.1 below. This formal setup shows several legislators from one parliamentary party group who are allocated to committees. This schematic depiction is, at first sight, applicable to all parliaments. The question arises, how autonomous are members in committees? How is the work in committees communicated and coordinated with other legislators from the same parliamentary party group who are not on the committee and how do legislators on the same committee work together?
The qualitative part of this dissertation gives more insight into these relationships of individual committee members and other actors in the legislature. Altogether, four relationships are covered in the analysis. These are deemed to be the most important ones for the purpose of this study. The ‘point of departure’ is the individual legislator in a committee. The analysis focuses specifically on a legislator’s relation (here denoted as MP1A) with:
Figure 3.1: Schematic depiction of relationship between one parliamentary party group and its members in committees
Source: Own depiction. For Parliamentary Party Group B and C (see Figure 1.3) only the individyal committee members are listed in this picture.
• ‘rank-and-file’ legislators from the same parliamentary party group on the same committee (here denoted as MP2A, MP3A)
• ‘rank-and-file’ legislators from the same parliamentary party group not on the same committee (MP4A, ..., MP11A)
• ‘rank-and-file’ legislators from other parliamentary party groups on the same committee (MP1B, MP2B, MP3B, MP1C, MP2C)
• the leadership of their own parliamentary party group (MP12A)
The theoretical framework sketches several broad approaches which can be distinguished to capture the relationship between individual legislators in committees and these four other actors/groups of actors. It should be noted that the theoretical framework is to a certain extent stretched for this part. The theories were not developed to account for this type of analysis. However, it is possible to deduce several clear ‘strategies’ which follow the predictions of each theory. With these rationales enough theoretical predictions are incorporated to allow for a comparison between parliamentary party groups and systems. None of these rationales dispute the influence of parliamentary party groups (their principal involvement in the organisation of parliaments has been discussed in the section on the theoretical framework, see p. 39) but rather make subtle differences with regard to how the individual elements relate to each other.
Table 3.2 clarifies the various rationales.41
Table 3.2: Prediction of each theory with regard to the functioning of committees
Prediction
regarding... Distributive Partisan/ ‘keeping
tabs’ Informational
legislators in committee
autonomous, no reporting requirements, largely independent from each other
autonomy and reporting requirements depending on
jurisdiction
little autonomy, clear reporting requirements on all issues
(primary) focal point of legislators
own committee, low amount of
collaboration
parliamentary party group leadership
parliamentary party group
leadership of parliamentary party group
Veto only in exceptional cases
Highly influential in
specific issues Umbrella function
highly autonomous committee members from parliamentary party group
←− autonomy −→
strong involvement of parliamentary party group
According to an informational logic of committee workings, unintended changes at the committee stage are undesirable and a major concern to other legislators from the same parliamentary party group. Subsequently, the rank-and-file members and the leadership of the parliamentary party group are expected to keep a close eye on everything that happens in committee. In terms of room for manoeuvre for the individual legislators on the committee, very little autonomy is granted to committee members. Even though they are assigned to the committee and may develop the initial position autonomously, they have clear reporting requirements. The positions that are taken in committees can be vetoed by the majority of the parliamentary party group who still remains the main actor. The parliamentary party group leadership has a veto right in all areas and can take over issues it deems to be of major importance, but it is still accountable to the parliamentary party group. Within committees, similar proceedings are maintained. Legislators are not autonomous in their decisions but need to coordinate and communicate their positions with other legislators from the same parliamentary party group. These legislators are a main gatekeeper for checking positions. Working together closely with legislators of the same parliamentary party group is encouraged, working together with legislators from other parliamentary party groups on the same committees is not.
41 Autonomy here refers to general independence from other actors at all stages during the committee proceedings.
On the other ‘extreme’, a parliamentary party group may give a high degree of autonomy to the committee members in their work (distributive rationale).
Being assigned to a committee means having ‘property rights’. According to this rationale the actual committee members are relatively independent in their work from the parliamentary party group. They are able to develop the positions taken in committee with a high degree of autonomy from their parliamentary party group colleagues on the same committee and other legislators who are not on the same committee. There are very few reporting requirements and limited possibilities for other legislators from the same parliamentary party group who are not on the committee to veto a position taken by committee members. Committees develop into relatively closed networks, in which only those who are actually on the committee have a say.
The relationship with legislators from other parliamentary party groups on the same committee is closer than those with legislators from the same committee who are not on the committee. The parliamentary party group leadership grants high levels of autonomy to committee members as well.
In terms of committee workings, the ‘keeping tabs’ perspective and the partisan theory have similar predictions. Both theories can be interpreted as a
“medium” way which highlights the conditional nature of monitoring and actions. Both are used in this study for a ‘middle ground’: some committees are more closely monitored, while others are not. They differ, however, in the reason on why these committees are special. The partisan theory highlights the importance of certain issues which are of major importance for the parliamentary party group at the next elections. The committees dealing with these issues are, therefore, assumed to be more closely monitored. The ‘keeping tabs’ perspective highlights policy disagreement between the parliamentary party group of a minister and ministers of the coalition partners to explain actions of the parliamentary party group. To provide evidence for this perspective, the interviews should uncover that legislators who are assigned to committees which deal with divisive issues between coalition partners are more closely monitored than others. With regard to the relationship between other legislators from the same parliamentary party group, committee members on these ‘high importance’ committees have to follow the parliamentary party groups‘ preferences closely. Those committees whose subject matter only concerns a small subgroup of legislators and are not of interest to the majority of the parliamentary party group are relatively autonomous. All in all, the role of the leadership is strengthened compared to the informational and the distributive theory. Decisions taken in committee and positions which are developed need to be communicated and coordinated with the leadership rather than the ’rank-and-file‘ legislators from the parliamentary party group.
The main task of the leadership is to ’protect’ the parliamentary party group from undesirable outcomes which might harm it in the long run.
Method of the Study: Large-n Cross-National Analysis of Committee Structures, Small-n Focused Comparison of Committee Workings
Following the main research question and sub-questions the study consists of three analyses. Each sub-question is tackled with a different methodology and case-selection. One of the primary goals of this study is to broaden our understanding of committee structures beyond the special case of the U.S.
Congress to cases which possess varying degrees of strong parliamentary party groups. The ‘baseline’ case selection therefore includes advanced democracies with a parliamentary system of government. Parliamentary party groups within legislatures of these countries have an incentive to be strongly organised due to the relationship that the legislature has with the executive branch. A strong internal organisation and cohesiveness allows them to collectively support the government or bring it down. In order to count as an advanced democracy a standard measure was employed (>0.8 on the Human Development Index, see United Nations Development Programme (2015)). These cases do not just include Western countries but also post-communist countries that exhibit the hallmarks of economic development and democracy, and includes France which is usually classified as semi-presidential. Altogether, 33 countries were analysed. In case of bicameral legislatures the dataset comprises the popularly elected lower chambers.42 Appendix 1 gives a more detailed overview of size and mode of election for each chamber.43All of these cases are characterised by the existence of (varying degrees of ) strong parliamentary party groups. The comparative analysis of committee systems (Chapter 4) includes all of the legislatures mentioned above. The focus of the subsequent chapters lies on
42 The following cases are included (LH = Lower House in case of bicameral legislatures):
Australia: House of RepresentativesLH; Austria: NationalratLH; Belgium: Kamer van VolksvertegenwoordigersLH; Canada: House of CommonsLH; Croatia: Hrvatski sabor; Czech Republic: Poslanecká snˇemovnaLH; Denmark: Folketing; Estonia: Riigikogu; Finland: Eduskunta;
France: Assemblée NationaleLH; Germany: BundestagLH; Greece: Vouli ton Ellinon; Hungary:
Országház; Iceland: Althing; Ireland: (Dáil Éireann)LH; Israel: Knesset; Italy: Camera dei DeputatiLH; Japan: Sh ¯ugiinLH; Latvia: Saeima; Lithuania: Seimas; Luxembourg: Chambre des Députés; Malta: House of Representatives; The Netherlands: Tweede KamerLH; New Zealand:
House of Representatives; Norway: Storting; Poland: Sejm Rzeczypospolitej PolskiejLH; Portugal:
Assembleia da República; Slovakia: Národná rada; Slovenia: Dr˘zavni zborLH; Spain: Congreso de los DisputadosLH; Sweden: Riksdag; Switzerland: NationalratLH; United Kingdom: House of CommonsLH. Singapore, although being a parliamentary republic in the Westminster tradition is excluded as it is a hybrid regime. Further excluded legislatures are the House of Assembly in the Barbados and the Liechtenstein Landtag. In both legislatures the committee system is not comparable because they only delegate ’housekeeping’ matters to committees.
43 The overview contains information on the size and electoral mode on lower chambers (if not unicameral), as well as the last date of the general election as of 2013 and the maximum length of the legislative period. All information reflects the status in Summer 2013. When compiling the list, the priority was to use first-hand sources like the parliamentary website or other official sources (e.g. National Electoral Commissions). Only if those sources did not provide the information other sources were used, primarily the data on parliamentary elections by the Norwegian Social Science Data Services (NSD).
