Beyond rules and resources: Parliamentary scrutiny of EU policy proposals Daniel Finke, Heidelberg University (
[email protected]) Annika Herbel, Heidelberg University (annika.herbel @uni-heidelberg.de)
Forthcoming in European Union Politics issue 16/4
Abstract In this paper we study the factors that shape parties’ motivation to invest time and other resources in scrutinizing European Union (EU) policy proposals. We distinguish between two different motivations to engage in scrutiny activities. First, parties use scrutiny mechanisms to directly influence the national position. Second, parties play the two-level game, which is widely discussed in the literature on EU politics, and use scrutiny to manipulate their negotiator’s domestic constraints. We argue that both arguments depend on a set of conditions, namely the government’s relative strength in Brussels, the transparency of the EU decision making process as well as the government’s relative strength and cohesion in the domestic arena. On the empirical side, we study scrutiny at the level of committees in the national parliaments of Finland, France, Germany, Ireland, Italy, Poland, Slovakia, and the United Kingdom over a 13-year period, during which 26 governments are covered. Our findings suggest that parties deploy scrutiny to shift the domestic constraint strategically, but only if such a shift can be communicated convincingly to the international bargaining partners. Moreover, our findings suggest that opposition parties use scrutiny to influence the position of a weak government.
From a constitutional perspective, the main reason for parliamentary oversight is to reduce the deputies’ information deficit and avoid shirking by the government. Accordingly, parliamentary oversight is most relevant for those executive decisions that are not subject to parliamentary approval, as is often the case in foreign affairs and European Union (EU) politics. Here, governments decide legislation in the Council of Ministers that is binding at the national level. Only in the case of directives does the parliament have the possibility of influencing implementation policies within a prescribed range. Therefore, the governments’ accountability decreases when an increasing number of policy areas are shifted to the EU level. The limited transparency of the Council’s decision-making process increases the information deficit of the domestic opposition, who find it hard to assess the set of politically feasible policy alternatives. From a normative perspective, this extension of governmental discretion calls for additional parliamentary scrutiny activities. Currently, our knowledge regarding why some EU law proposals are being scrutinized whereas others go unchecked is limited. The literature on parliamentary scrutiny of national legislation provides two prevalent arguments, both of which focus on the importance of political parties. First, compared with parties in the opposition, the government in general and the leading minister in particular possess superior information on any specific law proposal. Hence, the opposition has a motivation to reduce this information asymmetry by means of parliamentary scrutiny (Döring, 1995). Second, parliamentary scrutiny is used to reduce the information asymmetry between the party of the lead minister and the coalition partner (Martin and Vanberg, 2004). However, these approaches have been developed for domestic politics and neglect the opportunities and incentives available to parliamentary parties in EU politics. Both approaches share the assumption that the overall motivation for scrutiny is policy seeking, which presumes that either opposition parties or coalition partners are capable of changing the policy. This presumption appears all the more questionable with a view to EU politics. Therefore, the existing theoretical approaches have been adapted to account for parliamentary scrutiny of EU law proposals (Finke and Dannwolf, 2013). Here, we study the conditions under which parties are motivated to engage in parliamentary scrutiny. Theoretically, government’s capability to successfully deploy the veto threats implied by the two-level-games literature should depend on its relative power in Brussels. By contrast, the comparative literature argues that the use of scrutiny depends on the strength of parliament vis-à-vis government. Therefore, we analyze committee involvement in the national parliaments of small states with strong parliaments (Finland, Slovakia), a small state
with weak parliaments (Ireland), large states with strong parliaments (Germany, Poland, Italy) and large states with weak parliaments (the UK, France). Our data set covers a period of 13 years and includes 26 governments during this period. Our findings suggest that the seat share of the government vis-a-vis the opposition is a crucial condition for the observable scrutiny mechanisms. Specifically, we find that the opposition uses scrutiny to attack weak, noncohesive governments. Moreover, we find tentative support for Martin and Vanberg’s (2004) original argument according to which scrutiny is being used to keep the coalition partner in check. However, the adaption of their argument to the two-level nature of EU politics turns out to be the most powerful explanation for observed parliamentary scrutiny. Scrutiny is being used to shift the domestic constraint strategically, but only if the domestic parties are sufficiently represented in either the Council of Ministers or the European Parliament. This paper continues by briefly summarizing the existing literature on the role of national parliaments in EU politics. Thereafter, we derive and present our theoretical arguments, which are tested in the subsequent empirical section.
National parliaments and European Union politics The literature on the Europeanization of national political systems (for an overview, refer to Goetz and Meyer-Sahling, 2008) discusses the role of national parliaments during the formulation and implementation of EU law. Comparative empirical studies find that EU law rarely affects more than 30% of domestic legislation (Müller et al., 2010). Yet, this number barely indicates the importance of EU law for national politics. Firstly, in contrast to directives, EU regulations and decisions are directly binding and do not require transposition to national law. Secondly, although national parliaments may be involved during the implementation stage to help solve intra-coalition disagreement (Franchino and Høyland, 2009), the majority of EU directives are implemented via governmental decrees (e.g., König and Mäder, 2007; Raunio and Wiberg, 2010). This advancement in Europeanization has been accompanied by concerns about democratic legitimacy. These concerns were met by an increasing parliamentarization of EU politics. Besides the well-known empowerment of the directly elected EP, the role of national parliaments in European policy formulation has been strengthened in various ways (Auel, 2007; Norton, 1995). First, national parliaments have increased their cooperation on the
supranational level via institutionalized meetings (Krekelberg, 2001).1 Second, cooperation between national parliaments and the EP has improved the amount of information available to members. Examples of such mechanisms include bilateral committee meetings (Maurer and Wessels, 2001: 458–460), the establishment of offices of national parliaments at the European Parliament (EP) (Neunreither, 2005), and the participation of Members of the EP in European Affairs Committees (Raunio and Hix, 2000: 157). Third, as of September 2006, the European Commission began forwarding all its communications and proposals to the national parliaments (European Commission, 2008: 2). At the domestic level, the additional supply of EU documents has been met with significant improvements of parliamentary scrutiny systems (e.g., Raunio and Hix, 2000; Maurer and Wessels, 2001; Auel, 2007; Winzen, 2012). However, the formal powers of national parliaments differ across countries (Pahre, 1997; Bergman, 1997; Saalfeld, 2005). Whereas most parliaments are limited to issuing a non-binding scrutiny reserve, few have the right to issue a binding mandate that limits the minister’s discretion in the Council (Raunio, 2005: 322–323).2 Nordic parliaments, for example, have established stronger control procedures (Denmark has created a mandating system and a strong European Affairs Committee [EAC]) than southern parliaments (Winzen, 2010). The new Central and Eastern European (CEE) member states generally display stronger parliamentary control of EU affairs, despite variance within CEE member states (Karlas, 2011). Moreover, the effectiveness of parliamentary scrutiny systems depends on the available resources, such as the administrative staff, the size and composition of the EACs (Bergman et al., 2003), and the informal contacts to MEPs (Maurer and Wessels, 2001). Behavioral assessments of scrutiny activities are often based on aggregate indicators such as the frequency of EAC meetings, the opinions produced, the number of meetings with EP committees, and memoranda received by the government on its positions (Karlas, 2011; Bergman, 1997; Bergman et al., 2003). Nonetheless, our knowledge about the effect of these different institutions with respect to observable scrutiny activity is still limited (Benz, 2005: 519). Other authors point to the fact that an assessment of formal powers and resources does not suffice to evaluate the role of national parliaments in EU politics (Auel and Benz, 2005) or to assess the “real” degree of parliamentary involvement as some rights may be under-used (Hefftler et al., 2013). In addition, there is still no consensus over how much influence
Importantly, the “Conference of Community and European Affairs Committees of Parliaments of the EU” (COSAC) has been officially recognized in protocols of the Treaty of Amsterdam and the Lisbon Treaty. 2 For an overview of the member states, please refer to COSAC (2007: 14–19); Bergman et al. (2003); and Raunio (2005). 1
parliaments actually have over EU decision making via their governments (Auel, 2012: 6). Formal classifications may serve as useful starting points for the analysis of parliamentary scrutiny, but in practice unexpected patterns may emerge (Neuhold and Ruiter, 2010: 70). In response, a number of recent studies focus on parties’ motivation for investing time and other resources into scrutiny activities (e.g. Auel and Benz, 2005; Finke and Dannwolf, 2013).
Explanations for scrutiny behavior Theories of delegation and oversight were initially applied to the relation between Congress and the executive branch in US politics (e.g., McCubbins and Schwartz, 1984; McCubbins et al., 1989; Kiewiet and McCubbins, 1991). In parliamentary democracies, the oversight instruments have long been considered to be less powerful than in presidential systems (Strøm, 2000). Nevertheless, numerous studies point towards the existing information asymmetry between the government and parliaments in Europe (e.g., Saalfeld, 2000; Strøm, 2000; Müller, 2000) as well as within governments (Martin and Vanberg, 2004, 2011). All of these existing studies conceive of scrutiny as a means to directly influence the government’s policies.
Opposition control Students of comparative politics argue that parliamentary oversight mechanisms provide one important avenue for the opposition’s influence on policy making (e.g., Powell, 2000). From this perspective, the government’s policy drifts away from the median MP. The opposition’s incentive to counteract such “governmental drift” increases with its policy distance to the government. The goal is to pull the policy towards the position of the opposition. The mechanisms to achieve this goal differ, but it is generally easier to influence the position of a weak government than to influence the position of a strong government. Specifically, we identify two conditions which determine the strength of the government and, accordingly, the relevance of the “opposition control” argument for explaining parliamentary scrutiny. First, low party cohesion among the ruling parties raises the opposition’s chances to effectively use scrutiny to influence the government. Low cohesion increases the likelihood that some MPs representing the ruling parties may prefer shifting the policy proposal towards the position of the opposition.
Second, in minimal winning coalitions, every vote counts. Accordingly, few deviant members suffice to veto the government’s policy or, more relevant to EU politics, to adopt a resolution that requires substantial changes to the present governmental policy. Accordingly, the effect of the conflict between government and opposition should depend on (i) the cohesion among and inside the ruling parties and (ii) the relative strength of the government or the size of the coalition.
Coalition control Martin and Vanberg (2004) find that parliamentary scrutiny is being used to reduce information asymmetries among Dutch and German coalition partners. They conceive of ministers as agents who are mandated to implement the coalition agreement struck among party leaders. Yet, the responsible ministers hold private information on the policy proposals in their jurisdiction, which can lead to ministerial drift, i.e., the minister’s policy proposal deviates from the coalition agreement. The likelihood for, and severity of, ministerial drift increases with the dissent between the two coalition partners (Martin and Vanberg, 2004: 20). Accordingly, the incentive for scrutiny increases with intra-coalition dissent. This information asymmetry can be mitigated by a number of mechanisms, such as the nomination of junior ministers from each coalition partner who function as watchdogs; the cross nomination of committee chairs and the minister in the same jurisdiction; and most importantly, parliamentary scrutiny. According to the authors, effective parliamentary scrutiny requires strong parliaments endowed with the necessary resources (e.g., specialized committees) and rights (e.g., compel witnesses and rewrite draft bills). We agree that formal rights and resources are important to explain the frequency and extent to which scrutiny is being used in different national parliaments. However, here we are interested in the observed variation across legislative proposals. From our perspective, the relevance of the coalition control argument for explaining this variation depends on the following two conditions. First, after committee members have successfully detected ministerial drift, they must convince the guardians of the coalition agreement, namely party leadership. The message is most credibly delivered by MPs from the minister’s own party who have an interest in guarding the coalition agreement. Such intra-party opposition to ministerial drift is more likely if members’ preferences are heterogeneous, i.e., party cohesion is low.
Second, the deceived coalition partner will ask for concessions. Obviously, such retaliation is easier for powerful senior partners than for junior partners. Accordingly, scrutiny should be more likely if the leading minister belongs to a small coalition partner. Thus, the effect of intra-coalition dissent should depend on (i) the cohesion of the leading minister’s party and (ii) the relative strength of the leading minister’s party in the coalition.
EU strategy So far, our arguments elaborate on existing approaches to explaining parliamentary oversight in domestic politics, but they do not accommodate the two-level nature of EU politics. The closed nature of meetings in the Council of Ministers exaggerates the informational asymmetries between the government and the opposition. In contrast to domestic politics, the asymmetry extends to the effects of competing policy alternatives and to the set of politically feasible alternatives. More specifically, opposition parties hold limited information on the interests represented in the Council of Ministers. Considering these difficulties, Proksch and Slapin (2011) argue that only government parties can use the Council of Ministers to oversee and influence EU legislation, whereas opposition parties can only direct their influence to the EP, for example, by questioning the responsible commissioner. In a recent article, Finke and Dannwolf (2013) follow up on earlier case-study evidence by arguing that governing parties initiate parliamentary scrutiny to improve their bargaining leverage in the Council of Ministers. The underlying mechanism reflects the so-called Schelling conjecture (Milner, 1997): “In international negotiations, the ability of a negotiator to credibly say to his or her counterpart that ‘anything we sign here has to be ratified by my country’s legislature’ provides a bargaining advantage that this person would not otherwise have” (cf. Tarar, 2001: 320). In contrast to international treaties, EU legislation does not require formal ratification by national parliaments to become effective. Even when no direct ratification is necessary, displeased coalition partners have diverse means for ex post sanctioning of ministers, such as retaliating on other issues or, ultimately, withdrawing support for the government. Hence, we conceive of the coalition agreement as a domestic constraint. Thus, a reform skeptic (friendly) minister may credibly shrink the room for bargaining concessions by pointing towards an even more reform-skeptic (friendly) coalition partner. At its core, this argument is an adaption of the “coalition control” theory by Martin and Vanberg (2004) to the tow-level nature of EU politics. Whereas scrutiny in domestic politics aims at directly influencing policies, scrutiny in international politics aims at manipulating the negotiators’ domestic constraints.
Consider the case of a one-dimensional law proposal where the leading minister (m) is located to the left of the expected bargaining outcome in the Council (o), i.e., m < o (see Finke and Dannwolf, 2013). Under this scenario, the leading minister will point towards a coalition partner (c) who is even more skeptical of the expected bargaining outcome, i.e. if c < m < o. Specifically, the minister will claim that the costs of justifying the outcome towards the coalition partner increase with the bargaining outcome moving to the right and decrease with the bargaining outcome moving to the left. In this scenario, the minister has a motive to initiate parliamentary scrutiny as a means to credibly reveal this information to colleagues in the Council of Ministers.3 The Schelling conjecture: If m < o, the likelihood for parliamentary scrutiny increases with (m − c) increasing.4 The Schelling conjecture implies an information asymmetry between the leading minister and the international bargaining partners. However, the latter are not required to monitor the scrutiny activities in all national parliaments because the Schelling conjecture presumes that it is in the minister’s self-interest to alert the partners to a formal scrutiny procedure. Finke and Dannwolf (2013) also discuss the “flip-side of the Schelling conjecture” where the leading minister loses credibility for denying concessions once information on the coalition partner(s) has been revealed to the colleagues in the Council. If, for example, the preference constellation resembled m < c < o, the leading minister would be foolish to reveal the location of the coalition partner to the international partners. However, any political group located to the right of the expected outcome has an incentive to whistle-blow by placing the topic on the national parliament’s agenda. Importantly, this so-called “whistle-blowing scenario” assumes that the whistle-blower is capable of communicating the scrutiny activities to the international bargaining partners. The whistle-blowing conjecture: If m < o, the likelihood for parliamentary scrutiny decreases with (m − c) increasing.5 Please note that the two-level games argument is partly explorative: Ex ante, we cannot be certain under which conditions either the Schelling conjecture or the whistle-blowing conjecture prevails. We can, however, identify conditions under which the strategic use of scrutiny should generally be more likely. First, the two-level games strategy implies that 3
Our model ignores the location of the status quo for which we lack sufficient information. As a consequence, we assume that the win-set is infinitely large. This assumption can be justified by the results of the decision making in the EU project (Thomson et al. 2006; for a discussion see König and Junge 2010). 4 We are dealing with one-dimensional policy space, hence the assumption m
