The European Commission’s use of consultation during policy formulation: The effects of policy characteristics

Abstract

This article asks to what extent the European Commission’s use of consultation depends on policy characteristics of the proposal being formulated. It examines all consultations organized during a formulation process and looks at the number of open consultations (online consultations) as well as the number of repeat consultations (expert groups, policy fora) by the total number of consultations. Three types of use of consultation are conceptualized. A multinomial logistic regression is performed on a cross-sectional sample of 150 policy proposals adopted between 2010 and 2014. Data are triangulated from official documents. The empirical analysis indicates that the Commission’s use of consultation depends on the complexity, newness, and salience of a proposal. These findings are also discussed in view of the 2015 reform of Better Regulation.

Keywords

Bureaucratic politics consultation European Commission policy formulation rational choice

Introduction

The European Commission frequently organizes consultations in order to discuss its draft policies with various kinds of stakeholders—member state representatives, interest groups, and interested citizens. The Directorate-Generates (DGs) within the European Commission are known to seek advice from expert groups, stakeholder fora, to conduct online consultations, to organize seminars, etc. Recent literature witnessed that the Commission’s use of consultation varies strongly across policy areas and even across policy proposals (Quittkat and Finke, 2008; Rasmussen and Toshkov, 2013). It is, however, still a pending question how this variation can be explained. This article therefore asks to what extent the Commission’s use of consultation varies in response to the policy characteristics of a proposal being formulated.

Policy characteristics entail the basic attributes touching upon the complex, the legal or salient nature of a policy under formulation which are found increasingly relevant to understand EU policy-making (Bunea, 2013; Klüver, 2011; Klüver et al., 2015; Mahoney, 2008). As a part of policy planning, the Commission initially analyzes a policy problem and identifies concerned stakeholders. Policy characteristics enable the Commission to identify its precise needs and to design consultations in the most suited way to attain those needs. Hence, the Commission is expected to assess anew how to consult for each formulation process which would explain why we observe so much variation. Alternatively, the use of consultation might also vary because Commission DGs have developed diverging routines over time (March and Olsen, 1989).

This article studies consultation patterns which cover the entire range of consultation practices within the formulation process. This includes online consultations, expert groups, as well as seminars, workshops, or conferences which are institutionalized to a minimum degree (but informal practices like office appointments, lunch meetings, or even phone calls fall outside the scope of the analysis). It therefore considers each consultation practice as a single unit making part of the consultation pattern associated with that formulation process. Such a comprehensive approach is advantageous for two reasons. First, this article employs a higher level of analysis than studies in the field of bureaucratic politics, executive politics, and public administration which analyzed only part of all the consultations carried out by the Commission throughout the formulation process. This results in a complete picture about the relation between the Commission and stakeholders. Second, this view ties in more closely with the daily work practice inside the Commission as stakeholder consultations are carefully planned as part of the formulation process.

There is a consensus in the broader literature on consultation that the Commission consults stakeholders with two goals in mind: capacity-building and legitimization (Bouwen, 2009; Bunea and Thomson, 2015; Klüver, 2013; Princen, 2011). On the one hand, the Commission operates in an increasingly complex policy environment while possessing only limited in-house expertise. Hence, the Commission uses consultation to strengthen its problem-solving capacity by collecting expertise from stakeholders (Princen, 2011). On the other hand, the Commission is a nonmajoritarian institution and, for this reason, it seeks political support of stakeholders to legitimize its policy plans. The Commission therefore organizes consultation to gain attention for its policy proposals and to mobilize support from key constituencies, varying from the public at large to specific economic groups (Klüver, 2013). Yet, this does not explain why the Commission consults in a variety of ways.

This article argues that different consultation practices are less/more suited for capacity-building or legitimation based on their procedural design. It continues with the conceptualization of the Commission’s use of consultation and presents four hypotheses about the effect of policy characteristics on the use of consultation. The research design is presented next. The empirical analysis is based on a cross-sectional sample of 150 policy proposals, drafted by four Commission DGs and adopted between 2010 and 2014. Data are triangulated through inspection of official documents. The findings show that the Commission designs consultations in line with the complexity, newness, and salience of draft policies. Briefly stated, online consultations are the Commission’s preferred manner to consult about (technically) complex and/or new policies. Stakeholders are mainly consulted via expert groups and seminars about amending and/or salient proposals. To conclude, these findings are interpreted in view of the recent reform of Better Regulation.

Consultation patterns and the Commission’s use of consultation

The use of consultation is a novel concept describing how the Commission interacts with stakeholders in the formulation of policies. This interaction depends on the accessibility and repetition of consultation (see Table 1). Accessibility indicates to what extent access to consultation is restricted for stakeholders in a consultation pattern. Repetition indicates whether a consultation pattern mostly includes one-off or consecutive meetings with stakeholders. Examination of accessibility alongside repetition allows to deduce in what way the Commission intends to use consultation. The concept can take on three values: fixed use, custom use, and all-round use.

Table 1.

Capacity-building and legitimation based on the design of consultation.

	One-off meeting	Consecutive meetings
Open access	Capacity-building: collecting information	Capacity-building: collecting information
	Legitimation: gaining maximum attention/mobilizing broad support	Legitimation: gaining maximum attention/ mobilizing broad support
Restricted access	Capacity-building: collecting information + focused discussions	Capacity-building: collecting information + structural problem-solving
	Legitimation: mobilizing targeted support	Legitimation: mobilizing targeted support + consensus-building

Fixed use (open access—one-off meeting): Fixed use denotes that the Commission prefers organizing open consultations which cover a one-off meeting like most conferences or online consultations. First, the open character of consultation enables the Commission to draw maximum attention and to mobilize broad societal support for its policy plans. Public as well as economic support from specific interest groups is crucial to legitimize a policy proposal in the face of political opposition (Klüver, 2013). The reverse side of consulting a large and diverse audience is that the transaction costs to interact with participants increase along (North, 1990). Second, a short time span of consultation should make stakeholders reluctant to alter their policy positions. Taken together, fixed use aims at information collection but is hardly interactive in nature.

Custom use (restricted access—one-off meeting): Custom use indicates that the Commission mostly organizes restricted consultations which cover a one-off meeting like seminars or workshops. First, the restricted nature of consultation inevitably biases participation as the Commission invites only a limited number of stakeholders. Restricted consultations are less suited for legitimation as the Commission can only enhance political support among the selected audience. On the benefit side, a smaller and potentially more homogeneous audience limits transaction costs and permits the Commission to discuss more freely with participants. Second, the short time span suggests that seminars/workshops are more suited to discuss specific or technical affairs but less so to conduct political negotiations. Thus, custom use aims at informed discussions about specific policy aspects which can be round up within one and the same meeting.

All-around use (restricted access—consecutive meetings): All-round use denotes that the Commission primarily consults through expert groups or stakeholder fora to which access is restricted and which convene at multiple occasions. First, the Commission can interact more freely with participants when the access to consultation is restricted (see above). Second, this feature is strengthened as consultation takes place over consecutive meetings within the same framework. This gives a limited number of stakeholders the prospect of building up an enduring relationship with the Commission. In return for this privileged access, the Commission expects these stakeholders to adopt a more pragmatic stance and to participate constructively in a problem-solving or consensus-building process (Rasmussen and Toshkov, 2013). The all-round use is therefore most interactive and its design is suited for capacity-building as well as for legitimation.

Implausible use (open access—consecutive meetings): Finally, the Commission could, in principle, organize open consultations which consist of consecutive meetings. For example, the Commission measures popular opinion at multiple occasions through its default Eurobarometer surveys. However, these surveys focus on revealing general political attitudes and less on detailed opinions that are relevant for the formulation of one specific policy proposal. Evidently, the Commission could organize more than one online consultation but these constitute independent events. Given that my empirical data show that the Commission rarely consults through open and consecutive meetings within the context of policy formulation, this type of use of consultation is disregarded in the remainder of the article.

The Commission’s use of consultation in view of policy characteristics

I expect the use of consultation to reflect differences in policy characteristics presuming that the Commission’s internal capacities to formulate a policy as well as its political legitimacy also vary across policies. This section presents hypotheses about the effect of policy characteristics (settings, scope, newness, and salience) on the use of consultation. I control for one other legal characteristic, the bindingness of a proposal, as well as for two institutional characteristics, being administrative capacity and Commission DG.

I treat settings and scope as two distinct characteristics of ‘complexity’ in contrast with previous work presenting complexity as one catch-all concept which denotes how policy-makers cope with uncertainty in decision-making (Klüver, 2011; Neshkova, 2014). First, following Hall (1993: 278), I consider settings as the technical provisions constituting a policy, different from the overall goals steering that policy in a field and the choice of policy instruments applied to achieve those goals. Regulatory settings, like the maximum CO₂ emission level defined for cars, matter for stakeholders because they have redistributive consequences, in this example for car manufacturers who should adjust their production process to comply with regulations (Majone, 1996). The Commission is poorly capable to assess the adjustment costs of proposed settings as these differ for particular stakeholders (Rasmussen, 2015). As consulting the concerned stakeholders is essential for the formulation of effective policies, I expect the Commission to involve them closely through all-round use of consultation. The Commission can negotiate the calibration of settings with stakeholders in order to ensure their cooperation in sharing policy-relevant information and, later on, in complying with regulations.

H1: The more settings a policy includes, the more likely the Commission chooses all-round use (restricted access—consecutive meetings).

Second, scope indicates the extent to which policy effects are spread out over multiple policy fields (Mahoney, 2008). The Commission might not always foresee the policy effects created directly or indirectly through interaction with other existing policies. Such unintended consequences are, however, likely to arise in specific circumstances (Pierson, 2000). The likelihood for unintended consequences presumably increases along with the scope of a policy. The Commission can, however, minimize this risk by seeking maximum attention for its plans and collecting information from all interested stakeholders. In other words, the Commission is expected to consult broadly and only fixed use is suited for this purpose. At the same time, stakeholder submissions enhance the Commission’s insight in the variable costs and benefits which a policy with a wide scope might generate for diverse stakeholders. The Commission can therefore revise its proposal in order to enhance stakeholder support.

H2: The wider the scope of a policy, the more likely the Commission chooses fixed use (open access—one-off meeting).

Third, newness refers to the proposed legal change in a policy which is highest for new proposals (instead of amending ones) that aim to change the policy acquis substantially (instead of procedurally) (Bunea, 2013; Reh et al., 2013). Policy changes incur adjustment costs on stakeholders and require the Commission to legitimate its policy plans. New/substantial proposals incur the highest adjustment costs on a diverse range of stakeholders. I therefore expect the Commission to consult all interested stakeholders through fixed use. This way the Commission can gain a better understanding of the political views of diverse stakeholders in order to ensure that its proposal will eventually enjoy sufficient political support among the public at large and economic groups. The actual involvement of stakeholders in the formulation process remains limited to exchanging information.

H3a: When formulating a new/substantial proposal, the Commission most likely chooses fixed use (open access—one-off meeting).

The situation is quite different for amending proposals which only affect the stakeholders who ‘adapted to the existing policy path over the course of time’ (Hartlapp et al., 2014: 19). Amending/procedural proposals leave most of the policy acquis intact. The changeover from the old into a new situation should proceed smoothly for the majority of stakeholders (Alexander, 2001: 258). Moreover, I expect the Commission to possess higher internal capacity to formulate an amending proposal given its prior experience in drafting a legal basis. The use of consultation therefore primarily aims to ensure stakeholders support and is less aimed at enhancing problem-solving capacity. Custom use is most suited to discuss the specific changes that the Commission wants to introduce with a targeted audience of stakeholders. The analysis should point out whether these arguments also extend to new/procedural and amending/substantial proposals.

H3b: When formulating an amending/procedural proposal, the Commission most likely chooses custom use (restricted access—one-off meeting).

Fourth, salience is defined as the relative importance attached to a policy by the Commission (Warntjen, 2012). The Commission prioritizes policy problems on its working agendas, with salience being its ‘cognitive shortcut’ to choose which policy problems should be addressed by public policy (Oppermann and de Vries, 2011: 3–4). The Secretariat-General ensures that the responsible (or lead) DG coordinates the planning of all important policies prior to the drafting of a policy proposal with other DGs through the Impact Assessment (IA) procedure (Radaelli and Meuwese, 2010). Commission DGs are compelled to mutually agree on the main content of a policy under the IA procedure. The link between salience and IA is convenient for two reasons. First, the Commission’s perception of salience matters as draft policies almost never receive public attention prior to consultation. Second, salience can have the effect of a spotlight on decision-making and stir up political opposition which is why the lead DG needs to intensify the planning of a salient policy through the IA procedure, which focuses on defining a policy problem, identifying policy options, performing cost–benefit analyses. As regards the use of consultation, I expect the lead DG to consult a broad stakeholder audience in order to collect policy-relevant information as well as to gain insight in the policy positions of all concerned stakeholders and, eventually, to mobilize their political support. In other words, fixed use enables the Commission to substantiate its IA analysis while also legitimating its plans.

H4: The more a policy is considered as salient, the more likely the lead DG chooses fixed use (open access—one-off meeting).

Finally, bindingness indicates whether the Commission proposes a nonbinding instrument or a specific legally binding instrument to implement policies. The planning and drafting of a proposal including a nonbinding or binding instrument is identical. Existing work usually focuses on studying proposals for binding legislation, but these proposals frequently build further upon decisions that were already communicated in previous nonbinding documents. Both types of proposals can therefore be of equal substantive importance for the Commission as well as for stakeholders (Delreux and Happaerts, 2016: 151). Given that the Commission is held politically accountable for nonbinding initiatives as much as for binding proposals, it is expected to legitimate a nonbinding initiative through stakeholder consultation as likely as a binding proposal. Hence, the use of consultation is not expected to vary across binding and nonbinding proposals.

I control for which Commission DG was responsible to conduct consultation as well as for differences in administrative capacity. On the one hand, scholars noticed that the use of consultation varies across DGs without taking policy characteristics into account (Gornitzka and Sverdrup, 2008; Quittkat, 2011; Quittkat and Finke, 2008). Bureaucratic organizations like the European Commission are, however, highly conducive for the creation of standard operating procedures (March and Olsen, 1989). The sectoral DGs within the Commission could develop particular routines for consultation over time as a means of self-legitimation. This could also explain why the use of consultation varies in the first place across departments, instead of being policy specific. On the other hand, departments with more staff have a higher institutional capacity to process policy problems (Princen, 2011). By consequence, consultation would mainly serve the purpose of legitimation in such departments. For instance, Gornitzka and Sverdrup (2008) found that Commission DGs consult expert groups to supplement poor in-house expertise while Chalmers (2014) stressed that processing input from online consultations is a demanding process, especially for smaller departments.

Research design

Each formulation process linked with a Commission proposal classifies as one case. A cross-sectional sample is collected through EUR-Lex and includes proposals formulated by DG Climate Action, DG Communications Networks, Content & Technology, DG Environment and DG Internal Market and Services, which have been adopted by the College of Commissioners between 2010 and 2014. This makes a total of 260 cases with the Commission actually consulting stakeholders in 150 or 57.7% of the cases. The empirical analysis thus focuses on these 150 cases. The selected DGs have policy-making competences which were transferred to the EU level at different times. Given that the Commission developed new consultation practices through time, I expect a relatively young department such as DG Climate Action to be more inclined to use online consultation (which was recently developed) than an older department like DG Internal Market and Services. Alternatively, the latter might reach out more easily to previously established expert groups than the former. The remainder of this section discusses the operationalization of the variables (for details see the Online appendix).

Dependent variable

I identified consultation practices organized in relation with the 150 sampled formulation processes on the basis of Commission proposals, IA reports and I verified the completeness of this information with the Commission DGs. I classified each consultation as having open or restricted access and referring to a one-off meeting or to consecutive meetings within the same setting. I then calculated relative frequencies for accessibility and repetition in a consultation pattern. First, the relative frequency for accessibility is equal to the number of open practices divided by the total number of practices in a consultation pattern. Values below (or above) 0.5 indicate that a consultation pattern is restricted (or open) in nature. Values equal to 0.5 mean that a consultation pattern includes as many open as restricted practices. Second, the relative frequency for repetition is the number of practices covering consecutive meetings by the total number of practices. Values below (or above) 0.5 indicate that the Commission consults mainly through one-off meetings like online consultations or consecutive meetings (organized within the framework of an expert group). Likewise, values equal to 0.5 imply that a consultation pattern is balanced in terms of repetition.

Figure 1 depicts the resulting measures. The dots show the empirical distribution of cases in the research sample. The size of these dots indicates the number of consultation patterns lying at the same position, thus, having a similar composition. Overall, the use of consultation stands out as highly diverse. Consultation patterns situated close to the corners of the graph include either one practice or multiple yet similar practices. Consultation patterns lying in the middle of the graph consist of multiple yet diverse practices. The main interest is to find out why the Commission prefers to consult in any particular direction on the graph.

Figure 1.

Distribution of the use of consultation.

I recoded these relative frequencies into one categorical variable, i.e. the use of consultation. The cutoff lines correspond with the values of 0.5 in Figure 1. Every case on the vertical cutoff line is classified as all-round use to compensate for the fact that, for example, multiple meetings within one expert group are counted as only one consultation practice in a pattern. This underestimates the real contribution of such consultations. Consultation patterns on the remainder of the horizontal cutoff line are classified as fixed use. Especially online consultations require the Commission to carefully design as well as test a survey and later on to process a large amount of stakeholder submissions. The planning/conduct of online consultations is more labor intensive for the Commission than that of seminars (custom use). The recoding entails some loss of information although the resulting categorical measure is more comprehensible and captures the essential differences in terms of accessibility and repetition. There are 61 observations of fixed use, 36 of custom use, and 53 of all-round use. The upper-right half of the map is empty which reaffirms that open access combined with consecutive meetings was not observed in the research sample.

Figure 2 adds nuance by showing the use of consultation in relation to the absolute number of consultation practices observed within a formulation process. It shows that, as the number of consultations increases, fixed use and all-round use are observed less frequently. By contrast, custom use occurs more frequently when the Commission organizes more consultations.

Figure 2.

The Commission’s use of consultation in relation with the number of consultations organized.

Independent variables

The policy characteristics are operationalized hereafter. First, the relative importance of settings in each proposal is reflected in how pronouncedly that proposal treats indicator-, standard-, or target-setting (Van Ballaert, 2015). I counted the number of references in a proposal to the terms ‘standards’, ‘standardi’, ‘indicator’, and ‘target’ and weighted the total number of search hits by the text length of the respective proposal. Second, scope is reflected by the number of policy fields referred to by a proposal’s EUROVOC descriptors. The Publications Office of the European Union assigns EUROVOC descriptors to archive proposals according to their conceptual content. These descriptors cover all policy fields in which the EU institutions are active (Steinberger et al., 2002: 417). Third, a proposal can take on four values of newness: procedural/amendment, substantial/amendment, procedural/new, and substantial/new. The Commission did not consult stakeholders during the formulation of procedural/new proposals.¹

Fourth, salience is linked with IA, which is mandatory for all important policies. Presuming that concerned DGs will less easily make concessions on the most salient policies, I consider the number of formal meetings needed to find an interservice agreement on the main content of a policy (see IA reports) as a proxy for salience.² Admitted, this proxy only captures salience as perceived within the Commission early on in the formulation process. Fifth, bindingness has four values. Nonbinding documents cover green papers or Commission communications which are instruments of soft law. In ascending order of intrusiveness, legally binding instruments include a Commission proposal for a Decision, a Directive, or a Regulation (Chalmers et al., 2010: 98–103).

Furthermore, Commission DG indicates whether DG Climate action, DG Connect, DG Environment, or former DG Internal Market were responsible for the formulation of a proposal. Administrative capacity measures the absolute number of staff working within a Directorate (one level below the DG) (Chalmers, 2014: 9–10). Information is retrieved out of the annual activity reports of DGs.

Data analysis

Multinomial logistic regression is selected to analyze the use of consultation which is a nominal variable. Independent variables are related linearly with the dependent variable and collinearity statistics did not reveal problems with multicollinearity. The Durbin–Watson test indicates no autocorrelation between residuals. The analysis is performed on 137 cases as seven cases were excluded after residual inspection.³ Maximum likelihood is used for estimation. The multinomial model predicts the use of consultation significantly better than the null model (p < 0.001). Model fit is excellent given that McFadden R² equals 0.250 (McFadden, 1979). Five independent variables predict the use of consultation significantly (see the Online appendix). These are settings (p ≅ 0.000), newness (p = 0.023), salience (p = 0.009), Commission DG (p = 0.021), and scope (p = 0.040). In contrast, bindingness (p = 0.112) and administrative capacity (p = 0.161) cannot significantly predict the use of consultation.

The regression model in Table 2 estimates parameters separately for the likelihood of finding (1) fixed use relative to custom use, (2) all-round use relative to custom use and, finally, (3) all-round use relative to fixed use. The regression coefficients represent odds ratios, which indicate whether and how an increase in one measurement unit of an independent variable affects the likelihood for having one type of use of consultation. The interpretation for categorical variables requires choosing a referent category against which to compare each other category. Table 2 also provides the 95% confidence intervals for the odds ratios.

Table 2.

Multinomial logistic regression model for use of consultation.

Independent variables	(1) Fixed use relative to custom use	(2) All-round use relative to custom use	(3) All-round use relative to fixed use
Settings	3.192*** (1.717; 5.933)	1.792 (0.975; 3.296)	.562** (.375; .841)
Scope	1.755* (1.087; 2.833)	1.569* (1.015; 2.425)	.894 (.606; 1.318)
Newness (referent category = substantially new)
• Procedural/amendment	201 (.035; 1.142)	.401 (.076; 2.121)	2.001 (.417; 9.604)
• Substantial/amendment	101* (.015; .687)	1.066 (.219; 5.189)	10.592** (1.954; 57.423)
Salience	.607* (.397; .930)	1.017 (.729; 1.419)	1.675** (1.143; 2.456)
Bindingness (referent category = proposal for a Regulation)
• Nonbinding COM document	.317 (.049;2.077)	1.231 (.238; 6.367)	3.878 (.746; 20.155)
• Proposal for a Directive	4.017 (.838; 19.261)	1.145 (.277; 4.740)	.285 (.071; 1.150)
Commission DG (referent category = DG Internal Market)
• DG Climate Action	.510 (.055; 4.754)	2.739E-10 (2.739E-10; 2.739E-10)	5.369E-10 (5.369E-10; 5.369E-10)
• DG Connect	1.848 (.207; 16.497)	2.352 (.425; 13.023)	1.273 (.198; 8.168)
• DG Environment	.717 (.152; 3.381)	.327 (.080; 1.339)	.456 (.106; 1.969)
Administrative capacity	.988 (.971; 1.006)	.998 (.992; 1.003)	1.009 (.992; 1.027)

DG: Directorate-Generate.

Notes: Dependent variable—Use of consultation. Reference categories: Commission DG —DG Internal Market; Bindingness—proposal for a Regulation; Newness —substantial/new; N = 137. Coefficients represent odds ratios; 95% confidence intervals in parentheses; significance levels: ***p ≤ 0.001; **p ≤ 0.01; *p ≤ 0.05.

Policy characteristics significantly explain variation in the use of consultation. First, proposals with a higher degree of settings are most likely associated with the fixed use and least likely with the custom use (custom < all-round < fixed use). The empirical data refute H1. Second, proposals with a wider scope are most likely associated with all-round use or fixed use (custom < all-round ≅ fixed use). This only partially supports H2. Third, a substantial/new proposal is set as the referent category for newness. This means that fixed use is more likely (than custom use) to occur when formulating a substantial/new proposal compared to when the Commission formulates an amending/substantial as well as procedural/amending proposal. In similar circumstances, the lead DG is also more likely to prefer fixed use instead of all-round use. H3a and H3b are supported. Fourth, a relatively salient proposal is most likely subject to custom use or all-round use (fixed < custom ≅ all-round use). This refutes H4. Finally, bindingness cannot predict the use of consultation significantly.

Institutional characteristics only explain limited variation in the use of consultation. The likelihood ratio test indicates that Commission DG significantly improves the prediction of the use of consultation even though none of the regression coefficients is significant. Furthermore, administrative capacity cannot predict the use of consultation meaningfully.

Discussion of results

This article initially asked to what extent the Commission’s use of consultation is policy dependent. The analysis reveals that the Commission designs its consultation pattern in response to settings, scope, newness, and salience of a policy.

Complexity characteristics significantly explain the use of consultation. On the one hand, the Commission prefers fixed use regarding proposals with relatively many settings and is less keen than expected to negotiate directly with stakeholders. Given the redistributive consequences which settings might have, it is not hard to imagine that a large number of stakeholders ranging from associational groups to competing enterprises and national regulatory authorities all want to have a say in the formulation process. The Commission collects information from all these stakeholders through online consultations after which it can play out regulatory competition between economic groups by cross-validating their submissions. Overall, the Commission stands out as an administration that is highly capable to formulate technically complex policies as stakeholder involvement remains limited under the fixed use. More generally, this affirms a neo-pluralist view of Commission–stakeholder relations as specific economic groups are unlikely to ‘capture’ the Commission through open consultations.

On the other hand, the Commission usually opts for fixed use or all-round use when it formulates cross-cutting proposals. This means that online consultations are observed as often as expert groups on the same proposals, with the Commission’s preference for either of these consultation practices varying only slightly across cases. As hypothesized, the Commission collects information to minimize the risk for unintended side effects. However, consulting a captive stakeholder audience on repeated occasions implies that more intense cooperation between the Commission and stakeholders is needed on transversal policies to ensure policy effectiveness. The larger degree of stakeholder involvement can be interpreted as supporting evidence that cross-cutting policy problems are most challenging for the Commission despite the mechanisms that are in place within the Commission for interdepartmental coordination. More generally, this also raises questions about the extent to which the Commission organizes diverse consultations for strategic reasons, for example, by using recommendations from online consultation to pressure stakeholders thereafter in other fora or by granting privileged access to supportive actors.

The relationship between newness and the use of consultation conforms to prior expectations. The Commission frequently verifies through online consultation whether stakeholders agree if a new policy respects the legal principles of subsidiarity and proportionality. Stakeholders are then asked to confirm the political need for a new EU policy, hence, legitimizing the Commission’s initiative. Amending proposals are formulated after existing legislation has been evaluated together with concerned stakeholders. These evaluations usually take place in seminars or expert groups.

The saliency of a policy encourages the Commission to consult stakeholders by means of custom use or all-round use. The lead DG prefers to organize seminars (custom use) to discuss specific aspects of a salient policy with targeted audiences. Alternatively, the lead DG consults the same stakeholders repeatedly in expert groups (all-round use). More precisely, the lead DG consults stakeholders in parallel with the IA procedure. It de facto consults a broad audience at an early stage of IA about general aspects, with additional restricted consultations taking place when policy planning has progressed into the more advanced stages of IA. This observation raises new questions about the Commission’s strategic behavior and its possibilities to venue-shop between consultations. From a bureaucratic perspective, stakeholder consultations and interservice coordination appear to function as internal checks to each other: a low (or high) number of internal meetings tend to be associated with less (or more) consultations. It is unclear who strategically benefits most of this inside the Commission. The lead DG might build up expertise through consultation which is useful to resolve political conflicts with other DGs. By contrast, the other DGs are nowadays more involved in policy planning due to which the lead DG needs to consider their opinions more carefully than it once did.

Contrary to prior expectations, open consultations (fixed use) constitute a minority of the total number of consultations when the Commission formulates a salient policy. This is likely so because multiple open consultations increase public visibility. This risks exacerbating political opposition while not necessarily resulting in new insights. Instead, the Commission aims to rationalize salient policies through the IA. This is done by using the information collected in consultations in order to frame a proposal as being ‘evidence based’ and ‘objective’ in nature. As a result, the distinction between consultations aiming for capacity-building and for legitimation becomes less pronounced.

It is noteworthy that the use of consultation does not vary meaningfully depending on the bindingness of a policy. This nuances the view that the Commission consults stakeholders to strategically anticipate legislative decision-making as the use of consultation remains similar when there is no prospect of interinstitutional negotiations. It confirms that stakeholder consultation is potentially as important for every document adopted by the Commission.

There is only limited evidence that institutional characteristics also affect the Commission’s use of consultation. First, the lack of differences between the Commission DGs is surprising as the sample has been designed to compare consultation patterns between departments of varying age. One minor observed difference entails the all-round use of consultation as DG Connect, DG Internal Market, and to less extent also DG Environment choose the all-round use of consultation more frequently than DG Climate action. The latter became an independent Commission department in 2010 when there was already repeated public criticism of the use of expert groups. Since then an overall decrease could be noted concerning the creation of new expert groups (Metz, 2015) and this dynamic appears especially strong in new policy fields such as the one dealing with climate change. To be clear, DG Climate action does consult expert groups and the like but, in terms of relative frequencies, these consultation practices are balanced by, for instance, online consultations or seminars. Moreover, the stakeholder landscape in older policy fields is shaped more profoundly by expert groups as the Commission started establishing stakeholder fora decades ago (Broscheid and Coen, 2007; Quittkat and Finke, 2008).One can think of the policy fora created in the field of water policy (DG Environment), telecommunications policy (DG Connect), and procurement policy (DG Internal Market) to name but a few examples. Second, the number of staff working in a Directorate has no effect on the use of consultation. This is surprising but rather a consequence of consultation patterns being studied in terms of relative frequencies whereas the administrative workload is determined rather on the basis of the absolute number of consultations.

Conclusions

This article examined consultation patterns comprehensively in order to demonstrate that the European Commission’s use of consultation is policy dependent. The Commission can use consultation in one of three different ways. The Commission mainly organizes online consultations, conferences, or hearings (i.e. fixed use) in order to collect and assess information about complex policies or to gain wide attention for policies introducing substantial/new changes. The Commission mainly organizes seminars or workshops (i.e. custom use) regarding proposals that amend existing legislation and regarding salient policies. These consultation practices are most suited for the Commission to engage in focused discussions with stakeholders about specific policy aspects. The Commission relies mostly on expert groups or other stakeholder fora (i.e. all-round use) when preparing policy proposals with a wide scope, amending proposals and salient policies, although these findings are also contingent on the Commission DG in charge of consultation. The newest DG in the study sample, DG Climate action, is far less likely to consult expert groups than the older DGs. Expert groups are relatively versatile and enable the Commission to engage in problem-solving as well as to mobilize a targeted audience and build support for its policy plans.

The Commission announced to consult stakeholders even more systematically as part of its 2015 reform of Better Regulation. This adds to the relevance of this article which proposed an analytical framework suited to study multiple consultations. My findings suggest that, on the one hand, the Better Regulation reform formalizes administrative practices that were already in place considering that DGs already planned consultations based upon the characteristics of a policy between 2010 and 2014. The Commission now publishes roadmaps or inception IAs at the beginning of the formulation process to inform stakeholders about any planned consultations, among others. On the other hand, stakeholders can provide feedback on these consultation plans right away so that the Commission could still adjust its plans. It remains to be seen to what extent these reforms will make the Commission more prone to stakeholder pressure. Stakeholders with limited resources have reason to keep consultation as transparent as possible and will most likely try to further advance the fixed use of consultation. By contrast, economic groups or national authorities who dispose of relatively many resources will aim to maximize their involvement and their influence on policy outcomes either by asking to limit the number of participants or by asking to consult at multiple occasions. Future research on consultation should keep track of whether the Commission is responsive to stakeholders’ feedback and whether this feedback is balanced in terms of the interests represented (Bunea, 2017).

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Alexander

(2001) Institutions, path dependence, and democratic consolidation. Journal of Theoretical Politics 13(3): 249–269.

Bouwen

(2009) The European Commission. In: Coen

Richardson

(eds) Lobbying the European Union: Institutions, Actors and Issues, Oxford: Oxford University Press, pp. 19–38.

Broscheid A and Coen D (2007) Lobbying activity and fora creation in the EU: Empirically exploring the nature of the policy good. Journal of European Public Policy 14(3): 346–365.

Bunea

(2013) Issues, preferences and ties: Determinants of interest groups’ preference attainment in the EU environmental policy. Journal of European Public Policy 20(4): 552–570.

Bunea

(2017) Designing stakeholder consultations: Reinforcing or alleviating bias in the European Union system of governance? European Journal of Political Research 56(1): 46–69.

Bunea

Thomson

(2015) Consultations with interest groups and the empowerment of executives: Evidence from the European Union. Governance 28(4): 517–531.

Chalmers

(2014) In over their heads: Public consultation, administrative capacity and legislative duration in the European Union. European Union Politics 15(4): 595–613.

Chalmers

Davies

Monti

(2010) European Union Law: Cases and Materials, Cambridge: Cambridge University Press.

Delreux

Happaerts

(2016) Environmental Policy and Politics in the European Union, London: Palgrave MacMillan.

10.

Gornitzka

Sverdrup

(2008) Who consults? The configuration of expert groups in the European Union. West European Politics 31(4): 725–750.

11.

Hall

(1993) Policy paradigms, social learning, and the state: The case of economic policymaking in Britain. Comparative Politics 25(3): 275–296.

12.

Hartlapp

Metz

Rauh

(2014) Which Policy for Europe?: Power and Conflict Inside the European Commission, Oxford: Oxford University Press.

13.

Klüver

(2011) The contextual nature of lobbying: Explaining lobbying success in the European Union. European Union Politics 12(4): 483–506.

14.

Klüver

(2013) Lobbying in the European Union: Interest Groups, Lobbying Coalitions, and Policy Change, Oxford: Oxford University Press.

15.

Klüver

Braun

Beyers

(2015) Legislative lobbying in context: Towards a conceptual framework of interest group lobbying in the European Union. Journal of European Public Policy 22(4): 447–461.

16.

McFadden

(1979) Quantitative methods for analyzing travel behaviour on individuals: Some recent developments. In: Hensher

Stopher

(eds) Behavioural Travel Modelling, London: Croom Helm, pp. 279–318.

17.

Mahoney

(2008) Brussels Versus the Beltway: Advocacy in the United States and the European Union, Washington, DC: Georgetown University Press.

18.

Majone G (1996) Regulating Europe. New York: Routledge.

19.

March

Olsen

(1989) Rediscovering Institutions: The Organizational Basis of Politics, New York, NY: The Free Press.

20.

Metz

(2015) The European Commission, Expert Groups, and the Policy Process: Demystifying Technocratic Governance, Basingstoke: Palgrave Macmillan.

21.

Neshkova

(2014) Salience and complexity in supranational policymaking: The case of subnational interests. Governance 27(1): 9–36.

22.

North

(1990) A transaction cost theory of politics. Journal of Theoretical Politics 2(4): 355–367.

23.

Oppermann

de Vries

(2011) Analyzing issue salience in international politics. Theoretical foundations and methodological approaches. In: Oppermann

Viehrig

(eds) Issue Salience in International Politics, Abingdon: Routledge, pp. 3–22.

24.

Pierson

(2000) The limits of design: Explaining institutional origins and change. Governance 13(4): 475–499.

25.

Princen

(2011) Agenda-setting strategies in EU policy processes. Journal of European Public Policy 18(7): 927–943.

26.

Quittkat

(2011) The European Commission’s online consultations: A success story? Journal of Common Market Studies 49(3): 653–674.

27.

Quittkat

Finke

(2008) The EU Commission consultation regime. In: Kohler-Koch

De Bièvre

Maloney

(eds) Opening EU-Governance to Civil Society: Gains and Challenges, Mannheim: University of Mannheim, pp. 183–222.

28.

Radaelli

Meuwese

ACM

(2010) Hard questions, hard solutions: Proceduralisation through impact assessment in the EU. West European Politics 33(1): 136–153.

29.

Rasmussen

Toshkov

(2013) The effect of stakeholder involvement on legislative duration: Consultation of external actors and legislative duration in the European Union. European Union Politics 14(3): 366–387.

30.

Rasmussen

(2015) The battle for influence: The politics of business lobbying in the European Parliament. Journal of Common Market Studies 53(2): 365–382.

31.

Reh

Héritier

Bressanelli

et al. (2013) The informal politics of legislation. Comparative Political Studies 46(9): 1112–1142.

32.

Steinberger

Pouliquen

Hagman

(2002) Cross-lingual document similarity calculation using the multilingual thesaurus EUROVOC. In: Gelbukh

(ed.) Computational Linguistics and Intelligent Text Processing, Berlin: Springer, pp. 415–424.

33.

Van Ballaert

(2015) The politics behind the consultation of expert groups: An instrument to reduce uncertainty or a strategy to offset salience? Politics and Governance 3(1): 139–150.

34.

Warntjen

(2012) Measuring salience in EU legislative politics. European Union Politics 13(1): 168–182.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.26 MB

0.10 MB

0.02 MB