Lobbying influence: public consultations and EU policy making

The European Commission considers involvement of interest groups as a way to strengthen democratic legitimacy in the European Union (EU). In terms of a strategy for enhancing good governance, the Commission has taken a range of actions to increase civil society inclusion. Whether these measures really add up to democratic legitimacy, however, depends on the allocation of influence among interest groups. In this context, the purpose of this study will be to analyze the distribution of influence by means of a quantitative analysis of submissions via the Commission’s public consultations.


Introduction
The number of interest groups, active in the European Union, increased sharply from the mid -1980s onwards (Greenwood in Dür, 2008: 1212. As one of the goals of these groups is to influence policy outcomes, the question how successful they are in doing so is relevant for both academics interested in studying policymaking in the EU and those concerned about the validity of decision-making. Still, few attempts at evaluating interest group influence or power in the EU have been undertaken (Lowery et al. in Dür, 2008: 1212. This state of affairs is particularly astonishing given that both of the grand theories of European integration, neofunctionalism and liberal intergovernmentalism, ascribe a major role to interest groups (Grossman in Dür, 2008: 1212. Interest group activity exploded as a result of the continuing transfer of governing functions from member states to EU institutions and the introduction of qualified majority voting on the Single Market issues (Coen, 2007:3). In parallel, institutional demand for such activity was facilitated by the openness of the European Commission and European Parliament and the funding of EU groups by the European A VUZF University, Phd student,Sofia, e-mail: t.v.todorova@umail.leidenuniv.nl.
Commission. Thus, "EU interest groups were able to exert influence along the European policy process from initiation and ratification of policy at the Council of Ministers, agenda setting and formulation at European Commission led forums, reformulation of policy at the European Parliament committees, to the final interpretation, harmonization and implementation of regulation in the nation state. In following and accessing all points of the policy process, EU interest groups are important supply of information on the development and delivery of EU public policy and a potential source of legitimacy to policy-makers" (Coen, 2007:3).
In terms of a strategy for better governance, the EU has enforced various instruments to involve civil society with the goal to increased democratic legitimacy. This paper will look at one of those instruments, i.e. submissions in the Commission's online consultations and analyze the distribution of influence there. Empirical research shows that online consultations have become a regular instrument of consultation with a great variety of actors participating, e.g. business associations and firms, NGOs of international, European, national and subnational origin (Klüver, 2009:2). By identifying policy objectives of interest groups and measuring their success in attaining advocacy goals, it will be possible to draw conclusions on the "winners" of the decision-making process. Whether these actions really improve democratic legitimacy, however, will depend on the distribution of influence among interest groups involved.
As previously mentioned, the multi-layered structure of governance coupled with "high fragmentation" in European institutions offers various access points to the process of decisionmaking.
These institutional provisions complement with an increasing openness towards civil society (Kohler-Koch, 2005: 8). "Due to the constant criticism of the democratic deficit, the European Commission has started to consider interest group inclusion as a means to compensate for the representational deficit. Since the beginning of the 1990s, the Commission has therefore taken various initiatives to increase the participation of interest groups such as the White Paper on Governance or the Transparency Initiative" (Kohler-Koch/Finke, 2007in Klüver, 2008. Yet, although the initiatives from the Commission have offered wide access to interest groups, recent empirical evidence demonstrates that some groups have the ability to exploit this access more effectively than others and hence mobilize more effectively. In a case examination of a consultation on the Commission's proposal for a new European chemicals policy, Persson (2007) concludes, for instance, that business interest groups are significantly better represented than nongovernmental organizations (NGOs) (Klüver, 2008: 1). Also, Dür and de Bièvre (2007) confirm that businesses gained more success in influencing European trade policy than NGOs. There are two reasons why businesses exert more influence over (the content of) policy rules. First, according to Yackee (2006) policy rule makers are likely to be influenced by the number of comments coming from business interests. Second, "comments from business-related interests provide more information and signal a greater level of commenter expertise, causing agencies to respond to the requests made by business commenters" (Yackee, 2006: 130). In addition, business interests are likely to be able to provide higher quality comments than other kinds of commenters for at least another two reasons -business interests are more likely to have the capacity to understand the complex, technical data and studies that the policy makers may cite in support of its proposed rule and business interests are "likely better placed to respond to agency data with their own scientifically sound technical data" (Yackee, 2006: 131).
Although the issue of estimating influence comes first to mind when studying interest groups, only a few researchers have actually addressed the question (Dür in Klüver, 2008: 2). This finding is particularly surprising given that the justification of policy outputs as well as democratic legitimacy of the EU is significantly dependent on how much influence interest groups have and how influence is spread among them. In this context the objective of the current research is to examine the extent and influence of interest groups using online consultations on EU decision and policy making in year 2009, as this particular year shows the largest increase in number of registered representatives according to statistical data from the EU Register (860 entries in January, 2009, 2230entries in December, 2009). More specifically, the purpose is to point out the determinants that shape the advocacy process in these consultations with regards to two policy fields -Transport and Information society. Before going into the empirical findings and assessing the influence in those policy fields, some more information on their selection and character is necessary.
The sectors of Information society and Transport were chosen on the basis of information obtained from two initial field interviews with heads of lobbying associations from each sector, who offered their (lobbying) expertise. In both cases, the experts confirmed an interest in making an impact on a policy proposal. Other selection criteria refer to ISSN 2534-9228 (2020) VUZF review, 5(3) specific findings in lobbying literature relevant to the EU. Researchers claim, for instance, that lobbying is only successful in typically, low politics decision making whereas high politics issues are exceptional cases, where advocacy does not make much difference (Greenwood, 2003in Michalowitz, 2005; "that business interests are especially successful due to the financial resources they can employ for lobbying; that specific institutions, such as the European Parliament and individual committees are especially public interest group-friendly" (Schmitter, 2000in Michalowitz, 2005. Also, the two selected policy fields are considered sectors with issues of typical and largely low politics decision making and have been studied by other researchers as well (Michalowitz, 2005:11). This paper will start with an overview of those aspects of the literature on interest representation that have motivated the present study. Further, in addition to answering the question "Are certain interest groups indeed more successful in influencing than others?" the author will also address some essential subquestions: "How do interest representatives lobby?", "To what extent does lobbying affect EU sectoral policies?" and more important, "To what extent is that evident in decisions taken?", "Are there any correlations between issue characteristics and lobbying success?", "Which indicators are significant to measure interest involvement and why?". For the purpose of answering the main and sub questions above, this study will use as a theoretical model "attributed influence" -i.e. interest group respondents, both public and private actors, lobbyists and decision-makers will be asked to provide self-assessment on the extent of influence and will assess the influence of other groups -in addition to Mahoneyʼs (2008) issue context criteria, where influence differs according to the degree of conflict, the scope, public salience and complexity involved in policy issues of the respective decision-making process entered.

Material and methods
For the purpose of measuring policy positions of interest groups, their submissions in online consultations will be analyzed. In year 2000, the European Commission initiated online consultations as an instrument for providing comments. Interest groups typically react on an initial proposal before the final policy proposal is approved by the College of Commissioners and passed on to the Council of the European Union and the European Parliament (Klüver, 2009:6). The focus on organized interests targeting the European Commission via online consultations is a good starting point for this research, as online consultations constitute a straightforward form of access and all interest groups that try to influence policy proposals are covered. As other empirical studies accordingly show, a variety of actors participate in online consultations, e.g. companies, business associations, and NGOs of international, European, national and subnational origin (Quittkat & Finke, 2008). "Between 2000 and June 2007 544 online consultations were carried out" (Quittkat & Finke, 2008: 206). The researcher is aware that interest groups may also use other channels such as informal contacts or other consultation forms to influence policy proposals.
As a starting point, I categorized the four issues in terms of high and low complexity in accordance with Gormley's (1986) study on rule salience and technical complexity and based upon 1) the extent to which decision makers rely on scientific or technical information to carry out their responsibilities as evident from the consultation papers and 2) the extent to which relevant Transport and Information Society directorates recruit technically skilled personnel or contractors. Policy Issues were then assigned a value for the degree of technical complexity. The researcher believes that this measure has good validity and is also consistent with previous rankings of this sort (Bawn, 1995in Ringquist, Worsham, Eisner, 2003. Public consultations in two low-politics fields, Information Society and Transport, with expected higher level of influence (interest groups tend to influence technical issues more than issues of "high politics" (Smith 2000;Greenwood 2003: 20;Beyers, 2008) will be used and variation will be across the technical complexity of issues chosen in those fields listed below: high-complexity issues: 1) "Internet consultation relating to the preparation of a European road safety action programme, Green Paper TEN-T: A policy reviewtowards a better integrated trans -European transport network at the service of the common transport policy" (Transport); 2) "Information and Communication Technologies for a Low Carbon Society: Broadband guidelines on the application of EU state aid rules to public funding of broadband networks" (Information society); vs.
low -complexity issues: 1) "Driver training and traffic safety education" public consultation (Transport); 2) "Towards a Strengthened Network and Information Security Policy in Europe: Contributions to the EU Commission consultation on strengthening the competitiveness of the EU publishing sector" (Information society).
Policy positions of interest groups were extracted from their submissions in online consultations of the European Commission. Based on a consultation paper, which sets out the preliminary Commission position, advocacy groups have a chance to offer comments during a consultation period of 8 weeks before the proposal of final policy is decided upon. Even though it is possible that these submissions reflect "strategic" rather than "true" policy positions, this should not be a problem, as it is plausible to assume that there is no systematic variation of strategically over -or understating policy preferences across all interest groups, so that the revealed policy position can be taken as an alternative for the true policy position. Being also aware that there are other channels to exert influence, most interest groups trying to influence a policy proposal should be included by the analysis as online consultations constitute the easiest form of access and a wide variety of actors participate (Quittkat and Finke, 2008;Klüver, 2010). The conducted survey with registered organizations was small (N=22), however, an effort has been made to select respondent in such a way that the results would represent the views of a wide range of well-informed experts. Given the small size of the sample and the response rate 8.3 % (22 responses from 263 surveys sent and their distribution (68 % of the responses are from Transport policy field; see Figure 1), these survey results should not be considered representative.

Fig. 1 -Policy Field Response
Source: Author's own data Instead, the aims were rather to explore the opinions of a range of informed practitioners, as well as to generate interesting findings and thus stimulate further research using EU public consultations. 6 of the informants were government officials, 3 from NGOs, 5 from professional associations and another 5 from trade/business organizations (see Figure 2). Figure  2 also shows the distribution among interest groups in percentages, confirming that business interests and professional lobby associations are equally represented (50 %).
Furthermore, this thesis paper argues that lobbying success depends on factors such as degree of complexity, salience, scope and level of conflict of policy issues in question.
It should be noted, though, that the findings here will not indicate the totality of interest group influence because these groups may be working directly to influence rulemaking, as well as working indirectly through informal channels to influence bureaucrats (Chubb, 1983;Furlong, 1997in Yackee, 2010. What is interesting, however, in the case of EU lobbying is that instead of lobbying actors ensuring that certain decisions /directives are set on EU agenda and later implemented, interests' groups react on decisions already proposed by the Commission. In this context, this study will try to explore the dimensions of and hypothesize on EU decisions policy outputs as a result of the distribution of interests involved. For reasons already mentioned previously, the period in year 2009 of closed consultations has been studied, and a variety of sources, such as official documents, telephone interviews, internal archives of EU Register of Interest Representation and secondary sources have been included. Interviews were conducted with 1) respondents having considerable general knowledge of lobbying and 2) respondents with experience pertaining to particular cases. Altogether at least 24 (digitally and telephone) filled in survey responses were received. The two respondents who possessed considerable general knowledge were the staff heads of units. The interviews held with them were generally semi-structured, i.e. there was a questionnaire covering key questions, but there was also room for the respondents to go more deeply in some issues, depending on the specific interview situation.
Data regarding all lobbying via public consultations will be analyzed quantitatively whereas both the interviews and the variety of sources discussed are qualitative in nature. The dependent variable will be measured using cross tabs and correlation, in addition to a simple regression analysis. The author of this document is aware that measuring influence of interest groups comes with certain extent of bias that may vary according to the actors asked. Respondents can be biased when exaggerating or diminishing the extent of influence. In addition, the answers given might be deliberate: "minimizing the role of other actors to reduce their importance or inflating it to create a public response" (Dür, 2008: 10). As already mentioned, another possible limitation when measuring interest group influence refers to the assessment of perceptions of influence, rather than actual influence (Polsby, 1960quoted in Dür, 2008. In addition, if respondents are asked to provide an average across many issues, they might omit potential differences from one issue to another. Finally, limitations that affect all surveys are relevant: interviewer bias, ensuring a sufficient response rate and a tendency to avoid extreme values when respondents are asked to provide a numerical estimation on a pre-given scale (Dür, 2008:11).
Although relatively common in the United States, this method, "which measures perceptions of influence rather than influence as such", has been cautiously used in the context of the EU (Dür, 2008:9). Thera are exceptions -Janet Edgell and Kenneth Thomson (1999) studied 20 British interest groups who were asked to estimate their influence on the Common Agricultural Policy. Andreas Dür, and Dirk De Bievre (2007) examined interest groups on their influence on the EU's trade policy. According to Dür, the results attained from such studies, though, have to be treated with caution "because interest groups may have good reasons to either overestimate it -if they want to legitimate their existence vis-a-vis their members -or underestimate it -to avoid the creation of counter-lobbies that may affect their policy impact" (Dür, 2008(Dür, : 1224. This difficulty can only be solved if results are crosschecked with the assessments made by other groups and public actors and in our case, this becomes even more necessary due to the small sample size used.

Results and discussion
-Issue conflict Not only the salience of an issue, but also the degree of conflict over an issue strongly affects the ability of interest groups to influence policymaking (Mahoney, 2008). The degree of conflict is the dispersion of actorsʼ policy preferences over an issue. Highly conflictual policy issues create a difficult environment for interest groups in terms of successfully lobbying policy-making. If a policy issue is much contested, decision-makers are confronted with countervailing forces, which attempt to push the policy output in opposing directions. Interest groups should, therefore, find it very difficult to influence policy-making since they are fighting against a strong opposition.
By contrast, if the majority of interest groups share the same policy goal, it should be relatively easy for an interest group to influence policymaking since all actors are pushing the legislator into the same direction. It is important to note the difference between salience of a policy issue and the conflictuality of an issue. A policy issue can be highly salient, but all actors can have similar policy preferences so that it is easy for interest groups to shift the policy output towards their ideal point. However, if a policy issue is characterized by a high degree of conflict, actors with opposing policy preferences are trying to push decision-makers into different directions, which make lobbying success unlikely (Klüver, 2010:7). Hence, "the higher the degree of conflict over an issue, the more opposing interest groups are lobbying decision makers in one particular direction and thus the harder it is to shift the policy output" (Klüver, 2010:7).
The point when interest groups form advocacy partnerships with other institutions that share their interest or the degree to which interest groups have to struggle for their cause against a wide range of opposing interests is likely to affect the power of an opinion voiced to change a legislative act (Michalovitz, 2005:7). This matter is mostly studied by US scholars (Baumgartner andLeech, 1998 in Michalovitz, 2005: 8) and it remains to be established in how far European interest groups manage to get involved at the stage when European agenda is set. "Due to the complex system in the US composed of various federalstate and centralized member states, the ability of interest groups to set the agenda appears weaker" (Michalovitz, 2005: 8). However, if EU interest groups manage to influence positions and agendas before conflict can occur, this could also be seen as influence in connection to the factor conflict, since the degree of conflict is determined at the very beginning (Michalovitz, 2005: 8). Therefore, the likelihood of not attaining a goal increases as the conflict on an issue rises. A Pearson Chi square test below shows that in this study the relationship is almost significant at the .252 with 15 degrees of freedom. The crosstab table indicates that as advocates (75 %) rate an increasing level of issue conflict, in 66.7 % of the cases they have exerted moderate influence. Author`s own data; a. 24 cells (100,0%) have expected count less than 5. The minimum expected count is, 05.
-Issue Complexity Another major hypothesis in interest group studies is that interest group influence differs according to the complexity of policy issues. Complexity denotes the degree to which a given policy problem is difficult to analyze, understand or solve (Klüver, 2010:5). Policy-making is a very challenging task and decision-makers are increasingly lacking sufficient information about the impact of specific policy measures. Legislators are, therefore, gathering external information by widely consulting among interest groups to compensate for their lack of information (Klüver, 2010:5). Hence, decisionmakers demand information from private actors and by supplying this information, interest groups are able to influence the outcome of policy-making processes. Correspondingly, Austen-Smith (1993: 799-800 in Klüver, 2010 points out, that decision makers are regularly choosing policies without complete information on their consequences, in which case, information becomes valuable, and those who possess it are accordingly in a position to influence it. Yet, the necessity for external expertise varies from policy proposal to policy proposal. Some proposals may have an effect the entire internal market and may be highly technical, whereas other proposals may only relate a very small sector and simply constitute a small adjustment to existing legislation. If policy proposals are highly complex, the need of external expert knowledge is very high and legislators are particularly open for an exchange with interest groups. However, if a policy issue is of very simple nature, the decision-makersʼ demand for information should be very low and interest groups should find it more difficult to influence policy-making. The chances of being influential should then be particularly high if a policy proposal is very complex as in those cases, legislators are highly dependent on external information. Thus, the following hypothesis will be tested -the higher the complexity of a policy issue, the higher the probability that an interest group is able to influence the policy-making process. The crosstab, however, does not support this tendency, showing 46,2 % of advocates having moderate influence on issues with high complexity and 55,6 % responses of slight influence on issues with low complexity. The relationship is again not significant exceeding the level of significance with three degrees of freedom and Chi-square value .620. Source: Author`s own data; a. 6 cells (75,0%) have expected count less than 5. The minimum expected count is ,41.
In general terms, contrary to the theoretical expectations, scope, complexity and conflictuality of policy issues do not have a statistically significant effect on interest group influence. Thus, the ability of interest groups to influence policy formulation does not vary systematically with the scope, complexity or the degree of conflict of a policy issue. However, the salience of policy issue has a statistically significant positive effect on the ability of interest groups to influence policy formulation. Thus, the more salient the given policy issue, the lower the probability that an interest group is able to influence the content of the policy proposal issued by the European Commission. The regression analysis below further confirms the findings of the crosstabs. The standardized coefficient (-.105) of issue salience shows a statistically significant effect on interest group influence. The regression analysis, applied in order to test and assess the identified correlations of the dependent and independent variables for predicted causal relationships, confirms that within this set of variables, salience is best able to predict the outcome when the effects of the other variable are controlled for.

Conclusions
On the whole, this paper aimed at investigating the effect of issue characteristics on the ability of interest groups to influence policy-making. Drawing on recent literature, it argued that the complexity, the relative scope, the conflictuality as well as the salience of policy issues affect the ability of interest groups to successfully lobby decision-makers.
While recent literature has been stressing the importance of the issue context for understanding variation in interest group influence, EU studies testing these contextual effects have been lacking due to methodological difficulties to measure influence. Drawing on "attributed influence" approach to measure interest group influence, this paper modestly aimed to overcome the shortcomings of the literature by providing a small -N analysis of the effect of issue characteristics on the ability of interest groups to influence policy-making in the European Union. The empirical analysis stands out for its recent, quantitative nature exploring four policy issues in two typically "low politics" fields and confirming that issue salience accounts most for variation in interest group influence.
Some limitations of the research should also be noted, though, which have prevented the accomplishment of the initial objectives. During the research process the author encountered a number of obstacles, the majority of which were out of the researcher's area of influence -for instance, the small response rate, as well as respondents' willingness to provide information and/or providing own interpretation of the tested variables "scope", "salience", "issue conflict" and "complexity." In addition, this research was limited to time, restricting the quality of primary data gathered through field interviews. Overall, the limitations and obstacles of this research slowed down the research process and forced a pursuit of alternative solutions, without jeopardizing its validity and reliability.
Although this paper aimed to imply a less democratic structure of governance to resolve potential problems with influential lobbyists, this weakness of democratic cultures needs to be pointed out, and "it should be considered as a possibly unsolvable problem inherent in the freedom of speech, association and in governance" (Michalovitz, 2005:24).