Are You Influential? Interest Group Influence on Political ... · Are You Influential? Interest Group Influence on Political Agendas and Decisions ... Interest groups may be politically

1    

Are You Influential?

Interest Group Influence on Political Agendas

and Decisions

Helene Helboe Pedersen

Department of Political Science, Aarhus University

[email protected]

Anne Skorkjær Binderkrantz

Department of Political Science, Aarhus University

[email protected]

Paper prepared for the ECPR general conference in Glasgow, September 2014

FIRST DRAFT, PLEASE DO NOT CITE

2    

Abstract

Interest groups may be politically influential in many ways. Here, we distinguish between affecting

political agendas and influencing decision making. We investigate whether patterns of influence are

the same across different group types for these two types of influence. The analyses are based on

measures of perceived influence reported in surveys among domestic interest groups in Denmark

and the UK. This allows us to incorporate factors at the group level, the issue level and the country

level in a single analysis. We find different patterns of influence depending on the type of influence

in question. Economic groups – business groups and institutional groups – report to be most

successful in affecting political decisions, while citizen groups – identity groups and public interest

group – experience higher levels of success in agenda setting. Depending on the measure of

influence used, researchers may thus draw different conclusions about the distribution of political

influence across different types of groups.

3    

Interest groups are of political and scholarly relevance because of their influence in politics (Dür,

2007; Leech, 2010). Influence may come in many shades. Most notable are successful attempts to

affect political decisions, but influence may also occur earlier in the policy process. Groups voice

their concerns in the media; they lobby politicians and seek access to bureaucrats. In these pursuits

they may raise the awareness of the media to their issues, they may persuade politicians or

bureaucrats to pay attention to their concerns and they may affect the ultimate decisions made in

parliament or by ministries. In this paper we distinguish between influence on the public policy

agenda on the one hand and on public policy on the other. Specifically, we ask: “What factors affect

interest group influence on agenda setting and decision making?”

Investigating patterns of interest group influence is of central concern to debates about the

distribution of power between different societal groups. Among the important questions is whether

the same groups are dominant across different arenas and different stages of policy processes

(Baumgartner and Jones, 1993: 35). Evidence suggests variation in the political strategies used by

groups with some types of groups relying more on insider contacts to decision makers, while others

place more emphasis on outsider strategies such as holding press briefings or contacting reporters

(Beyers, 2004; Binderkrantz, 2008a; Dür and Mateo, 2013a; Kriesi et al., 2007). It has also been

demonstrated that the representation of groups varies across different political arenas (Binderkrantz

et al., 2014; Halpin et al., 2012; Salisbury, 1984: 74-75).

Here, we argue that different patterns of success may also be present depending on whether

we focus on agenda setting influence or decision making influence. We hypothesize that different

group types have distinct advantages in regard to different stages of the political process. Economic

groups such as business interests are particularly advantageous in the later stages of the process,

where decisions are made and implemented. Citizen groups are relatively more influential when it

comes to raising the political awareness of their causes – that is setting the agenda. These

expectations are based on a resource exchange model emphasizing variation in the goals and

resources of different types of interest groups (Bouwen, 2004; Braun, 2012: 7). Groups pay

different attention to affecting political agendas and decisions and the value of group assets also

differ across arenas. In turn, this affects their ability to access different arenas and affect agenda

setting and decision making respectively (Binderkrantz et al., 2014).

While our main focus is on the distribution of influence between different types of groups

we also incorporate the material resources possessed by groups, the policy portfolio of groups and

4    

the national setting in the analysis. The contribution of this paper is the investigation of variation in

interest group influence across different stages of the policy process. We analyze this based on

surveys in two countries characterized by different patterns of interest group intermediation.

Denmark is traditionally classified among the most corporatist countries in Europe, while the UK is

a prime example of a pluralist country (Siaroff, 1999). Our dependent variable is based on self-

reported measures of influence.

The measurement of group influence has continually been debated (Dür, 2008), and while

we acknowledge the challenges of relying on group reports on their own influence we maintain that

studies based on measures of perceived influence may provide us with insights about factors

shaping different patterns of influence. While measures of perceived influence may be problematic

in terms of determining absolute levels of influence, groups are more likely to be able to evaluate in

which phases of the decision–making process they find themselves to be most influential – just as

they are able to evaluate which strategies are the most important to them.

We now proceed to a discussion of the theoretical argument, followed by a section on

research design discussing the pros and cons of relying on self-reported influence. The analysis

focuses on comparisons of interest group influence on agenda setting and political decisions.

Explaining influence: Groups, issues, and political systems

Influence can be wielded at any of the various stages from the formation of the political agenda to

the eventual implementation of public policies (Bernhagen, 2011: 559). Political influence may be

defined as success in promoting or changing political decisions to the benefit of the group interests

and/or success in gaining attention for issues of interest to the group. Political influence is thus a

multidimensional phenomenon both related to agenda-setting and decision-making.

The political role of groups can be seen as an “influence production process” (Lowery and

Gray, 2004) and studies have focused on different stages of this process. Central for organizing the

literature is a distinction between strategies, access and influence (Binderkrantz, 2014). Strategies

are the overall approaches groups adopt in seeking political influence, and can be observed

empirically as combinations of specific activities or tactics (Berry, 1977: 212). Access requires that

groups have successfully passed the threshold to for example parliament (Hansen, 1991). Influence

is only obtained when groups are successful in affecting political decisions – including decisions

about which issues should be on the political agenda (Schattschneider, 1975 [1969]).

5    

Considerable progress has been made in theorizing and mapping the lobbying tactics and

strategies used by groups in search of influence and to some extent also the access gained by

groups. A series of studies have documented the wide action repertoire of interest groups (Beyers,

2004; Binderkrantz, 2005; Dür and Mateo, 2013a; Kriesi et al., 2007). Thus, we know that groups

prioritize different strategies focusing primarily on insider or outsider strategies. Studies of group

access have shown that group representation varies across different political arenas or venues with

differences in publicity and decision-making capacity (Binderkrantz et al., 2014; Bouwen, 2004;

Salisbury, 1984). The question is if we can find similar patterns in groups reporting of their actual

influence in the different phase a political process.

Factors at three different levels may potentially affect the influence reported by groups: the

group level, the policy area level and the system level (Dür, 2007). We discuss these factors in the

following sections.

Different groups; different patterns of influence

Interest group influence may be seen as the result of an exchange of resources between interest

groups and gate keepers in politically relevant arenas. At the most basic level, groups supply

decision makers with relevant resources and in return gain influence (Bouwen, 2004; Hall and

Deardorff, 2006; Hansen, 1991; Woll, 2006). Depending on the relative emphasis groups place on

affecting agenda setting and decision making and on the resources possessed by groups, different

patterns of influence may be expected.

While all groups are likely to include different goals in their portfolios, the balance between

goals may vary with some groups emphasizing insider access and the ability to strike deals on

policy concessions, while others find it more attractive to attract public attention to their causes

(Binderkrantz et al., 2014; Maloney et al., 1994). In terms of resources we distinguish between

insider resources and outsider resources. Insider resources consist of information and expertise of

relevance to the policy process on the one hand, and external control – e.g. of members – of

relevance to the political fate of policies on the other. Outsider resources are related to the

representation of causes with broad public appeal or the ability of groups to provide reporters with

stories of news value (Binderkrantz et al., 2014).

The match of goals and resources is crucial for influence, and different patterns of influence

may be expected depending on the group type in question. Interest groups vary both in terms of the

type of interest they represent, the members they organize, and their organizational resources. Here

6    

we define interest groups as membership organizations seeking political influence without running

for public office (Jordan et al., 2004). The characteristics of an interest group will impact on its

political influence because group characteristics influence group goals (Binderkrantz and Krøyer,

2012; Salisbury, 1984), their choice of strategy (Beyers, 2004; Binderkrantz, 2008b), and their

ability to gain access to different political arenas (Binderkrantz et al., 2014).

Interest groups represent a wide variety of interests. We distinguish between seven types of

groups(For a discussion of different categorizations of groups see: Baroni et al., 2014), which can

be compressed into the two more comprehensive categories of economic groups and citizen groups

as listed in table 1. Economic groups – business groups, institutional groups, trade unions and

professional groups – all organize members related to societal production in either the private or

public sector. In contrast, citizen groups organize members based on non-vocational factors and

include for example patient groups, environmental groups and hobby groups (Berry, 1999).

Table 1: Group Categorization

Group label Description

Business group Organizing and representing interests of firms and industries

Institutional group Represent (public) institutions providing services such as education, health, or culture

Trade union Organizing employees and negotiating salaries and work conditions on behalf of their members

Professional group

Organizing employees to help their carriers or professional development, but not negotiating salaries and work conditions

Economic groups

Identity group Organizing citizens with similar interests they try to promote such as students, elderlies or single parents

Leisure group Organizing citizens in relation to their hobbies or personal lives such as sports associations or religious communities

Public interest group

Organizing citizens concerned with specific causes not specifically related to their own personal situation such as environmental or humanitarian organizations

Citizen groups

The different types of economic groups all represent well-defined membership groups, and we

expect them to be mainly concerned with promoting decisions that will benefit their membership.

These groups are also well-endowed with insider resources due to their representation of members

related to societal production (Binderkrantz et al., 2014). This allows them to provide decision

makers with relevant information for example about the likely consequences of political decisions

(Bernhagen, 2011; Bouwen, 2004). Many of these groups also control crucial parts of society’s

economy. If business interests are not met, they may move their business to another place or simply

close down (Lindblom, 1977). Trade unions negotiate salaries and may mobilize their members for

7    

strikes (Rokkan, 1966). Institutional groups organize institutions crucial for public production of

services such as hospitals or schools. In conclusion, we expect economic groups to emphasize

decision-making influence and be relatively more likely than citizen groups to gain it due to their

possession of insider resources.

Citizen groups are on their side likely to be relatively more concerned with affecting agenda

setting. First, many citizen groups promote diffuse interests and are therefore in constant need of

publicly promoting their cause and organization not only to gain political influence but also to

survive as an organization (Dunleavy, 1991). Second, even though citizen groups may also be

concerned with specific political decisions, their purposes are often relatively broad including for

example raising the general awareness of a specific disease or a humanitarian cause. Citizen groups

also have relatively strong outsider resources due to their representation of broadly appealing causes

and causes of often great news value such as poor treatment of elderly, child abuse or humanitarian

disasters. In conclusion, citizen groups are thus more likely than economic groups to be agenda-

setting influential.

Groups’ material resources may also impact on their ability to gain influence (Binderkrantz

et al., 2014; Eising, 2007). If your organization has more staff to write press statement or money to

launch a big campaign the chance of being successful in setting the agenda increases. Similarly, if

your organization is able to hire professional staff that can provide, systemize and communicate the

crucial expertise demanded by the decision-makers and you have the staff to monitor the political

process and build and maintain networks with key actors, your organization stand a better chance in

influencing public policy. So, as Eising (Eising, 2007) has stated, an almost trivial expectation is

that material resources are positively related to influence in the agenda setting phase as well as in

decision making.

The impact of group type as well as material resources may however depend on the policy

area, where groups seek their influence and the political system they act within. The next two

sections discuss these potentially moderating effects.

Moderating impact of policy portfolio

Interest groups vary in the set of policy areas they seek to affect and the policy context may be

crucial for the ability of groups to affect politics. Some areas may be marked by fierce competition

and conflict between different types of groups, while groups operating in other areas may

experience less conflict intensity (Binderkrantz et al., 2014; Klüver, 2011; Lowi, 1964). Also, the

8    

degree of institutionalization of political processes varies across different policy areas even within

the same political system. Some areas may be characterized by stable patterns of interaction

between selected groups and public decision makers, while others may mirror pluralist patterns of

interest intermediation (Rhodes and Marsh, 1992).

Dür and Mateo (2013b) argue that the effect of group type on choice of strategy depends on

the policy area the groups are active in. Specifically, they argue that distributive policy will make

economic groups (business groups and professional groups) more likely to use insider strategies not

to call attention to issues where they have the possibility to obtain concentrated gains and not face

strong opposition from other actors. However, on regulatory issues they are more likely to face such

opposition and thus pushed into using outside strategies (Dür and Mateo, 2013b: 665). While Dür

and Mateo find the expected effects and their hypothesis regarding choice of strategy is supported,

we may not be able to transfer their argument directly to the question of influence. In respect to

influence, the most important impact of policy may instead relate to the relevance of group

resources within different policy areas.

Here, we focus on the distinction between policy areas related to the labor market and

societal production such as industry and agriculture. In these areas, institutionalized interaction

between groups and decision makers is likely to be most pronounced and the advantage of

economic groups particularly strong (Rhodes and Marsh, 1992). We therefore expect that economic

groups with a relatively high emphasis on policy areas related to societal production are more

successful in affecting both agenda setting and decision making.

Country differences: corporatism vs. pluralism

Considerable scholarly attention has been devoted to discussions of country level differences in

state-society interaction. Different institutions may enhance or lower the access interest groups

enjoy to policymakers (Dür, 2007: 4). Of particular interest is the distinction between corporatist

and pluralist countries originally introduced by Schmitter (Schmitter, 1974). The definition and

operationalization of corporatism has been continually discussed, but a main institutional feature of

corporatist systems is the privileged integration of specific groups into public decision making. In

corporatist countries, interest groups are thus tightly included in the preparation as well as the

implementation of public policy (Öberg et al., 2011). In pluralist societies, groups are considered

more distanced from the state. Rather than being institutionally included in public decision making

9    

groups are portrayed as being in open competition to affect public policies (Baumgartner and

Leech, 1998).

Different patterns of interest group influence may be expected when we compare corporatist

and pluralist systems. First, economic groups are likely to benefit from a corporative system. They

provide the relevant resources for negotiations and are able to strike political agreements on behalf

of their members. Hence, we expect corporative groups to be generally more influential in

corporatist countries than in pluralist ones.

Second, the importance of material resources may also vary across systems. In corporatist

systems, some groups obtain a privileged position and thus superior access to affecting political

decisions. While the privileged groups are certainly likely to possess material resources, other

criteria may also be at play and these may to some extent offset the importance of resources. In

pluralist systems, resources may therefore be even more crucial than in corporatist systems because

no such offsetting institutions are present (Christiansen, 2014). Hence, as a final expectation we

expect material resources to matter more in pluralist than in corporatist systems. Our expectations

are summarized in table 2.

Table 2. Expectations

Group expectations

E1 Economic groups perceive of themselves as more influential on decision making than citizen groups

E2 Citizen groups perceive of themselves as more influential on agenda setting than economic groups

E3 Groups with more material resources perceive themselves as more influential both regarding to agenda setting and decision making

Policy area expectation

E4 Economic groups perceive themselves as more influential both in regard to agenda setting and decision making in policy areas related to societal production

Country expectations

E5 Economic groups perceive themselves as more influential both in regard to agenda setting and decision making in corporatist than in pluralist countries

E6 Material resources affect perceived influence on both agenda setting and decision making more in pluralist than in corporatist countries

Research design

To test the expectations outlined in table 2 we draw on surveys among domestic interest groups in

the UK and Denmark. This enables us to test the possible variation between corporatist and pluralist

10    

countries in patterns of interest group influence: while the UK is among the most pluralist European

countries, Denmark is usually considered relatively corporatist (Siaroff, 1999).

Identifying relevant and comparable interest group populations across countries is

challenging. We restricted the study to associations – that is groups with members or supporters and

included only groups identified in political arenas and identified groups in a wide range of sources

including for: 1) The UK: written and oral evidence to parliament, meetings with ministers,

appearance in national newspapers (the Guardian and the Telegraph), and responses to selected

government consultations, 2) Denmark: letters to parliament, membership of government

committees, appearance in national newspapers (Politiken and Jyllands-Posten), and responses to

consultations on bills.

We obtained response rates of 26 percent (UK) and 61 percent (Denmark)1. Response rates

among UK groups varied between 23 for leisure groups and 31 for institutional groups, while the

rates of response for Danish groups varied from 46 for professional groups to 76 for institutional

groups. The low response rate among British groups means that we need to be particularly cautious

about the possibilities for generalizing to the broader set of UK groups. In regard to obtaining

statistically reliable estimates it is, however, important to notice that the absolute numbers of

respondents are relatively similar in each country (520 for Denmark and 477 for the UK).

All groups in the survey population have been categorized into group types following the

INTERENA coding scheme. After all groups were coded we draw a sample of 100 groups of each

country population and made a reliability test. We obtained satisfactory kappa scores 0.906 for the

Danish groups and 0.755 for the British groups. The distribution of group types in the survey data is

shown in table 3.

                                                                                                                         1 For Denmark the survey was also distributed to groups who were not identified in documentary sources. To facilitate comparison these are not included in the analysis. The response rate reported here is for groups identified in documentary sources and therefore differs from the response rate reported in publications based on the full survey (Binderkrantz et al., 2014).

11    

Table 3. Group types in the UK and Denmark

Denmark United Kingdom

Trade union 20.6 3.8

Business group 26.0 21.4

Institutional group 8.5 10.7

Professional group 6.5 20.1

Identity group 15.0 11.7

Leisure group 7.5 5.7

Public interest group 16.0 26.6

N 100 (520) 100 (477)

A particularly noticeable difference between the British and Danish group distributions is the larger

share of trade unions in Denmark and the larger share of professional groups in UK. This indicates

the differences in the organizational structure of the labor market, where Danish workers are among

the most organized reflecting the corporative tradition of labor market regulation, whereas

employees in UK are typically organized in more narrowly defined professional groups focused on

career counselling and professional development.

In the survey, the groups were asked about their material resources, their activities on

different policy areas as well as their success in relation to different objects. Hereby the survey

provides measures of the independent as well as dependent variables.

Two dimensions of influence

The question of how to measure interest group influence is highly debated (Dür, 2008: 560; Leech,

2010), but despite the challenges in studying influence, recent studies have begun to tackle the issue

of group influence empirically (Baumgartner et al., 2009; Dür, 2008; Klüver, 2009; Klüver, 2011;

Mahoney, 2009). Here, we focus on information about perceived influence on agenda setting and

decision making. Rather than focusing on success in specific cases, we are interested in the relative

success of different group types across the set of issues they are involved in. This allows us to

investigate the effect of factors at different levels and to obtain information from a large number of

groups operating in different countries. Further, we do not have to rely on publicly expressed group

preferences in for example official documents issued by groups as indicator of their preferences.

12    

Finally, such an approach provides less of a need to rule out the effect of other forces such as public

opinion and counteractive lobbying (Dür, 2008).

Instead, the main challenge is the reliance on groups as informants. Two issues are relevant

here: First, groups may strategically over- or underestimate their own influence (Dür, 2008: 565).

As examples, Dür cites the need of associations to stress the relevance of their work to members

and the fear of the creation of counter-lobbies if it becomes known that a group is influential (Dür,

2008: 565). These concerns are, however, more pressing in studies where the influence of specific

groups will be reported than in anonymous surveys. Also, in comparing the factors affecting

different types of influence this issue is less pressing than in studies of absolute levels of influence.

Second, groups may not have reliable information about their level of influence since the evaluation

of lobbying effects is not only complicated for scholars but also policy participants (Pedersen, 2013:

30). While this is a concern, the comparison of group estimates of different types of influence is

likely to be more valid than their evaluations of their overall level of success.

However, it needs to be acknowledged that we study perceived rather than actual influence

and that our conclusions should ideally be tested in studies relying on other measurements of

influence. In order to validate our measure of self-reported influence we have analyzed the

correlation between survey responses on perceived influence and quantitative measures of group

access to the media, influence in consultations and agenda setting influence in parliament (for the

Danish case where these data were available). These analyses result in high levels of correlations

between survey questions asking about for example success with making MPs ask questions in

parliament and actual questions asked by MPs to ministers (See also: Pedersen, 2013).

The two measures of perceived influence are based on a set of questions asking groups

about different aspects of influence. The questions were introduced as follows: “We would like you

to indicate how often within the last year your organisation’s work has led to various outcomes.

Please indicate how often your organisation’s work was significant in the following ways”. Groups

had the opportunity to respond “very often”, “fairly often”, “occasionally” or “never”. Table 4

shows the factor loadings of a rotated principal factor analysis2.

The factor analysis indicates the expected two dimensional structure of agenda-setting and

decision-making. Some items clearly belong to one of the factors. Influencing public opinion, the

media agenda or the interest group agenda are clearly related to factor 1, which we call agenda-

setting. In contrast, making government launch a proposal or change already suggested proposals or

                                                                                                                         2 We accept the four point scales as approximately interval scales and use a varimax rotation

13    

decisions are clearly related to factor 2, which we term decision-making. Other items are less

clearly related to one of the factor. Here we have been guided by factor loadings as well as

theoretical justification. Most importantly, we have added the items of making other parties launch

a proposal in parliament to the index of agenda-setting as it loads equally on both factors and seem

to fall in the same category of activities as making politicians take up an issue or ask a question to

ministers.

Based on this we construct two additive indexes imputing missing data by means of index

items, if the group responded on at least four of the six items in each index (alpha 0.85 for Agenda-

setting and 0.88 for Decision-making). The indexes where recoded on a 0-100 scale (Agenda-

setting mean=39.20 (std.=20.22) Decision-making mean=30.05 (std.=30.05)), with higher values

indicating groups perceiving themselves to be more influential.

Table 4. Principal factor analysis, factor loadings

How often within the last year has your organization’s work led to:

Factor 1 Agenda-setting

Factor 2 Decision-making

Changes in public opinion 0.567 -

Media have taken up an issue 0.714 -

Other interest groups have taken up an issue 0.634 -

Public servants have taken up an issue 0.439 0.467

Politicians and parties have taken up an issue 0.542 0.338

Government has launched proposals in parliament - 0.510

Other parties have launched proposals in parliament 0.374 0.389

Legislative proposal was changed by the ministry - 0.840

Legislative proposal was changed in parliament - 0.751

Questions asked to the minister in parliament 0.422 0.369

Changes to orders or circulars - 0.715

Changes to decisions in government ministries - 0.642

Note: factor loadings smaller than 0.3 are not reported, N=904

Material resources and policy area

Measures of material resources and policy focus are also obtained from the survey. As a measure of

material resources we use the number of full time staff in the organization working on political

14    

relevant tasks such as contact to civil servants, politicians or journalists, generating analyses or

monitoring the political process. We use this measure rather than the budget or total number of

staff, since the way the money are spent within an organization is important to its lobbying success

and this measures have proven to be the strongest in other analyses (Binderkrantz et al., 2014;

McKay, 2011). The number is log-transformed and recoded on a scale from 0-100.

Regarding policy areas we asked the groups to indicate how active their organization was on

19 different policy areas3 on a four point scale ranging from very active to not at all active. Our

expectation is that economic groups will be especially influential on public policy decisions on

areas closely related to the regulation of private production. These are 1) labor market policies, 2)

industry and consumer policies, and 3) agriculture and food production policies. We therefore

construct a measure of how active a group is on these areas relative to its overall activity across the

full set of policy areas. Specifically, groups get a score of 3 if they are very active on a policy field,

2 if they are somewhat active, 1 if they are a little active and 0 if they are not active at all. We

summarize a group’s score to have a measure of its overall activity. We further summarize the

group’s score on the three selected policy areas and divide this score by the overall activity of the

group. Hereby, we have a measure we call ‘CorpPolicy’ which ranges from 0 to 1. Overall activity

potentially ranges from 0 to 57, in the data the mean score is 12.90 (min=0, max=50). The mean

score of CorpPolicy is 0.15 (std.=0.15).

Agenda-setting and decision-making influence according to interest groups

A first question of interest relates to the level of perceived influence reported by groups. Figure 1

shows the distribution on the two indexes of agenda-setting and decision-making influence. On

average, groups find themselves more successful when it comes to affecting the political agenda

than in regard to political decisions. Also, while the distribution on the agenda setting index

approximates a normal distribution, it is evident that a large number of groups report low levels of

decision-making influence, while relatively few groups achieve a score of more than 50 on the

index – roughly corresponding to reporting to experience the different types of successes ‘often’.

                                                                                                                         33 Labor market policy, Urban and housing policy, Research, technology and communications policy, Defence and security policy, Industrial and consumer policy, EU policy, Refugee and immigrant policy, Religious policy, Local government and regional policy, Culture and sports policy, Agriculture, fishery and food policy, Environment and energy policy, Monetary, fiscal and tax policy, Law and order/justice policy, Social affairs and families policy, Health policy, Traffic and infrastructure policy, Education policy, Foreign affairs (excluding EU)

15    

Figure 1: Distribution of agenda-setting and decision-making influence

16    

It is also interesting to note that the mean levels of influence are not equal across the two countries.

The Danish groups score on average 2 points higher than British groups on decision-making

influence (Chi2=4.1, p=0.042) and 0.1 point lower on agenda-setting influence (Chi2=9.2,

p=0.002). This may signal a higher level of influence on decision-making in a corporatist country

like Denmark, but the differences are small and given the relatively low response rates of the UK

survey we are reluctant to draw such a conclusion.

Figure 2 shows the means of decision-making and agenda-setting influence across group

types. Business groups, institutional groups, and trade unions find that they are able to influence

decisions the most, whereas public interest groups perceive themselves as more influential in

agenda-setting. Also, identity groups are rather successful in term of agenda-setting whereas leisure

groups and professional groups tend to be least influential. Only the differences across groups on

decision-making are significant (Chi2=17.52, p=0.008). Overall, this descriptive finding offers

some support for E1, that economic groups report to be more influential on decisions, while citizen

groups are more influential on agenda-setting, but with variation within these broader group types

as evident in the figure.

Figure 2. Mean influence on agenda and decisions across group types

17    

Again, the pattern is not exactly the same across the two countries. In Denmark, differences across

group means in agenda-setting are close to being significant (Chi2= 11.37, p=0.078), and clearly

business groups, institutional groups, and trade unions report the highest levels of decision-making

influence. In the UK, differences in agenda-setting are not significant and in regard to decision-

making differences are only borderline significant (Chi2=12.76, p=0.047). Here, trade unions are

not more influential on decisions than identity groups. So, differences across groups tend to be more

pronounced and economic groups relatively stronger in Denmark, which offers some support to E5.

Regarding policy area, we find no indication of interaction. Figure 3 shows that the

differences in mean influence on decision-making and agenda-setting across economic and citizen

groups are almost similar independently of whether they are relatively highly engaged in policy

areas related to societal production or not, and none of the mean differences across group types are

significant. For both group types, they tend to be more influential both regarding agenda-setting and

decision-making if they are highly active on policy areas related to societal production. But the

increase is not higher for economic groups as expected (E4).

Figure 3. Mean influence across group type and activity on policy areas related to societal

production

18    

Finally, in order to test our expectations in a multivariate analysis including the impact of material

resources measured as political staff, we run OLS-regressions for agenda-setting and decision-

making influence respectively. We present four models for each dependent variable in order to

facilitate the interpretation of results. In the first model, we only include factors we expect to have a

direct impact on the levels of influence. This is group type and staff. In this first model we show

differences across the disaggregated group types with business groups as reference category. In the

other three models we only differentiate between economic and citizen groups to simplify the

analyses. In the second model, we include the country dummy and interaction terms to test the

expected moderating impact of country on the impact of staff and group type. In the third model, we

exclude the country and include policy area and interaction terms to test the expected moderating

impact of policy focus on the impact of group type. Finally, in the fourth model we include all

variables in a single model. Table 5 presents the results for perceived agenda-setting influence.

Table 6 presents the result of perceived decision-making influence.

In model 2-4 in table 5 and 6 we see that adding the moderating variables does not increase

the explanatory power substantially and we find only very few significant effect. For agenda-setting

the impact of group type and staff is stable across all three models. Economic groups are

significantly less influential on the agenda compared to citizen groups and staff increases groups’

ability to be influential. Only in model 4 we see that staff has a slightly stronger effect in UK

supporting E6, and we also find that groups tend to be less agenda-setting in the UK compared to

Denmark even though the effect only shows after controlling for policy area.

For decision-making the impact of staff is stable and positive across all models supporting

E3, but group type tends to become insignificant after controlling for country. This is however a

consequence of differences with-in the category of economic groups which are evident from model

1 in table 6. These results are stable also after controlling for country. Similarly to the results for

agenda-setting we find that groups tend to be less influential in UK and staff matters more for

influence in UK than in Denmark, but again this moderating effects only appears after controlling

for policy area. Finally, we find no support of E4 regarding policy area. It does not seem to matter

for decision-making influence whether or not groups are relative more active on policy areas related

to societal production.

19    

Table 5. Explaining agenda-setting influence ---------------------------------------------------------------------------- (1) (2) (3) (4) ---------------------------------------------------------------------------- Trade Union 0.591 (1.932) Institutional -1.966 (2.143) Professional -3.952* (2.009) Identity 6.801*** (1.939) Leisure -2.891 (2.477) PIG 9.075*** (1.688) Staff (log) 0.518*** 0.508*** 0.545*** 0.500*** (0.0347) (0.0434) (0.0341) (0.0427) Economic -8.045*** -6.509*** -8.168*** (1.607) (1.682) (2.091) UK -3.542 -6.588** (2.259) (2.343) UK*Economic 1.771 3.802 (2.357) (2.431) UK*Staff 0.117 0.154* (0.071) (0.071) CorpPolicy 5.633 2.966 (6.974) (7.093) Corp*Economic -2.938 -0.161 (8.627) (8.733) Constant 27.24*** 34.26*** 31.60*** 34.08*** (1.386) (1.389) (1.312) (1.621) ---------------------------------------------------------------------------- N 927 927 877 877 adj. R-sq 0.252 0.234 0.239 0.243 ---------------------------------------------------------------------------- Standard errors in parentheses * p<0.05, ** p<0.01, *** p<0.001

20    

Table 6: Explaining decision-making influence ---------------------------------------------------------------------------- (1) (2) (3) (4) ---------------------------------------------------------------------------- Trade union -6.996*** (2.024) Institutional -1.108 (2.232) Professional -5.639** (2.101) Identity -3.658 (2.043) Leisure -6.088* (2.606) PIG -6.023*** (1.770) Staff (log) 0.546*** 0.502*** 0.535*** 0.497*** (0.036) (0.045) (0.036) (0.045) Economic 1.280 3.907* 1.765 (1.677) (1.750) (2.176) UK -6.255** -8.617*** (2.362) (2.441) UK*Economic 2.670 4.525 (2.454) (2.528) UK*Staff 0.133 0.148* (0.074) (0.074) ‘CorpPolicy’ 6.737 2.452 (7.268) (7.374) ‘CorpPolicy’* Economic -6.973 -3.269 (8.981) (9.068) Constant 22.77*** 19.90*** 16.24*** 19.77*** (1.454) (1.457) (1.370) (1.699) ---------------------------------------------------------------------------- N 920 920 872 872 adj. R-sq 0.220 0.213 0.218 0.227 ---------------------------------------------------------------------------- Standard errors in parentheses * p<0.05, ** p<0.01, *** p<0.001

21    

In sum, we conclude that group type and staff are the most important factors for explaining

perceived influence. To give a better impression of the impact of group type we rerun model 1 in

table 5 and 6 including the country dummy as control due to the significant effects found in table 5

and 6 and plot the distance between predicted mean influence for each group type and the overall

mean for agenda-setting as well as decision-making influence. The plot is shown in figure 4.

The plot shows that – as expected – business and institutional groups on average report to be

more influential on decision making than other groups type. However, the logic does not seem to fit

all types of economic groups. Trade unions and professional groups score below the mean –

indicated by the zero-line. In general, citizen groups report to be less influential on decisions, but

identity groups are actually quite successful even though not as much as business and institutional

groups. So, regarding decision-making the subset of economic groups that control the production

means seem to be the most successful, while the groups that organize employees do not seem to

benefit from the insider resources we expected them to have.

Figure 4. Distance between predicted means and aggregated means of influence across group types

22    

Regarding agenda-setting we clearly see that identity and public interest groups report to be more

influential than the average group. This supports our expectation (E2). All economic groups score

below the overall mean, and among citizen groups only leisure groups perform below average. So,

the figure shows that interest groups vary in their ability to influence the political agenda and

political decisions. Citizen groups – public interest groups and identity groups – find that they are

successful in influencing the political agenda making different political actors pay attention to their

causes. Economic groups – more specifically business and institutional groups – find that they are

capable of influencing the political decision either by being successful in amending bills, proposals

or administrative decisions.

Conclusion

Political influence comes in many shades. In this paper we set out to investigate whether the groups

that dominate insider games of politics and affect political decisions are the same, who are

successful in attracting the attention of the media as well as decision makers to their causes. We

relied on measures of perceived influence based on surveys among domestic interest groups in

Denmark and the UK. While the merits of such measures may be debated, they do provide us with

valuable insights into the relative success experienced by groups when it comes to different types of

political influence.

Our findings clearly demonstrate that while economic interest – notably business groups and

institutional groups – find themselves more successful than the average group in terms of affecting

political decisions, citizen groups – identity groups and public interest groups – report higher levels

of success in shaping political agenda. These findings mirror previous studies of interest group

access to political arenas (Binderkrantz et al., 2014) as well as studies of group choices of strategies

(Beyers, 2004; Binderkrantz, 2008b; Dür and Mateo, 2013b). They also support the idea of interest

group influence as a result of a resource exchange between groups and gate keepers in political

arenas. In this exchange, some groups have political goals and resources that lead them towards

affecting decisions, while others have a better stand in more public arenas.

An interesting finding is also that little variation is present in the patterns of influence

reported across the two countries. It is customary in the interest group literature to point to large

differences in patterns of state-society interaction across countries, but this study indicates that such

differences may be less important than usually acknowledged. This may be related to a gradual

23    

equalization due to the process of de-corporatization reported to have characterized different

countries in the last couple of decades (Öberg et al., 2011).

Given the differential patterns of influence it is relevant to ask: what kind of influence

matters the most? It is tempting to value influence on public decisions higher than influence on the

political agenda. After all, talk is cheap, and what matters in politics is whether issues on the agenda

are eventually subject of legislation. Before drawing this conclusion at least two points must,

however, be considered: First, when asked, many interest groups report to be highly interested in

raising the awareness of their causes. For groups, agenda setting is thus an important priority.

Second, while some gains may be achieved in short-term political processes, there is reason to

believe that major political changes take place over an extended period of time. Here, it may be

more important to gradually change the public and political understanding of political issues than to

gain influence on specific regulations under consideration. In the long run those who are successful

in affecting the political agenda and the framing of political issues may therefore ultimately be more

powerful than those who are able to achieve small changes to proposed regulations.

For researchers, perhaps the most important lesson is that studies of different aspects of

group influence may lead to very different pictures of the distribution of influence across different

types of groups. Here, we have relied on relatively rough measures of perceived influence on

agenda setting and decision making. Further studies may use more fine-grained operationalizations

of group influence in different stages of the decision making process and thus be able to give a more

detailed picture of the influence of different types of groups.

References

Baroni L, Carroll BJ, Chalmers AW, Muñoz Marquez LM and Rasmussen A (2014) Defining and classifying interest groups. Interest Groups and Advocacy 3(2): 141-159.

Baumgartner FR and Leech BL (1998) Basic Interests. the Importance of Groups in Politics and in Political Science. Princeton, N.J.: Princeton University Press.

Baumgartner FR and Jones BD (1993) Agendas and Instability in American Politics. Chicago: The University of Chicago Press.

Baumgartner FR, Berry JM, Hojnacki M, Kimball D and Leech BL (2009) Lobbying and Policy Change. Who Wins, Who Loses, and Why. Chicago: The University of Chicago Press.

Bernhagen P (2011) Who Gets What in British Politics - and How? An Analysis of Media Reports on Lobbying around Government Policies, 2001-7. Political Studies: 1-21.

24    

Berry JM (1999) The New Liberalism. the Rising Power of Citizen Groups. Washington, D.C;: Brookings Institution Press.

Berry JM (1977) Lobbying for the People: The Political Behavior of Public Interest Groups. Princeton: Princeton University Press.

Beyers J (2004) Voice and Access. Political Practices of European Interest Associations. European Union Politics 5: 211-240.

Binderkrantz A (2008a) Different Groups, Different Strategies: How Interest Groups Pursue Their Political Ambitions. Scandinavian Political Studies 51(1).

Binderkrantz A (2008b) Different Groups, Different Strategies: How Interest Groups Pursue Their Political Ambitions. Scandinavian Political Studies 31(2): 173-200.

Binderkrantz A (2005) Interest Group Strategies: Navigating Between Privileged Access and Strategies of Pressure. Political Studies 53: 694-715.

Binderkrantz AS and Krøyer S (2012) Customizing Strategy: Policy Goals and Interest Group Strategies. Interest Groups and Advocacy 1(1): 115-138.

Binderkrantz AS, Pedersen HH and Christiansen PM (2014) Interest group access to the administration, parliament and media. Governanve Forthcoming.

Binderkrantz AS (2014) Legislatures, Lobbying, and Interest Groups. In: Martin S, Saalfeld T and Strøm KW (eds) The Oxford Handbook of Legislative Studies. : Oxford: Oxford University Press, 526-542.

Binderkrantz AS, Christiansen PM and Pedersen HH (2014) A privileged position? The influence of business interests in government consultations. Paper.

Bouwen P (2004) Exchanging goods for access: A comparative study of business lobbying in the European Union institutions. European Journal of Political Research 43: 337-369.

Braun C (2012) The Captive or the Broker? Explaining Public Agency-Interest Group Interactions. Governance: An International Journal of Policy, Administration, and Institutions.

Christiansen PM (2014) The Paradox of Decorporatization. Why Less Corporatism Increases Inequality in Interest Group Access. Paper for Presentatino at the ECPR General Conference, Glasgow.

Dunleavy P (1991) Democracy, Bureaucracy and Public Choice. Economic Explanations in Political Science. New York: Harvester/Wheatsheaf.

Dür A (2008) Measuring Interest Group Influence in the EU: A Note on Methodology. European Union Politics 9(4): 559-576.

Dür A (2007) The Question of Interest Group Influence. Journal of Public Policy 27: 1-12.

25    

Dür A and Mateo G (2013a) Gaining access or going public? Interest group strategies in five European countries. European Journal of Political Research 52: 660-686.

Dür A and Mateo G (2013b) Gaining access or going public? Interest group strategies in five European countries. European Journal of Political Research 52(5): 660-686.

Eising R (2007) Institutional Context, Organizational Resources and Strategic Choices: Explaining Interest Group Access in the European Union. European Union Politics 8: 329-362.

Hall RL and Deardorff AV (2006) Lobbying as Legislative Subsidy. American Political Science Review 100(1): 69-84.

Halpin D, Baxter G and MacLeod I (2012) Multiple Arenas, Multiple Populations: Counting Organized Interests in Scottish Public Policy. In: Halpin D and Jordan G (eds) The Scale of Interest Organization in Democratic Politics. Data and Research Methods. Houndmills, Basingstoke: Palgrave Macmillan, 118-140.

Hansen JM (1991) Gaining Access. Congress and the Farm Lobby, 1919-1981. Chicago: The University of Chicago Press.

Jordan G, Halpin D and Maloney W (2004) Defining Interests: Disambiguation and the Need for New Distinctions. British Journal of Politics and International Relations 6(2): 195-212.

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

Klüver H (2009) Measuring Interest Group Influence Using Quantitative Text Analysis. European Union Politics 10(4): 535-549.

Kriesi H, Tresch A and Jochum M (2007) Going Public in the European Union. Action Repertoires of Western European Collective Political Actors. Comparative Political Studies 40(1): 48-73.

Leech BL (2010) Lobbying and Influence. In: Maisel SL and Berry JM (eds) The Oxford Handbook of American Political Parties and Interest Groups. , 234-567.

Lindblom CE (1977) Politics and Markets: The World's Political-Economic System. New York: Basic Books.

Lowery D and Gray V (2004) A Neopluralist Perspective on Research on Organized Interests. Political Research Quarterly 57(1): 163-175.

Lowi TJ (1964) American Business, Public Policy, Case-Studies, and Political Theory. World Politics 16(4): 677-715.

Mahoney C (2009) Brussels Vs. the Beltway. Advocacy in the United States and the European Union. Washington, D.C.: Georgetown University Press.

26    

Maloney WA, Jordan G and McLaughlin AM (1994) Interest Groups and Public Policy: The Insider/Outsider Model Revisited. Journal of Public Policy 14(1): 17-38.

McKay A (2011) Buying Politics? The Effects of Lobbyists' Resources on Their Policy Success. Political Research Quarterly Preview.

Öberg P, Svensson T, Christiansen PM, Nørgaard AS, Rommetvedt H and Thesen G (2011) Disrupted Exchange and Declining Corporatism: Government Authority and Interest Group Capability in Scandinavia. Government and Opposition 46(3): 365-391.

Pedersen HH (2013) Is measuring interest group influence a mission impossible? The case of interest group influence in the Danish parliament. Interest Groups & Advocacy 2(1): 27-47.

Rhodes RAW and Marsh D (1992) New directions in the study of policy networks. European Journal of Political Research 21(1-2): 181-205.

Rokkan S (1966) Norway: Numerical Democracy and Corporate Pluralism. In: Dahl RA (ed) Political Oppositions in Western Democracies. New Haven: Yale University Press.

Salisbury RH (1984) Interest Representation: The Dominance of Institutions. American Political Science Review 78(1): 64-78.

Schattschneider EE (1975 [1969]) The Semisovereign People. A Realist's View of Democracy in America. US: Thomson Learning.

Schmitter PC (1974) Still the Century of Corporatism? Review of Politics 36: 85-131.

Siaroff A (1999) Corporatism in 24 industrial democracies: Meaning and measurement. European Journal of Political Research 36: 175-205.

Woll C (2006) Lobbying in the European Union: From sui generis to a comparative perspective. Journal of European Public Policy 13(3): 456-469.