Strategic Framing & The Policy Image: The Tobacco Products Directive 2014/40/EU

(1)

Directive 2014/40/ EU

Research Question: What are the effects of strategic framing by stakeholders on the

decision-making process in regards to the Tobacco Products Directive 2014/40/EU?

MASTER THESIS

The Institute of Public Administration Faculty of Governance and Global Affairs

Leiden University, The Netherlands January 12th 2020

Jemila Benchikh (s2087030) Supervisor: Dr. Caelesta Braun

(2)

2

Acknowledgements:

I would first like to thank my mom, who supported me emotionally and financially during my time here in the Netherlands. I am forever grateful for the sacrifices you have made and there are no words to express how thankful I am to have you as my mom. In addition, I am grateful for the support of my dad, my Auntie Martha, my family in Los Angeles and in Besançon. You pushed me and motivated me to reach my goal even when I myself had lost sight of my purpose. Thank you for everything you have all done for me. I would be nothing without your love and support. I would like to thank my supervisor, Dr. Caeleste Braun at Leiden University, who was extremely helpful, answered all my emails and met with me whenever she could even though she was busy. Dr. Braun’s vast knowledge in the field helped guide this thesis in the right direction. Furthermore, her work on interest group lobbying truly inspired my thesis and I would not have been able to complete my research without her. Thank you for all your support, your guidance and most importantly your patience.

Thank you to my amazing group of friends: Ariunaa, Emma, Christof, Thomas, Ryan, Rocia, Eimear, Giulia, Michaela and Aurélien for being there for me when I needed them most. Whether that was answering my frantic phone calls, sending me motivational messages, lifting my spirits or offering to help read my thesis and provide feedback. You were there for me and gave me strength during this difficult period of my life. Thank you for your friendship, love, understanding and patience.

Lastly, I would like to thank Leiden University for given me the opportunity to study in The Netherlands. It was a difficult journey to get to this point however, I know it was all worth it. I have grown academically, professionally and personally and I could not have done it without this institution.

(3)

3

List of Tables and Figures:

Figure 1: Infographic representation of before and after EU rules on Tobacco Products ... 10

Figure 2: Flow Diagram of the Advocacy Coalition Framework. ... 21

Figure 3: Supply and Demand model. ... 23

Figure 4:Diagram showing the various venues the tobacco industry lobbies. ... 33

Figure 5: Timeline of the Tobacco Products Directive. ... 47

Figure 6: February-July 2013 timeline of Parliamentary Committee Decision-Making Process relating to the TPD. ... 52

Figure 7: Pie Chart showing stakeholders meetings with IMCO and ENVI. ... 53

Figure 8: Actor position scale relating to TPD labelling and packaging requirements. ... 55

Figure 9: Bar chart Committee support for pictorial and text health warnings. ... 57

Figure 10: Actor position scale relating to TPD Nicotine Concentration(mg) for NCP's. ... 59

Figure 11: Pie Chart demonstrating Parliamentary Committees position relating to E-Cigarette Classification... 61

Figure 12: Pie Chart demonstrating Parliamentary Committees position relating to menthol regulation. ... 63

Figure 13: Actor position scale for all TPD stakeholders. ... 65

Table 1: Demand of resource good. ... 24

Table 2: Access good provided by organization. ... 25

Table 3: Types of Frames. ... 28

Table 4: Factiva search results ... 37

Table 5: Major differences and similarities between Commission proposal & final TPD outcome. ... 48

Table 7: Position of EU venues relating to regulations of ingredients ... 62

Table 9: Case study protocol for Collecting Data (H1) ... 80

Table 10: Case Study Protocol for Analyzing data (H1) ... 80

Table 11: Case study Protocol for Collecting Data (H2) ... 81

Table 12: Case Study Protocol for Analyzing data (H2) ... 81

Table 13: Operationalization of the Traditional Tobacco Lobbying arguments ... 81

Table 14: Operationalization of Harm Reduction arguments ... 84

Table 15: Stakeholder response to consultation ... 85

(6)

6

List of Abbreviations:

ACF Advocacy Coalition Framework

AGRI Agriculture and Rural Development

CoR Committee of the Regions

ENVI Environment, Public Health and Food Safety

EC European Commission

EESC European Economic and Social Committee

EP European Parliament

EPHA European Public Health Alliance

EU European Union

EK Expert Knowledge

FCTC Framework Convention on Tobacco Control

IA Impact Assessment

IAB Impact Assessment Board

IASG Impact Assessment Steering Group

ITRE Industry, Research and Energy

IDEI Information about the Domestic Encompassing

Interest

IEEI Information about the European Encompassing

Interest

IMCO Internal Market and Consumer Protection

INTA International Trade

JURI Legal Affairs

MEP Member of Parliament

NCP Nicotine Containing Products

NGO Non-Governmental Organizations

RYO Roll-Your-Own

SCENIHR Scientific Committee on Emerging and Newly

Identified Health Risks

SFP Smoke Free Partnership

TPD Tobacco Products Directive

WHO World Health Organization

WTO TRIPS World Trade Organization Agreement on Trade

(7)

7

1.0-Introduction and Research Question:

Tobacco consumption is a serious public health concern as it is the single most preventable cause of death and disease (WHO, 2019). The groundbreaking 1964 report from the US Surgeon General, Luther Terry, concluded that “cigarette smoking is a health hazard of sufficient

importance in the United States to warrant appropriate remedial action.” (Mendes, 2014). This report was the catalyst for tobacco control through regulatory action yet tobacco consumption has remained a worldwide issue. State of the art scientific evidence consistently links smoking tobacco to cardiovascular, respiratory disease, coronary heart disease and various types of cancer yet there are still 1.1 billion smokers as of 2015 (WHO, 2019). In Europe, for example, tobacco consumption is some of the highest in the world with about 28% of adults smoking, with the rate of smoking amongst the highest for adolescents (WHO, 2019). As a result, tobacco consumption is responsible for the death of 8 million people each year and is considered to be one of the biggest preventable public health concerns in the world (WHO, 2019).

In response to the growing worldwide public health concern, the World Health Organization (WHO) in 2005 created the Framework Convention on Tobacco Control (FCTC). The FCTC currently has 168 signatories and is the largest evidence-based treaty concerning tobacco with a goal to reduce worldwide tobacco consumption through substantial regulatory action. Despite global efforts and scientific knowledge tobacco policies are slow to change, one explanation is the tobacco industry’s incredible lobbying power. The tobacco industry is reputed for denying the scientific evidence behind tobacco consumption and uses their extensive resources in order to influence/ curb legislation at the local, national and international level. For example, Philip

(8)

8 Morris attempted to sue the Australian government due to a 2012 law stating that cigarettes be sold in generic packaging and with graphic health warnings.

In regards to the European Union (EU), the first EU tobacco regulations began in the late 1980’s banning tobacco advertisement and sponsorship on television. These initial regulatory actions have since expanded to include tax measures, smoke free environments and regulations on tobacco ingredients. The EU directive that addresses tobacco regulations is known as the Tobacco Products Directive (TPD) and from 2001 until 2014 the policy had remained unchanged. The FCTC and the EU being a signatory to the treaty formally began the review process in 2010 with the intention of repealing TPD 2001/ 37/ EC. The entire legislative process took over 5 years and led to significant regulatory change in public health (see figure 1).

Despite the knowledge regarding the dangers of tobacco, the status quo is a powerful point of reference and moving away from this “conventional wisdom” is quite difficult. Those defending the status quo will mobilize in order to defend their interest. To understand the reform process and the life of a policy, scholars have focused on lobbying strategies, lobbying coalitions,

institutional factors and framing as unit of analysis (Baumgartner & Jones, 2009; Bouwen, 2004; Klüver, Braun & Beyers, 2015; Sabatier, 2009). Scholars such as Baumgartner and Jones (2009) focus on the stability of the policy through framing strategies, therefore, the policy image. The manner in which the policy image is understood within the policy environment (i.e. interest groups and decision makers) is how the issue will be regulated. The status quo is maintained by controlling the policy image. The status quo plays a role in the policy image as this is how the policy develops overtime (Baumgartner, 2013). On the other hand, Sabatier (2009) focuses on the role that advocacy coalitions play in maintaining/ overturning the status quo and their ability

(9)

9 to push their preferences onto the agenda. These theories have been widely applied within the field yet it is unclear what factors single-handedly have the biggest effect (Baumgartner & Jones, 2009; Sabatier, 2009). When studying the reform process, it is important to take into

consideration the context and the interaction between stakeholders and institutions. In order to understand the reform process, this study will draw upon literature relating to the policy image and the theory of access. The policy image focuses on the importance of ideas and how the status quo develops as a result of the decision-making process meanwhile, the theory of access is a framework that explains the access that interest groups experience as a result of their relationship with the institutions (Baumgartner, 2013; Bouwen, 2004). In addition, this study will focus on the effects of interest group framing strategies on the policy making process and the policy image. The theory of access and strategic framing help shed light on the stickiness of the status quo, as each venue is receptive to stakeholder framing strategies as a result of their own

particular interest and background (Baumgartner & Jones, 2009; Bouwen, 2004). The goal of this research is to provide an in-depth analysis of EU tobacco health policy using the TPD

2014/40/EU as a case-study as it is a prime example of the strength of the status quo that is often discussed in Public Administration. This case-study will shed light on the “stickiness” of the status quo using strategic framing and the theory of access as a framework.

(10)

10

Figure 1: Infographic representation of before and after EU rules on Tobacco Products1

Previous studies done on the TPD have demonstrated that despite heavy presence from the tobacco industry, they were largely unsuccessful (Hiilamo & Glantz, 2015; Peeters, Costa, Stuckler, McKee & Gilmore, 2015). Even though the industry was largely unsuccessful, there was a significant departure from the European Commission (EC) proposal to the final 2014 outcome(Peeters et al., 2015). The question then becomes why did the process take so long to

1 Commission, European. (2016) EU Rules on Tobacco Products. Retrieved from https://ec.europa.eu/health/sites/health/files/tobacco/docs/tobacco_infograph2_en.pdf

(11)

11 implement? And why was there a significant departure from the Commission proposal to the final act? Despite public knowledge regarding the effects of tobacco consumption and intense public health initiatives on a worldwide level and European level, why does reform prove to be so difficult? In order to study the outcome of the TPD, this research will draw upon interest group literature, the theory of access, the punctuated equilibrium framework and strategic framing in order to explain the result. The research design of this study will be based on document analysis, media analysis and process tracing as the “deductive theory-testing side of process tracing examines the observable implications of hypothesized causal mechanisms within a case to test whether a theory on these mechanisms explains the case” (Bennett & Checkel, 2015, p. 8). The value added of this study to the literature will be in its ability to outline the major developments of the TPD but more importantly shed light on the nature of the status quo in relation to tobacco health policy within the EU decision-making process. Past research such as Hiilamo and Glantz (2015) and Peeters et al. (2015), mainly focus on the EC proposal and the consultation period. These studies are valuable to the topic of tobacco lobby influence in the EU context but are limited in their scope and only provide a shallow understanding of framing effects on the EC proposal. The main goal of this thesis will be to understand the development of the policy image by analyzing stakeholder strategic framing and drawing upon the theory of access in order to provide a more complete understanding of the final 2014 TPD outcome. Thus, the research question is as follows:

What are the effects of strategic framing by stakeholders on the decision-making process in regards to the Tobacco Products Directive 2014/40/EU?

(12)

12 The study intends to answer this research question by developing testable hypotheses based on the theoretical concepts by reviewing relevant literature on the role of interest group lobbying strategies, the policy image, the theory of access and framing effects. The third section explains the methodology including a research design, data selection, collection and analysis followed by the operationalization of the major concepts. The fourth section outlines the major findings and discussion of the general results. The fifth section ends with the conclusions, reflections and limitations and lastly with the implications and future research.

2.0- Literature Review and Conceptual Framework on the Decision-Making

Process:

This chapter will analyze the state of art literature to understand the reform process of a policy. Four main aspects relating to the research problem will be discussed: the literature on interest groups lobbying, the status quo and the impact on the policy image, institutional

behaviors in regards to the decision-making process and lastly framing effects. The first section will begin by discussing the wider literature on interest groups. The second part will discuss the literature on the institutional factors that play a major role in the decision-making process such as the status quo and the policy image. The third section will develop the theoretical framework by further examining the policy image and the relationship between stakeholders and institutions. Three explanatory factors will be discussed: venue receptiveness, framing strategies and media framing strategies. The purpose of this chapter is to develop testable hypotheses for the main research question and to identify the relevant works of literature on providing insights about the

(13)

13 possible effects of institutional factors on interest group lobbying strategies and ultimately the decision-making process.

2.1- Literature on Interest groups:

Various terms are used in the literature to refer to groups that have an interest in the policy formulation. This study will follow the definition put forward by Beyers, Eising and Maloney (2008) when discussing interest groups. The authors divide interest groups into three main characteristics: Organization, Political interest and Private status (Beyers, Eising, & Maloney, 2008; Klüver, 2011). The concept of organization refers to the nature of the group rather than referring to a broad social movement or social group (Beyers et al., 2008). Political interest refers to a group that is interested in influence and shaping the policy outcome and will be driven by this intention (Beyers et al., 2008). Lastly, private status which refers to the fact that the group is not political in that it is not seeking office and it is neither a private institution (Beyers et al., 2008). The authors further divide this into associations and companies, associations are based on membership meanwhile companies are not (Beyers et al., 2008). This definition by Beyers et al. (2008) is specific to groups that are pursuing a particular interest; therefore, this study will also use the term stakeholder as it refers to a wider group of companies, organizations, agencies and individuals. The term stakeholder will be used interchangeably throughout the research as it is all encompassing term to refer to an actor that has a stake in the issue.

One key question in Public administration is understanding who is influential and what is their success rate. Many studies have attempted to answer this question citing various

(14)

14 institutional factors such as: scope of conflict, salience, media coverage, access, lobbying

coalitions etc… (Baumgartner et al., 2009; Boräng & Naurin, 2015; Bouwen, 2004; Klüver et al., 2015; Sabatier, 2009). Furthermore, scholars often refrain from using the word impact or

influence as it implies there is a direct clear causal link between lobbying activities and outcome.

Certain scholars prefer to use terms such as success or frame congruence (Boräng & Naurin, 2015; Klüver, 2013; Kurzer & Cooper, 2013). Yet, even with the vast amount of case studies scholars are unable to say with certainty what factors make interest groups successful (De Bruycker, 2017). However, despite all the ambiguity scholars either look at broader legislative proposals (Baumgartner et al., 2009; Bernhagen, Dür, & Marshall, 2015; Klüver, Mahoney & Opper, 2015) or particular case studies (Baumgartner, de Boef & Boydstun, 2008; Kurzer & Cooper, 2013). The benefit of studying multiple policies is to see their development through time, meanwhile focusing on particular cases allows researchers to study the in-depth factors that may have had an impact on interest group success rate. The major take away from the literature is that interest groups’ end-game is align the status quo with their preferences. Research on interest group lobbying strategies is too vast to make a definitive conclusion on who wins and who loses in the policy-making game. However, the following sections will discuss some of the factors that scholars believe play a crucial role when discussing the policy making and interest groups success rate.

Within a policy environment there are multiple actors attempting to influence the policy debate and can be divided into two separate camps: diffuse vs. economic interests. Scholars highlight that in order to understand which side is more successful, it is important to take into consideration the contextual characteristics (Klüver, 2013; Klüver et al., 2015a; Kurzer &

(15)

15 Cooper, 2013). Context is important as it provides a better understanding as to how the policy environment functions between actors and institutions. Within Public Administration research scholars are often attempting to definitively state which interest group is in a position of power. However, a lack of contextual understanding provides a shallow understanding of the policy process. Despite these criticisms, Baumgartner et al. (2009) argue that there are essentially two sides to the policy debate, one side is trying to keep the status quo, meanwhile, the other side is pushing for policy change. Certain scholars argue that when business is unopposed then they are more likely to have an advantage in the policy formulation stage (Hojnacki et al., 2015;

Bernhagen et al., 2015; Klüver, 2011). However, this is not always the case and business even if unopposed is likely to see less frame congruence (Baumgartner et al., 2009, p. 213). Bernhagen, Dür and Marshall (2015) state that business actors are usually on the defensive and do not see success as they are trying to keep the status quo or are trying to limit additional regulations. As can be seen, the literature is quite varied, Kurzer and Cooper (2013), on the other hand, use their EU food labelling case to make a definitive argument on a particular piece of legislation in order to speak to the wider phenomenon of interest group lobbying success. The authors state that there are three main aspects that lead to success: framing opportunities, strength of lobbying coalition and the identity of the rapporteur (Kurzer & Cooper, 2013). The EU food labelling case is an inspiration to this particular study as it analyzes context to understand the political forces at play within the EU decision-making process (Kurzer & Cooper, 2013). How the status quo develops through time is important to understanding the winners and losers relating to a policy.

(16)

16 As stated, it is important to take into consideration the context of a policy as it allows researchers to understand the variables at play. One way to understand the context of the policy is through the policy image. The punctuated equilibria theory developed by Baumgartner and Jones (2009) addresses the stasis and the change that is often associated with policy-making. The punctuated equilibria framework differs from the advocacy coalition framework, which will be discussed later, as it encompasses both stability and change within its framework (Sabatier, 2009). According to Baumgartner and Jones (2009) research, most policies remain stable and it is very rare to see drastic changes unless affected by exogenous factors such as a crisis which usually uproots the status quo as a result of policy failure. The status quo is a very powerful as this is how an issue is defined (policy image) and is generally thought of as the best point of reference, therefore deviating from this point is quite difficult (Baumgartner, 2013). When studying the status quo, the main unit of analysis is the policy image and the effect of ideas on the status quo may vary depending on the venue (Baumgartner, 2013; Daviter, 2007; Klüver et al., 2015b). Thus, studying the policy image is studying the process of ideas through the venues. A policy in one venue can be perceived as an agricultural issue but, in another venue, may be perceived as a trade issue (Boräng & Naurin, 2015). How the status quo is defined within an environment may determine the subsequent policy choices (Daviter, 2007; Klüver et al., 2015b). The goals and preferences of a venue will differ compared to another as each change in venue is an opportunity to change the rhetoric surrounding a policy which will ultimately affect its

development overtime (Baumgartner & Jones, 2009). Interest groups are political strategic actors and as a result, will venue shop in search of a venue that is more receptive to their message (Baumgartner & Jones, 2009; Daviter, 2007). The punctuated equilibria framework highlights

(17)

17 the importance of framing and its potential to be used as a tool to maintain the status quo

(Baumgartner et al., 2008). The dominant frame is the status quo as this is how the policy image is defined. Status quo defenders will push hard and will argue against any changes stating that deviating from the norm can be quite risky (Baumgartner, 2013). Although the status quo is a powerful point of reference, it can be discredited as a result of various factors, which provides a window of opportunity that allows for changes in the policy (Baumgartner, 2013). Most

punctures in the equilibria are a result of new knowledge entering the policy environment which shift the status quo (Baumgartner, 2013; Baumgartner et al., 2009). Multiple studies have been conducted measuring the policy image, for example, a study conducted on the death penalty in the US demonstrated that the robustness of the policy was due to the dominant frame

(Baumgartner et al., 2008). The power of rhetoric as well as incremental changes were

instrumental in reframing the issue which led to a major shift in regards to the US death penalty policy (Baumgartner et al., 2008). One way to measure the policy image is through frame

congruence which can be defined as: “the degree to which lobbyists preferred frames correspond to the perceptions of the policy proposals among policymakers” (Boräng & Naurin, 2015, p. 501). The EU food labelling case demonstrated that the Commission’s proposal was more in line with the NGO’s frames however, after the legislative process, the issue had seen increased congruence with the food industry as a result of venue change (Kurzer & Cooper, 2013).

2.2.1- Alternative explanations affecting the status quo:

The policy image and the strength of the status quo has an immense effect on the development of the policy. However, this would only provide one facet of the policy. Scholars have outlined various other factors that affect the policy development such as: salience, media

(18)

18 coverage, scope of conflict, interest group resources and policy type (Bernhagen et al., 2015; Boräng & Naurin, 2015; Klüver, 2013; Yackee, 2006). Public Administration scholars often cite salience as affecting the policy-making process and is defined as the amount of attention a policy will receive from stakeholders, the media and the public (Klüver et al.,2015a). The salience of a policy is quite important because the more salient a policy is, the more scrutinized the policy will become by the public and by invested stakeholders (Klüver et al., 2015a). A highly partisan issue will likely involve many stakeholders which could play a part in the delay of the policy due to the high stakes. Salience is often accompanied by intense media coverage which can generate increased public attention and adds an extra dimension of salience to the policy (Boräng & Naurin, 2015). Interest groups use the media as an indirect lobbying tool and is often exploited in order to promote their interest in relation to the policy (Boräng & Naurin, 2015). However, media coverage is dependent on the lobbying groups actions and if their frame is accepted by the media (De Bruycker & Beyers, 2015, cited in Boräng and Naurin, 2015). Of course, salience can be determined by various other factors such as the policy type and the characteristics of the policy will determine lobbying strategies, the resources that are mobilized and the potential influences from various interests' groups in shaping the outcome of the policy (Klüver et al., 2015a).

Scholars also refer to the scope of the conflict which differs from salience in that it refers to actors who are normally outsiders joining in on the conversation as it is affecting their

interests, and the participation of these outsiders expands the scope of the conflict (Boräng & Naurin, 2015, Klüver et al., 2015a; Yackee, 2006). This expansion ultimately affects the equilibrium as it is bringing new players. Based on the policy type this may lead to major

(19)

19 opposition and divide interest groups into separate coalition groups (Klüver et al., 2015a). A policy may be salient but may not face major opposition within the policy environment (Klüver et al., 2015a). Fighting amongst lobbying camps can lead to less overall success as there is no unified front ultimately leading to compromises and no clear indication of a winning side (Klüver et al., 2015a). Another factor that is discussed in the literature is the resources of an interest group. The European Union has multiple venues for lobbying opportunities; therefore, resources can translate into greater access and more lobbyist to push their interest to EU policy makers (Hojnacki et al. 2015; Klüver, 2013).

To sum up this section on the literature regarding interest groups, it is clear that

endogenous and exogenous factors all work together to determine the outcome of the policy. The policy image is a process that is determined by its contextual factors. This section outlines the major factors that come into play when discussing the status quo. Klüver et al. (2015a) state that in order to properly understand interest group lobbying strategies it is not possible to do it without understanding the policy environment and the context which determine the extent of these explanatory factors.

2.3-Theoretical Framework: The Advocacy Coalition framework, The Theory of Access, Framing:

2.3.1- The Advocacy Coalition Framework & The Theory of Access:

The previous section outlined the major theories and ideas relating to how a policy evolves. The policy image as a concept is quite broad and relates generally to the overall development of a

(20)

20 policy. Various factors were discussed in the previous section, one common belief is venue receptiveness plays a major role in the development of the policy image. Measuring the degree of venue receptiveness enables a further understanding of the interplay between stakeholders and institutions. This next section will discuss the major elements relating to measuring venue receptiveness and develop a theoretical framework in order to further understand the processes relating to the status quo.

In 1988, Paul Sabatier (2009) developed the Advocacy Coalition Framework (ACF) which is one of the most applied theories in Public Administration. The Advocacy Coalition Framework can best be seen through the diagram (see figure 2).

(21)

21 Figure 2: Flow Diagram of the Advocacy Coalition Framework.2

The ACF has five main assumptions:

▫ A central role of scientific and technical information in policy processes ▫ A time perspective of 10 years or more to understand policy change ▫ Policy subsystems as the primary unit of analysis

▫ A broad set of subsystem actors that not only include the traditional actors but all levels of government

▫ Perspective that policies and programs are best thought of as translations of beliefs (Sabatier & Jenkins, 1999, cited in Sabatier, 2009)

The ACF focuses on the power of lobbying coalitions and their relationship with policy brokers and highlights the interaction between the subsystem and the broader political

movements (Weible et al., 2009). One assumption that the ACF has, that is relevant to this study, is that the status quo is determined by the dominant lobbying coalition which plays a major role in the outcome of the policy (Sabatier, 2009). Lobbying coalitions share a set of values that drive their interest and preferences and not being part of the dominant coalition could be detrimental to the interest groups preferences (Klüver, 2011; Kurzer & Cooper, 2013; Sabatier, 2009). The ACF is similar to the punctuated equilibrium theory in that they both focus on the status quo defender however they differ in certain respects. The ACF highlights the importance of the dominant lobbying coalition, therefore, the power of the coalition, in its ability to maintain or shift the status quo (Sabatier, 2009). Meanwhile, the punctuated equilibrium theory focuses

2 Weible, M. C., Sabatier, A. S., & McQueen, K. (2009). Themes and Variations: Taking Stock of the Advocacy Coalition Framework. Policy Studies Journal, 37 (1), 121-140. Doi: 10.1111/j.1541-0072.2008.00299.x

(22)

22 mainly on the status quo defender's ability to maintain the policy image which determines the outcome (Baumgartner & Jones, 2009). This research will heavily draw on certain assumptions laid out by the ACF such as the role of information, the system of actors within the environment and on the belief of policies and programs. The literature review has already discussed system of actors within a policy environment and to some extent on the belief of policies and programs. The following section will further develop the role of information/ access within institutions. This research will not exclusively rely on the ACF as it requires studying policies based on a 10-year period, which is not the intention of this study.

The ACF heavily discusses the relationship between coalition groups and the resources they provide institutions which prove critical to its political success (Sabatier, 2009). This concept was first put forward by Pfeffer and Salancik (2003) who discuss the resource

dependency model. The resource dependency theory states that organizations are dependent on their environment for survival (Pfeffer & Salancik, 2003). Heike Klüver (2013) draws upon the concepts put forward by the ACF and the resource dependency theory but focuses on interest groups rather than lobbying coalitions and develops an exchange of goods model between interest groups and the European Commission to showcase the interaction between the two (Pfeffer & Salancik, 2003; Sabatier 2009). The assumption is that institutions have a task of creating sound policies however, they are merely policy makers and do not have the ability/ information to create such policies. Therefore, institutions will interact with stakeholders based on an information supply chain (Klüver, 2013). Klüver (2013) focuses on information supply which focuses on stakeholder’s ability to provide information to an institution. However, Pieter

(23)

23 Bouwen (2004) focuses on access, therefore, the extent at which stakeholders are able to gain access to EU institutions.

Figure 3: Supply and Demand model.3

The theory of access developed by Pieter Bouwen (2002) seeks to understand the behavior of institutions vis-à-vis business interest. Bouwen (2002) analyzes interest groups access to EU institutions rather than attempting to measure influence mainly due to the constraints that scholars have had in the past, which has been discussed previously. Bouwen (2002) clarifies that access does not necessarily translate into influence, however, many scholars state that access to EU policymakers is one of the most important avenues towards influencing public policy (Beyers & Braun, 2014; Baumgartner et al., 2009). The main concept underlying the theory of access is the relationship between the EU and interest groups which is based on an exchange relation which draws upon the supply and demand model and the resource dependency model (See figure 3) (Beyers & Braun, 2014; Bouwen, 2002; Pfeffer & Salancik, 2003; Klüver,

3 Bouwen, P. (2002) Corporate Lobbying in the European Union: the logic of access. Journal of European Public Policy, 9(3), 365-390. Doi:

(24)

24 2013). Pfeffer and Salancik (2003) argue that each organization is dependent on another for their resources and must operate in this manner in order to survive. Bouwen takes the

resource-dependency theory, the supply and demand model to develop his theory of access by placing it in an EU context. The EU demands certain goods which he has classified into three different types (Bouwen, 2004):

Venue Demand of Resource Description

European Commission

Expert Knowledge (EK) The expertise and

technical know-how

required from the private sector to understand the market

European Parliament

Information about the European Encompassing Interest (IEEI)

The IEEI relates to the aggregated needs and interests of a sector in the EU internal market

Council of

Ministers

Information about the domestic Encompassing Interest (IDEI)

The IDEI relates to the aggregated needs and interests of a sector in the domestic market

Table 1: Demand of resource good.4

Bouwen (2004) argues that in order to gain access to an EU institution, the actor must provide goods which will then grant the actor access to the related institution. Bouwen (2002) further divides the actors into the supplies that they are able to provide:

(25)

25

Organization Best provided access good Individual Firm Expert Knowledge (EK)

European Association Information about the European Encompassing Interest (IEEI)

National Association Information about the Domestic Encompassing Interest (IDEI)

Consultant Expert Knowledge (EK)

Table 2: Access good provided by organization.5

Bouwen (2002) takes Heike Klüver’s argument (2013) further as he takes into consideration all potential lobbying venues within the EU. The conclusion of his research is that the European Parliament and the European Council will interact with a variety of actors but will interact more with European associations' as they provide IEEI meanwhile the Council of Ministers will interact more with national associations’ as they provide more IDEI. To sum up, the EU venue has different functions and will interact with various stakeholders in relation to their function. Which will ultimately translate into access which is based on the degree of access goods that stakeholders provide.

To relate back to the original goal this research, which is to understand why reform takes so long and why policies see relatively small change. The policy image is determined by the strength of the status quo which is affected by the dominance of certain ideas/ frames based on

(26)

26 the policy-making process. Bouwen (2004) and Baumgartner and Jones (2009) are essentially discussing the same concept but through different perspectives. Where Bouwen (2004) is talking about access goods, Baumgartner and Jones (2009) is talking about venue receptiveness. In order to understand the policy image, there is a need to measure stakeholder access which is

determined by the degree of venue receptiveness. The various EU venues provide an opportunity for interest groups to lobby and employ framing strategies based on access but also based on the degree of venue receptiveness which can ultimately translate into a change in the status quo. The interaction between stakeholders and institutions can lead to change in the policy image.

Therefore, driven by the goal of the research and the research question followed by the assumptions outlined in the literature review, we can deduce the following hypothesis:

Hypothesis1: The degree of venue receptiveness to stakeholder framing effects the policy

image.

2.3.2- Framing:

Framing is a lobbying strategy that is used by interest groups and a major factor that is discussed in the ACF and the punctuated equilibrium theory. The theory of access does not fully address the concept of framing as it mostly focuses on the institutional behavior of the European Union. Framing does not make a causal link between influence and policy outcome; however, it is assumed throughout the literature that frames do have an effect on policymakers and thus the policy-outcome (Baumgartner et al., 2009). As discussed earlier linking frames to influence is a difficult task, therefore, this study will focus on frame congruence. The next section will focus

(27)

27 on framing, the different types of frames within the policy environment and lastly media framing strategies.

Framing is a concept that is applied to a variety of fields not just public administration. Defining the concept of framing is difficult however, most scholars refer to the definition put forward by Robert Entman (1993) “to frame is to select some aspects of a perceived reality and

make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/ or treatment recommendation for the item described” (52). Not all messages from interest groups are considered framing as

they may simply be conveying their interest or the group that they are representing (Citation needed). Entman (1993) argues that framing performs any of these four functions: Define problems, diagnoses causes, makes moral judgements and suggest remedies. Entman (1993) provides a strong base for understanding the concept of framing however, Baumgartner and Mahoney (2008) take framing and apply it directly to the life of a policy. The authors argue that there are two levels of framing which are vital to understanding policy issues and its

development through time and space (Baumgartner & Mahoney, 2008). The first level is the individual level- the micro-level, individual frames have a goal of influencing political elites so as to reframe the issue but this is not always the main goal (Baumgartner & Mahoney, 2008). As Baumgartner et al. (2009) argue, it is very difficult to reframe an entire policy debate therefore, individual frames can succeed and have smaller gains. The second level of framing is the collective issue definition, the overall debate and how it develops through time. The collective issue definition is the collective mix of political elites, advocacy coalitions and the media which come together to define the policy (Baumgartner & Mahoney, 2008). Framing is crucial to the

(28)

28 debate because it describes what is at stake within the problem definition and ultimately

influences political choices (Daviter, 2007). As discussed previously, many scholars focus on the degree of interest groups success in relation to policymakers, however, this thesis will focus on frame congruence. The framing strategies of a lobbyist can push the status quo and the legislative process towards the personal preferences of the lobbyist (Daviter, 2007).

2.3.3- Interest groups & Media Frames

According to interest group framing literature, there are two types of frames that actors can employ: institutional frames and policy frames. This can then be further divided into generic and specific frames which creates a fourfold typology (see table 3):

Institutional Policy Generic Relates to general

rule system of the EU

Arguments about cause and effects of the policies

Specific More detailed rule system of the EU political system

Normative, causal or remedial policy aspects

Table 3: Types of Frames.6

This typology outlines the major frames that interest groups employ within the EU which can be employed based on different contextual factors as described in section 2.2.1 (Eising & Rozbicka, 2015). The policy type, the degree of salience can determine the frames that are most present

6 Eising, R. R., D., & Rozbicka, P. (2015). Institutions, policies, and arguments: context and strategy in EU policy framing. Journal of European Public Policy, 22( 4), 516-533. Doi: 10.1080/13501763.2015.1008552

(29)

29 within a policy debate. Furthermore, research on framing makes a further distinction between ideal and self-regarding frames (De Bruycker, 2017; Boräng and Naurin, 2015). Ideal-regarding frames are defined as: frames that refer to the ideals of the speaker for example: environmental protection, safety, etc... Self-regarding frames are defined as those that represent the interest of the group or the group that they are representing (Boräng and Naurin, 2015). In regards to this case study, the macro level will present ideal and self-regarding frames. In order to identify strategic framing within a policy environment there is the deductive and inductive approach. Through the deductive approach, Hiilamo and Glantz (2015) have outlined the main self-regarding frames present throughout the TPD consultation which have been classified into two categories: Traditional tobacco industry lobbying arguments and harm reduction arguments. More details regarding the 7 frames and the inductive approach can be found under section 3. Based on the theoretical assumptions, it is important to identify stakeholder frames and measure the level of frame congruence in order to understand the policy image.

The collective issue cannot be understood without understanding how the media portrays the policy and how this influences the overall policy image (Baumgartner et al., 2009). The media has a wide audience and the manner in which a story is framed affects the audience and how it is perceived (De Vreese & Semetko, 2001). How information is categorized, the words used, the actors present or omitted affects how the policy image is understood among the audience and policy makers (De Vreese & Semetko, 2001). Table 3 outlining the major interest group frames, is a typology that is present in De Vreese and Semetko’s (2001) research on media framing. A policy that is highly salient can lead to intense media coverage which, as discussed previously, can be a tool used by interest groups in order to promote their preferences or used as

(30)

30 a tool to shift the attention to issues that are beneficial to their cause/ interest. Thus, there is a need to measure the degree of stakeholder frames in the media and to measure how the policy image is portrayed within the media as this will provide an indication as to how the policy develops over time. Therefore, based on the research question and the theoretical presumptions as outlined by the literature on interest group framing strategies and media framing studies leads to the development of the following hypothesis:

Hypothesis2: The degree of stakeholder media framing strategies effects the policy image.

3.0- Methodology:

The goal of the research design is to provide a framework to answer the following question:

What are the effects of strategic framing by stakeholders on the decision-making process in regards to the Tobacco Products Directive 2014/40/EU?

Furthermore, the goal of this research is answering the following sub-questions: Why did the process take so long to implement? Why does reform prove to be so difficult in relation to the TPD? The previous chapter outlined the state-of-the art literature relating to interest groups framing strategies, the policy image, the theory of access. The main theoretical assumptions were outlined and led to the formulation of the two main hypothesis which will direct the structure of this research. In order to answer the research questions, the hypotheses will be tested using the TPD as a typical case design. The first part of the chapter will outline the research design, the

(31)

31 case selection and an explanation on the data collection and analysis followed by the

operationalization. 3.1- Research Design:

The main goal of this research is to analyze major patterns to determine stakeholder framing strategies and venue receptiveness to the policy outcome. This study follows a single case research design as the main explanatory variables at stake lie within the impact of framing attempts by various stakeholders and congruence with the final outcome. Toshkov states: “Single-case studies examine multiple pieces of evidence about a single unit. The analysis is within rather than across cases. Instead of measuring few variables for many cases, we make many observations related to a single case.” (2016, p.285). This study is X1/Y centered as it will

be investigating a particular outcome, thus connecting a particular cause with a particular outcome (Gerring, 2007). This research will employ a typical case which serves as an

exploratory role (Gerring, 2007). A typical case is used as it probes for causal relationships and is most useful for theory testing which is used to confirm or disconfirm the generated

hypothesis(Gerring, 2007).

The first part of the research design justifies the case selection due to the high level of interest groups and because of the final outcome. Followed by a description of the data selection method, which relies on official documents, stakeholder consultations and on published media articles, to uncover stakeholder framing strategies, the degree of venue receptiveness and media framing strategies to understand the development of the TPD policy outcome.

(32)

32

3.1.1- Case selection and justification:

In order to answer the research question and the test the major hypotheses based on the theoretical presumptions a typical case will be selected. In order to select a case, the case must be related to a supranational tobacco regulation in which the status quo has remained quite prevalent using the process as a case. The TPD 2014/40/EU was chosen as the most representative of this phenomenon and of the major concepts discussed within the literature review.

Interest groups will fight to retain the status quo in order to maintain their interest. Part of their strategy is to gain access to political venues and draw attention to an issue that aligns with the stakeholder’s preference. Gaining access to venues can be seen through the open

consultations, documents relating to stakeholder meetings, committee amendments and media reports. The multi-level governance system of the EU provides multiple opportunities which stakeholders can gain access to and relay their interest/ preferences. Intense lobbying by the tobacco industry has been widely documented. For example, the WHO published a report in 2009 called the Tobacco industry interference with tobacco control in which they state that the industry uses their vast resources to influence tobacco control through direct and indirect lobbying methods (see figure 4).

(33)

33 Figure 4:Diagram showing the various venues the tobacco industry lobbies.7

The WHO report (2009) confirms that any legislation towards tobacco control would include large amount of resources and mobilization in order to curb any further regulations. Although, the FCTC was a major global effort to combat tobacco consumption via regulatory measures, multiple countries around the world and in Europe have already proposed radical public health initiatives. The TPD was chosen specifically because it is a massive public health initiative involving a very profitable industry. Furthermore, the entire legislative process took a total of 5 years and involved many amendments, stakeholder consultations, debates, committee opinions etc... The study done by Costa, Gilmore, Peeters, Mckee and Stuckler (2014)

demonstrated that there was a significant departure from the Commission proposal to the final

7 World Health Organization. (2009). Tobacco industry interference with tobacco control (NLM Classification: HD 9130.6). Geneva: WHO

(34)

34 Act signed in 2014. From the Commission proposal until the final outcome, lobbyist would have had multiple venues at which they could have gained access and lobbied their interests. In addition, the TPD is an interesting case as Europe has amongst the highest levels of tobacco consumption more specifically among adolescents therefore, the industry has a strong monetary power (WHO, 2019). The TPD case speaks to the wider issue at large which is the interaction between large industries and (supra) national institutions in a public health context.

3.2- Method of data selection:

The method of data selection will be threefold: Stakeholder consultation, document analysis and media analysis. The stakeholder consultations provide direct insight into the tobacco lobbyists and the non-profit organizations (NGO’s) preferences and frames. Insight into

stakeholder frames is important to measure the potential effect that stakeholders had on the TPD. Furthermore, understanding the stakeholder frames will provide the tools to measure frame congruence and the development of the policy image. With the document analysis we can trace the decision-making process within and between the venues. The data collected from the

stakeholder consultation and the document analysis will allow the researcher to study the degree of venue receptiveness and frame congruence based on the output rhetoric of the venue

documents. Lastly, the media analysis provides insight into how the media portrayed the issue and which stakeholder frames were more present within the media. This data will provide the information to measure the impact of interest group framing strategies on the policy image and the TPD final outcome.

(35)

35

3.2.1- Selection of empirical material for analysis:

The selection of the empirical material for testing the hypothesis has to match the following characteristics. First the document had to have been released after 2010 as this was when the official review process began for the TPD. Discussions relating to the Security and Traceability features which is part of the TPD were held after the final outcome in 2014, thus documents published after 2014 were not included. Due the nature of the TPD and the EU, it was imperative to find empirical material from a wide variety of sources, this included looking at documents which were produced by the Commission, the Parliament and the Council. After careful research and reviewing, not all documents were deemed relevant, and all documents can be found through the EUR-LEX database, the European Parliament procedural file and the public register of Council documents. Furthermore, empirical material consisted of primary documents from Philip Morris International (one of the biggest tobacco companies in the world) and Smoke Free Partnership News (one of the largest pan-European tobacco control advocacy groups). Searching for documents outside of the obvious sources is part of Bennett and Checkel’s (2015) 8 best practices for process tracing: “Casting the net widely for alternative explanations” (p.108). The result of data collection was 47 total documents all in the English language.

3.2.2- Selection of empirical material for analysis- Textual media sources:

The selection of the empirical material for textual media sources has to match the following characteristics. First, newspaper articles for analysis needed to be between the period of 2010-2014 and had to have been about the TPD. The media analysis was done through

(36)

36

Factiva database, which is a Reuters and Dow Jones joint venture, the database comprises full

text articles of international newspapers and journals. The Factiva database is free of charge and is authorized for Leiden University students. The search criteria for the database include

keywords, sources and timeframe. The keywords used were (“EU” or “European Union”) AND (Law OR Directive) AND (Tobacco OR E-cig* OR Plain Packag* OR Menthol or Cross-border OR traceability) which would provide the most comprehensive search. These keywords were chosen as they relate to the major points of the TPD. Seeing as the unit of analysis is an EU Directive, the focus will be on newspapers that focus on European issues and did not include national newspapers or journals. The following newspapers were included in the search: The Guardian, BBC, Politico Europe, The Financial Times, Agence Europe and the Telegraph. All newspapers were in English. A more extensive search would have included newspapers and journals in other European Union languages however, this was not in the capacity of the

researcher. As discussed previously in the document analysis, the timeframe was set from 2010-2014 to include the most relevant articles and to see the development of interest group framing and media framing overtime. The Factiva search produced the following results:

(37)

37 Table 4: Factiva search results

Although, the results produced a large number of articles, not all articles were relevant. Only articles that were deemed relevant and non-repetitive were included in the final media analysis. This resulted in a total of 69 relevant articles.

3.2.3- Stakeholder consultation:

To properly test H1 and H2 it is necessary to analyze the stakeholder responses from the

European Commission consultation which was held from September 24th 2010 to December 17th

2010. According to the European Commission, they received 85 000 responses via the online form (European Commission Health and Consumers Directorate- General, 2011c). The report states: “no previous consultation launched by the European Commission had ever registered such significant participation” (European Commission Health and Consumers Directorate- General, 2011c, p. 5). The target group of the consultation was citizens, businesses, non-governmental organizations and national authorities. The consultation was done in survey format and allowed for personalized comments, participants were first asked to review a document which outlined extensive details and information regarding each question. Firstly, each respondent was asked to read the report relating to the TPD which identified the six main areas of concern defining the problem (European Commission Health and Consumers Directorate- General, 2010). Secondly, each respondent had to identify themselves with name, country, email and affiliation: citizen, government, NGO or industry. Within each section, there were three questions. The first question asked the respondents to agree or disagree with the problem definition. The second question asked the respondent to choose which policy option they agreed with the most. The

(38)

38 final question provided an “additional information” box which allowed respondents to provide personalized responses relating to the problem definition. Following the end of the consultation, the European Commission published a report regarding the results which proved beneficial for this study as it clearly outlines which stakeholders responded to which questions. Therefore, part of the stakeholder consultation data was already provided for by the Commission report. The results of the data will be further discussed in the analysis section. Due to the high amount of data, it was not in the researcher’s capacity to review all responses. In order to have a more manageable sample size it was decided to focus on the responses exclusively from the industry and the NGO’s, this in itself was a sizeable amount of data with almost 4,000 pages. Many responses were not in English, therefore, the responses only in English were reviewed. All data can be found on the Commission website dedicated to the stakeholder consultation:

https://ec.europa.eu/health/tobacco/consultations/tobacco_cons_01_en. 3.3- Data analysis:

The method of analysis for this research will be content analysis which will analyze the primary documents, the stakeholder responses and the media analysis. The first step will be to identify the stakeholder frames. The second step would be to identify the frames in the

documents and verify frame congruence with stakeholder frames to the government’s behavior and determine level of venue receptiveness. The third step would be to conduct an analysis of the media and verify frame congruence and analyze for the explanatory factors. The focus of this study is on the policy image and will draw upon process tracing in order to capture the

(39)

39 research does not include regression analysis or advanced quantitative research however, does include limited statistical analysis to further strengthen the conclusion.

Seeing as the goal of this research is to understand the decision-making process in

relation to the TPD through interest group framing strategies. Content analysis would be the best research technique to study this phenomenon. Krippendorff (2004) characterizes content analysis as a method of inquiry into symbolic meaning of messages. Content analysis is a scientific research technique that is used to process data into their context, the process must be objective, systematic and more importantly replicable (Krippendorff, 2004). Content analysis can be used when analyzing documents, newspapers, letters, artwork etc... As discussed, this thesis will be looking at the messages of stakeholders and the decisions made by policy makers. The crucial aspect of content analysis is the idea that each message has a purpose that may or may not be obvious to the receiver. Krippendorff (2004) develops his argument by stating that text does not have a single meaning and can be received in multiple ways. A text is a two-way street, the author combines words, sentences and paragraphs that can be interpreted by the reader in another manner unintended by the writer (Krippendorff, 2004). Essentially the message may not be shared by all parties and the content can differ depending on the receivers cultural and sociopolitical background (Krippendorff, 2004). The purpose of content analysis is to yield inferences from verbal, symbolic or communicative data by distilling it into content-related categories (Krippendorff, 1982, cited in Elo & Kyngäs, 2008).

(40)

40 3.4- Data collection:

In order to conduct the research, the first step would be to do an identification of frames. There is the deductive and the inductive approach, both are appropriate for this research. The deductive approach is based on operationalization done by previous studies, theories or models (Elo & Kyngäs, 2008). Pre-defined frames have been outlined in the research study done by Hiilamo and Glantz (2015) which have identified two issue-specific frames: Traditional tobacco lobby arguments and harm reduction arguments. Hiilamo and Glantz (2015) further divide this into 7 frames: Illicit Trade, Employment & Economic growth, EU regulation, denormalization, smokeless tobacco products being less harmful and lastly, informing consumers on health risk levels. In essence, the deductive approach is applying these filters to the empirical material. The second approach, the inductive approach, looks at the specific and then the general (Elo and Kyngäs, 2008). This would capture the frames that may not have been previously identified by previous research. The inductive approach would be based on the definition of frames as outlined by Robert Entman (1993) and Eising and Rozbicka (2015) typology (see literature review).

In order to test H1: The degree of venue receptiveness to stakeholder framing effects the

policy image. H1 requires an identification of industry and NGO frames through the stakeholder

consultations via the inductive and deductive approach. This data would then be used in relation to the venue documents to verify frame congruence and determine the degree of venue

receptiveness to stakeholder framing methods. Pieter Bouwen’s (2004) research on interest group access within the EP states, “it is common wisdom that one of the most important ways for business interest to influence the legislative process is convincing MEPs to amend the text under

(41)

41 review” (p. 482). Bouwen’s (2004) studies access by interviewing MEPs, however, this was not possible for this research. One way to go about measuring venue receptiveness is measuring the number of tabled amendments, measuring congruence with stakeholder frames and lastly, analyzing stakeholder meeting documents with venues. For a detailed description on the data collection and analysis procedure for H1 (see Appendix A).

In order to test H2, the process is slightly different. H2: The degree of interest group

media framing strategies effects the policy image. We gather the data generated from H1 (the

stakeholder frames & the document analysis) and we do a media analysis to inquire as to how the media reported on the TPD throughout the years and identify the presence of stakeholder frames. Followed by this, we draw upon process tracing to see the development of the policy image over time. According to the literature, a policy image is the overall mix of rhetoric to characterize a policy. The media analysis and process tracing of the documents will provide an overall understanding of the outcome variable being the TPD 2014/40/EU. For detailed description on the data collection and analysis procedure for H2 (see Appendix B).

3.5- Operationalization:

In regards to H1, venue receptiveness, stakeholder frames and the policy image need to be operationalized. Due to the multi-level governance system, political decisions go through multiple venues. The European Parliament, for example, has about 20 committees which prepare legislation and provide opinions with the ability to recommend amendments. Due to the lack of interviews, venue receptiveness in this case, relates to the extent which venues are in line with stakeholder frames which will be operationalized through Boräng and Naurin (2015) concept of

(42)

42 frame congruence. Based on Pieter Bouwen’s (2004) research, this would entail measuring the number of tabled amendments from each EU venue and outlining each actors position in order to assess venue receptiveness to certain stakeholder frames. As discussed previously, the deductive approach and the inductive approach will be used to operationalize stakeholder frames. For the deductive approach, the frames have been derived from the research by Hiilamo and Glantz (2015). The researcher categorized two general frames and further divided them into 7 frames. The operationalization of the 7 major frames will be done by searching for key words relating to each major frame as outlined by previous research (Hiilamo & Glantz, 2015). For a more

detailed description on general frame and their associated keywords (see Appendix D). H2, unlike H1, will be focusing on the macro-level. The literature stipulates that to understand the policy image at the macro level it is important to understand the collective issue definition which includes a media analysis. Interest group framing strategies will be

operationalized by identifying stakeholder frames within media reports (industry & NGO). The overall policy image will be operationalized by analyzing the dominant frames present within the media which will provide one aspect of the policy image and identifying the presence of the explanatory factors. For a more detailed description of the operationalization of media frames (see Appendix C).

3.6- Reliability and Validity:

According to Krippendorff (2004), the research design in regards to generating data must perform three main actions: stability, reproducibility and accuracy. Stability refers to the

(43)

43 order to increase reliability. This has been addressed by analyzing the media, documents and stakeholder responses and then returning to the data in order to verify if the same conclusions can be drawn. It would have been better to have had an objective researcher perform a stability test however, this was not in the capacity of this particular research. Therefore, stability cannot be fully addressed. Reproducibility according to Krippendorff (2004), is addressed through “intra-coder” reliability, however, this would require to have several researchers with different socio-cultural backgrounds to analyze the data in the hopes of generating the same results to increase reliability. As I was the only one reviewing the data, this would rule out accuracy. Having additional reviewers would have reduced inconsistencies or disagreements present within the data. Even though, there are no additional reviewers to reproduce the data, it can be

addressed by providing a well-structured and detailed case study protocol and clear instructions on how to gather the data and how to analyze the data generating process. As a result, when it comes to reliability in regards to this study, it would not be possible to rule out subjectivity.

In relation to validity, Krippendorff (2004) outlines three types of validity related to content analysis: face, social and empirical validity. Face validity is simple in that it is the extent to which the research is common sense (Krippendorff, 2004). Social validity relates to the societal contribution of this research, to what extent is this research relevant or beneficial to society or to the academic community at large (Krippendorff, 2004). Face and social validity have been addressed and discussed in the introduction. As for the last concept, empirical validity, Krippendorff (2004) divides this into the subcategory of content validity which is then further divided into two subcategories: semantic and sampling validity. Sampling validity is “the degree to which a population is accurately represented in the sample” (Krippendorff, 2004, p. 321). To