Radical change in natural resource policy in the Netherlands

Radical change in natural resource policy in the Netherlands

Which causal factors account for the decision to stop extraction in the Groningen gas field?

Agustín De Julio Pardo - s2152479 11/01/2019

Master thesis Public Administration – Public Management track Thesis advisor: Dr. N.A.J. van der Zwan

Table of contents

Introduction p. 3

Chapter 1: Theoretical framework p. 6

Chapter 2: Methodology p. 15

Chapter 3: Nature of political contest p. 22

Chapter 4: Expert opinion and policy venues p. 27

Chapter 5: Civil society response p. 30

Chapter 6: Socio-political circumstances p. 36

Chapter 7: Analysis p. 45

Conclusion p. 51

Bibliography p. 54

Introduction

In 1952, an operation to discover oil fields in Harkstede, province of Groningen, accidentally found low-calorie natural gas reservoirs (Whaley, 2009). An increasing number of these were discovered soon thereafter. During that decade, follow-up drillings confirmed an unexpected finding: the separate small gas fields in Groningen were indeed all connected and formed the biggest gas field in Europe, the Groningen field (Whaley, 2009). This gas field has an

extension of 900km2, a volume of an estimated 2800 billion m3 of gas available for

extraction, and is located in a densely populated area in the province (De Waal, Muntendam-Bos, Roest, 2015, 130), (NAM, 2018). The operation and extraction rights were awarded by the state to the Nederlandse Aardolie Maatschappij, NAM, a conglomerate owned in equal parts by Shell and ExxonMobile (Whaley, 2009).

This particular policy has had a number of consequences for the country and for the province. While the gas extraction proved to be a considerable source of income to the Dutch state, and it funded to a great extent the Dutch welfare state (Ibidem), it has also had dire consequences for the inhabitants of the province. Firstly, extraction caused soil subsidence in the province and seismic activity (De Waal et. al, 2015, 131). Both these phenomena were researched a priori and expected by experts. Unforeseen was the acceleration of subsidence from the 1970s onwards and the exponential increase in earthquakes starting from 1991 (Ibidem). The NAM reports a grand total of 1272 induced earthquakes in the province of Groningen from 1991 to December 2018 (NAM, 2018).

As a consequence of the seismic movements in the province of Groningen, buildings and houses have been damaged (De Waal et al, 2015, 130). The ones most severely damaged are regularly bought by the state or the NAM and demolished for security reasons (Het

Verdwenen Groningen, 2018). These buildings are often of cultural or historical significance. Structural damage combined with fear for future earthquakes and a loss of faith in national policy has caused a decrease in property value of 9.5% from 2008 to 2015 alone (Durán, Elhorst, 2018, 19). Social consequences of the extraction policy include a sharp decline in the perceived safety of inhabitants, a sharp increase in physical and mental health issues

(Groninger Perspectief, 2018, 20), and a sharp loss in confidence in the ability of the national government and other political actors to effectively solve the problem (DvhN, 2018).

The government implemented a number of policy solutions in order to alleviate these problems. They include compensation schemes for owners of damaged property and for decrease in value (Durán, Elhorst, 2018, 2) and production caps. Ensuing a heated societal debate, Minister of Economics and Environment Eric Wiebes declared the intention of the Ministry to stop extraction in the year 2030 (Wiebes, 2018). This was done in an official letter to the second chamber of Dutch parliament. This outcome is remarkable for a number of reasons. Firstly, the policy solution of stopping extraction altogether was not in the public discussion nor was it proposed by any stakeholder before. Secondly, the decision came from a government and a party that was seen as reluctant to tackle the policy problem at hand. Lastly, it was not accompanied by a group of policies detailing how the outcome would be achieved. All of these factors are dealt with further in the present study, but motivate the following research question:

Which causal factors account for extractive policy change in the Groningen gas field of the Netherlands?

A study of this phenomenon is relevant for a number of reasons. Firstly, the radical nature of the policy solution proposed by the government and its abruptness makes it a deviant case worthy of study. This will be done through the use of policy change theory. As it will be outlined within the study, the proposed policy solutions by a number of stakeholders such as political parties and expert groups never came close to proposing full cessation. A closer look at the case could shed light on the reasons behind such an extreme solution. This case study presents a combination of factors that led to the 2018 Kamerbrief. By answering this

question, it can be assessed if policy change theory can indeed account for such an outcome, and future research can test if the combination of factors can account for other cases of (radical) policy change. Secondly, the on-going political debate about the issue makes this research relevant. The 2018 Kamerbrief by Minister Wiebes about the decision to stop extraction expresses political will and is an official statement. The specific policy on the ways in which this will be achieved is still being developed. Specifics, such as the

reinforcement policy of private and public properties in the area, or the rate and speed of the decrease in production are being developed and contested in the political arena (RTVNoord, 2018). It is thus, very much, a policy in development. A clearer understanding of the

mechanisms through which such a decision was made could thus aid the drafting of new policy.

Literature on gas extraction in Groningen has mostly focused so far on either technical and geological research, or economic research. As studied in the following chapters, the NAM, KNMI and SodM had, for a prolonged period, a primacy in research on the issue. Topics studied included seismic activity and soil subsidence, among others. Most recent literature studied topics such as the economic impact of extraction on land value, and only very recently, matters of societal perception. This research attempts to contribute to the literature from a public administration perspective by using policy change theory. This includes political and social considerations, areas which were not dealt with in the same of depth as the aforementioned topics in the literature about the Groningen extraction policy.

The present research uses a process tracing methodology to identify a series of causal factors that contributed in varying degrees to the occurrence of the outcome. The research presents a causal configuration of factors that lead to the outcome, not necessarily interacting in a sequential manner. Through the use of policy change theory, expectations are derived, and then empirically tested for their causal inference.

In this manner, the present research is composed of the following chapters: Firstly, an overview of policy change theory is presented, from where expectations are derived which motivate the research. Secondly, the methodology through which the empirical data is obtained is discussed in greater detail. The empirical chapters of this research are composed of four chapters. Firstly, the distribution of costs and benefits of the policy issue are

presented, from which the nature of the policy contest is discussed. Secondly, the effect policy venues and expertise had in destroying the policy monopoly around gas extraction is assessed. Thirdly, the civil response to the policy issue is discussed. An analysis chapter follows, that runs theoretical tests to assess the nature of the contribution of each particular factor to the outcome. Lastly, the policy issue is placed in Downs’ issue-attention cycle, and the effects of focusing events are discussed, in order to assess the socio-political

circumstances in which the outcome occurred. An analysis follows in which the causal importance of factors is discussed.

Chapter 1: Theoretical framework

The present section will outline the main debates in policy change theory in order to derive a number of theoretical expectations to be tested by the ensuing research. Different schools of thought emphasise different mechanisms whereby policy change emerges. Baumgartner and Jones, within their classic theory of punctuated equilibrium, define policy monopolies as a supremacy on ideas and understandings about a policy, established by powerful policy entrepreneurs (2009, 6). Policy change, thus, occurs when these monopolies are challenged by non-status quo policy entrepreneurs who manage to change public opinion and dismantle the established monopolies. This happens when policy images, a set of values and beliefs about a certain policy (Baumgartner and Jones, 1991, 1045), change in the collective imaginary. Mechanisms of positive or negative feedback can either strengthen or jeopardise the existing stable monopolies, raise issues to the arena of macro politics, or keep them safely within specialist policy subsystems (Timmermans and Scholten, 2006, 1107). Shared

amongst the scholarship is the notion that policy exists in long periods of stability and short, violent outbursts of change (Baumgartner and Jones, 1993, 5).

The specific characteristics of the Dutch political system are of particular relevance to this research. Most of the scholarship on policy change was developed in a US-setting, including the work of Baumgartner and Jones, a cornerstone in the literature on which many other scholars built. Although the literature can be used to make valid inferences about policy change around the world, the differences between the political realities in the Netherlands and the US make it necessary to adjust the theory to local realities. The Dutch political system is characterised by its famous poldermodel, a decision making process based on compromise and consensus-building between government, employers and trade unions (Van Dyk, 2006, 409). Lijphart, in his seminal 1968 book Verzuiling, pacificatie en kentering in de Nederlandse politiek, attributes these characteristics to a traditional pillarisation of Dutch society in which separate groups or pillars (28), divided along either religious or social-economic lines, that live to a certain degree independently from each other have to devise appropriate mechanisms to co-govern. The separation of these societal pillars would logically lead to division and difficulty to govern, yet the cooperation between the elites of different

pillars lead to co-governance arrangements based on cooperation and the peaceful resolution of disputes (Idem, 99).

Van Dyk (2006) claims that both scholarship and actors in the field display a broad consensus about the necessity of finding a consensus since the 1982 Treaty of Wassenaar (411). There are concrete institutional expressions of this phenomenon. Firstly, Dutch post-war

governments are invariably coalitions, either broad, minority or narrow based, of two or more parties (Lijphart, 1968, 15). These governments’ policies are informed by coalition deals that define their agendas and constraint their scopes of action (Timmermans, Scholten, 2006, 1108). Secondly, Dutch policy making is defined by an institutionalised corporatism, with formal channels created in order to provide different stakeholders with the ability to participate in the process (Ibidem): who gets a seat at the table becomes of paramount importance in this setting. These differences with the American system thus affect policy making and call for extensions of the theory.

The Issue-Attention cycle

In his seminal 1972 study, Anthony Downs presents the issue-attention cycle. It is composed of five stages which might lead (or might not) to policy change. Within this framework, it is the stage in which a public debate finds itself that determines the likelihood of the demise of a given policy monopoly and the occurrence of policy change.

Firstly, in the pre-problem stage, a disadvantageous social phenomenon exists but it still has not garnered much attention from the public. Some niche expert or pressure groups might already know of the phenomenon at this stage (Downs, 1972, 39). The second stage is that of alarmed discovery and euphoric enthusiasm. In it, because of an event, the public suddenly realises the magnitude of the problem and gears up to tackle the problem, accompanied by a feeling of optimism and a certainty that society can effectively solve the issue at hand (Ibidem). At the third stage, that of realising the costs of significant progress, society in general gradually loses confidence in their ability to successfully solve the policy problem at hand. A better understanding of the now ubiquitous social phenomenon leads to a widespread consciousness about the financial and societal costs that would be encompassed in a potential solution (Idem, 40). The fourth stage is that of gradual decline of intense public interest. The discouragement that comes with the deeper understanding of costs associated to a policy

issue makes societal attention effectively wane. Boredom sets in, and some groups feel threatened by the prospect of being personally disfavoured by solutions (Ibidem). Lastly, in the post-problem stage, the policy issue starts to be replaced by another policy issue and remains in a limbo of sorts, not being fully irrelevant, not being intensely discussed. The main difference between this stage and the very first one is that during the stages in which attention was sharply focused on the issue, institutional arrangements might have been created to deal with it. The formal nature of such arrangements make them persistent and still deal with the policy issue, even when public attention has faded (Idem, 41).

The issues this theoretical model of the issue-attention cycle applies to are problems that affect a minority, benefit a majority and have no intrinsically exciting characteristics that would sustain public interest by themselves (idem, 42). Policy change is thus more likely in the stages of alarmed discovery and realising the costs of significant progress, where policymakers are being scrutinised by society at large, instead of the professionals and experts that deal with the policy issue on a daily basis.

Policy venues

During the different stages of the issue-attention cycle, a variety of actors exert influence on both public opinion and decision-making processes. Their actions are often decisive in

pushing issues to specific stages in the cycle. Timmermans and Scholten (2006) maintain that it is through policy venues – institutional sites where actors pose policy problems and

possible solutions – that agenda setting is influenced, and policy change potentially emerges (1105). In these venues, policies are presented, amended, supported by some and eventually adopted legally (Ibidem). Parliament, the executive branch and the judiciary are all formal examples of policy venues. The media, universities and expert groups could very well be examples of more informal policy venues, given they perform the same function but exist (to varying degrees) independently of the government and have different kinds of access to formal policy making. As these venues are used to either attack or support certain policy monopolies, policy entrepreneurs strategically attempt to control them.

Timmermans places special emphasis on one kind of policy venue: the scientific, where expert knowledge is traded and informs and motivates policymakers at the time of drafting policy (2006, 1006). The output produced by these venues is of special importance given the

knowledge and expertise these actors provide, which is simply unavailable elsewhere. In the most practical conception of the use of scientific knowledge by policymakers, problems are identified first, then scientific knowledge is produced/ Issues are re-defined in light of the newly produced knowledge and policymakers take decisions accordingly (Weingart, 1999, 154). Scientific knowledge can have various impacts: it can depoliticise an issue and keep it in specialist policy subsystems, thus protecting and legitimising the status quo. Alternatively, scientific policy venues can also be utilised by pro-change policy entrepreneurs who find the venue’s positions advantageous to initiate positive feedback processes and instigate policy change (Timmermans, 2006, 1107).

A shift in power between policy venues, either by the appearance of new venues, or by changing positions in venue’s existing policy images are thus fertile ground for policy change. Politicians and other policy entrepreneurs will actively seek for venues that back their positions in order to legitimise them. At the same time, entrepreneurs who find themselves in disadvantageous venues, where they have difficulty advancing their interests and opinions, will seek for and flock to other venues that are more amenable and will present less resistance to their agendas (Kaunert and Leonard, 2012, 1397).

Two factors determine greatly the rate of success of these policy entrepreneurs: the institutional framework within which they operate and their individual strategies (Baumgartner and Jones, 1991, 1045).

The relationship between expertise and policy-making in the Netherlands has some particular characteristics due to its political system and the way research is organised and funded. The Dutch corporatist model ensures participation and influence from a selected number of stakeholders, effectively forming an elite with primacy in the policy cycle (Timmermans, 2006, 1108). Research and policy-relevant knowledge creation is also highly institutionalised and often funded by the Dutch state (Ibidem). This characteristic could potentially influence the independence and the quality of the research produced by institutions, but at the time ensures the production of knowledge and indicates that their research is in fact used in the process – it is unlikely that a government would fund these activities in order to ignore their findings.

Both Baumgartner and Jones, and Timmermans lay emphasis on the rationally bounded actions of policy entrepreneurs themselves. While Timmermans sees actors as rational units attempting to generate the biggest impact possible through selecting venues, Baumgartner and Jones zoom in on their possible courses of action regardless of the venue. These include the use of symbols, rhetoric and policy analysis (Baumgartner and Jones, 1991, 1045). Especially interesting from a theoretical point of view are the former two, given their presence in the case. Symbols are representation of greater concepts, highly dependent on personal interpretation (Stone, 1988, 157). Effective symbols are rooted in selfhood, non-political in nature and bind the non-political with the identitary, albeit subconsciously (Cohen, 1979, 87). Likewise, rhetoric and narrative stories are the means through which entrepreneurs attempt to protect or dismantle policy monopolies (Stone, 1988, 158). The effective use of the latter by entrepreneurs thus is of vital importance for policy change. The action of informal policy entrepreneurs that are exogenous to formal policy makers is of particular interest given through their actions they manage to influence public opinion and the political process. The aforementioned Dutch corporatist system has implications for individual policy

entrepreneur action. As stated above, selected stakeholders (traditionally unions, employers and government but not exclusively) are formally involved in decision making. The inclusion or exclusion of actors in the process is thus vital to determine their influence. It also means that individual entrepreneurs that are excluded from the formal process can either attempt to join the favoured groups included in it, or find non-institutional venues and strategies to still be able to influence policy. There are multiple examples of the latter kind that will be studied in the following chapters.

Focusing events

The issue-attention cycle sheds light on the likelihood of change given the circumstances in which a policy contest exists, but does not explain fully why certain policy changes happen at a given time. Within policy change theory, a number of scholars have studied and

emphasised the importance of focusing events in agenda setting and change. These are sudden, unexpected, uncommon and (potentially) harmful phenomena that affect a defined community and are gotten to be known to both policy makers and the wider society at the same time (Birkland, 1998, 54). These sudden, dramatic events give actors the possibility of changing radically the policy image held by society at large, and thus further explain the

timing of policy change (Jensen, 2011, 145). As opposed to long-term problematics, focusing events have the ability to push issues to the agenda immediately, and amplify the effects of critical voices (idem, 62).

Focusing events have a clear relationship with the stages of the cycle. Connecting this

concept to the issue-attention cycle can mean that focusing events can prompt dormant issues that are in the post-problem stage and launch them right to the centre of public discussion, that is, an earlier stage in the cycle. In agreement with Downs, focusing events are also phenomena of concentrated, visible harm to a limited geographical area or community (Birkland, 1998, 54). These events galvanise the population and prompt group responses to apparent policy failures (Idem, 55). Contests between pro-change and pro-status quo actors will ensue, each trying to co-opt strategic policy venues from which agenda setting can be influenced. Should there be an organised pro-change community prior to the focusing event, it would manage to capitalise on it to a higher degree (Idem, 72).

A contemporary addition to the scholarship on focusing events is its application to natural disasters, terrorist attacks and other kinds of dramatic focusing events. These focusing events encourage actors (especially politicians) to get involved in the public arena on issues they might had not been involved in before (Jensen, 2011, 145). This creates heightened pressure on policy monopolists, who are often seen by the wider public as the very root of the problem (Ibidem). This very influx of actors, with relatively less ties to existing power structures, less responsibilities, and who are completely external to the existent policy monopolies lead more often than not to radical policy change, in a response to the dramatic circumstances (Ibidem). Birkland (1997) claims that certain characteristics of natural disasters prompt higher political responses (53). The more concentrated the damage is, the more graphic it is and thus the easier it can be visually communicated to the public, and the deadlier a natural disaster is, the more likely the underlying policy goes to the mainstream political arena (Ibidem).

The findings posed by Birkland are of interest given they outline the relationship between politics and natural disasters, yet his study is about tectonic movements rather than induced earthquakes, the object of this present study. The earthquakes discussed in this paper are direct consequences of policy making rather than chance or exogenous forces. They happen in a politicised setting, with actors eager both to stop and continue the activities causing the earthquakes, making it a theoretically interesting case. It conforms a deviation from the norm,

as natural disasters often have a less direct relationship with policy making and politics in general.

Nature of political contest

Another part of the scholarship zooms out from individual action and focusing events and study the contingent characteristics of policies and their likelihood to bring about change. Baumgartner and Jones (1993) emphasise the importance of the nature of political contests in order to determine if an issue will become salient societally and to identify what the possible outcome of said context could be, by using James Wilson’s classical policy contest matrix (Baumgartner, Jones, 1993, 41). It is, according to Stone, the distribution of costs and benefits that determine saliency and consequently policy change. People are more likely to organise and mobilise around what is perceived to be a bad than a good, and the perceived intensity with which they would be affected by a certain policy both determine the outcome of a contest (Stone, 1988, 238).

Fig. 1, Wilson’s policy contest typology (Stone, 1988, 239).

From this typology, four distinct contests arise. Two pairs are broadly similar with regard to outcomes. Type 1 contests have both diffuse costs and benefits, and fail to mobilise public opinion. They are likely to remain out of the political agenda, and safely in the hands of policy experts and professionals. These have the tendency to expand gradually, as policymakers are eager to distribute benefits to their constituencies and weak opposition arises (Idem, 239). Type 4 contests are similar to the latter given both costs and benefits are concentrated, leading to an equal contest. These tend to become long-contested issues that often end up in stalemates, with camps alternating small victories between themselves. In contests with unequal cost/benefits, issues leave policy subsystems and go into the

mainstream political debate. In Type 2 contests, benefits are concentrated and costs diffuse. The group that stands to win from such a policy will swiftly organise and influence the debate greatly, while the diffuse group that stands to lose will struggle to find motivation to

organise given they do not stand to lose much from the contest (Idem, 239). Type 3 contests, with concentrated costs and diffuse benefits are extremely similar to the last contest

discussed. In them, pro-change movements organise and mobilise to dismantle a detrimental policy monopoly.

One of the theoretical challenges that come with this classic typology is the difficulty with which real-life political issues can be fit into such an ideal scheme, Valuable inferences can be made from its application, but the typology by itself cannot explain complex and evolving political phenomena in full. Deborah Stone, in her 1988 book The Policy Paradox, famously applies a constructivist lens to policymaking, shedding light on the aforementioned

challenges. Stone emphasises the importance of how policy challenges are perceived rather than their intrinsic characteristics, and underlines the active role policy entrepreneurs play in portraying policy contests as a certain typology (Stone, 1988, 242). Regardless of the

characteristics of the case themselves, pro-status quo actors will emphasise the benefits a certain policy brings, and pro-change actors will emphasise the costs, in an attempt to sway the perception of the public. Creating the illusion of an unequal contest, in the spirit of David and Goliath, is one strategy often pursued by both camps, in order to rally support or

dismantle opposition.

In the same vein, Stone argues that certain interests are commonly perceived as good

interests, while others are perceived as bad ones (Idem, 245). Personal values are indivisible from policies and thus normative judgment is unavoidable in the subjective arena of public policy. Weak, collective and social interests are generally seen as virtuous, while

individualistic, material, economic interests are perceived as morally inferior to the latter. This argument naturally complements that of group and identity formation: collective and social goals are seen as morally superior and this helps the galvanisation of public opinion and mobilisation of individuals.

Empirical expectations

From the discussion of political attention, agenda setting and saliency, a number of empirical expectations can be derived from theory. Given unequal contests in the Wilsonian matrix yield policy change as a result, the Groningen case is expected to be such a contest. Stone’s (1988) distinction between good and bad interests suggests that the moral superiority of the

many disadvantaged, against the inferiority of the few advantaged will eventually cause policy change in this issue. Given Downs’ claim that the intensity of the social debate during the alarmed discovery and euphoric enthusiasm stage defines the likelihood of policy change, change is expected to happen in the Groningen case in periods of high societal and media attention. This is expected to concur with focusing events, which increase the intensity of the social debate. These focusing events force policy entrepreneurs (even reluctant ones) to express opinions on the policy issue at hand. It is expected that even pro status-quo actors that supported the extractive policy would be forced to express their opinions on the policy issue after earthquakes or other focusing events even if not doing so would suit their interests better. The seriousness of dramatic focusing events threatens these actors, given a failure to express an opinion would compromise their standing: moral, social or otherwise. Negative expert reports, new policy venues and available policy alternatives are also expected to encourage policy change. Likewise, it is expected that a strong civil response to the policy issue will galvanise public opinion and reaffirm group feelings, and help to ultimately bring about the demise of the policy. Together, these expectations form an array of causal factors that lead to the outcome, the Kamerbrief. The following empirical chapters will explore these and their causal importance will be discussed in the analysis chapter of the study.

Chapter 2: Methodology

In the present study, policy change theory will be used in order to study the case of energy policy change in the Dutch province of Groningen. It is a single case study, in which it will be attempted to uncover a causal configuration of factors that lead to an outcome: the

kamerbrief1 from Minister of Economic Affairs and Climate Policy Eric Wiebes presented to Parliament in March 2018, indicating the political decision to stop gas extraction in 2030. Research design

This research will attempt to obtain findings about the mechanisms through which the outcome occurred through a configurational study of the case. Through the use of this approach, either additive causal chains or interactive causal configurations are discovered. Causal conjunctions are configurations of factors that, converging at a certain moment in time, produce a given outcome (Blatter and Haverland, 2012, 94). Causal chains, on the other hand, identify an initial necessary or sufficient condition that leads to another such condition, that in the end leads to the outcome of interest (Ibidem).

In the case of this study, it will be concluded that a multiplicity of mechanisms, with only some sequentially related to the other, have made the outcome possible, making it a hallmark example of a process-tracing, configurational study (Blatter and Haverland, 2012, 81). While some of the factors evidently result in others in this particular case (gas extraction leads to earthquakes for example), other factors such as elections and the ability they have to bring issues to the agenda are not sequentially connected to the aforementioned outcome, yet still important for the outcome to occur. In studying causal configurations, it is of vital

importance to identify and control for all the relevant factors in order to avoid spuriousness and ensure internal validity, and then assess the importance of each factors’ contribution to the outcome. This is done by a number of theoretical tests, which identify necessary or sufficient conditions for an outcome to occur (Gerrits, 2006, 183).

1 _{A Kamerbrief, literally chamber letter, is an official document from a ministry or individual public} servant about punctual topics of interest to the whole chamber. Kamerbrieven have no binding power and are thus often a statement of political intent.

The four causal factors to be studied are the nature of political contest, expert opinion and policy venues, civil society action, and placement in the issue-attention cycle and focusing events.

The identified factors will be tested in the analysis section of this study through the use of a number of tests pertaining to process tracing methodology. These tests assess the degree of causality of factors, especially in single case studies where counterfactual analyses are not possible (Punton and Welle, 2015, 1). These analytical tests are used to determine the sufficiency and necessity of given factors for an outcome to occur. In these impact evaluations, four kinds of factors can be found. Straws in the wind are nor necessary nor sufficient for an outcome to occur, but slightly raise or lower confidence in a hypothesis (Idem, 3). Hoops are necessary factors but not sufficient for the outcome to occur. They discard hypotheses should the test be failed. Smoking guns are sufficient but not necessary and can greatly increase confidence in the importance of a factor (Idem). Lastly, doubly decisive factors are both necessary and sufficient for an outcome to occur (Idem), but are rarely found in complex configurational studies. The factors studied are studied through these tests in the analysis chapter of the study.

Casing and periodisation

In order to ensure the reliability, external validity and comparability of results, a periodisation has to be specified and search parameters have to be defined. This study will focus on the period between August 16 2012, date of the Huizinge earthquake and March 29 2018, date of the aforementioned Kamerbrief. The rationale behind this periodisation is comprised by a number of motivations, namely, the increasing amount of earthquakes, the rising media and societal attention, and the 2012 earthquake in Huizinge. These three factors converging motivate the periodisation.

The increased amount of earthquakes during the 2012-2018 period is the first motivation for the chosen periodisation. Fig. 1-3 (under) shows the amounts of (induced) earthquakes per year at the national level. Three different measures were taken to ensure objectivity and confirm the trend, as all stakeholders (NAM, Groninger Bodem Beweging and the state) could very well be interested in showing biased or inaccurate data in order to serve their interests. Figures roughly correspond to one another and show a clear trend, of ascending

occurrence of earthquakes from the mid-90s and a clear growth tendency until the 2017 peak. Data was collected up until October 2018, which means that 2018 can still confirm the ascending trend, should more earthquakes occur before the end of the year.

Fig. 2: NAM (2018). Number of earthquakes in the Groningen gas field.

Fig. 3: Number of earthquakes per year, Groninger Bodem Beweging (2018).

Secondly, Figure 4 shows the number of hits on Lexis Nexus for the search terms specified under. These stringent terms were used on purpose: to show the rising trend in media attention and justify the casing and it was not used to gather information for the research proper. It is clear from this search that from 2012 onwards, the policy issue gained traction in national and local media outlets, thus being the period where the mechanisms behind policy change were most active.

Fig. 5: Number of hits on Lexis Nexus for (Groningen and aardbeving and gaswinning and nam) per year.

Lastly, policy change literature is concerned with focusing events that bring issues to the fore of societal discussion and give actors the possibility to influence agenda setting (Birkland, 1998, 54). A clear case of a focusing issue was the 2012 earthquake in Huizinge, which brought the issue to national discussion. This particular earthquake was the strongest

measured to that date, the first earthquake above 3.0 in the Richter scale, and it prompted the national meteorological institute, the KNMI, to officially state the relationship between increased gas production and the increased occurrence of earthquakes (Dost, Kraaijpoel, 2013, 3). This is of special importance given the (veiled) interest of the government to protect the status quo. This will be further discussed in the empirical chapters that follow.

Data collection

This study will make use of primary sources such as official documents, legislation, and secondary sources, both academic and media produced. The different sections of the study will make use of different source material. For the study of the socio-political circumstances that allowed for the outcome to happen, both official documents from government sources, institutions and other actors will be used, together with media documents. The institutional arrangements governing extraction policy will make use of official and archival documents, especially when examining regulatory frames and ownership rights. As for the role of civil society in policy, secondary sources will be used, media will take a prominent role and official documents published by social movements will be examined.

The main methodological difficulties when conducting this research had to do with the difficulty of establishing a systematic way of using databases to gather information. Much of the source material on this topic is local, like the coverage of social movements, and often unavailable in databases. A number of measures have been taken to overcome these difficulties. Firstly, the inclusion of arguments or source material is to be exclusively

motivated by the theory presented in the following chapter. This is done to focus the research and avoid the inclusion of material that interesting as it may be, is of little theoretical

relevance to the academic debate. Secondly, for media-based sources, the Dutch newspaper database LexisNexis Academic will be used to gather information. The search parameters used for LexisNexis will be detailed per chapter in appendix 1.

For the first chapter concerning the nature of political contest, financial information publicly available by the NAM and statistics provided by the Dutch state (Centraal Bureau voor de Statistiek) is used in order to determine the cost and benefit distribution of the policy. For the chapter on expert opinion and policy venues, public reports published by different Dutch governmental organs and associated institutions is used. For the chapter on civil society movements, a number of official founding documents are used, retrieved from the NGOs’ own websites. This chapter is complemented by the use of media sources. Lastly, for the placement of the issue in the issue-attention cycle, official documents published by the NAM, SodM, and a number of media sources are used. The second part of this chapter presents an analysis of focusing events, for which statistics of earthquakes are used from

government sources the NAM are used together with media statistics retrieved from the LexisNexis database. For the last part concerning the 2017 national election, party manifestos are used. This gives the researcher the unique opportunity of having formal party documents with concise, often numerical policy proposals. These constrain actors and form somewhat binding expectations from the electorate to the parties, influencing their courses of actions and determining the likelihood of policy change. Said manifestos were directly retrieved from the party’s websites.

Validity/Reliability

The internal validity of this study is ensured by the control and study of all factors derived from policy change theory. In single case studies it is important to analyse every factor derived from the theory in depth, in order to rule out alternative causal explanations that could turn the study invalid.

Causal configurations and the findings resulting from single case studies are often difficult to generalise (Blatter and Haverland, 2012, 81). Despite this, the configurations of sufficient and necessary conditions these studies identify can have greater applicability than one case (Gerrits, 2006, 178), meaning such a study can still have external validity. Within-case causal configurations result in a number of factors that apply to that very study but are not

necessarily sufficient or necessary conditions for outcomes from other cases to occur (Blatter and Haverland, 2012, 82). The configuration identified in this study can be further tested against different cases of (natural resource) policy change, at single case, medium or large N settings. It could therefore be generalised to a number of cases, and potentially have

theoretical power should it apply to these.

The aforementioned methodological difficulties when collecting data on local news outlets or social movements mean that some of these sources had to be sought for individually and their use is not the result of a systematic search. Here, a trade-off occurs between validity and reliability: the inclusion of these sources make the study internally valid, as with their use, all potentially explanatory variables are included in it, yet its use is to a certain degree

Empirics

Chapter 3: Nature of policy contest

The first empirical chapter will deal with the cost and benefit distribution that resulted from the gas extraction policy. This factor lays the groundwork and context in which the other factors operate. In Wilson’s typology of policy contests, outcomes are determined by a distribution of costs and benefits (Baumgartner, Jones, 1993, 41). Institutional arrangements that distribute these in unsustainable manners, are destined to be replaced, following his logic. In the case of the province of Groningen, although it is straightforward to identify where the costs of the extraction policy lay, it is a much more intricate matter to identify how the benefits are distributed. This section will provide two applications of the Wilson model: one with concentrated benefits, and the other with diffuse benefits. A discussion of which outcome would theoretically emerge and a comparison with the real outcome will ensue. Lastly, the extent to which the distribution of costs and benefits allowed for by the institutional arrangements caused the outcome will be discussed.

Determining the costs in the case of Groningen is rather straightforward. Extraction in the Netherlands occurs mostly in the province of Groningen, with small scale extraction sites in the provinces of Friesland, Drenthe and, recently, Overijssel (NAM, 2018). Damage to property is exclusive to Groningen only, making the costs extremely concentrated: a whole country benefits from the tax profits, while only the inhabitants of Groningen suffer from the seismic movements caused by extraction. Durán and Elhorst (2018) estimate the loss in property value in the period 1993-2015 to be an average of 9.3%, with figures as high as 22.2% in the municipality of Loppersum. The perceived safety of inhabitants in the province has significantly decreased and this led to an increase in physical and mental illness in the province (Gronings Perspectief, 2006, 47).

The beneficiaries of the policy, on the other hand, are harder to pinpoint: it can be argued that society at large benefits from extraction, or that the extracting parties benefit from it.

Following this logic, two scenarios will be presented: one with concentrated benefits, and other with diffuse benefits.

Fig. 6: Unequal contest A: concentrated costs, diffuse benefits.

In the first conception of beneficiaries, A in the graph above, one can think of the population at large – a diffuse group - being a beneficiary from extraction. Tax is collected from

resource sale, jobs in the sector are created, profits are made from sale and other positive externalities emerge. From this, the Dutch state was able to reinvest earnings into areas like raising standards of living, education and health (Whaley, 2009), effectively funding the Dutch welfare state. Following this logic, the benefits of extractive policy are very diffuse.

Fig. 8, earnings from gas extraction and percentage of government earnings. CBS, (2018).

Figure 8 shows the percentage of the total earnings of the Dutch government that come from gas extraction. In the years 2012 and 2013, the share of gas earnings was as high as 5%, which indicates the paramount importance of this particular industry to the capabilities of the Dutch state. With production caps in place and media and societal attention sharply focused on the policy issue, extraction was limited and so were the earnings resulting from it. A sharp descent is evident from the data in the period 2013-2017 (CBS, 2018). It can be claimed, thus, that the population at large is a beneficiary from extractive activities in the north of the Netherlands. The scale in which they benefit is to be had into account, as citizens only reap the benefits of extraction after government decides where and how to allocate its resources, and that if conceived per capita, these figures become relatively small. This indicates that, under this conception of the policy contest, benefits can be seen as diffuse, creating an unequal contest, and thus policy change is expected to happen.

Equal contest B: concentrated benefits and costs

Under the second view, B in the figure below, the businesses that perform the extraction are seen as the main beneficiaries. They make most of the profit and amass considerable political power later used for influencing policy making. In this case, the NAM, the conglomerate of Shell and ExxonMobil not only own the exclusive extraction rights to the hydrocarbon

deposits, but the hydrocarbons themselves (Whaley, 2009). This means economic benefits are extremely concentrated.

The NAM presents annual financial statements to the Dutch Chamber of Commerce (NAM, 2018). These documents are not readily accessible to the public. From the NAM’s press releases about their financial statements, however, some information that illustrates the level of profitability can be taken. In 2017, the NAM reported a gross revenue of 3.4 billion Euro and profits to the tune of 249 million Euro (Ibidem). This was mostly due to the smaller gas fields around the main Groningen/Slochteren field, despite the increasing costs of

remuneration to citizens, building reinforcements and production caps set (Ibidem). The full statement for 2017 was not disclosed to the public. In 2016, the NAM did publish their financial statement to the public, and reported 35 billion gross profit and 526 million in profits (NAM, 2017). It is outside of the scope of this study to provide with a comprehensive study of the NAM’s financial situation, yet these examples show that even in adverse

situations, the operation of the gas fields in the north of the Netherlands is still a highly profitable activity.

Considering the NAM as a beneficiary that profits almost exclusively from extraction, the resulting policy contest would result in a stalemate result, given both camps are well

organised and articulated and these contests result either in non-action or in victories for one camp followed by victories for the other, with little progress either way (Stone, 1988, 239). This conception of the beneficiary does not explain the policy change that indeed transpired in the end.

Deborah Stone’s (1988) considerations on the normative power of arguments gives interesting insights on this conception of the beneficiary, and contradicts the predicted outcome from the matrix. Some interests, according to this scholar, simply hold a higher moral standing than others. People are more likely to get engaged in a debate and mobilise

with the threat of a potential bad, or loss of welfare, than a potential good (Stone, 1988, 238). This could explain why in the end, the expected outcome of this contest did not happen. Likewise, in the process of interest definition, interests that represent the general public are seen as intrinsically virtuous, yet weak, and deserving of protection (Idem, 244). Concepts as the general will, the people, and the social are often associated to these interests. On the other hand, special interests, held by select groups of powerful individuals are seen as strong interests that need no special protection, and are not as morally strong as those the whole of the population hold (Idem). These are broadly connected to the elites, and are usually economic. It is remarkably simple to draw the distinction between what the virtuous and the what the special interest in the case of Groningen after reading this chapter. The fact that this definition between the virtuous and the special is so clear cut could provide an alternative explanation to why policy change happened, even if the concentration of costs and benefits could not fully account for it.

Chapter 4: Expert opinion and policy venues

According to Timmerman and Scholte’s (2006) study of (expert) policy venues, these are of paramount importance in order to positively or negatively influence certain policy

monopolies. This section will sketch an overview of the role of institutional venues, changes in attitude in the stipulated period and important instances of expert opinion influencing the agenda.

Expert reports

Two main trends can be identified in the development of expert opinion around the gas extraction in Groningen. Firstly, expert opinion went from being produced exclusively by extractors to being increasingly produced by independent actors. Secondly, its content became increasingly negative towards the policy image. The very first expert reports published on the policy issue were the own research of the NAM. In 1973, research was carried out on the subsidence of the soil in extraction areas, which happened to be occurring at a lower rate than expected (De Waal, Muntendam-Bos, Roest, 2015, 132). This was

complemented by further scientific research on the phenomenon, but stayed firmly within the scientific policy subsystem (Ibidem). In 1990, Toksöz and Walsh, independent scholars from MIT reassessed the NAM subsidence measurement model. This indicated both the beginning of the involvement of third-party experts in the production of knowledge and one of the earliest processes of positive feedback, openly criticising NAM policy. From then on, aided partly by the first earthquakes like that in Middlestum in 1991 (M=2.4) (Ibidem), there was an increase of scientific research on the topic.

Focusing events such as earthquakes also prompted the publishing of expert reports. The 2012 report by Dost and Kraaijpoel from the National Meteorological Service, KNMI, was a direct response to the 2012 earthquake in Huizinge. Its main scientific contribution is a correction in the magnitude report, as the earthquake was in fact of a higher magnitude than reported, and an update of the current hazard analysis (Dost, Kraaijpoel, 2013, 19). The report, though, links explicitly the correlation between higher production and higher incidence of seismic movements (Idem, 3), which is of significance, especially when made

by KNMI researchers, thus the public service. Other expert reports, like the Reassessment of the probability of higher magnitude earthquakes in the Groningen gas field by Muntendam-Bos de Waal from the KNMI, updated current prediction models for maximum magnitudes of earthquakes. The 2016 Advies Winningsplan Groningen produced by the SodM was

presented directly to the then Minister of Economic and Climate affairs Henk Kamp, and adviced to limit production to 24Nm3 (SodM, 2016). It is this report that informed the policy solutions presented by political parties at the general elections of 2017.

Lastly, the SodM, after the 2018 Huizinge earthquake presented an advisory study, Advies Groningen-gasveld n.a.v. aardbeving Zeerijp van 8 januari 2018, urging the authorities to immediately stop production at the Loppersum clusters, avoid great fluctuations in extraction and to limit production to 12Nm3 (SodM, 2018, 4). This damning report shows the radical change in attitude from scientific venues, and is a clear example of positive feedback, whereby the policy monopoly on extraction is challenged and put into question. New policy venues

Limitedly rational policy entrepreneurs with defined interests will partake in venue shopping, a search for policy venues from where they can influence monopolies the most effectively (Kaunert and Leonard, 2012, 1397). The apparition of new policy venues changes power dynamics dramatically, making entrepreneurs with affinity to the new venues flock to them, and making policy change more or less likely according to the scope of action of said venue. In this case, the creation of the National Coordinator for Groningen in 2015 is a clear instance of a new institutional policy venue from which knowledge and feedback were created. On the 1st of May 2015, the Ministry of Economics decreed the creation of the National Coordinator for Groningen and the Government Service for Groningen, its administrative branch (Kamp, 2015). Its main task was to implement and shape the Aardbevingsbestendig en Kansrijk Groningen Programme from the Ministry: by the creation and sharing of knowledge, the advancement of education programmes in order to implement it, and by acting as a platform through which the main stakeholders could communicate and cooperate (Ibidem). These included different government levels, such as the provincial government of Groningen, twelve municipalities of the earthquake areas and civil society actors. The position of National Coordinator, upon its creation, went to PvdA ex-Commissary for the Queen Hans Alders (Nationaal Coordinator Groningen, 2015).

By creating a separate government agency, the national government and the Ministry of Economics yielded a significant amount of power over the policy area: the NCG regularly publishes often critical independent reports and reports carried out by universities, most notably the Rijksuniversiteit Groningen (NCG, 2018). By creating this institutional policy venue, the power balance shifted in the policy area: there is now space even within formal government for open critique and thus for positive feedback, which can destabilise and eventually bring about the demise of a policy monopoly.

To sum up, in the early years of extraction, knowledge creation and scientific research was mainly carried out by those directly involved in the process, with much of it produced directly in-house by the NAM. Slowly but surely, other independent actors joined in the production of said knowledge, such as professional researchers and government agencies. With time, these independent researchers started producing critical reports towards the policy monopoly of extraction. Due to the special weight that scientific input carries, these positive feedback mechanisms caused the monopoly on extraction to be weakened. Evidently, the existence of scientific and expert information on a certain policy issue will affect it, and influence the likelihood of change, but it is not sufficient to singlehandedly cause it or necessary: the complete cessation solution that came to materialise was indeed never suggested by experts. Expert opinion and the creation of new policy venues are rather facilitating factors increasing the likelihood of change but neither necessary nor sufficient to account for it.

Chapter 5: Civil society response

This present chapter will provide an overview of the most relevant civil society actors in Groningen, their stages of development during the stipulated time period, their actions, and the importance of their contribution to the outcome. The threefold consequence of civil society action, namely giving legitimacy to the cause, galvanising support and expanding the issue to the national discussion will be discussed in this chapter.

Civil society actors

Civil society movements attempted to boost media attention and bring the debate around gas extraction to the mainstream during the stipulated period, albeit with differing rates of success and scopes of action. To make up an exhaustive list of all social actors present in the policy discussion is outside of the scope of this study, but this section will sketch a rather broad overview of salient actors in order to facilitate the later analysis of their actions. One of the most highly developed institutions in the field is the Groninger Bodem Beweging, an organisation founded in Loppersum in 2009 by Jelle van der Knoop and Hilda Groeneveld (GBB, 2018). Its goal is that of assisting inhabitants of the province who were personally affected, directly or indirectly, by the national extraction government (GBB, 2018).

Motivated initially by the reluctance of the provincial government to conduct formal research on the consequences of extraction after the 2006 earthquake in Westeremden, a citizen workgroup was formed and then officially established as an NGO in 2009 (GBB, 2018). Its activities are comprised of four pillars. Firstly, they advocate for the interests of members and other people personally disadvantaged by providing juridical assistance, consulting the

National Coordinator of Groningen, trying to get media attention, contacting politicians and organising meetings, among other activities (GBB, 2018). Secondly, the GBB organises marches, demonstrations and peregrinations to The Hague, where government sits to express their discontent (Ibidem). A paramount example of these are the Fakkeltochten, massive torch marches through the city of Groningen. The group also provides information to a wider group than its member base by keeping an online inventory of informative documents

independent research in order to substantiate their policies and to generate new independent knowledge (Ibidem).

Other, more specialised organisations include Recht voor Groningen and Houd Groningen Overeind. Recht voor Groningen is a Loppersum-based NGO that focuses on the juridical aspect of the gas extraction. Its main goal is to act for the safety, health and wellbeing of the Groningers who are affected by the extraction, especially by carrying out fundraising actions in order to raise funds for lawsuits against the state or, more often, the NAM (Recht voor Groningen, 2018). Houd Groningen Overeind is an educational NGO that compiles news articles, written accounts and cultural artifacts about the case (HGO, 2018). One of their main activities is to report earthquakes as soon as they automatically appear on the National

Meteorological Service webpage on social media platforms such as Facebook and Twitter (HGO Facebook page, 2018), but they also organise demonstrations such as pickets to national politicians visiting the province (RTV Noord, 2018) and collaborate with the other groups. Fascinatingly enough, the website of the NGO Houd Groningen Overeind, hosts two links called ‘Us’ and ‘Them’, listing all the names of organisations they perceive to be on their side and organisations that work against their cause (HGO, 2018). This is an overt example of the disenfranchisement felt by these groups with the rest of the country. It is interesting to note how not only private institutions, like the NAM, are listed under the heading ‘Them’, but also all government instances.

The Groninger Gasberaad is an organisation which shows the importance of the previously discussed new policy venues, and evidences the access new entrepreneurs have to formal channels after the creation of such venues. This umbrella organisation serves as a forum for social actors, where their positions and views get aggregated, formalised and then presented to governmental authorities, such as the aforementioned National Coordinator for Groningen (Groninger Gasberaad, 2018). These actors include representatives for nature and the

environment, churches, entrepreneurs, industry, the agrarian sector, among others. The Gasberaad attempts to act both at the institutional level, by shaping policy, and at the individual level, by providing individuals with legal advice (Ibidem).

Lastly, one of the most militant, if informal, foundations in the field is Ons Laand Ons Lu (Our Land, Our People in the local dialect). Founded by Jan Holtman, Jannie Knot and Pieter Douma, the now legally founded association organises civil disobedience activities, like

blockades of NAM facilities (DvhN, 2018), or protracted hunger strikes (DvhN, 2018). Holtman and Knot, in turns, held the hunger strike in Appingedam, in front of the Centrum Veilig Wonen in 2018. Most famously, in a richly symbolic act, Ons Laand Ons Lu declared the Unique Republic of Groningen (Republiek Uniek Groningen, abbreviated RUG to mimic the local university’s name) on King’s Day 2018. This was done by holding a demonstration and a musical act as a protest against the royal family’s connection to Shell and an expression of detachment from the rest of the Kingdom (DvhN, 2018). Ons Laand Ons Lu routinely bring their protest message in long peregrinations to Den Haag and sit at debates in

Parliament as to compensate the lack of representation (AD, 2018). Other social movements active in the policy area also include Van Graan Naar Banaan, Onze Klei, and Schokkend Groningen.

Policy entrepreneur action

Policy entrepreneurs, such as the ones identified above, aggregate and articulate the political will of a part of the population in order to exert influence on agenda setting and policy making (Stone, 1988, 230). The individual actions of policy entrepreneurs can be of great importance for policy change. Although policy entrepreneurs are often conceived as formal players that play a role within policy making, there is no reason why private individuals that attempt to influence a policy with no formal institutional roles cannot also be considered as such. In the present case, making this extension of the theoretical notion allows to include characters from fields other than public administration and politics in the analysis, which is especially interesting given the high amount of cultural and artistic entrepreneurs that mobilised to bring about the demise of the existing policy image and monopoly.

National and local personalities – informal policy entrepreneurs – played an important role in raising awareness of the issue, especially beyond the region. An effect of paramount

importance of this individual policy entrepreneur action was its ability to bring the local policy issue to a national discussion. In short, it contributed to bring the issue from a policy subsystem controlled by public servants, experts and those directly related to extraction activities to the whole country, society at large and the political sphere. Groninger musician and comedian Freek de Jonge unofficially became the public face of the anti-extraction movement, by writing opinion pieces, attending national talk shows, leading the

a song by Gronings troubadour Ede Staal to create a battle cry for the movement (Freek de Jonge, 2018) (DvhN, 2018). De Jonge made a number of appearances on national media, and most notably, he appeared on the national talk show De Wereld Draait Door in 2017

championing the cause and singing the aforementioned song with a host of Gronings personalities (DWDD, 2017). De Jonge also led the campaign for a petition calling for political action in 2017, which garnered more than 200000 signatures and was personally presented to premier Mark Rutte by de Jonge in June of that year (RTVNoord, 2017). Gronings comedian and late night talk show host Arjen Lubach dedicated an episode of his national TV show Zondag Met Lubach to the Groningen case, calling for policy reform, especially in the compensation mechanisms to those affected by the earthquakes (Volkskrant, 2017). Lubach himself has been cited as a highly influential personality, and has notably campaigned via his programme for a host of political issues, leading to massively supported petitions to governernment (NRC, 2017). Further contributions were also made by musicians such as Pé and Rinus, Marlene Bakker, and Erwin de Vries who, through either television appearances or songs recorded especially for the occasion, championed the anti-extractive cause (DvhN, 2018).

The most salient instances of collective policy action are the Fakkeltochten of 2016, 2017 and 2018. The increasing number of attendance attests for the growing pressure on policy makers. The 2017 march attracted 4000 protesters (RTVNoord, 2017). The 2018 march was to a great extent motivated by the Zeerijp earthquake, and gathered approximately 10000 protesters (Volkskrant, 2018). They were endorsed by the University of Groningen, the

Hanzehogeschool, the University Medical Hospital and the FC Groningen, attended by national political figures such as the leader of GroenLinks, Jesse Klaver and the leader of the Christen Unie, Gert-Jan Segers (ANP, 2018). These massive marches were not only widely reported nationally, helping to bring the issue to the national mainstream (Volkskrant, 2018), but also replicated in the city of The Hague, which is traditionally seen as the political centre of the country, and the city of Appingedam (Milieudefensie, 2018).

Local political figures, through their action, also lent legitimacy to the anti-extraction cause by showing their dissent from their national parties’ policies. The mayor of Loppersum Alfred Rodenboog and the former National Coordinator for Groningen Hans Alders from traditional parties such as the CDA and PvdA respectively, pushed hard-line anti-extractive

policies and supported civil society movements while their parties were conciliatory and moderate at the national level (DvhN, 2018). A clear example of this is the aforementioned extraction levels proposed by the national parties CDA and PvdA on the one hand, and the participation of mayor Rodenboog in advisory meetings of the Groninger Bodem Beweging (GBB, 2018), and the resignation of Hans Alders as National Coordinator for Groningen and his subsequent membership of the aforementioned NGO (DvhN, 2018).

Use of symbols

Symbols are cultural expressions that are existential, non-political in nature but used in political ways, hold different meanings to different spectators and are powerful at eliciting emotional responses in those exposed to them (Cohen, 1979, p87). They act as a manner of connecting individual identity with collective feelings. The organically produced, local symbols tend always to be more effective than the imposed in binding irrational sentiment with rational decision-making (Cohen, 1979, p100). Gronings, the local dialect of the province, is a clear example of a symbol used to strengthen, legitimise the cause and galvanise opposition to the policy monopoly. It is naturally apolitical, though strongly associated with both the cause and naturally regional identity.

The January 2018 Fakkeltocht was of great symbolic significance as well. It saw an estimated 12000 protestors march through the city of Groningen, making it one of the biggest protests at a national level since the 1980s (RTVNoord, 2018). Imagery and local culture were emphasised at this event: the torches lit to protest extraction and weak government response to its challenges contrasted with the almost-anthem of Gronings resistance, ‘t Het is Nog Nooit zo Donker West’, the protest song penned by the famous traditionalist singer Ede Staal. This present chapter has outlined and determined the extent to which social movements were active in changing the policy image and dismantling the existing policy monopoly about gas extraction in the North of the Netherlands. The effect these movements had on the issue were that of incensing public opinion, lending legitimacy to the cause and bringing it from a local to a national political discussion. A strong social response to any given policy is helpful in creating a climate in which policy change can occur, but can hardly be defined as a necessary condition for it. It is clear that the social response has facilitated the process of dismantling of the policy monopoly, by creating constant positive feedback and providing actors with new

policy venues. Theoretically, this factor by itself could be a sufficient cause for policy change. The strong, identity based response to the policy monopoly from the inhabitants of the province exacerbated the perception of a crisis and successfully helped bringing the policy issue to the agenda.

Chapter 6: Socio-political circumstances

The last causal factor to be analysed is that of the socio-political circumstances during the stipulated time period. In this section, the social and political climate in the Netherlands in general and in Groningen in particular will be analysed.

Placement in Downs’ issue-attention cycle

Downs’ issue-attention cycle, as outlined in the theoretical chapter of this study, remains an awkward fit with the Groningen gas case but yields interesting results when applied. The very first problem with the use of the model to the case is that of applicability: according to

Downs, policy issues must not be intrinsically exciting in order to suit the issue-attention cycle (Downs, 1972, 42). This means that public attention cannot be held indefinitely by the issue at hand. Issues that are intrinsically exciting can thus linger in the collective debate and policy solutions that solve these are more likely to be found. The post-problem stage in which issues are forgotten and not fully solved could not be reached under those permanently

alarming circumstances. It is thus not possible to claim that the case of Groningen follows the issue-attention cycle in a linear fashion. The simple fact that earthquakes occur periodically bringing the topic to national discussion attests to this.

What can be observed in the case are sporadic reboots, in which the issue goes from a

dormant state in the post-problem stage, to an alarmed discovery state, prompted by focusing events. The case of gas extraction in Groningen is a drawn-out policy issue with its origins in the late 1950s, when the deposits were discovered (NAM, 2018). Policy is still being drafted as of the time of this study (RTVNoord, 2018), (NRC, 2018), among others. It is thus a protracted policy issue that has already gone through all the cycle stages. The pre-problem stage happened, between discovery of the deposits and the start of the public discussion about the costs and benefits of said policy. From 1959 to 2012, the issue rested mostly on a

scientific policy subsystem, in which technical information and studies were carried out by the state, extractors and experts, typical of this stage (Downs, 1972, 41), (De Waal, 2015, 131). Examples of this are SodM, NAM and academic reports about ground subsidence (NAM, 1973), (Toksöz and Walsh, 1990). Only these groups with direct stakes in the production were active in the policy field.