Institutional crises in governmental organizations
An in-depth study about the Groningen gas extraction case
Master thesis
Written by:
Erlineda C.J. Knoopers
Student number: s2100584
Supervisor:
Dr. Joery Matthys
Second reader:
Dr. Gabriele Landucci
Word count:
± 19.000 words
including in-text references
Leiden University
Table of Content
1.0 Introduction ... 1
1.1 Occurrence of institutional crises ... 1
1.2 Research question ... 3 1.3 Case study ... 3 1.4 Academic relevance... 4 1.5 Societal relevance ... 4 1.6 Reading guide ... 5 2.0 Theoretical framework ... 6 2.1 Crises ... 6 2.2 Institutional crises ... 7
2.3 The stream theory ... 9
2.3.1 The policy stream ... 10
2.3.2 The opinion stream ... 10
2.3.3 The political stream ... 11
2.3.4 The aggregation of the three streams ... 11
3.0 Methodology ... 13
3.1 Research design ... 13
3.2 Case study selection ... 14
3.3 Data collection ... 15
3.4 Data analysis ... 17
3.5 Operationalization ... 18
3.6 Validity and generalization ... 19
4.0 Analysis ... 21
4.1 Gas extraction in Groningen ... 21
4.2 Indicators institutional crisis... 24
4.3 Indicators policy stream ... 28
4.4 Indicators opinion stream ... 33
4.5 Indicators political stream ... 39
4.6 Intertwining of the three streams ... 44
5.0 Conclusion ... 46
5.1 Discussion ... 46
5.2 Answer research question ... 46
5.3 Limitations ... 47
Bibliography ... 48
1.0 Introduction
1.1 Occurrence of institutional crises
In April 2014, the administration of the State of Michigan decided to change the water supply for the city of Flint from the Detroit water system to the Flint River (Phillips, 2016). This decision seemed an easy way to save a lot of money for the American State of Michigan. However, the water quality was lower and Governor Rick Snyder and his administration did not improve the quality after complaints of the residents. The decision to change the water supply caused many health problems among the population, and was most likely the reason for several casualties (CNN, 2018; Phillips, 2016).
The problems of the water supply made the residents of Flint question the integrity of the Snyder administration. Trust in the administration decreased and the situation became national news. As a result, the situation almost created an institutional crisis for the administration of Governor Snyder.
The switch of the water supply in took place in 2014 (CNN, 2018; Phillips, 2016). Immediately after this, residents of Flint started to report that the water had a deviant smell, color, and taste. People also complained about health problems, like hair loss and rashes (Phillips, 2016). In the summer of the same year, it was announced that the water contained unhealthy bacteria and people were advised to boil the water before consuming (CNN, 2018; Phillips, 2016).
A committee that was formed to address the water issues claimed that the water quality was improved and met all safety standards (Phillips, 2016). Despite this claim, people continued to report the health issues of their children. Extremely high levels of lead were found in tap water of homes in Flint (CNN, 2018). Later test results of Virginia Tech showed that 40 percent of the homes in the city contained high levels of lead. Doctors found high levels of lead in the blood of children and advised to stop the consumption of the water from the Flint River (CNN, 2018; Phillips, 2016). First, public officials continued to claim that the water was safe, but shortly after, it was admitted that the water in the city of Flint contained above-average levels of lead (CNN, 2018). As a result, the city of Flint switched back to Detroit’s water system, and health services were provided (CNN, 2018; Phillips, 2016). Governor Snyder also apologized for the harm that has been done to the residents.
The lead-contaminated water had poisoned the residents of Flint for months (CNN, 2018). At least twelve people had already died as a result of Legionnaire’s decease, most due to the high levels of lead in the water (CNN, 2018; Phillips, 2016). The Mayor of Flint and the Governor of Michigan declared a state of emergency. The United States President Barack Obama declared Flint a federal emergency area. This means that the city of Flint received governmental aid from the Obama Administration to replace the pipelines in the city. At the beginning of 2017, the levels of lead in Flint were within standards (CNN, 2018).
Many lawsuits and official complaints have been filed against government officials, Governor Snyder, the State of Michigan, the city of Flint, and state organizations (CNN, 2018). In 2017, at least
13 people have been officially charged for their wrongful behavior in, or contribution to the Flint water crisis (CNN, 2018).
A year and a half after the switch to the Flint River, the Governor apologized for what has happened to the city and its residents (Phillips, 2016). His testimony about his unawareness of health complaints is being questioned and some people, including famous politicians, asked for this resignation. Nevertheless, he has not been forced to step down as governor. At different governmental levels, funds and financial support has been promised to the residents of Flint (CNN, 2018).
The government is expected to keep people safe, protect them and decrease the risk of citizens getting harmed, both mentally and physically (Kuipers & ’t Hart, 2014). This expectation is based on the fact that the public gave part of its autonomy to the government, trusting them to be protector of the inhabitants. The use of violence, for instance, can only employed by appointed governmental departments. The public’s trust is essential for the government to function properly. If people lose that trust, they might start taking matters into their own hands.
The government is also expected to protect a society’s core values (Boin, Kuipers, Bos, & Groenleer, 2000). One way to do that is by implementing rules and regulations. For instance, decreasing the risk that people can endanger others can be facilitated by implementing safety measures and legislation. The core value of having clean and safe drinking water was endangered by the administration of the State of Michigan in the Flint water crisis (Phillips, 2016). Additionally, government officials did not seem honest about the safety risks. This implies that government officials did not honor rules and regulations about water safety.
If values are not respected and the responsible parties are not acting accordingly, the public, the media, and the political opposition will start debating the performances of the policy sector. In the worst case, the legitimacy of a whole sector is being questioned, and the policy sector experiences an institutional crisis by losing its trust (Boin et al., 2000).
A uniform definition of an institutional crisis is not given in literature but there are characteristics that can be identified. Increased public and media attention and decreased trust in the policy sector are some examples of characteristics of an institutional crisis (Boin, ’t Hart, & McConnell, 2009; Boin et al., 2000).
Boin et al. (2000) argue with their stream theory that an institutional crisis happens when three streams come together and intertwine: the policy stream, the opinion stream, and the political stream. When these three streams intertwine, the whole sector’s institutional structure can be put under scrutiny and trust in the institution will be lost. Consequently, the survival and the direction of the organization or policy sector are uncertain: an institutional crisis has started. The stream theory is one of few theories about institutional crises. Moreover, this theory is derived from Kingdon’s well-known multiple stream
approach for policy change (Kingdon, 1984). Both these reasons make the institutional stream theory interesting for empirical testing.
1.2 Research question
The preceding leads to the following research question:
Can the stream theory of Boin et al. (2000) explain the emergence of institutional crises in governments in regard to safety-related issues?
1.3 Case study
The empirical case that is chosen for this study is the Groningen gas extraction. In 1959, a gas field in the ground of Groningen, a province in the Netherlands, was discovered, which later appeared to contain 2800 cubic meters of natural gas (EenVandaag, 2014c; Havermans, 2015a). Natural gas is used by both households and industries to generate heat.
Nederlandse Aardolie Maatschappij (NAM), a joint venture between Shell and Exxon Mobil, was appointed by the Dutch government to exploit the gas. The responsible person from government was the former minister of Economic Affairs Jan Willem de Pous. The government decided to transport the gas to houses in the Netherlands. Using the gas for this purpose generated the most profits for both the Dutch state and NAM (Bokkum, 2018b). Between the start of the exploitation of the Groningen gas field, in 1963, and 2018, 75 percent of the gas has been extracted (Berg, 2018a; ‘Voorspelbare bevingen? Negeren en wegmoffelen’, 2013).
The extraction caused earthquakes in the province of Groningen. In 1987 the first light earthquakes were registered by the Koninklijk Nederlands Meteorologisch Instituut (KNMI). The first earthquake noticeable for people struck in 2003 (EenVandaag, 2014c). In 2012, an earthquake with the magnitude of 3,6 on Righter’s scale struck the village Huizinge (Ellenbroek, 2013b). This was the heaviest earthquake that has happened as a result of the gas extraction in Groningen.
Over the years, the earthquakes have damaged buildings and in extreme cases they made houses uninhabitable (Takken, 2014). Many damage reports have been filed by victims but many people were not, or insufficiently compensated. Most importantly, the safety of the residents of Groningen has been jeopardized when they still have to live or work in unsafe buildings. The trust of the Groningen residents in both NAM and the minister Henk Kamp decreased (Ellenbroek, 2013b; Luyendijk, 2013e).
Media reported about the issue for several years and public dissatisfaction about the gas extraction increased (Heemelaar, 2018). In politics, many debates about the topic have taken place and many politicians condemned the way the government has dealt with this matter so far (Trouw, 2005).
For a long time, only small institutional changes were made but the conditions for the residents of Groningen did not improve. In 2018, six years after the first severe earthquake, the minister of Economic Affairs decided to reduce the gas extraction as much as possible, and eventually to completely stop the gas extraction in Groningen in 2030 (van Santen & van der Walle, 2018).
1.4 Academic relevance
This research topic is of academic relevance because there is a gap in the literature, and because the stream theory (Boin et al., 2000) is neither further developed, nor it has not been tested yet.
There is a gap in crisis management literature about the direct reason that lead to the emergence of an institutional crisis. Most of the current crisis management literature about the reasons why crises happen describe the process of the development of underlying problems within organizations or policy sectors. However, not much has been written much about the direct processes within society that could lead to institutional crises. Therefore, this research might add valuable insights to the current literature. The stream theory (Boin et al., 2000) has not been described extensively by the authors, or by others scientists. The amount of literature about this theory is limited and very specified. This thesis could be helpful to extend the theory, because here it is more elaborately described and applied. The theory could also be adjusted in order to strengthen it.
This research will test if Boin et al.’s (2000) stream theory is applicable for safety-related issues. The theory has not been tested before and so this research might add to literature on the stream theory.
Lastly, the stream theory is one of a few crisis theories that is derived from an already known and acknowledged theory, Kingdon’s stream theory for organizations (Kingdon, 1984). Therefore, testing this theory means that a possible link could be established between organizational theories and crisis management theories.
1.5 Societal relevance
As mentioned before, governments are expected to protect their people. Their actions should not in any way jeopardize the safety of their constituents. Extensive knowledge about crisis management could help governments and organizations to better execute their tasks of protecting people. Research on crisis management is therefore of societal relevance.
Governmental institutions have the responsibility to operate efficiently since they are paid by public money and because they spend public money on their projects. The guarantee of this efficiency and the prevention of institutional crises can save a lot of time and money, which makes this research important.
This research could also help to forecast whether a situation is about to evolve into an institutional crisis. Then, appropriate steps could be taken to prevent that from happening.
Lastly, public institutions can only operate when they have the trust of the people and society. Institutions need to keep their legitimacy in order to be able to function fully. If people start to disobey or rebel against governmental institutions, the government might lose its power and the whole constitution might be jeopardized. It is important for institutions that they do not experience institutional crises.
1.6 Reading guide
This thesis has four more chapters: Theoretical framework, Methodology, Analysis, and Conclusion. In the chapter Theoretical framework, the main concepts will be explained and developed based on institutional crisis literature. As the title explains, methodologic choices will be justified in Methodology. The operationalization, the internal, and external validity of the research will be discussed in this chapter as well. Analysis will include all relevant data that has been gathered based on the methodology. Finally, in Conclusion, the answer to the research question and the consequences of the answer will be formulated.
2.0 Theoretical framework
2.1 CrisesNatural disasters, technological breakdowns, aircraft accidents, mass shootings, industrial accidents and sudden political instability can all be called crises. A clear-cut definition of the concept of crisis is not given in the literature but all crises have three features in common: threat, uncertainty, and urgency (Boin, ’t Hart, Stern, & Sundelius, 2016; Boin, Kuipers, & Overdijk, 2013; Rosenthal & Kouzmin, 1997). In this description of crisis, threat relates to risk of the continuation of people’s daily life (Boin et al., 2016; Roux-Dufort, 2007). This means that people’s lives are at risk or that their livelihoods are threatened. During crises, there is uncertainty about the cause of the crisis and the possible consequences of the decisions that could be made. Lastly, the threat must be addressed with urgency in order to limit the harm that is being done to people. The time pressure under which decisions must be made in crisis situations frequently results in extraordinary measures (Roux-Dufort, 2007).
Other characteristics that distinct crises from accidents are the scale, the unreadability, the total interdependency, the instantaneity, and the in-depth dislocation (Topper & Lagadec, 2013). First, the number of people or the amount of nature that gets affected is enormous during a crisis; whole countries, continents or even the entire planet can be affected (Topper & Lagadec, 2013). Also, accidents are difficult to predict and map but most crises are too complex to map due to the many and unexpected shocks it causes. Thirdly, the increased interdependence and global networks of present day make the consequences of crises global and widespread. Also, information is shared in real time nowadays, which means that the consequences could be noticed in other places of the world right away. Lastly, the consequences of a crises do not move slowing in a logic and pre-set pattern. Instead, they jump from one place to another due to the interconnectedness of the world.
Crises can be considered to be circular processes that consist of multiple stages (Roux-Dufort, 2007; Turner, 1976). There are different theories about the number of stages a crisis process can have. On the one hand, a four-stage process consists of ‘warning signals, [an] acute stage, amplification, and resolution’ (Roux-Dufort, 2007, p. 109). On the other hand, a six-stage process consists of the ‘notionally normal starting point, incubation period, precipitating event, onset, rescue and salvage – first stage adjustment, and full cultural readjustment’ (Turner, 1976, p. 381). The difference is that the four-stage process does not include a normal four-stage like the six-four-stage process does. A similarity between the two theories is that the ‘warning signals’-stage of Roux-Dufort (2007) can be considered to be the same as the ‘incubation period’ of Turner (1976). Furthermore, the ‘acute stage’ is the same as the ‘precipitating event’. Also, ‘amplification’ and ‘onset’ are similar. After all, ‘resolution’ is similar to the two last stages: ‘rescue and salvage – first stage of adjustment and full cultural readjustment taken together’. So, if the two different processes are being compared, many similarities can be recognized. Therefore, these processes have been combined and made into one integral process. A crisis process
where both processes of Roux-Dufort (2007) and Turner (1976) are combined would be visualised as Figure 1.
Figure 1. Adjusted crisis process (Roux-Dufort, 2007; Turner, 1976)
In the adjusted crisis process, the normal stage entails the period in which organizations can operate in their daily, routinized way (Boin et al., 2000). During the normal stage, the public’s perception of the organization is in line with the way an organization operates (Turner, 1976). During the incubation period, operations start to be inconsistent with the public’s beliefs and the internal dysfunctions of an organization accumulate (Roux-Dufort, 2007; Turner, 1976). These events could either happen unnoticed, or the people within an organization do not realize the possible consequences of them yet (Turner, 1976). During the triggering event, these dysfunctions have manifested themselves (Roux-Dufort, 2007). This is an event that is noticeable for the people outside an organization and it probably attracts the attention of the public and the media (Boin et al., 2000; Turner, 1976).The amplification stage is the stage when the direct consequences of the triggering event are noticeable (Roux-Dufort, 2007; Turner, 1976). A crisis is stabilized when appropriate adjustments are made or when there is a resolution for the future, both belong to the adjustment stage. A crisis period is ended by a new ‘normal’ or a new understanding of the organization’s operations.
It is important to acknowledge that a crisis is not only the triggering event but a process that begins before the triggering event and ends when a new stability is found (Boin & ’t Hart, 2000; Roux-Dufort, 2007). The acknowledgement of this is crucial because it is only possible to investigate prior and after events when it is believed that there is a relation to the triggering event.
2.2 Institutional crises
The terms policy sector and public institution will be used interchangeably as they comprise the same sort of organizational structure. Both concepts describe a network of public or semi-public
1. Normal stage 2. Incubation period 3. Triggering event 4. Amplicication stage 5. Adjustment stage
organizations, individuals and/or organization departments with specific rules and practices that address societal problems or issues (Alink, Boin, & ’t Hart, 2001; Boin & Christensen, 2008).
A public institution or policy sector is different from a public organization in that it has been founded informally (Boin & Christensen, 2008). A public institution has managed to add value to important public values and gained both societal and political legitimacy because of that. Furthermore, internal as well as external stakeholders value the effective way the institution operates. Also, its institutional performances have shown to be consistent and high. This is important since an institution has many different interests to look after and many factors that complicate the achievements of its goals. Examples of these different interests and complicating external factors are personal perceptions of employees and critical citizens, and defective control systems and a quick changing environment.
A policy sector or public institution consists of internal and external actors (Alink et al., 2001; Ansell, Boin, & Kuipers, 2016; Boin & ’t Hart, 2000). The internal actors perform institutionalized and routinized governmental or semi-governmental operations, with specific rules, regulations, values, and goals (Ansell et al., 2016; Boin & ’t Hart, 2000; Boin & Christensen, 2008). At the centre of the policy sector are political officials and public administrators who are responsible for the performances of the sector (Alink et al., 2001; Ansell et al., 2016; Boin & ’t Hart, 2000). External actors are the clienteles, NGO’s, and lobbyists, who are also pressure groups that can have an effect on a policy sector. Nonetheless, they are less important than internal actors. The mass media and politicians are external actors that could have a big impact on institutions (Boin et al., 2000). All the external actors together form the environment of the policy sector (Alink et al., 2001; Ansell et al., 2016; Boin & ’t Hart, 2000).
Current literature does not provide a standard definition for institutional crises yet, however, multiple characteristics can be found that describe the occurrence of an institutional crisis (Boin et al., 2000). Just like during other kinds of crises, urgency, uncertainty, and threat are present during institutional crises (Boin et al., 2016, 2000). However, the difference is that institutional crises threaten the continuation and survival of institutions or policy sectors.
A developing institutional crisis is often reinforced by a triggering event (Boin & ’t Hart, 2000; Boin et al., 2000). This is an incident that generates a lot of public and media attention and brings to light possible decreasing performances of a policy sector (Boin & ’t Hart, 2000; Kuipers & ’t Hart, 2014). The nature of a triggering event differs from time to time. It might be a whistle-blower, a fatal accident, a press release or investigation results (Boin & ’t Hart, 2000, p. 20). The public and the media will consider the triggering event as the symbolization of long-term malpractices of the policy sector (Ansell et al., 2016; Boin & ’t Hart, 2000; Boin et al., 2000). This is in contrast with periods before the triggering event when the policy sector’s actions were taken for granted.
appropriate by most of the people in the environment of a policy sector. Also, in order to call an issue a crisis, the term must stick with the general public (Boin et al., 2000). There are cases when there is no significant change in a policy sector’s performances, although, a crisis did occur because the environment perceived it as one (Ansell et al., 2016). This happened, for instance, in the 1990’s in the Netherlands. A seemingly high number of prison escapes became publicly known as a result of extensive media coverage. The media attention had caused public and political dissatisfaction about the prison system and, consequently, an institutional crisis occurred. However, the number of prison escapes was not increasing, to the contrary: the number of prison escapes was declining for decades. This example shows that the perception of the public can be leading in the occurrence of an institutional crisis. Furthermore, the media have an important role in the development of an issue into a crisis. Media are in many cases the reason why the public is paying attention to certain issues (Boin et al., 2000).
Another way to look at decreasing support, is to describe support as the expectation from the environment that the policy sector operates appropriately (Boin & ’t Hart, 2000; Boin et al., 2000; Useem, Pacholke, & Mullins, 2017). When incidents happen, the gap widens between at the one hand, the expectations of the public and the media, and, at the other hand, the performances of a policy sector; a performance deficit develops (Ansell et al., 2016). This could also be described as an environment that questions the institutional integrity of a policy sector (Boin & ’t Hart, 2000; Boin et al., 2000). Due to decreasing support of the environment, the routinized operations of a policy sector are being questioned and a disruption of operations occurs (Boin & ’t Hart, 2000). In that case, the status quo of a sector is unstable and insecure. As a result, the institution’s authority is declining and the legitimacy decreases (Boin & ’t Hart, 2000; Boin et al., 2016). This decreasing legitimacy comes with ‘media scrutiny, public criticism, political controversy and calls for reform’ (Boin & ’t Hart, 2000, p. 10).
Based on the previous, in this thesis, institutional crises will be defined as critical moments for institutions or policy sectors during which the institutional survival and direction is endangered and uncertain (Ansell et al., 2016; Boin & ’t Hart, 2000; Boin et al., 2016, 2000).
2.3 The stream theory
‘Crises are […] the result of multiple causes, which interact over time to produce threats with devastating potential.’ (Boin et al., 2016, p. 9). This statement debouches into Boin et al.’s (2000) stream theory because it describes that an institutional crisis of a policy sector is an accumulations of three factors. Based on Kingdon’s (1984) flow theory, Boin et al. (2000) explain that an institutional crisis is the intertwining of the policy stream, the opinion stream, and the political stream. When all streams are stable, they exist separately from each other and have their own characteristics and dynamics.
2.3.1 The policy stream
The policy stream of a public institution is the collection of the day-to-day operations (Boin et al., 2000). People within a public sector act based on the sector’s policies which describe the rules and practces of an institution (Alink et al., 2001). All these operations will result in the performances of policy sectors. In most cases, the policy stream of a public institution is stable (Boin et al., 2000). This means that the institution performs close to how it desires to function. The ideal performances could be describes in future policy plans, the mission statement or expressed by the institutional leaders (Boin & Christensen, 2008). When an institution performs like this, there is no reason for the people within the organization to change or improve the institutional operations (Boin et al., 2000). People within the policy sector are satisfied with its performances.
A policy stream could, however also decrease or increase (Boin et al., 2000). An increased policy stream means that the sector performs better than expected or desired. This could for instance happen when revenues or output of an institution increase. On the other hand, a decrease in this stream means that the institution performs not as desired. In other words, the performances are lower than expected. Also, a decrease in the policy stream could be the occurrence of incidents like fires, accidents, fraud or an increased safety risks. A late response, a wrong response, or no response to these kinds of incidents are also insufficient performances (Boin et al., 2009, 2000). The decreased policy stream is recognized by the people within the policy sector and changes are made to improve the institutional performances.
2.3.2 The opinion stream
The opinion stream is the public’s opinion about an institution or policy sector (Boin et al., 2000). Normally, the reputation of a certain public institution is relatively stable. This means that people do not often think or talk about the institution and have no striking good or bad associations (Boin & Christensen, 2008). In other words, the way the institution operates is taken for granted. The public is under the impression that the performances of the institution are somewhat in line with its expectations of the organization. They have trust in the institution and are satisfied with it.
If the opinion stream decreases, the public’s perception of a policy sector is negatively influenced by the information it has about an institution (Boin & Christensen, 2008; Boin et al., 2000). When an incident was man-made or could have been prevented easily, the public will hold the policy sector responsible for the occurrence of the incident (Boin et al., 2000; Kuipers & ’t Hart, 2014). Consequently, other actions and operations of the policy sector will be reviewed and questioned (Boin et al., 2000). The public will reconsider its positive opinion about the sector which causes a change in societal morals (Boin et al., 2000). Its opinion about the policy sector will then be predominantly negative. This results in a loss of trust in the sector and a dissatisfaction about the sector in general (Boin & Christensen, 2008; Boin et al., 2000). This could eventually result in an overall consensus that the policy sector must change its institutional operations (Boin et al., 2000).
The opinion stream increases when the public pays more and positive attention to a certain institution (Boin & Christensen, 2008). People might praise the public institution for the way they operate and its performances. People do not suggest that the institution should change or that the performances should or could improve. In other words, people speak and think mostly in a favourable way about a policy sector.
2.3.3 The political stream
The last stream is the political stream. This stream is the attention that is being paid to an organization by politicians (Boin et al., 2000). The attention will be expressed during debates between different political parties. During debates they have the chance to express their opinion officially and publicly. Also, those moments allow for asking the responsible minister about the policy sector’s performances. Moreover, the responsible minister and political parties could be held accountable of the actions and performances of the policy sector.
When the political stream is neutral, politicians will not or rarely put the performances of an organization on the political agenda (Boin & Christensen, 2008). When they do, they might just ask some questions about the progress or developments of the organization (Boin et al., 2000). Thus, in those times, there is political tranquillity about the institution.
This stream decreases when politicians ask the minister questions about the reasons why certain decisions are made or why certain events have occurred (Boin et al., 2000). Their opinion about the performances of the policy sector is negative and the politicians start to question the minister’s ability to make the right decisions (Boin & Christensen, 2008; Boin et al., 2000). Consequently, they ask the minister if he or she is able to make changes, improve the performances, or make different decisions because they are dissatisfied or disappointed by the performances (Boin & Christensen, 2008). In the worst case, it is possible that motions of distrust will be submitted or requests for parliamentary inquiries will be filed (Boin & Christensen, 2008; Boin et al., 2000).
The political stream has increased when politicians speak predominantly positive about a certain institution or the performances of that institution during political debates (Boin & Christensen, 2008). During the debates the minister is complimented for the performances or the performances of the policy sector are being taken as an example for other policy sectors.
2.3.4 The aggregation of the three streams
The accumulation of the three streams is the reason for the occurrence of an institutional crisis (Boin et al., 2000). These streams do not, however, always intertwine in a similar or consistent way. There are a few characteristics of the intertwining of the streams. To start, the policy stream is often the offset of an institutional crisis because the performance of an institution are the reasons for public, media and/or political attention (Boin et al., 2000). However, as explained before, sometimes a perceived decrease of a stream can be the offset as well (Boin et al., 2009, 2000).
Moreover, the opinion stream regularly precedes the political stream (Boin et al., 2000). This is a natural cause because politicians represent the citizens and they should, therefore, respond to their worries and questions.
Boin et al. (2000) point out that the decrease of the streams is dependent on circumstantial factors. To start, the decrease of the policy stream depends on previous performances of a public institution. When multiple incidents have happened in the past, people are inclined to quickly distrust the institution. If a policy sector had a trustworthy reputation or no incidents have happened before, sectors have some credit to make mistakes. If an incident occurs that is completely out of the sector’s control, the public will, presumably, not hold the policy sector completely responsible (Boin et al., 2000; Kuipers & ’t Hart, 2014).
Furthermore, the opinion stream is dependent on the understanding of the public about a certain topic. A reason why incidents do not always develop into institutional crises is that not everything is understood by the general public (Boin et al., 2000). Media have an important role in creating public understanding. Therefore, the media could be crucial in the development of crises. This has two implications. On the one hand, many complicated issues do not evolve into crises because journalists are not able to describe the issue in a way most people understand it. On the other hand, some incidents get a lot of media attention and evolve into a national issue while there is not much changed in the policy stream (Boin et al., 2009, 2000). However, people overestimate the importance of it due to all the media attention.
Also, political or personal gain could be a reason why politicians pay attention to an issue (Boin et al., 2000). For instance, political opposing parties could be willing to challenge coalition parties to make them look inadequate (Boin et al., 2009, 2000). Specific political viewpoints might be a reason to ignore an issue to not endanger one owns position (Boin et al., 2000). When politicians do not benefit from paying much attention to a certain issue, it could take longer before a subject gets politicized. Ministers and political parties can also feel pressured to respond to an issue when the issue affects many people in society.
Political interest is an important factor in the emergence of institutional crises because the decrease of the political stream is often the one following the two others (Boin et al., 2000). If the policy sector’s continuity is at stake, the issue can be called an institutional crisis (Ansell et al., 2016).
3.0 Methodology
3.1 Research design
The research is designed based on the research question:
Can the stream theory of Boin et al. (2000) explain the emergence of institutional crises in governments in regard to safety-related issues?
Figure 2. Boin et al.’s (2000) stream theory
If the policy, opinion, and political streams are present in the selected case, it is expected that the stream theory (Boin et al., 2000) is applicable to governmental safety-related issues. If the streams are not present, this research is not able to indicate that the stream theory (Boin et al., 2000) is applicable for governmental safety-related issues.
The aim is to answer the research question with a qualitative longitudinal case study design. The case study that has been selected to test the stream theory (Boin et al., 2000) is the Groningen gas extraction case because there can be several components could be identified that could indicated that the case is an institutional crisis.
An institutional crisis is contextual a very complex phenomenon that develops under complex circumstances and under certain conditions, therefore, qualitative research approach is chosen for this study (Ansell et al., 2016; Boin & ’t Hart, 2000; Boin et al., 2000). In order to understand the phenomenon and to be able to fully describe it, it is important to have an in-depth description of the case and its environment (Boin et al., 2000; Yin, 2006). Based on the explanatory research question, a longitudinal single within-case study will be performed.
A case study design allows for explanatory research and holistic investigation, this research design is, therefore, appropriate for this research question. Patterns and causal processes are of interest, which could not have been found when a quantitative research design is chosen (Mahoney, 2007). More
Policy stream Opinion stream Political stream
importantly, the context is also taken into account, which is crucial to be able to say something about possible causal mechanisms.
As mentioned, explanatory research makes it possible to identify causal mechanisms. If these mechanisms are identified, it is a possibility to change the social world that has been studied (Toshkov, 2016). A within-case study research design is chosen because the causal mechanisms within one case are of interest (Mahoney, 2007). The stream theory (Boin et al., 2000) suggests that the intertwining of the three streams leads to an institutional crisis and, therefore, this study will test if there is a reason to believe that this causal mechanism might exist in governmental safety-related issues (Boin et al., 2000). Also, case studies allow for the investigation of links that are established over time (Keller, 2012; Mahoney, 2007; Yin, 2006). This study will be a longitudinal case study: all data used for the analysis comes from sources between 2010 and 2018. The longitudinal timeframe of the study is also the reason why there has been chosen for a single case study: investigating multiple cases would not have been feasible.
Lastly, case study research provides the possibility to investigate different kinds of data. In this research, for instance, news broadcasts, newspaper articles, parliament meetings, and governmental inquiry reports will be analyzed. Multiple different kinds of sources help to triangulate information.
The method to analyse data will be a content analysis of written documents and videos that consisted of both visuals and audio. The requirement for documents to be analysed based on the principles of content analysis, is that they consist of a message. Content analysis has the goal to identify underlying meanings in texts or videos and to find patterns within the same text (Pashakhanlou, 2017). This method is most applicable to the explanatory research question because it allows for the separate, independent analysis of the three streams. Above all, the analysis provides descriptive data.
3.2 Case study selection
The first criterion for choosing a case was that a public safety-related problem or hazard was created, not prevented, and/or not immediately resolved. The second criterion was that the responsible entity was a governmental or semi-governmental organization or department. Another criterion is that the crisis should have happened between 2000 and 2018 because of the availability of digital documents.
The cases that met the criteria were the floods in New Orleans (Boin et al., 2000), the Groningen gas extraction (Es, 2012), the Flint water crisis (Phillips, 2016), and the wrongly early-release of inmates in Washington (Useem et al., 2017). However, the New Orleans, Flint, and Washington cases show less organizational changes than the Groningen case. This means that they match the given definition of institutional crisis less than the Groningen case.
Therefore, the unit of analysis of this thesis is the Groningen gas extraction case (Keller, 2012). The case seem to be an institutional crisis because the case has many characteristics of a crisis (Boin et
al., 2000). This means that this case has a good chance to prove that the theory is correct (Rohlfing, 2012).
The Groningen case is also interesting to investigate because institutional crises do not happen often in the Netherlands. In the Netherlands, a culture of willingness exists to corporate and make compromises in order to meet all parties’ wishes in disputes (Nieuwsuur, 2012). Hence, it is exceptional that the dissatisfaction and feelings of unsafeness of people in a whole region were not taken into consideration for a long time by the responsible policy sector, gas exploiter Nederlandse Aardolie Maatschappij. Also, safety issues in particular created by governmental or semi-governmental organizations are uncommon and so, the prevalence of the safety problems make this case very unique. The units of observation are the documents and data provided by the policy sector and by the most important actors in the environment of the policy sector. The documents and videos that were analysed, were non-randomized, purposive sampled (Keller, 2012). Randomized sampling would not have been able to provide the research with the most relevant information. All data is selected based on its relevance. Its relevance will be based on the form of publishing, the author(s) and publisher(s), the date on which the data was released, the title and subject of the content, and the size of the data. A specification of the criteria for the data can be found in the next sub-chapter.
In 2012, a heavy earthquake hit the province of Groningen. This earthquake had a big impact on the residents of the province. Also, the public and media attention about the consequences of gas extraction increased. Events just before the 2012 earthquake could have influenced the reaction of people, in that sense, data from 2010 onwards was considered to be useful for the longitudinal study. All data that will be selected for this research dates from January 1, 2010 until March 31, 2018. The data has to be either a video with visuals and sound or a written document.
3.3 Data collection
Policy documents, inquiry reports, newspaper articles, news broadcasts, documentaries, and political debates about the gas extraction were used to investigate if more attention was being paid to the Groningen case. These documents was chosen because policy document and inquiry reports give insights into the internal processes of a policy sector. Especially the external view of inquiries provides an objective and neutral documentation of the operations of a policy sector. Moreover, newspapers, news channels, and documentaries report on the one hand, about issues the public should know about, and on the other hand, about issues that the public is discussing. Because of that, these media are good sources for topics that the public is discussing. Furthermore, during political debates, politicians openly question the procedure of certain issues. Therefore, the issues discussed in debates could be considered representative for political attention.
Key words in Dutch are expected to be most relevant because most document and videos about the case are in Dutch. Also, the issue has not reached many other countries. Therefore, it is expected that Dutch documents and videos will be most relevant.
The key words ‘Gaswinning Groningen’ (Gas extraction Groningen) were used most when searching for documents to analyse because most documents about the gas extraction contain these words at least once. Furthermore, there are no other Dutch words that describe this topic better. These key words were either inserted in the searching tool on the website of the mentioned data sources. Lastly, the collected documents dated between January 1, 2010 and March 31, 2018.
The search tool on the website of the Ministry of Economic Affairs was used to gather data about gas extraction because it was expected that the ministry publishes most of the information on this website. With the key words ‘Gaswinning Groningen’ and a date range the search resulted in 342 hits. It should be noted that some hits included multiple documents for instance a letter from the minister, a report and an addendum.
The website of Nederlandse Aardolie Maatschappij (NAM) was consulted for relevant data about the policy sector’s operations and future plans. This search resulted in 646 hits when the main key words were inserted. However, it was not possible to insert a date range. Also, one hits could contain multiple documents.
The research council Onderzoeksraad voor Veiligheid (OvV) was consulted. Four results were found when inserting ‘Gaswinning Groningen’. Only one result included multiple documents.
For the collection of relevant newspaper articles, the newspaper database LexisNexis was consulted. The selected newspapers were: Trouw, NRC Handelsblad, and De Volkskrant. These newspapers are qualified as legitimized newspapers in the Netherlands. For newspapers the word ‘Aardbeving’ (Earthquake) is added to the key words because the reactions from the residents will mostly be about the earthquakes and the safety related problems. The number of newspaper articles that is being published between 2010 and 2018 was 245 .
The Nederlandse Pubieke Omroep (NPO) was consulted for the collection of documentaries and televisions programs about the gas extraction. The NPO is financed by the government and broadcasts non-commercial television programs. The news broadcasts that will be used are those of the Nederlandse Omroep Stichting (NOS), Nieuwsuur, EenVandaag, De kennis van nu, 2Doc, Zembla, Andere tijden, Kruispunt, Brandpunt, and De monitor. Television programs will be selected based on multiple search terms because the searching tool is only able to find them based on just their title and the short description that has been given. The terms ‘Aardgas’ (Natural gas) and ‘NAM’ will be added to the list because broadcasts about those topics might mention the Groningen case. In total, 1034 videos were found.
The website of the House of Representatives (Tweede Kamer der Staten-Generaal) was used to find broadcasts of political debates. The statements made by politicians about the Groningen case are of interest. Political debates will be have the addition of term ‘NAM’, because that organization is a part of the policy sector the politicians might talk about. Forty-three videos of debates were found.
Data source Medium Search terms Number
of found documents
Policy documents - Ministerie van Economische
Zaken
- Nederlandse Aardolie Maatschappij
- Gaswinning Groningen 988
Inquiry reports - Onderzoeksraad voor Veiligheid - Gaswinning Groningen 4
Newspapers - NRC Handelsblad - Volkskrant - Trouw - Aardbeving gaswinning Groningen 245 Television programs - Nieuwsuur - De Kennis van nu - 2Doc - Zembla - EenVandaag - Andere tijden - Kruispunt - Brandpunt - De monitor - NOS Journaal - Gaswinning Groningen - Aardgas - Aardbeving Groningen - NAM 1034
Table 1. Used key words for data collection.
3.4 Data analysis
Content analysis is a suitable way of doing a longitudinal case study in a limited time to conduct the study because the studied documents are from a bigger time frame than the time frame from study itself. Moreover, documents have advantages over interviewing people because it prevents the occurrence of cognitive biases, especially those that arise after some time.
The used approach is deduction. The indicators were constructed based on the concepts described in the theoretical framework. Based on these categories and labels were made before the data analysis.
The content of the documents will be read, watched, and/or listened to, and relevant words, word groups, and terms will be labelled and categorized. Based on the number of labels and the kind of labels, a document will be considered relevant for the research or not.
3.5 Operationalization
The first concept was being operationalizes is institutional crisis. It is operationalized by statements or reports of the minister of Economic Affairs in which he declares that the gas extraction has to be reduced, or has to be terminated.
The minister is the responsible person for making the final decisions about the gas extraction in the Netherlands. His decisions are influencing the future of the whole policy sector because the reduction or the termination of gas extraction will directly and negatively affect the entire sector. This means that uncertainty about the direction and continuation of the policy sector occurs, which has been defined before as an institutional crisis. In other words, the minister’s statements that confirm the reduction or termination of the gas extraction will cause uncertainty about the sector’s future. Those statements can, therefore, be considered the indicators for institutional crises.
The second operationalized concept is the policy stream. The gas exploiter NAM is responsible for the daily operations of the policy sector because it executes the gas extractions. This means that the company has a direct influence on the performances of the sector. A neutral policy stream includes the continuation of the natural gas extraction and also doing this without harming people or nature. A decrease in the policy stream means that the daily operations of the company harm people. In this case is that endangering people’s safety.
The first research indicator for a decreasing policy stream is be the occurrence of a triggering event because a triggering event is often the event that exposes the decreasing institutional performances. Furthermore, statements or reports made by the policy sector about increased public safety risks are indicators of a decreasing policy stream. This shows that the policy sector is aware of decreasing performances and publicly acknowledges it. Also, inquiry reports stating a increased safte risk for inhabitants is a clear indicator of decreased perfomances. Lastly, statements about the ambition to change the performances regarding public safety are indicators of a decreased policy sector. If the policy sector publicly communicates about wanting to change its operations, it incicates that the policy sector acknowledges not to be satisfied with the performances.
The opinion of the inhabitants of Groningen form the opinion stream in the Groningen case. First, this stream could be identified by statements of inhabitants from the region about their dissatisfaction about the institutional performances. Negative opinions are for example the expression of distrust in the policy sector, disappointment or disbelieve of the occurring events. People can also state that they hope for improvements and different behavior. Also, the inhabitants can also voice their dissatisfaction about the situation that has been created by the policy sector. For example an unsafe, insecure, or unpleasant environment. The second indicator for the opinion stream will be the opinion of the media since they
is likely to take it on. The way that media are writing and reporting about certain issues are predictors of the opinion of the public. Negative reactions about the gas extraction and distrust in the policy sector will indicate that the opinion stream is decreasing.
The attention paid by politicians from both the governing and the opposition parties are the political stream in the Groningen gas extraction case. The political stream decreases when politicians start debating and questioning the current actions taken by the public institution regarding the natural gas extraction in Groningen.
Politicians could, for instance, publicly question the performances of the policy sector ask the minister to justify his decisions. Furthermore, they could object to the way the sector is acting or submit a motion of distrust to demonstrate that they do not have any faith that the minister is able to perform his duties accordingly.
Concept Indicator Data source
Institutional crisis - Statements or reports by minister that confirm reduction or termination of the gas extraction
- Reports Ministry of Economic Affairs
The policy stream - Occurrence publicly noticeable event - Statements from NAM or inquiry reports about changed public safety risks
- Statements from NAM about changing its operations regarding public safety
- Reports NAM - Reports Ministry Economic Affairs - Inquiry reports - Newspapers - NPO
The opinion stream - Expressed distrust in policy sector by the inhabitants of the province of Groningen - Negative opinion about the policy sector by public media
- Newspapers - NPO
The political stream - Politicians debating policy sector - Tweede Kamer der Staten Generaal Table 2. Operationalization of the three stream concepts
3.6 Validity and generalization
The internal validity of this research is being assured by triangulation. This means that the data is being validated by other documents from different sources.
The main criticism of case study research is that the results cannot be generalized (Yin, 2006). When the population of a case study is being identified correctly, this could be rejected. The goal is, therefore, not to generalize for a population with a wide variety of characteristics, but to generalize for a specific group with specific characteristics.
The Groningen gas extraction case can be generalized for possible institutional crises that were in the population for the case selection (in short): a public safety problem of hazard that was created by a governmental organization between 2000 and 2018. However, the generalization can be broadened to institutional crises that happened outside this time period, because not much will be different about institutional crises in history or the future. This research can be generalized because many different kinds of crisis management literature have been used and the consulted literature has been considered to be valid and grounded.
4.0 Analysis
4.1 Gas extraction in Groningen
In 1959, the first natural gas drillings in the province of Groningen were executed by Nederlandse Aardolie Maatschappij (NAM) (EenVandaag, 2014c; Havermans, 2015a). NAM is a joint venture of Shell and ExxonMobil (Kleinnijenhuis, 2018; Schwartz, 2018). The natural gas that the company was looking for was methane. Methane arises when plants and trees decay in the ground (Calmthout, 2014). This happened 300 million years ago under the surface of the province of Groningen. The gas is stored in a 200 meters thick layer of sandstone, underneath an one kilometer thick layer of halite and a two kilometers thick layer of rock (Calmthout, 2014; Knip, 2014).
The gas from Groningen differs from gas from other countries because it has low calorific value (Berg, 2018f). This means that the temperature of the gas is lower than the temperature of high calorific gas (Engels, 2013).
Natural gas is used in houses to heat up water or to cook (‘Aardbeving sterker bij vergrote gaswinning’, 2013; Luyendijk, 2013a). In industries, this gas is used to heat up water or other materials.
In 1963, the drillings for natural gas began in Slochteren, a small village in the Dutch province of Groningen (Berg, 2017b; Calmthout, 2014; EenVandaag, 2014c; Ellenbroek, 2013b; Giebels, 2015; Havermans, 2015a, 2015b; ‘Voorspelbare bevingen? Negeren en wegmoffelen’, 2013). The company started the gas drillings because it was expected that a small natural gas reservoir was going to be found (EenVandaag, 2014c; Schwartz, 2018).
After a few years of extraction, the reservoir appeared to have a volume of approximately 2.800 billion cubic meters of natural gas (2Doc, 2017; EenVandaag, 2014c; Havermans, 2015a; Schwartz, 2018). The gas field was located below the municipalities Appingedam, Bedum, Bellingwedde, Delfzijl, Eemsmond, Groningen, Haren, Hoogezand-Sappemeer, Loppersum, Menterwolde, Oldambt, Pekela, Slochteren, Ten Boer and Veendam (Kamp, 2014b).
From 1964 on, the gas was exported to other countries, including Belgium and France (EenVandaag, 2014c; Havermans, 2015a; Schwartz, 2018). In the 1970’s, the state’s share of the gas revenues was 85 percent (Schwartz, 2018). The Dutch state gains approximately 220 million euros from one million cubic meters gas (Berg & Giebels, 2014). The first few years of the extraction the Dutch state gained between two and four billion euros (Nieuwsuur, 2015a). This money went to social services, safety and public administration, and infrastructure (EenVandaag, 2014c; Nieuwsuur, 2015a).
In 1985, the government acknowledged that the ground under Groningen was lowering (DeKennisvanNu, 2015). The relation between the lowering ground and the damages on houses,
however, was questioned. People could not see how the gas extraction could cause the damages on houses. The lowering of the ground was gradual and was hard to notice.
The Royal Netherlands Meteorological Institute (KNMI) started from 1987 registering light earthquakes in the ground of the province of Groningen (EenVandaag, 2014c). These earthquakes were not noticeable for people, but were measured with meteorological equipment.
In the years after, the frequency and the intensity of the earthquakes increased (EenVandaag, 2014c). The earthquakes were not being linked to the gas extraction at first. However, in 1991, the KNMI did not rule out the possible relation between the gas extraction and the earthquakes (EenVandaag, 2014c; Havermans, 2015a).
The ministries of Economic Affairs and Infrastructure and Water Management appointed in 1991 a board to investigate the causes and consequences of earthquakes (Onderzoeksraad voor Veiligheid, 2015; Zembla, 2014). This board concluded in 1993 that the gas extraction could be a cause of the earthquakes in the region. NAM admitted in the same year that there might be a link between the gas extraction and the earthquakes (Berg, 2015b). The company acknowledged that an earthquake with a maximum magnitude of 3.3 on the Righter Scale was possible but it also claimed that the chance of such an earthquake was diminishable small (Zembla, 2014).
Induced earthquakes are earthquakes that occurred as a result of human activity (Onderzoeksraad voor Veiligheid, 2015). Earthquakes are caused by the changes in pressure in the sandstone layer (Knip, 2013, 2014). The extracted gas causes a decrease in volume in the sandstone layer and, thus, a lower in the sandstone layer. The weight of the layers on top cause the compaction of the sandstone layer. When compaction occurs along a fault line, an earthquake can occur.
In the province of Groningen, the earthquakes are triggered relatively low under the surface (Onderzoeksraad voor Veiligheid, 2015). Also, the structure of the ground allows for further oscillation. Moreover, the province is located on a fault line of earth’s crust, which is the reason why the earthquakes are noticeable above the ground.
In 2003 the first noticeable earthquake struck the village of Loppersum (EenVandaag, 2014c). The earthquake had a magnitude of 3.0 on the Richter Scale. At first glance, the biggest problem that the earthquakes caused seemed to be the damages on buildings (Ellenbroek, 2013b; Es, 2013e; Luyendijk, 2013d).
In 2013, both the KNMI and Department for Governmental oversight on Mining and Ernergy production (SodM) reported that the maximum magnitude earthquakes is higher than was predicted before. Instead of 3.9 on the Richter Scale, the maximum magnitude of earthquakes that occurred could have a magnitude of four or five (Ellenbroek, 2013b; Es, 2013e; Luyendijk, 2013b). The chance that such an
by SodM on one out of 14 (Schreuder, 2013). The only way to reduce the number and the magnitude of earthquakes was to extract as few as possible, and as slower as possible.
SodM qualified the gas extraction region in 2018 as an unsafe area and indicated it with the highest risk level. The safety level for the inhabitants is much lower than it is for citizens in rest of the country (Berg, 2018c, 2018e, 2018f).
The Gas Builing is the corporation of all the companies and organizations involved in the exploitation, transportation, distribution, and sale of natural gas in the Netherlands (Onderzoeksraad voor Veiligheid, 2015). The parties within the Gas Building are fully aware of the interests of each other (Onderzoeksraad voor Veiligheid, 2015). Their aim is to serve all these interests through overall consensus during decision making. The culture within the Gas Building is one that tries to extract as much as possible. Also, they want to attract as little as possible attention from people outside the organisation because the involved parties do not want to jeopardize the gas extraction in any way.
The Gas Building is made up of less than ten people who are closely connected (Onderzoeksraad voor Veiligheid, 2015). Also, most people in the collaboration have had previous positions within organizations that are closely connected to the gas extraction in Groningen. This interconnectedness of people makes it difficult for contribution and discussion of critical opinions (Onderzoeksraad voor Veiligheid, 2015). An investigation in the Gas Building concluded that the organization could improve itself by acting more like a semi-public organization and by being more transparent (Onderzoeksraad voor Veiligheid, 2015).
Officially, NAM is the license holder for the gas extraction and, therefore, responsible for the safety of involved people (Onderzoeksraad voor Veiligheid, 2015). However, the existence of the Gas Building makes the responsibility a shared one. However, all involved parties and people considered NAM to be the only responsible party.
The gas extraction in the province of Groningen, as a whole, is the responsibility of the ministry of Economic Affairs (Onderzoeksraad voor Veiligheid, 2015). Within the ministry, multiple departments are responsible for different aspects of the gas extraction. This distribution of tasks makes it possible that counter opinions are brought in and discussed. Safety, however, was not one of the topics that had a priority in one of the departments.
Eighty-eight percent of the gas revenues goes to projects in the west of the country and only one percent goes to projects in the province of Groningen (EenVandaag, 2014c; Nieuwsuur, 2015a). Between 1959 and 2014, the Dutch state earned approximately 265 billion euros from gas extraction. The revenues from the gas extraction are on average about ten percent of the Dutch state’s income (Onderzoeksraad voor Veiligheid, 2015).
4.2 Indicators institutional crisis
Concept Indicator
Institutional crisis - Statements or reports by minister about reducing or stopping the gas extraction Table 3. Indicators institutional crisis
The village of Huizinge was struck by an earthquake with the magnitude of 3.6 on the Richter Scale on 16 August 2012 (NAM, 2016; Onderzoeksraad voor Veiligheid, 2015). This was the heaviest induced earthquake ever in the province (NAM, 2012a; Onderzoeksraad voor Veiligheid, 2015).
Right after the earthquake, the public debate started about reducing of completely terminating the gas extraction in the Groningen gas field (Es, 2013c). The then minister of Economic Affairs, Henk Kamp, regretted all that was happening in the province. However, he believed that it was too premature to reduce the gas extraction in the Groningen gas field (‘Aardbeving sterker bij vergrote gaswinning’, 2013; EenVandaag, 2013; Kamp, 2013; Luyendijk, 2013a). He argued that reducing the gas extraction would reduce the state’s revenues which had big consequences for the whole county’s budgets. Minister Kamp also stated that seven million Dutch houses had to be provided with natural gas. Moreover, contracts with importing gas countries restricted him to reduce the extraction (Luyendijk, 2013a, 2013c; Zuidervaart, 2013). Therefore, minister Kamp decided to ignore SodM’s advice to reduce the gas extraction and he let the gas extraction continue.
As a result of the occurring earthquakes, minister Kamp and NAM did want to reduce the earthquakes and increase the safety of the residents in the province by investigating what risks existed (‘Kamer wil onderzoek naar risico bevingen’, 2013). The minister Kamp was planning on making a decision about the future of the gas extraction after he received the results of eleven investigations that were being conducted (Ellenbroek, 2013c; Es, 2013d; Luyendijk, 2013a). Despite the statements about possibly reducing the extraction in the future, minister Kamp allowed NAM to increase the extracted amount of gas extraction in 2013 (Zembla, 2014).
As a result of the Huizinge earthquake in 2012, minister Kamp experienced pressure to reduce the gas extraction in the province of Groningen. The minister responded to the pressure by investigating the grounds, gas extractions and earthquakes in the region. He was firmly convinced that reducing the extraction would not be a suitable solution for the moment. Thus, there was no sign of an institutional crisis in 2012 and 2013.
advised in 2014 to stop the gas extraction in Loppersum because the gas extraction caused a high risk of heavy earthquakes (Es, 2014; Luyendijk & Niemantsverdriet, 2014).
In the same year, Minister Kamp went to Loppersum to inform the local government about the new plans for the gas extraction in Groningen (Luyendijk & Sterk, 2014). His motivation for wanting to improve the safety in the province of Groningen was the earthquake that occurred at the beginning of 2014 (Kamp, 2014b). The most important change meant that NAM could, from then on, extract a maximum of 42.5 billion gas each year from the Groningen gas field (Luyendijk & Niemantsverdriet, 2014; Luyendijk & Sterk, 2014). Also, the extraction in five wells around the village of Loppersum would substantially be reduced (‘De maatregelen: Groningen: 1,2 miljard Groningen’, 2014). The argument to not further lower the amount was that it would not be able to supply the minimum required amount to the houses in the country (‘Kamp kan maar drie jaar vooruitkijken’, 2014). The minister explained that he followed the advice from SodM by constructing these plans but that there still would be a substantial risk of earthquakes and, therefore, consequential safety risks (EenVandaag, 2014d). He, however, believed that the people’s trust in the region would be restored by the measures he took (Es, 2014; Luyendijk & Niemantsverdriet, 2014).
Furthermore, minister Kamp decided to open a 1.2 billion-euro fund to improve the living conditions in the whole region (Luyendijk & Niemantsverdriet, 2014; Luyendijk & Sterk, 2014; Nieuwsuur, 2015a). This money was, among other things, meant to enforce damaged buildings and prevent other buildings from becoming unsafe (‘De maatregelen: Groningen: 1,2 miljard Groningen’, 2014). Lastly, more and better tests would be conducted to register changes in the ground (‘Kamp kan maar drie jaar vooruitkijken’, 2014).
Later on, the minister announced that he decided to audit around 11.000 buildings and enforce 3.000 houses (Kamp, 2014b). Also, new building guidelines were made to make sure that new buildings were build safe for possible coming earthquakes. Investigation into the best ways of building safe constructions continued.
As a result of the continued pressures and SodM’s advice, the minister decided to reduce the gas extraction in the region. Also, the extraction from specific dwells were going to be reduced substantially. These decisions of the ministry showed that there was uncertainty within the policy sector which could be identified as an institutional crisis.
Minister Kamp decided in 2015 to reduce the gas extraction in the province again (Berg, 2015a). From then on, NAM was allowed to extract a maximum of 39.4 billion cubic meters gas each year. A few months later, he decides that the first half of the year, only 16.5 billion cubic meters was going to be extracted (EenVandaag, 2015a; Luyendijk & Niemantsverdriet, 2015a). The amount that was going to be extracted the second half year, would be decided later (Luyendijk & Niemantsverdriet, 2015a). This
way, minister Kamp was hoping he could reduce the yearly amount with almost 4 billion cubic meters of gas.
Although SodM advised to extract substantially less than 33 billion cubic meters, the minister was planning on extraction a maximum of 30 billion cubic meters because he believed that that was enough to promise safety for the inhabitants of Groningen (Nieuwsuur, 2015a). He explained that his decision took the safety of the residents in Groningen into account, as well as the minimum amount that is needed to heat up Dutch homes. After a ruling of the highest court of the country minister Kamp was forced to reduce the gas extraction to 27 billion cubic meters (Nieuwsuur, 2015b).
It was made clear by the ministry that a new factory was going to be built to transform imported gas into gas for Dutch houses (Nieuwsuur, 2015b). Furthermore, less gas was going to be export other countries. Minister Kamp apologized in 2015 for the lack of interest for the safety of the Groningen residents (Berg, 2015c).
In 2015, the OvV published the report that the minister asked for in 2014. The OvV investigated whether the safety of the residents of the earthquake region have been taken into account during the decision making process of the gas extraction (Luyendijk, 2015a; Onderzoeksraad voor Veiligheid, 2015; ‘Raad: gasbolwerk moet hervormd’, 2015).
The OvV argue that the government is responsible to oversee multiple interests, therefore, the fulfilment of these interests should be strictly separated (Onderzoeksraad voor Veiligheid, 2015). Another task is the supervision of the information gathering of the exploiting companies.
The OvV had multiple recommendations for the minister to improve the decision making process in the future (Onderzoeksraad voor Veiligheid, 2015). First, the OvV believed that different ministries should be involved in the decision making process about activities in the ground. Second, the interests of the residents should have beem taken into account by involving local government. Lastly, the SodM should have an independent place in the ministry of Economic Affairs.
Another investigation, concluded that the gas companies are very powerful and the ministry of Economic Affairs did not show enough resistance to the companies (EenVandaag, 2015b). The administrators were not assertive or knowledgeable enough to resist the powerful gas companies.
Minister Kamp reduced the gas extraction multiple times during 2015. The maximum amount that NAM was allowed to extract went from 42.5 to 27 million cubic meters of natural gas. This extreme reduction had most likely an impact on the policy sector. The year 2015 could, therefore, be
