Struggle against water

0

Struggle against water

Research on the governance model used in water-disaster

management in the Netherlands

Amber van Schie, s1118692 Master: Crisis and Security Management

Supervisor: Dr. J. Matthys Second Reader: Dr. E. Devroe

1 Table of Content

List of Tables and Figures... 3

Chapter 1: Introduction ... Fout! Bladwijzer niet gedefinieerd.4 1.1 Research goals and questions ... 6

1.2 Relevance of the research: academic and societal ... 7

1.2.1 Academic relevance ... 7

1.2.2 Societal relevance ... 8

1.3 Structure ... 9

Chapter 2: Theoretical Framework ... Fout! Bladwijzer niet gedefinieerd.11 2.1 Crises, safety and security ... 11

2.2 Governance as a concept ... 13

2.2.1 Waves of reform ... 14

2.2.2 State structure of the Netherlands ... 15

2.3 Forms of governance ... 18

2.3.1 Hierarchy ... 19

2.3.2 Market ... 20

2.3.3 Network ... 22

Chapter 3: Research Design ... Fout! Bladwijzer niet gedefinieerd.26 3.1 Case Study ... 26

3.2 Data collection, Reliability and Validity ... 29

3.3 Operationalisation ... 31

Chapter 4: Analysis ... Fout! Bladwijzer niet gedefinieerd.39 3.1 The struggle against water... 39

4.1 Evacuation of Rivierenland (1995) ... 42

4.1.1 Goals ... 45

4.1.2 Normative Basis ... 47

4.1.3 Role of Politicians and Civil Servants ... 48

4.1.4 Means of Communication ... 49

4.1.5 Method of Conflict Resolution ... 51

4.1.6 Degree of Flexibility ... 52

4.1.7 Amount of Commitment among the Parties ... 53

4.1.8 Tone or Climate ... 54

2

4.2 Quay rupture in Wilnis (2003) ... 56

4.2.1 Goals ... 58

4.2.2 Normative Basis ... 59

4.2.3 Role of Politicians and Civil Servants ... 60

4.2.4 Means of Communication ... 62

4.2.5 Method of Conflict Resolution ... 62

4.2.6 Degree of Flexibility ... 63

4.2.7 Amount of Commitment Among the Parties ... 64

4.2.8 Tone or Climate ... 65

4.2.9 Actor Preferences or Choices ... 65

4.3 Comparison of analysis ... 67

Chapter 5: Conclusion... Fout! Bladwijzer niet gedefinieerd.71 5.1 Conclusion ... 71

5.2 Reflection on results ... 72

Chapter 6: Bibliography ... Fout! Bladwijzer niet gedefinieerd.74 Academic: ... 74

Books: ... 75

Newspapers:... 77

3 List of Tables and Figures

Tables:

Table 1 – Graphic representation of phases and steps of a crisis……….…12

Table 2 – Governance models………..24

Table 3 – Operationalisation table………31

Table 4 - Summary of the analysis………..70

Figures: Figure 1 - Graphical and simplified representation of Dutch state structure……….. 15

Figure 2 – Graphic representation of models of governance………...25

Figure 3 – Graphic representation of government structure during the water-disaster in Rivierenland……….44

Figure 4 – Graphic representation of relationships between government actors during the water-disaster in Rivierenland……….……47

Figure 5 – Graphic representation of government structure during the water-disaster in Wilnis………...……58

4

Chapter 1

Introduction

One of the most important tasks of a government is to provide safety and security for the people living in the governed territory (Rijksoverheid, 2012; Garland, 1993, p448). Crisis and security management has changed much in the past thousand years though (Helsloot & Ruitenberg, 2004; Rosenthal & Kouzmin, 1993; Crawford, 2006). Humans once used to be truly self-reliant and every man was for himself, creating a ‘state of nature’ without notions of right or wrong, or justice and injustice (Hobbes [1651] in Mizzoni, 2010, p65). However, this state of nature with unlimited freedom caused constant conflict and strife, resulting in people not getting the security and stability they want (Hobbes [1651] in Mizzoni, 2010, p65). Therefore people entered mutually beneficial contracts with others such as leaders or clans, exchanging freedom and loyalty for safety and security (Hobbes [1651] in Mizzoni, 2010, p65-66). According to Hobbes, humans enter these social contracts because they realize that their chances of getting what they want are higher when they work together (Hobbes [1651] in Mizzoni, 2010, p66). After people entered these mutually beneficial social contracts with others, they were no longer self-reliant for their security (Hobbes, 1651, p90-92). Leaders used armies to protect their citizens, and later cities and villages in the Netherlands were often protected by a voluntary city guard or ‘schutterij’ (Sickesz, 1864, p3). In the years that passed, many types of governance have been used and dismissed.

Since the Second World War, crisis and disaster management was based on military ideas and the military doctrine of C3: Chaos, Command and Control (Helsloot & Ruitenberg, 2004, p104). This implied that crises and disasters cause chaos, which can only be resolved with command and control (Helsloot & Ruitenberg, 2004, p104). According to this model, only military command structures are able to cope, and civil institutions are left out (Helsloot & Ruitenberg, 2004, p104). Since the nineteen-seventies, though, there has been a shift from first the C3 model to civil defence operations with aid organisations, and then from civil defence to disaster management (Helsloot & Ruitenberg, 2004, p104). Where citizens were seen as passive and helpless in the past, today resilience and citizen response have become more important (Chen et.al., 2013, p130).

5 The King’s speech in 2013 stated that society was becoming a ‘participatiestaat’ due to the fact that citizens are more articulate and independent, and the necessity of the government to retreat (Rijksoverheid, 2013). The trend of a smaller government and less governmental interference could also be seen in the Netherlands where the Dutch government started the decentralisation in many policy areas, and asks of those who can, to take responsibility for their own life and environment (Rijksoverheid, 2013; Ministerie van Binnenlandse Zaken en Koninkrijksrelaties [BiZa], 2013, p8). Examples of this in the area of safety and security are ‘Buurt Bestuurt’, which means that the neighbourhood governs itself in cooperation with the local municipality and police and decide what issues should be resolved, ‘Burgernet’ where citizens are being used to help the police solve recently committed crimes, and ‘BIN’ which is a network consisting of locals, the police and the local municipality that aims to improve safety, security and liveability by using communication though e-mails (http://www.rotterdam.nl/buurtbestuurt; https://www.burgernet.nl/; http://www.ccv-veiligheidsbeleving.nl/praktijkvoorbeelden/buurt-informatie-netwerk-bin).

This corresponds with the trend of the ‘network society’ in which we rely more on horizontal networks empowered by new information and communication strategies, instead of vertical hierarchical social structures (Castells, 2000, p5; Dupont, 2004, p76). States are no longer the central power-holding institutions in all policy areas because of a loss of legitimacy and credibility due to communication, information, scandals and global flows of wealth (Castells, 2000, p14). States, however, have not disappeared but are becoming ‘network states’ made of “a complex web of power-sharing, and negotiated decision-making between international, multinational, national, regional, local, and non-governmental, political institutions” (Castells, 2000, p14). To regain legitimacy, a process of devolution of power, decentralizing responsibilities and resources has been started (Castells, 2000, p14). This process could not have been possible without the use of new information and communication technologies (Castells, 2000, p14). This is also happening in the Netherlands, and the Ministry of Internal Affairs presented a two-year plan in 2013 called ‘Vision on an Open Government’ that sketched a situation in which the government should share more information with its citizens in order for them to be more active (Ministerie BiZa, 2013, p8). The government has become more horizontal instead of vertical in many areas, and there are pleas for a ‘do-democracy’ where citizens become more active and involved in governing (Ministerie BiZa, 2013, p8). Even though there is a tendency towards a ‘do-democracy’ and network-society in many areas, examples such as ‘Burgernet’ and ‘Buurt Bestuurt’ show that

6 citizens are not being used in every policy field and are often solely used for information-purposes (http://www.rotterdam.nl/buurtbestuurt; https://www.burgernet.nl/; http://www.ccv-veiligheidsbeleving.nl/praktijkvoorbeelden/buurt-informatie-netwerk-bin). Redefining the relationship between the government and its citizens is needed and new forms such as networks should be and are considered, but this does not mean that a network form of governance is fitting in every policy area (Ministerie BiZa, 2013, p8).

One specific type of disaster is a well-known threat in the Netherlands: floods (Rijksoverheid, 2014). And also in the case of water-disaster management, the government is trying to influence its citizens into becoming more resilient and prepared but is also looking at a possible collaboration with citizens in the future (Rijksoverheid, 2014; Opdam & Van Kessel, 2013). Jan-Hendrik Dronkers, director of Rijkswaterstaat, states that the Netherlands is the safest delta in the world, however the threat of flooding is still there because nature is unpredictable1. The sea-level rises due to climate change and the Netherlands often experiences heavy rainfall, and the land is not always capable of processing that amount of water in a short period of time1. Several cases, of which the flooding in 1953 in Zeeland is generally well-known, show that the Dutch government and its citizens should always be prepared for a water-disaster1. The Netherlands and water have a complicated relationship: water offers us many benefits, but the costs are often high1.

1.1 Research goals and questions

This research aims to help fill the knowledge gap about the coordination between government parties in water-disaster management in the Netherlands, and could be the beginning of more research on the role of government actors, private actors and citizens in water management. The main focus of this research is to explore the relationship between the different government actors active in the field of water-disaster management, and investigate the governance model that is being used by the Dutch government. This research will be guided by the following main question:

“What governance model best describes the coordination between

the Dutch government actors in the field of water disaster management?”

7 The governance models will be described in the theoretical framework, as is the concept of governance itself. But to answer the main question, a structural analysis of two cases regarding water-disaster management has been conducted by looking at the relationship between the government actors involved.

1.2 Relevance of the research: academic and societal

1.2.1 Academic relevance

The study of public administration focuses, amongst other things, on creating an effective and efficient government by creating new policies (Rutgers, 2004). Through all the years, there have been many ways of looking at the government and many ways of governing (Denhardt, 2011). The Dutch state changed from a facilitating ‘Verzorgingsstaat’ to a ‘Participatiestaat’ where citizens had to share responsibilities with the government, there was a trend from government to network-governance, and public administration changed from being influenced by economic concepts to a new public service that “is built on the idea of the public interest, the idea of public administrators serving citizens and indeed becoming fully engaged with those they serve” (Denhardt, 2011, p197). Thus there has been much research about governance, public policy and civil participation and its benefits for the government, but also some about using existing and natural networks, collaborations and relationships with citizens for government purposes (Warner & Frerks, 2010; Helsloot & Ruitenberg, 2004; Dynes, 1994).

The Dutch government wants to know whether cooperation with other actors and working in networks is possible in all parts of the security-chain, but also what instruments should be used to coordinate this and how accountability and responsibility fit into this (Opstelten, 2014). The commitment and cohesion is high in the Netherlands, and the many social collaborations and organisations that exist in our community can and should be used according to some scholars (Rijksoverheid, 2013; Warner & Frerks, 2008; Van Caem, 2008). However, before research into implementing new forms of governance is possible and how this should be achieved, there is a need for research into the current coordination between government actors in each policy field (Slomp, 2012, p7). The world around us is always changing, including the government and society itself, and by looking at the past and present we can learn from our successes and mistakes (Slomp, 2012, p7). For example as Lynton Caldwell said “Its relevance to the theory and practice of administration depends upon its

8 instructive value. Its ultimate justification depends upon its success in utilizing historical experience to contribute to the solution of administrative problems” (Caldwell, 1955, p458). Water-disaster management policies and plans are the result of floods and other disasters, and this combination of looking at the past and looking at water-disaster management policies and plans themselves is important to be able to cope with disasters in the future (Slomp, 2012, p11). It is important to look at the way how and reason why these policies and plans were created in order to understand how they work in the present (Wagenaar, Kerkhoff & Rutgers, 2011, p13).

Furthermore, this research will add to the ever-growing pile of knowledge about public policy and public administration. It is important to know how the policies and plans for water-disaster management were created and how they function during a disaster for evaluation purposes, but also for general knowledge purposes and future use (Wagenaar, Kerkhoff & Rutgers, 2011, p13). There seems to be a trend towards network-governance in the world, but is this also happening in the Netherland; or is there a tendency towards a hierarchical model of governance? These are questions that can be answered through this research, and these answers can also benefit other research involving the Dutch government and Dutch governance models. Research focusing on discovering the governance model used in Dutch water-disaster management is new, and the results can prove useful when looking at (disaster) management in other areas.

1.2.2 Societal relevance

Every civilian wants to feel and be safe, which is not always the case because of a high crime rate in the neighbourhood, or because of events such as the terrorist attacks in Paris on January 7th 2015 and November 13th 2015 (Huys, 1999; Klompenhouwer, 2015; Nieuws.Marokko, 2015; Rosman & Van Mersbergen, 2015). Safety and security are costly and difficult matters that governments cannot always solve on their own and complete elimination of feelings of unsafety is not possible (De Koning, 2012). In the case of water-disaster management, for example, the government creates a National ‘Water-plan’ every six years and implements different prevention methods, but completely eliminating the threat is impossible (Schreuder, 2014; Rijksoverheid, 2015). But it is still important for a government to improve the feelings of unsafety in society and research the possibilities in order to do so, because safety and security is an important task of the government and the ability to live without fear is fundamental to the exercise of political and other rights (Goldsmith, 2003, p3).

9 Therefore, governments should research the way in which it governs, for society may benefit from either using networks, collaborations and relationships with citizens; a hierarchical structure that focuses on procedures and laws; or a market form of governance that is very flexible and has output-based efficiency (Steyaert & Kwekkeboom, 2013; Powell, 1990, p301). It depends on the policy field what governance model suits best, because “countries and policy areas differ and require co-ordination mechanisms tailored to suit their specific needs”, and it is important to look at what governance model in being used in each policy field and whether this has been used with good reason (Organisation for Economic Co-operation and Development [OECD], 2005, p273).

Dutch citizens are, according to research done in 2016, not afraid of floods because they believe that they are well-protected by the Dutch government and believe they will be safe in the years to come (Ipsos, 2016). This research was set up because of the 100-year anniversary of the Zuiderzee-disaster in 1916’, and shows that 66% of the Dutch population underestimate the severity of the threat that floods pose (Ipsos, 2016, p1). The Dutch people are not prepared for a flood and don’t know that 60% of the Netherlands is susceptible to flooding (Ipsos, 2016, p1-4). However, this attitude and this lack of knowledge is incorrect because the threat of floods is only increasing due to climate change, tectonic subsidence and urbanisation (Rijksoverheid, 2015). A disaster like the flood in 1953 would mean billions of euros of damage, thus prevention of a water-disaster is very important for the Dutch society (Schreuder, 2014). Therefore the Dutch society benefits from research into water-disaster management, because this will ultimately lead to better water-disaster management and as such better protection for the Dutch society.

1.3 Structure

The first part of this thesis has introduced the subject of disaster management, and specifically the field of water-disaster management. The Netherlands is both blessed and cursed by water; it has given us economic advantages and has made us experts in flood-control and water-protection such as dams, but the water has also cost us many lives and resulted in a lot of damage. There has been a trend from government to network-governance, and due to all these changes and the fact that the Netherlands is always threatened by the water, it is useful to look at the coordination between the different actors of the Dutch

10 government involved in the field of water-disaster management and discover the governance model that is being used in this specific field.

In order to answer the main question posed in the introduction, it is important to first explore several theories concerning the concepts of safety, security, crisis and crisis management. These concepts will be thoroughly discussed and conceptualised in the following chapter, which will lay out a theoretical framework. Furthermore, this chapter will discuss several theories about the three main governance models: market, hierarchy and network and the current state-structure of the Netherlands. This chapter will introduce research by academics such as Baldwin, Chen, Garland, Milasinovic & Kesetovic, and Powell, and offer a clear graphic representation of the several aspects and characteristics for each governance model.

As with all research, choices had to be made in order to create focus and these will be explained in the chapter of research design. Furthermore, this chapter will explain how the research question will be answered, how the necessary data was gathered and how this will be used in the analysis. This chapter will also include the operationalisation and offers a graphic representation of the operationalisation and the units of analysis.

The fourth chapter is split into three sections: two sections about the two chosen cases and a third section about the general results of analysis. Both sections about the cases start with a general overview of the cases, namely the evacuation of Rivierenland in 1995 and the quay rupture in Wilnis in 2003, which is followed by an in-depth analysis of the gathered data and combine this with the information in the second chapter about the governance models. This combination of all the data and information will lead to an answer to the main-question. The graphic representation of the governance models in the operationalisation table on page 32 to 38 will provide the structure for the analysis.

Finally, the sixth and final chapter will conclude this research, answer the main research question and offer some recommendations for further research.

11

Chapter 2

Theoretical Framework

The modern world is increasingly being hit by crises and disasters (Rosenthal & Kouzmin, 1993, p1). Crises are defined as events that invoke danger, insecurity, a sense of emergency and alarm, and have a large impact (Rosenthal & Kouzmin, 1993, p1). Crisis management is therefore needed in all phases of a crisis; prevention, preparation, alleviation and recovery (Milasinovic & Kesetovic, 2008, p180). Whereas the state used to be responsible for everything, nowadays many states have recruited the help of private, public and civil entities in all these phases (Chen et.al., 2013, p135-136).

New crises are one of the many reasons for administrative reform, but are not the only ones, because advances in science and technology, research on administrative reform and modernization also draw attention to new programmes, management ideas and techniques (Pollit & Bouchaert, 2011, p6&14). Each state is different and has developed different, and as such they have different forms of governance, though there are some similarities (Pollit & Bouchaert, 2011). Thus there are many ideas about governing and as such there are many types of public management reform, though these can all be subdivided in three main models: market, hierarchy and network (Pollit & Bouchaert, 2011, p21; Kenis & Provan, 2007).

This chapter will, first, include the conceptualisation of safety and security, and disaster management. Theories by scholars such as Baldwin, Chen, Garland, Milasinovic & Kesetovic, Pollit & Bouchaert, and Powell will be discussed to provide the definition of these concepts and to provide an overview of the changes in governing in the past years. Finally the three main models for governance, market, hierarchy and network, will be discussed.

2.1 Crises, safety and security

Security can be defined as “the absence or low probability of threats or damage to acquired values” or “the condition of being protected against danger or loss, achieved through the mitigation of adverse consequences associated with the intentional or unwarranted actions of others” (Baldwin, 1997; Talbot & Jakeman, 2009, p5). Whilst security focuses on intentional acts, safety concerns protection against danger or loss by unintentional actions, such as natural disasters (Talbot & Jakeman, 2009, p5). Thus a safety or security risk is “any event that could result in the compromise of organizational assets such as loss, damage, risk or

12 harm” (Talbot & Jakeman, 2009, p7). However, whilst these will be the definitions used in this research, the concepts of safety and security remain subjective and difficult to define (Baldwin, 1997).

This difficulty with defining is also the case for the concept of crisis, because a crisis is largely perceptual (Canel & Sanders, 2010). “A crisis can be viewed as the perception of an event that threatens important expectancies of stakeholders and can impact the organization’s performance. Crises are largely perceptual. If stakeholders believe there is a crisis, the organization is in a crisis unless it can successfully persuade stakeholders it is not” (Coombs, 2010, p6). Thus, social scientists have been using the term crisis a lot in the past years and in many different ways (Milasinovic & Kesetovic, 2008, p168). It has been used to point out a decisive moment in time whereupon the development of a thing or situation could proceed in a positive or negative manner (Milasinovic & Kesetovic, 2008, p168). Yet, the concept of a crisis is mostly used for negative events, or events that are unexpected, unwanted, uncertain, unpredictable and unthinkable (Milasinovic & Kesetovic, 2008, p170). Most scholars use definitions that include concepts such as unpredictable, negative consequences, threat, damage, small probability, unclear causes and effects, and acute decision making (Milasinovic & Kesetovic, 2008, p170-171). So, the main elements of a crisis are a time pressure, a threat to a social system or organization, uncertainty and pressure to make decisions (Milasinovic & Kesetovic, 2008, p171). Therefore a crisis is “a serious threat to the basic structures or fundamental values and norms of a social system, which, under conditions of time pressure and very uncertain circumstances, demands the bringing of critical decisions” (‘t Hart in Milasinovic & Kesetovic, 2008, p171).

A crisis can be divided into three different phases: the pre-crisis phase, the actual crisis, and the post-crisis phase (Coombs, 2010, p8). The pre-crisis phase includes signal detection, prevention and preparation for a crisis, for example on an individual scale by having smoke detectors in your home or on a national scale by having policies concerning safety rules or action plans during crises or the sirens that are located in each neighbourhood (Coombs, 2010, p8). During the actual crisis it is all about response: recognition of the signals, processing these signals and respond to them by taking immediate action (Coombs, 2010, p8). The third and last phase, post-crisis, includes all actions after the situation has returned to normal such as providing follow-up information to stakeholders, cooperating with investigations and learning from the crisis (Coombs, 2010, p8). These three phases can be divided into five important steps: (1) prevention, (2) preparation [& signal detection], (3) response, (4) recovery and (5) learning (see figure 1)(Coombs, 2010, p22). Crisis

13 management focuses on the first phase of a crisis, pre-crisis, and the first and second step of the cycle, and as such includes all actions preparing and readying for crises in order to be able to properly respond to any crisis (Coombs, 2010, p17).

Table 1: Graphic representation of phases and steps of a crisis

2.2 Governance as a concept

There have been many changes in and around the government in the past several decades, for example new topics on the political agenda and new perspectives on how the government should govern (Nelissen, 2002, p5; Pollit & Bouchaert, 2011). New topics and problems require adaptations in governing, and as such new visions on governance have emerged (Nelissen, 2002, p5; Pollit & Bouchaert, 2011). The government has been renewed with perspectives that disassociate themselves from a monocentric approach by a single actor and plea for cooperation between multiple parties; which includes phenomena like deregulation, decentralization, retreating of the government, governance from a distance, privatization, liberalization, support enhancement, interactive policy and improvement of client orientation (Nelissen, 2002, p5-6). This change has often been described as a shift from government to network-governance (Bellamy & Palumbo, 2010). Network-governance is “a new way of steering society” and it is a popular but imprecise term (Rhodes, 1996, p652). Governance is not a synonym for government, “rather governance signifies a change in the meaning of government, referring to a new process of governing; or a changed condition of ordered rule; or the new method by which society is governed” (Rhodes, 1996, p652-653).

The concept of governance is also being used to describe the way of governing of a government, and as such the three main governance models: market, hierarchy or network (Rhodes, 1996, p653). These structures for allocating resources, exercising control and coordination are not intrinsically good or bad; it is up to the government to choose which structure works best in which conditions (Rhodes, 1996, p653; Pollit & Bouchaert, 2011, p214-215). All states are different in many ways, for example cultural, political, structural, functional, ethnic, and administrative differences (Pollit & Bouchaert, 2011, p48-49). Research by Pollit & Bouchaert in 2011 shows some of the differences between fifty different

14 states in the world, and also shows the type of governance affiliated with each state (Pollit & Bouchaert, 2011, p50). However, a state might be generally unitary and legalistic, some specific subjects might need another type of governance and other forms of governing are tested because choosing the right form of governance depends on the conditions (Rhodes, 1996, p653). Therefore not only the view on governance in general has changed, but also the views on which governance structure is to be used in the field of crisis management have changed as well (Rhodes, 1996, p653). The state is responsible for the safety and security of its inhabitants, and can enforce this due to its monopoly on violence (Hobbes, 1651, p106). However, instead of this more hierarchical structure, the state could also use a market structure and use security officers from the private sector to provide safety and security (Devroe & Terpstra, 2015).

2.2.1 Waves of reform

Thus governance is a difficult concept and the views on governance have changed a lot in the past (Rhodes, 1996, p653). Before the 1960s, administrative reform was seen as a technical or legal issue instead of a political or economic matter, meaning it was just a question of organizational and procedural changes with little international debate (Pollit & Bouchaert, 2011, p5-6). During the years that followed, three waves of administrative reform caused many changes in states all over the world (Pollit & Bouchaert, 2011, p6). The first wave of administrative reform was caused by high modernism and advances in science and technology; and both held the “promise of a more rational designed set of public policies and institutions” (Pollit & Bouchaert, 2011, p6). The dominant discourse involved rational and hierarchical planning and cost-benefit analyses, which were all possible due to the advances in science and technology (Pollit & Bouchaert, 2011, p11).

Administrative reform is described as waves of reform since the new ideas of reform overlie, but do not replace or sweep away everything of the last wave (Pollit & Bouchaert, 2011, p12). This can be seen in the second wave that was initiated by the global economic disturbances and the belief that the governments had become constraining on both citizens and employers, but also overloaded, ineffective and unaffordable (Pollit & Bouchaert, 2011, p6). This resulted in the creation of new fragmented public sector organizations, but the existing organizations from the first wave were not erased (Pollit & Bouchaert, 2011, p6). The government had to become more business-like and efficient to save money, but also “oblige public bureaucracies to act more responsively towards their citizen-users” (Pollit &

15 Bouchaert, 2011, p6). To fulfil these goals, more single- or few-purpose organizations were created and less of the large multi-purpose forms (Pollit & Bouchaert, 2011, p7). These single- or few-purpose organizations had explicitly defined sets of goals and targets to make them more efficient and affordable, but there were some coordination problems and problems with political accountability (Pollit & Bouchaert, 2011, p7). There were many different public sector organizations and trying to get them to collectively pursue same policy objective, has proven to be rather difficult (Pollit & Bouchaert, 2011, p7). These coordination problems also resulted in problems with political accountability since the ministers are responsible for the actions and policies of these organizations, however these organisations have proven to be rather difficult to control and coordinate (Pollit & Bouchaert, 2011, p7).

Reform remained important in the years that followed since not everything was going well during the second wave, but the main vision on administrative reform changed (Pollit & Bouchaert, 2011, p7). The third wave introduced the concepts of partnerships, joined-up government, trust, transparency, but still keeping an eye on the main concepts from the second wave: efficiency and quality (Pollit & Bouchaert, 2011, p7). The cause for this change of vision is unknown, but it is possible that the problems with the fragmented public sector organizations was one of the triggers (Pollit & Bouchaert, 2011, p7). Besides joining-up and inter-service coordination, e-government or new information technologies became an important concept (Pollit & Bouchaert, 2011, p7). E-government applications could “revolutionize public sector productivity, provide citizens with faster, better information and access to services, and even usher in a new wave of participatory democracy” (Pollit & Bouchaert, 2011, p7). However, whilst the first and second wave clearly had dominant models, the third wave only offers ideas and concepts like transparency and responsiveness and thus has not changed a lot (Pollit & Bouchaert, 2011, p8).

2.2.2 State structure of the Netherlands

The history of the state structure of the Netherlands is very long and starts in the Middle Ages, a period where the sovereign King was the most important and powerful person in a state (Wagenaar, Kerkhoff & Rutgers, 2011, p31-32). Concepts such as centralisation and the creation of a state became important from the 14th century onward when the Habsburg Monarchy started to change the state structure in the Netherlands (Wagenaar, Kerkhoff & Rutgers, 2011, p32-33). Unrest and dissatisfaction caused uprisings in the Netherlands and the States-General, that used to be an advisory body, became more important and took more

16 responsibilities (Wagenaar, Kerkhoff & Rutgers, 2011, p92). These actions resulted in the birth of the Dutch Republic, which lasted until the French attacked the Netherlands in 1795 and the Netherlands became part of the French monarchy (Wagenaar, Kerkhoff & Rutgers, 2011, p97). Several years later, in 1814, the Dutch monarchy was reinstated and that was the start for the Dutch decentralized and unitary state as we know now (Wagenaar, Kerkhoff & Rutgers, 2011, p149-150).

A decentralised unitary state, such as the Netherlands, is a state where the central government can delegate certain responsibilities and tasks to for example provinces and municipalities (Breeman, Van Noort & Rutgers, 2010, p17). However, it is also a state where the government has to act according to the laws of that specific state and where citizens have fundamental rights such as right to freedom of religion and freedom of expression (Breeman, Van Noort & Rutgers, 2010, p16). Other important characteristics of the Dutch state are the ‘trias politica’ or separation of power which means that the legislative, executive and judiciary powers are separate institutions; it is a constitutional monarchy which means that the King is the ruler but is bound by the Constitution; there is a parliament that is chosen directly by the people; and there is ‘functional government’ which means that there are government bodies that have restricted responsibilities such as the water boards (Breeman, Van Noort & Rutgers, 2010, p17).

17 Figure 1 on page 16 shows a simplified version of the state structure of the Netherlands. As stated above, the Netherland is a constitutional monarchy and the Dutch Constitution states that the central government is headed by the King and ministers together (Breeman, Van Noort & Rutgers, 2010, p34). Whereas the King used to have great power and influence on policy, nowadays he has a more symbolic role (Breeman, Van Noort & Rutgers, 2010, p35). The ministers are each responsible for different policy subjects and, through individual ministerial responsibility, are responsible for all actions of their civil servants (Breeman, Van Noort & Rutgers, 2010, p36). Government policy is the responsibility of the government and the States-General; both the government and the Second Chamber can propose a bill, though it is most commonly proposed by the ministers themselves (Breeman, Van Noort & Rutgers, 2010, p39). The proposal is checked by the Council of State and then sent to the Second Chamber where it can be amended before it is sent to the First Chamber for the final vote (Breeman, Van Noort & Rutgers, 2010, p39). Between the first draft of the bill and the final vote in the First Chamber, there are also social organisations that can be asked for advice or social organisations that try to influence policy making by lobbying (Breeman, Van Noort & Rutgers, 2010, p39-40).

Since the Netherlands is a decentralised unitary state, the central government is the most important government body, but there are several responsibilities that are delegated to lower government bodies such as the provinces (Breeman, Van Noort & Rutgers, 2010, p77). Their most important responsibilities concern the organisation and management of the living environment, and they supervise the municipalities and water boards (Breeman, Van Noort & Rutgers, 2010, p77). As can be seen in figure 1, the governmental organisation of the provinces consists of two main bodies: the States-Provincial and Provincial Executive (Breeman, Van Noort & Rutgers, 2010, p78). The States-Provincial has the highest authority in the province and are chosen by directly by the Dutch people and their most important responsibilities are setting the provincial budget, controlling and monitoring the Provincial Executive, and setting the main course for provincial policy (Breeman, Van Noort & Rutgers, 2010, p79). The Provincial Executive, on the other hand, is led by the Royal Commissioner and is responsible for daily management of the province (Breeman, Van Noort & Rutgers, 2010, p81). The Royal Commissioner has many responsibilities such as directing, and coordinating, but he is also responsible for appointing mayors and has a coordinating role in the field of public order, safety and security (Breeman, Van Noort & Rutgers, 2010, p81). The twelve provinces also discuss issues and cooperate in the ‘Inter-Provincial Meetings’

18 [IPO] and often continue to cooperate in smaller collaborations for specific issues (Breeman, Van Noort & Rutgers, 2010, p82).

The lowest government body in the Netherlands are the municipalities (Breeman, Van Noort & Rutgers, 2010, p89). Even though the municipalities in the Netherlands are mostly performing executive tasks delegated by the central government, they are very important since they are the closest to and most visible for citizens (Breeman, Van Noort & Rutgers, 2010, p89). Municipalities have a lot of freedom regarding internal organisation and policies; though the basis of the internal organisation is the same (Breeman, Van Noort & Rutgers, 2010, p90-91). The municipal council is chosen directly by the local population and is responsible for monitoring the mayor and his City Council Members; they, on the other hand, are responsible for the daily management of the municipality (Breeman, Van Noort & Rutgers, 2010, p95). The mayor him- or herself leads this group, but is also responsible for representing the municipality at local festivities and government affairs (Breeman, Van Noort & Rutgers, 2010, p98). But more importantly, he or she is responsible for public order, safety and security in the municipality (Breeman, Van Noort & Rutgers, 2010, p98).

Besides the central government, the provinces and municipalities, there are many more government bodies that are crucial for governing the Netherlands (Breeman, Van Noort & Rutgers, 2010, p126). The Dutch water boards are among those important government bodies, since they are responsible for the quality and quantity of water in the Netherlands (Breeman, Van Noort & Rutgers, 2010, p132-135). The water boards are led by a dike warden and is chosen by members of the water board, who are in turn chosen by the local population (Breeman, Van Noort & Rutgers, 2010, p136). There have been a lot of changes in the internal structure of the water boards and the awareness of the importance of proper water management has increased (Breeman, Van Noort & Rutgers, 2010, p137). This has led to more coordination and cooperation on a regional and local level (Breeman, Van Noort & Rutgers, 2010, p137).

2.3 Forms of governance

There has been a huge amount of public management reform over the last three decades that resulted in many new ideas and models (Pollit & Bouchaert, 2011). Even though these models have been tried out, the results are difficult to pin down since there are both good and bad things attributed to them by both supporters and critics (Pollit & Bouchaert, 2011, p214). However, the three waves of administrative reform have shown that there are three basic

19 forms of governance: market, hierarchy and networks; each with its own characteristics, advantages and disadvantages (Rhodes, 1996, p653). The main differences are summarized in table 2 on page 25. These are stylized models of the three main governance models, and they are not a perfect description of reality but enable us to understand the organisation of reality (Powell, 1990, p301).

2.3.1 Hierarchy

Many governments have started with some form of hierarchical structure, where the political and managerial leaders set the goals and decide strategies, and have returned to a model where the state is the main facilitator of solutions (Pollit & Bouchaert, 2011, p20 & 118). A hierarchy or the traditional bureaucracy has some virtues that are very important to the government, such as clear accountability, predictability, probity, continuity and close attention to the law (Pollit & Bouchaert, 2011, p23). Another important aspect of a hierarchy is the important role of the representative democracy on every level of government: central, regional and local (Pollit & Bouchaert, 2011, p118). Whereas hierarchies used to be very stiff and formal in the past, new forms of hierarchy are more modern and professional (Pollit & Bouchaert, 2011, p119). This professionalization means that bureaucrats are not simply following the laws and procedures and becoming an expert in just one area, they are becoming more oriented to meeting the needs of the citizens they are serving (Pollit & Bouchaert, 2011, p119).

Hierarchies are especially useful for transactions that involve uncertainty about the outcome, that occur frequently and require transaction-specific investments (Powell, 1990, p296-297). Whilst hierarchies often have the inefficiencies of bureaucratic organization, they are cheaper than market structures because actors within the organisation cooperate and exchange information (Powell, 1990, p297). And because actors in hierarchies share information, know one another, have a history of previous interactions and have firm-specific knowledge, the actors are more interdependent than in a market (Powell, 1990, p302). Relationships matter in hierarchies, but in the case of interorganizational exchange it is all about one’s position within the formal hierarchical structure of authority (Powell, 1990, p302). Therefore communication and exchange is also shaped by concerns with career mobility and considerations of personal advancement in the hierarchical structure of authority (Powell, 1990, p302).

20 In the case of a hierarchy, it is the management or higher officials that decides the course of the organisation (Powell, 1990, p303). The management lets employees work under their regime of administrative procedures, determines their roles, tasks and positions and establishes an authoritative system of order (Powell, 1990, p303). A hierarchy is therefore especially well-suited for mass production and distribution because of the clear departmental boundaries, clear lines of authority, reporting mechanisms and formal decision-making procedures (Powell, 1990, p303). Because employees work closely together, it is a social institution with its own expectations, routines and detailed knowledge (Powell, 1990, p303). The strength of a hierarchy is therefore its reliability and accountability; it can produce goods and services of the same quality in large amounts and it documents well how resources have been used in order to produce them (Powell, 1990, p303).

However, a hierarchy has reliability and accountability in times of stability, and as such their liabilities are exposed in times of crisis or fluctuations in demand (Powell, 1990, p303). Communication in hierarchies occur through routines, which makes them less flexible than markets or networks (Powell, 1990, p300). Furthermore, a hierarchical structure with a vertical and fragmented division of tasks offers some difficulties as well (Bovens, ‘t Hart & van Twist, 2007, p279). But besides conflicts between the managers and their subordinates, there may also be conflict between bureaucrats and professionals or experts, policy design and implementation, or between the different governmental layers (Bovens, ‘t Hart & van Twist, 2007, p279-280). Hierarchies often have the same organisational culture and methods, but this does not mean that all layers and actors have the same way of thinking and a municipality, for example, appreciates the interference of the central government (Bovens, ‘t Hart & van Twist, 2007, p280). So because of the liabilities in communication, the lack of flexibility, and the formal and bureaucratic climate; hierarchies are often seen as rigid and impractical (Powell, 1990, p303).

2.3.2 Market

The market form of governance offers choice, flexibility and opportunity and is a form of non-coercive organization (Powell, 1990, p302). “Markets are a spontaneous coordination mechanism that imparts rationality and consistency to the self-interested actions of individuals and firms” (Powell, 1990, p302). When looking at the internal organisation of market oriented governments, we see that markets are open to everyone and the participants are free of any future commitments, which does result in weak bonds between actors and

21 causes them to be self-interested and non-cooperative (Powell, 1990, p302). Local, ethnic and trading cultures and varying regimes of state regulation structure markets, whilst contracts determine the relationships between actors (Powell, 1990, p299). Markets do also have “powerful incentive effects for they are the arena which each party can fulfil its own internally defined needs and goals” (Powell, 1990, p302). But because actors do not have to rely upon one single actor, there is unconstrained and fast social interaction and the means of communication is the price of the product or service (Powell, 1990, p302). None of the actors have to rely upon one another for direction, and there is no need for system wide control or governance which gives it a high level of flexibility (Powell, 1990, p302).

However, actors that are active in markets do not invest in long-term relationships or trust with other actors, simply because the value of the goods is more important than the relationship itself, and there is a low amount of commitment due to them clearly specifying the benefits in transactions and bolstering agreements by the power of legal sanction (Powell, 1990, p301). Actors in markets are self-interested, non-cooperative and do not learn from each other because they do not transfer knowledge (Powell, 1990, p302-303). Therefore there is a lot of competition between the actors, for example between the different ministries and municipalities (Pollit & Bouchaert, 2011, p9). The standard strategy due to this competition is to bargain and create indebtedness and reliance, because the only means of communication is the price (Powell, 1990, p302). Prices are a simplifying mechanism and as such they do not capture the complicities of the complex and dynamic exchange of goods and services (Powell, 1990, p302). As exchanges become more frequent and complex, the costs increase and the need arises for other methods of structuring and monitoring exchange (Powell, 1990, p303). Therefore a market structure is only useful in exchanges that are straightforward, non-repetitive and those that require no transaction-specific investments like money, time and energy (Powell, 19990, p297).

Markets and business-like models were common during the first wave of reform, since there was a need for efficiency and consumer-responsiveness (Pollit & Bouchaert, 2011, p21; Bovens, ‘t Hart & van Twist, 2007, p260; Rhodes, 1996, p655). Especially the model of New Public Management (NPM), that claimed that entrepreneurial government was necessary and inevitable (Pollit & Bouchaert, 2011, p9; Rhodes, 1996, p655; Muller & Mekel, 2011, p9). The main goal of this model was to improve the public sector by importing business concepts, techniques and values and as such offer more quality and choice to the citizens (Pollit & Bouchaert, 2011, p9-10; Muller & Mekel, 2011, p9). The most common practices in NPM were an emphasis on performance such as measuring the output; a

22 preference for small, specialized and disaggregated organizational forms over large multifunctional forms; using contracts instead of hierarchical relationships for coordination; market type mechanisms such as competitive tendering and performance related pay; an emphasis on treating service users as customers; and an application of generic quality improvement techniques (Pollit & Bouchaert, 2011, p10; Muller & Mekel, 2011, p9). Osborne and Gaebler designed ten principles that NPM-governments should pay close attention to: a steering government; a community-owned government that supports citizen participation; a government that introduces competition in the provision of services; a government led by a goal or mission instead of rules and procedures; a results-oriented government; a customer-driven government; an enterprising government that focuses on both on collecting and spending money; an anticipatory government that invests in prevention; a decentralized government; and finally a market-oriented government (Osborne & Gaebler, 1992). Thus, NPM can be summarized into three main concepts: disaggregation, competition and incentivization (Dunleavy et.al., 2006 in Pollit & Bouchaert, 2011, p10).

Thus, markets use business concepts and mechanisms to improve the government, since advocates state that the traditional government is too big and rigid (Pollit & Bouchaert, 2011, p23). Management is a key skill if the government wishes to behave as a business, and as such become more efficient and effective, and consumer-responsiveness is one of the main goals (Pollit & Bouchaert, 2011, p23; Muller & Mekel, 2011, p9). Citizens needs are important, they are the customers and should be the primary focus of the government (Bovens, ‘t Hart & van Twist, 2007, p260).

2.3.3 Network

Hierarchies are rigid, markets are free; network forms of governance are much lighter and more flexible than hierarchies, but do offer more organisation than markets (Powell, 1990, p303; Pollit & Bouchaert, 2011, p20). Whereas markets and hierarchies all have actors that are self-interested and selfish, advocates of networks believe that actors participating in a network are altruistic and think of the common good (Muller & Mekel, 2011, p21-22). Networks have become more popular because they are flexible and thus fit well in a modern world that is becoming more complex every day (Pollit & Bouchaert, 2011, p20; Chen et.al., 2013, p140). Advocates of networks believe that problems cannot be solved by organisations on their own and as such, hierarchies are no longer useful and are replaced by horizontal networks (Koppenjan & Klijn, 2004, p3). “Good governance is said to entail the steering of

23 society through networks and partnerships between governments, business corporations and civil society associations”, because a complex modern society can only be effectively governed through complex networks of actors with different backgrounds (Pollit & Bouchaert, 2011, p21-23).

In a network “transactions occur neither through discrete exchanges nor by administrative fiat, but through networks of individuals engaged in reciprocal, preferential, mutually supportive actions” (Powell, 1990, p303; Chen et.al., 2013, p140). Networks are more complex than hierarchies with familiar paternalism or markets that have explicit criteria, but in essence the actors within a network agree to cooperate and forego “the right to pursue their own interests at the expense of others” (Powell, 1990, p303; Pollit & Bouchaert, 2011, p23). Actors are dependent on other actors for resources, but find that the pooling of resources has many gains (Powell, 1990, p303; Pollit & Bouchaert, 2011, p123). There are no individual units, but only units in relation to another (Powell, 1990, p303). By joining a network actors do not only pool their resources and benefits, but also their burdens (Powell, 1990, p303; Osborne, 2010, p9).

The relationships between the actors take considerable effort to establish and sustain, because there is a need for trust and flexibility to change if need be (Powell, 1990, p303; Pollit & Bouchaert, 2011, p11). There is knowledge about other actors that can be used for everyday communication but also for problem solving in times of conflict (Powell, 1990, p304). But the most important concept in networks in times of good or bad is reciprocity, which is the idea that all good acts must be repaid with good acts (Powell, 1990, p304; Chen et.al., 2013, p140; Rhodes, 2012, p3). However, reciprocity does not always mean good for good, but can also imply ill for ill (Powell, 1990, p304). Still, in most cases, reciprocity is a positive concept that enables a secure and stable relationship between actors in a network and encourages actors to search for innovative ways of accomplishing tasks, promote learning, exchanging information and creating trust (Powell, 1990, p305; Chen et.al., 2013, p140). Other important concepts for relationships in networks are reputation, friendship, interdependence and altruism, for complementarity and accommodation are the cornerstones for success (Powell, 1990, p304; Chen et.al., 2013).

Information is thicker in networks than it is in markets, and freer than in hierarchies (Powell, 1990, p304). Networks exchange qualitative information as know-how, technological capability, a spirit of innovation, a particular approach or style of production, or a philosophy, and this is only possible due to the open-ended relational features in networks (Powell, 1990, p304; Kenis & Provan, 2007, p231). Cooperation emerges out of mutual

24 interests and allows actors to transmit and learn new knowledge and skills from other actors (Powell, 1990, p304; Kenis & Provan, 2007, p239). And because there is a high level of trust in networks and the actors exchange information, this type of governance is extremely useful in circumstances where there is a need for efficient and reliable information (Powell, 1990, p304). “The most useful information is rarely that which flows down the formal chain of command in an organization, or that which can be inferred from shifting price signals. Rather, it is that which is obtained from someone whom you have dealt with in the past and found to be reliable” (Powell, 1990, p304).

Still there are some liabilities, starting with the fact that networks are often not as open as markets are (Powell, 1990, p305). Networks can be organised in different ways, for example they might be governed by another institution or govern themselves, and they might have direct contact with other networks or only contact with the main actor of the network (Kenis & Provan, 2007, p233-234). So it is possible that the main actor does not want any new actors in the network, or because of repeated trading and communication within the same group of actors, networks restrict access and limit the flow of newcomers and their knowledge and skills themselves (Powell, 1990, p305; Kenis & Provan, 2007, p233-234). Furthermore, each point of contact in a network can cause both harmony and conflict, it all depends on how a network is governed (Powell, 1990, p305; Kenis & Provan, 2007, p240). Actors may have the same goal, but not always the same road towards that goal (Powell, 1990, p305; Kenis & Provan, 2007, p240).

25 Forms

Key Features Hierarchy Market Network

Goals Professionalism,

efficiency,

responsiveness whilst the state remains a distinctive actor with its own rules,

methods and culture

Efficiency, consumer-responsiveness by injecting business-like methods Better informed government, flexibility, less exclusivity, self-organising networks, effectivity, legitimacy Normative Basis Employment

Relationship, hierarchical Contract – property rights, performance indicators, targets, competitive contracts, quasi-markets Complementary Strengths, network of independent or dependent stakeholders, partnerships Role of politicians

and civil servants

Politicians: traditional takers of authoritative decisions; Civil servants: professional implementers of laws and politicians’ decisions Politicians: strategic goal setting; Civil servants: autonomous managers or entrepreneurs held accountable through performance frameworks Politicians: forgers and guarantors of compromise deals between multiple stakeholders; Civil servants: network managers and partnership leaders or negotiators Means of communication

Routines Prices Relational

Methods of Conflict Resolution Administrative fiat – supervision Haggling – resort to courts for enforcement Norm of reciprocity – reputational concerns

Degree of Flexibility Low High Medium

Amount of Commitment Among the Parties

Medium to high Low Medium to high

Tone or climate Formal, bureaucratic Precision and/or Suspicion

Open-ended, mutual benefits

Actor Preferences or Choices

Dependent Independent Interdependent

Mixing of Forms Informal organization

Repeat transactions Status hierarchies Market-like features: profit centers, transfer pricing Contracts as hierarchical documents Multiple partners Formal rules

Table 2: Governance models

Powell, W.W. (1990), ‘Neither Market Nor Hierarchy: Network Forms of Organisation’, Research in Organizational Behavior, 12, pp: 295-336.

26

Chapter 3

Research Design

As stated in the introduction, the main goal of this research is to investigate the coordination and collaboration between government parties involved in water-disaster management. This means that the Dutch government is investigating new possibilities for the future, but this research will focus on the current situation by using the following research question:

“What governance model best describes the coordination between

the Dutch government actors in the field of water disaster management?”

Figure 2: Graphic representation of models of governance

3.1 Case Study

This research has used two cases in the field of water-disaster management, in order to find an answer to the main research question. A case study such as used in this research has been used for “in-depth examination of a single instance of some social phenomenon” (Babbie, 2010, p309). And in this case, the examined cases provide in-depth information about the coordination between the Dutch government actors in the field of water-disaster management.

Disaster

management

model

Contract, haggling, high degree of flexibility, low amount of

commitment, suspcision

Market

Employment, routines, administrative, supervision, low

degree of flexibility, formal, bureaucratic

Hierarchy

Complementary strengths, relations, reciprocity, reputational concerns, medium

degree of flexibility, mutual benefits, interdependent

Network

27 However, because one or more cases are studied specifically, it is more difficult to generalize the results (Babbie, 2010, p309). Other types of research, such as experiments or field research, require the simulation or occurrence of a water-disaster in order to examine and observe the case. However, water-disasters are difficult to simulate, and are unexpected and unwanted events; the chance of being able to observe during a water-disaster is very low. Hence the choice for a case study, consisting of two water-disaster cases in the Netherlands. A case study focusing on cases in the past offers the advantage of using existing research and documents, and the phase of coordination between the Dutch government actors in these cases has generally ended.

Case studies can be single or multiple, and can be used for both descriptive and explanatory purposes (Babbie, 2010, p309). This research uses a multiple case study, since using multiple cases makes comparison possible whereas a single case study remains very descriptive (Babbie, 2010, p309). There are several reasons that these cases specifically have been chosen for this research, starting with the fact that the chosen cases had a lot of impact on the victims, the government and water-disaster management (Omroep Gelderland, 2015; Gemeente de Ronde Venen, 2004). The first case of the evacuation of ‘Rivierenland’ in 1995 caused the evacuation of 250.000 people and 1 million cattle, which makes it one of the largest evacuations in the history of the Netherlands (Omroep Gelderland, 2015; Waterschap Rivierenland, 2015). In 2016, twenty years after the disaster, the local authorities organised several commemorations and activities to remember the disaster because inhabitants of Rivierenland still remember the day they had to leave their houses because of the risk of a flood (Visscher, 2015; Gemeente Neder-Betuwe, 2015). The commemorations and the fact that the case of Rivierenland led to a lot of changes in water-disaster management shows the impact of this specific case (Visscher, 2015; Gemeente Neder-Betuwe, 2015). But also the second case in Wilnis (2003) caused significant damage, resulted in mass media attention and caused the government to rethink certain policies and water-disaster management methods in the Netherlands (Omroep Gelderland, 2015; Gemeente De Ronde Venen, 2004). The most well-known disaster concerning water in the Netherlands that had a large impact on the Netherlands and water-disaster management, is the North Sea flood of 1953 that killed 1835 people and caused 72.000 people to flee the disaster area (Zeeuws Archief). This case has not been included in this research because of the lack of government records and many things have changed in disaster management since that time. (Watersnoodmuseum, 2003, p14). Furthermore, the Second World War had just ended, which meant that the dikes were in bad

28 shape, and the government changed a lot in the years afterwards (Watersnoodmuseum, 2003, p13; Wagenaar, Kerkhoff & Rutgers, 2011, p223). So, it has only been included in the introduction of the analysis because of its historical significance for the Netherlands and water-disaster management in the Netherlands.

Both cases had a large impact on the Netherlands and water-disaster management, but these cases have also been chosen because they differ in several other aspects. Firstly the case of Wilnis was local disaster, whilst the case of Rivierenland was a more regional disaster (Gemeente de Ronde Venen, 2004; RCC Tiel, 1995). Secondly, because Wilnis was a local disaster and Rivierenland was a regional disaster, they differ in scale when looking at amount of victims, damage and people involved in decision-making during the disaster. Furthermore, the case of Rivierenland concerns the largest disaster in the Netherlands in the past years and no physical damage to houses or infrastructure (RCC Tiel, 1995). In Wilnis, on the other hand, it was a real disaster instead of just a threat (Gemeente de Ronde Venen, 2004). This difference is important because decision-making is different in both situations; the authorities in Wilnis had to act right away whilst the authorities in Rivierenland had time to wait for more information (Gemeente de Ronde Venen, 2004; RCC Tiel, 1995). So finally, the last important difference between these cases is the fact that Wilnis was unexpected and the authorities had to act right away, but the authorities in Rivierenland were informed of the rising water levels days before (Gemeente de Ronde Venen, 2004; RCC Tiel, 1995).

So, the choice for these more recent cases is first and foremost rational, since these recent cases give a better image of the current situation regarding the relation between the Dutch government actors in the coordination in water-disaster management in the Netherlands. Furthermore, there are no other relevant cases for this research. For example, But the choice for more recent cases is also practical, since there is more information and this information is easier to access than a case that happened before the 1990s. Furthermore, both cases have been mentioned in the news in the past year. In the case of Wilnis, the Court in The Hague decided in 2014 that the water board was not responsible for the damage caused by the quay rupture, because they could not have known about the damage the draught had caused (Banning, 2014). And the Evacuation of Rivierenland happened exactly twenty years ago last January, and as such there have been several commemorations, educational programmes and a conference about the changes that followed this case (Nieuwsblad Geldermalsen, 2015; RTV Utrecht, 2015; Gemeente Neder-Betuwe, 2015).

29

3.2 Data collection, Reliability and Validity

The most important sources for this research are policy documents, evaluations or reports, newspaper articles. Most of the sources have been found on the internet and in the Regional Archive of Rivierenland (Tiel) and Regional Archive of Vecht & Venen (Breukelen), and have been subjected to content analysis. This specific type of research is used for analysing human communication in all forms, for example spoken or written, and is useful for answering questions like ‘who said what, to whom, why, how and with what effect’ (Babbie, 2010, p333). This allows for the researcher to describe the situation and analyse the tone and underlying purpose of the different documents, which is needed for answering the main question. In the case of Rivierenland, there were a lot of documents that could be found in the archives and this resulted in the selection of 378 documents of which 474 pages were selected. The selection of the documents that would be used for this research, or sampling, was conducted by looking at the documents and deciding whether they contained information about government actors and the relationships between them. Using the list of key features as shown in the operationalisation table, documents were scanned to see if they showed any of the characteristics of these key features. Finding documents about the quay-rupture in Wilnis was much more difficult, since a lot of government documents have not been made public yet. Therefore, the case of Wilnis was researched by using 8 different documents. Even though this number is far lower than the case of Rivierenland, there are several reasons why this is not a reason to deem them less valuable. The documents about Rivierenland often discuss the same information, show information that was not important to this research after all, and there were more documents since there were more actors involved. The documents about Wilnis, on the other hand, were not repetitive, and since the disaster was smaller, there was only one evaluation.

The researcher has also looked into the option of interviewing government officials that were involved in the disaster-management process during these disasters, however due to several reasons this was not possible within the time given for this research. Other methods of research such as direct observations during the disaster and executing an experiment were not possible for this type of research, since crises are unexpected and doing observations cannot be planned and results from an experiment would not be reliable. A method of research that has been considered, was doing a survey. This survey could consist of questions asking the different government actors about their experiences during the disaster, their opinion about