Master Thesis Ns Stakeholder Management

MASTER THESIS

NS STAKEHOLDER MANAGEMENT

A STUDY OF THE COMPLEX RELATIONSHIPS BETWEEN DUTCH RAILWAYS AND DECENTRAL GOVERNMENTS WHEN DECIDING ON URBAN, REGIONAL PUBLIC TRANSPORT PLANS

October 1st 2010

Thesis Project Pieter Guldemond Master Systems Engineering, Policy Analysis & Management (SEPAM) Delft University of Technology, Faculty of Technology, Policy & Management Supervision

Prof.mr.dr. Hans de Bruijn Dr.ir. Caspar Chorus Faculty of Technology, Policy & Management Faculty of Technology, Policy & Management [email protected] [email protected] Professor Second supervisor

Dr.ir. Bauke Steenhuisen Ir.drs. Adriaan Roeleveld Faculty of Technology, Policy & Management NS Business Development [email protected] [email protected] First supervisor External supervisor

October 2010

Pieter Guldemond Student number 1194658 [email protected]

Acknowledgements

This research is performed at Dutch Railways (in Dutch: Nederlandse Spoorwegen, NS). NS provided a fruitful working environment to perform my research. It was a great pleasure to work at the Business Development department in Utrecht. Without any exception everybody was very interested in my research and enthusiastic to share their experiences and ideas. I built up very good personal relationships with some of my colleagues, in particular with Adriaan Roeleveld. His ability to link scientific theories to practical NS situations is impressive. His fast and clear analyses were indispensable for this research. In addition, his humor and optimism made it a pleasure to cooperate.

The contribution of Bauke Steenhuisen to a successful round-off of this thesis cannot be emphasized enough. Bauke and I spent many hours discussing this research. His great commitment to my research and fascination for the way a company like NS works is unique. We had many lively discussions concerning not only the theoretical concepts, but also the practical findings of the interviews. His writing skills lifted this thesis to a higher level. I also want to thank Hans de Bruijn en Caspar Chorus. Even when they were occupied by many other activities, they found time to give me useful feedback.

I would also like to thank my graduation committee as a whole for the flexibility they showed when I announced the possibility to become vice mayor in Delft, before my graduation project was finished. Without their support it would not have been possible to both start my new job and finish my graduation.

To conclude, I want to thank Marjanne Poppen for her mental support; Bas Bakker for his encouragement; and Bas de Vrind, Boris Gooskens, Lian Merkx and Marte Guldemond for their support in finishing this thesis.

Delft, October 1st 2010, Pieter Guldemond

Preface

Since the successful dissertation of Bauke Steenhuisen about ‘competing public values’ (2009), NS gained interest in the thinking of the Faculty of Technology, Policy & Management (TPM). Especially Adriaan Roeleveld, working at the Business Development department, wanted to know more about how theories about process management and the multiple principals problem could contribute to the daily management of NS. This interest was formulated into the hiring of a TPM graduate. NS asked if it was possible to investigate the relationship with decentral governments in order to formulate recommendations for improving the cooperation. That formed the starting point of this research.

This master thesis addresses different types of audiences. People with a professional background will probably be interested in practical recommendations to improve their relation management in order to enhance their business result. In contrary, people with a scientific background are probably less interested in practical implications for NS, but more interested in the connections among the case studies, its findings and scientific theories. Serving all audiences forced me to make trade-offs. I chose to present a theoretical perspective, because it is possible to transfer more practical knowledge orally to NS. I made appointments with NS about a future cooperation to translate thoughts into practical steps.

Summary

Research problem (chapter 1)

Since 1993 liberalization and decentralization took place in the Dutch railway sector. As a consequence, the national government is not the only governmental body that governs the railway sector. Instead, multiple levels of government are currently involved in organizing public transport: national government, provinces, city regions and municipalities. At first sight governments essentially share the same mission: improving public transport. But when looked closer, the realization of this value reveals possible conflicts. Different scope and interests of governments can result in conflicting demands.

In practice we see that the mutual cooperation of NS and governments is far from a textbook example of allied actors in fruitful cooperation. The focus of this research will be on the relationship between NS and decentral governments. More specifically, the realization of cooperation between NS and decentral governments is investigated. This realization starts with desires of decentral governments or NS, and ends with a decision. Execution of projects falls outside the scope. Next, it is chosen to focus on urban regional transport plans, as in urban areas multi-layered governments are present. The more levels of governments are involved, the more complex a project is. A study of complex projects that are performed can provide insight in the cooperation between NS and decentral governments.

This results in the following research question:

How does NS cooperate with multi-layered governments to reach decisions on new urban regional transport plans and how can this cooperation be improved?

Main theories (chapter 2)

Based on several scientific theories an analytical framework is constructed.

•Number of decentral governments involved? •Multiple principals with conflicting interests? •Level of market dominance by one party? Multiple •Level of information asymmetry? principals problem •Strategic behavior recognized?

•Consensus about facts? •Consensus about norms? Wicked •How is the problem defintion shaped? problems

•How was the process prepared? •Four core elements taken into account (openness, speed, content & core values)? •Policy window present? Process •Sense of urgency present? management

The first part of this framework consists of the multiple principals problem. The second part focuses on the problem definition. The third part concerns insight in the way process management is applied. It is demonstrated that for many decisions on how to choose position in the relationship with decentral governments, the elements of process management might be useful for NS. The use of a process design can contribute to an improvement of the relationship between NS and decentral governments.

Research methods (chapter 3)

For this research it is very important to get deep understanding of how the real life cooperation between NS and decentral governments takes place. Therefore two case studies are performed. To select cases, three selection criteria are used: complexity of the project, multi-layeredness of involved governments and geographical spread. This resulted in the selection of two projects: - Programme Accessibility North Wing (‘PBN’) in the metropolitan region of Amsterdam - City Region Rail in the Arnhem-Nijmegen area

People involved in the cooperation between NS and decentral governments have unique knowledge about this cooperation, which cannot be found in documents. Therefore, 27

6 people closely related to the decision making of the two cases are interviewed. The protocol used for the interviews is based on the analytical framework.

It has to be prevented that the research depends on information from NS only. This independence is safeguarded by involving many people with a different background. Therefore an extensive study is done in which semi structured interviews are held with people from NS, decentral governments and other organizations; internal NS documents and external documents are studied; and observations are performed by visiting conferences.

Case study findings (chapter 4 and 5)

Case 1: PBN

The first case concerns investments in railway infrastructure and the opening of new stations, with a total budget of 171 million Euros. This budget is provided by the Ministry of Transport, with the constraint that decentral governments and NS agree on the investment programme. Decentral governments and NS started to negotiate about this programme in March 2005 and eventually signed a deal in June 2006.

In this case NS has to deal with well-organized, multi-layered governments. The multiple governments impose contradicting demands to NS, so NS can be considered an agent serving multiple principals. On top of that, NS was internally divided, as the regional board pursued different values than the central board did. The negotiations were done by regional representatives of NS, but their mandate to negotiate was very limited.

Decentral governments identified different bottlenecks in the area from the railway sector. So there was no consensus about the facts. The two parties also differed about the vision how the identified bottlenecks could be solved best, so there was also no consensus about norms. The area where the negotiations took place has an interesting characteristic in the perspective of strategic behavior: the absence of rail competitors for NS in this region. NS has a high level of market dominance and a high level of information asymmetry, which forms a barrier to a fruitful cooperation with the decentral governments.

The two conflicting visions resulted in a difficult, long and formal process of negotiations. NS did not define a clear strategy for these negotiations. NS was not able or not willing to give insight in their calculations and their tradeoffs to support or reject certain projects. This attitude feeds the perception of NS acting like a monopolist. NS proposes certain solutions, based on their information gathering. But these solutions are not recognized by decentral governments as they do not have insights in the used information.

There was a sense of urgency for both the railway sector and the decentral governments to improve the public transport in and around Amsterdam. In terms of Kingdon (1984) it seems like the streams of problems, policies and politics came together, creating a policy window.

Case 2: City Region Rail

The main goal of the project is to run four stopping trains every hour on all trajectories in the Arnhem-Nijmegen region. The project consists of five investments in infrastructure: three new stations and two capacity related investments, with a total budget of 55 million Euros.

NS has to deal with divided, multi-layered governments. In general, the municipalities support the city region, which closely cooperates with NS. In contrast, the province has a difficult relationship with NS and the city region. During the interviews it became clear that the city region functions as a central actor. Again, NS faces many principals in this case. NS cooperates in a very informal way with their partners in this area. This makes it difficult to reconstruct the process.

In this case a clear disagreement about the problem definition is found. According to the decentral governments the major problem is the regional accessibility. Therefore, the focus should be on filling up the gap of travel distances between 10 to 40 kilometers. NS on the other hand, wants to facilitate urban growth. In their vision it is not relevant to classify movements into categories of travel distance. This discussion about definitions shows that there is no clear problem definition. Therefore, this can be labeled an unstructured or wicked problem.

There are multiple rail operators active in this area. According to the decentral governments, the presence of other train operators caused a mentality change at NS. However, NS representatives confirm that in the new situation they are challenged to show the decentral governments what they have to offer.

There was a sense of urgency at both the railway sector and the decentral governments to improve the public transport in the city region. Governmental actors in this area are very keen on improving public transport. In terms of Kingdon (1984) it looked like the problem, policy and politic stream came together, creating a policy window.

Conclusions (chapter 6)

Based on the case studies, the research question can be answered:

How does NS cooperate with multi-layered governments to reach decisions on new urban regional transport plans and how can this cooperation be improved?

In general it can be concluded that NS did not react proactively on the increasing influence of decentral governments. There is no clear strategy about how to cooperate with multi- layered, decentral governments. There are no guidelines that describe how to choose position in the faced tradeoffs, resulting in random choices. The relationship between NS and decentral governments is strongly dependent on the strategy of individual NS employees. This worsens the multiple principals problem and restrains the cooperation to fully exploit the potential of involving all committed interests of decentral governments. As a consequence, it is not clear to external partners if and how NS will cooperate in urban, regional transport plans. The second part of the research question (improvement of cooperation) is discussed in the recommendations.

Based on the two cases, five specific conclusions and recommendations are formulated.

1. Two cases, different environment The cases reveal some important differences, summarized in the table below.

PBN City Region Rail Relationship between decentral United Divided governments Kind of cooperation Formal Informal Course of negotiations Clear defined process Process without clear beginning or ending Forming of solutions Confrontation of two already Sharing of ideas (dynamic) defined visions (static) Market situation One market party (NS) Three rail companies (NS, Veolia, Syntus) Dominant persons Civil servants are in charge Politicians are in charge

2. NS internal organization does not match the increasing influence of decentral governments NS is faced with the upcoming power of multi-layered, decentral governments. From the case studies the impression arises that NS was surprised by the increasing involvement of the decentral governments. The playing field has changed, but the central player, NS, did not adapt their behavior to the new field. It can be concluded that NS did not react proactively on this development. There is no clear strategy about how to cooperate with multi-layered, decentral governments. However, currently a shift is noticeable from reactive to proactive behavior when it comes to the attitude of NS towards decentral governments.

3. NS is facing the multiple principals problem Due to the presence of multi-layered governments, NS is facing multiple principals with partly conflicting interests. For NS it is impossible to comply with all principals. In practice, the first reaction of NS is to strictly follow the norms of the Ministry of Finance, which means that no investments are done in projects that are not profitable enough. Then, most of the time decentral governments are disappointed and a discussion is started about the contribution and impact of a certain project. Eventually, a deal is negotiated on which all actors agree.

4. Process design and process management is needed to deal with wicked problems NS has not chosen one strategy to deal with the decentral governments. When zooming in on the processes where NS deals with decentral governments it can be concluded that these processes are under designed. NS does not pay enough attention to four important characteristics of process management: speed, protecting core values, content and openness.

5. Limited information results in a ‘black box’ of decision making Despite the attempts of NS to deal with decentral governments, NS is still seen as a company acting like a monopolist. This is mainly caused by the lack of transparency when it comes to the explanation of decisions. One of the biggest frustrations of the decentral governments is the limited sharing of two sources of information: - Models that NS uses to calculate the value of certain investments . These models are not comprehended and thereby not recognized by decentral governments. - Statistical information about number of travelers, extra travelers when opening a station, etc. This information is seldom shared, which makes it difficult for decentral governments to understand decisions.

Recommendations (chapter 7)

Based on the five conclusions, five recommendations are formulated. Concrete examples related to the cases are displayed in text boxes in chapter 7.

1. Deal with the diverse environment Decision making in a political arena can always pass of capricious. Relationships that are built up and appointments that are made, can be brusquely disturbed by elections or unexpected political developments. Therefore it is recommended to stick to the global vision. But it is also very important to be adaptive to the local political situation. The three streams of Kingdon (1984), problems, policies and politics, can be helpful to determine whether efforts are needed to influence the local agenda. Developments in the political arena should be closely watched, so the global vision can be framed in a way that it appeals to the ruling political view. It is recommended to develop a flexible attitude when dealing with decentral governments in this politicized setting. Three dimensions need specific attention: united vs. divided governments, formal vs. informal cooperation and monopolistic vs. competitive market.

2. Acknowledge multi-layered governments as influential partners It is recommended to NS to acknowledge the new position of multi-layered decentral governments, because these governments can support NS in three ways: - They have budget available to invest in regional public transport; - They have extensive knowledge about the regional situation; - They can influence the central government and public opinion, which can have a positive effect on the opportunities for NS to obtain the central rail network concession after 2015.

In order to build up relationships with decentral governments, NS needs to get rid of the image of acting like a monopolist. Therefore, a change of attitude at all levels of the company is needed. Case studies show that the NS regional directors are very aware of this change in behavior and they already started the change in the right way. It is recommended to continue the course of building up relationships with decentral governments.

3. Share the multiple principals problem This research confirmed that NS has to serve multiple principals with partly competing interests. This automatically results in difficult tradeoffs for NS that have to be made at many levels. The most important tradeoff consists of realizing a certain public value versus the costs of such an investment. It is therefore recommended to share this multiple principals problem with the relevant governments. NS can try to make governments sharer of this tradeoff by giving more insights in the difficulty of choices. The tradeoff is then moved to the institutions that are responsible for creating this tradeoff, which makes it possible to

11 discuss whether the conflicting demands lead to a desirable outcome. If not, the demands of the governments might be adjusted, leading to a more desirable outcome. Sharing the multiple principals problem does not mean to give away influence and say to the environment of an agent, but to gain control and insight over the most desirable trade-off to make a cooperative platform of multiple principals and their agent.

4. Be prepared: pay attention to the four core elements of process design

- Content It is recommended to create more space to collectively determine a vision. This can be done by first defining a global vision internally, without fixing the details. Next, it is important to construct a good composition of the delegation that negotiates. NS has the tendency to focus on the content and to underestimate the characteristics of a good process. This means that in negotiations not only technical expertise, but also process expertise is needed. - Openness If NS participates in a project with multi-layered decentral governments, it is important to make an analysis of the governments involved. This can be done by making an actor analysis. Such an analysis can contribute to find the right allies; give insight in the agenda of stakeholders; and to create package deals. - Speed When there is no incentive to cooperate, an incentive needs to be constructed. Sometimes these incentives are already present, for example by a shared sense of urgency to solve a certain problem. In that situation it is important to clearly frame the sense of urgency, using terminology that appeals to the involved actors. It is important that representatives of NS have a mandate when they participate in a process. This means that representatives have certain room to negotiate, without discussing with their supervisors. - Core values When participating in a process, NS should on beforehand decide on what will be the bottom line and how tradeoffs between important values will be made.

5. Invest in more insight in decision making When NS wants to continue the course of opening up to the external environment, more insight into the use of their models and information is needed. Efforts should be made to make understandable what the effects on the national rail network are, for example by showing simulations. The effects on the national rail network should be made visible instead of just told.

A last recommendation is based on the recent collapse of the Dutch cabinet. The Secretary of State of the Ministry of Transport, Ms. Tineke Huizinga, took the initiative to write a vision about the development of regional public transport. However, due to the political developments the writing of this vision is postponed. This could be an opportunity for NS to

12 take the lead in developing a vision collectively with the decentral governments on the development of regional public transport. When process management is applied well, this can contribute to a stronger position for NS. When NS takes the lead, the chance of steering the process into the desired direction increases. If decentral governments are involved in a proper way, this can contribute to the image of NS as a constructive, open partner. This image might even contribute to the awarding of the concession of the central rail network after 2015.

Table of contents

Supervision ...... 2 Acknowledgements ...... 3 Preface ...... 4 Summary ...... 5 Research problem (chapter 1) ...... 5 Main theories (chapter 2) ...... 5 Research methods (chapter 3) ...... 6 Case study findings (chapter 4 and 5) ...... 7 Case 1: PBN ...... 7 Case 2: City Region Rail ...... 8 Conclusions (chapter 6) ...... 9 Recommendations (chapter 7) ...... 11 Table of contents ...... 14 Chapter 1. Introduction ...... 17 1.1 Improving public transport together ...... 17 1.2 A common public value? ...... 18 1.3 Dealing with the multi-layered public sector ...... 19 1.4 Deciding on urban, regional transport plans: from desires to decision ...... 22 1.5 Research question and thesis outline ...... 23 Chapter 2. Theory ...... 24 2.1 Multiple principals problem ...... 24 2.2 Wicked problems ...... 26 2.3 Process management ...... 27 2.3.1 Why use process management? ...... 27 2.3.2 Core elements of a well-designed process: openness, protection core values, speed & content ...... 30 2.3.3 Timing of the policy process ...... 31 2.3.3 Possible critics to process management ...... 32 2.4 Analytical framework ...... 33 Chapter 3. Research approach ...... 34 3.1 Investigate real life cooperation ...... 34 3.1.1 Looking for the right research strategy ...... 34 3.1.2 Multiple cases, holistic design ...... 35 3.1.3 Possible critics to case study ...... 36 3.2 Case study selection ...... 37 3.3 The art of interviewing ...... 38 Chapter 4. Case 1: Programme Accessibility North Wing ...... 41 4.1 Introduction ...... 41 4.2 Multiple principals problem ...... 42

4.3 Wicked problems ...... 44 4.4 Process management ...... 46 4.4.1 Process preparation ...... 46 4.4.2 Openness ...... 47 4.4.3 Core values ...... 48 4.4.4 Speed ...... 49 4.4.5 Content ...... 50 4.4.6 Timing of the process ...... 51 4.5 Rounding up ...... 52 Chapter 5. Case 2: City region rail Arnhem-Nijmegen ...... 53 5.1 Introduction ...... 53 5.2 Multiple principals problem ...... 54 5.3 Wicked problems ...... 57 5.4 Process ...... 57 5.4.1 Process preparation ...... 57 5.4.2 Openness ...... 59 5.4.3 Core values ...... 59 5.4.4 Speed ...... 60 5.4.5 Content ...... 60 5.4.6 Timing of the process ...... 60 5.5 Rounding up ...... 62 Chapter 6. Conclusions ...... 63 6.1 Two cases, different environment ...... 63 6.2 NS internal organization does not match the increasing influence of decentral governments ...... 65 6.3 NS is facing the multiple principals problem...... 66 6.4 Process design and process management is needed to deal with wicked problems .... 67 6.5 Limited information sharing results in a ‘black box’ of decision making ...... 68 Chapter 7 Recommendations ...... 69 7.1 Deal with the diverse environment ...... 69 7.2 Acknowledge multi-layered decentral governments as influential partners ...... 71 7.2.1 Three reasons to involve decentral governments ...... 71 7.2.2 Improve internal organization ...... 72 7.3 Share the multiple principals problem ...... 73 7.4 Be prepared: pay attention to the four core elements of process management ...... 74 7.4.1 Content ...... 74 7.4.2 Openness ...... 75 7.4.3 Speed ...... 76 7.4.4 Core values ...... 77 7.5 Invest in more insight in decision making ...... 77 7.6 Make use of the political vacuum ...... 78

7.7 Further research ...... 79 Chapter 8. Reflection ...... 80 8.1 Defining a roadmap ...... 80 8.2 Learning from the research ...... 81 8.3 Reflection on literature ...... 81 8.4 Personal reflection ...... 82 References ...... 84 Appendix A. Organizational structure of NS ...... 87 Appendix B. Long list projects NS - decentral governments ...... 88 Appendix C. Interview protocol ...... 90 Orientation ...... 90 Multiple principals problem ...... 90 Wicked problems ...... 90 Process management ...... 90 To conclude ...... 91 Appendix D. Lists of interviewed people ...... 92 Orientating interviews (NS internal) ...... 92 Interviews case study 1: PBN ...... 92 Interviews case study 2: City Region Rail ...... 93 Validation presentations (NS internal) ...... 93 Appendix E. Example of actor analysis ...... 94 Appendix F. Visualization of the relationship between NS and ProRail ...... 95 Appendix G. Roadmap for a successful cooperation between NS and decentral governments when deciding on urban, regional public transport plans ...... 96 Appendix H. Disclaimer for using this roadmap ...... 97

Chapter 1. Introduction

In this chapter an introduction to the problem is presented. First, recent developments in the Dutch railway sector are described (1.1). In paragraph 1.2 it is demonstrated that decentral governments and public transport companies at first sight share a common public value, but in practice their scope and interests differ. Paragraph 1.3 introduces the dilemmas of multi-layered governments. The scope of this research is determined in paragraph 1.4. The choice to study how NS cooperates with decentral governments in two cases of urban regional new transport plans is motivated. This leads to the formulation of the central research question and the thesis outline in paragraph 1.5.

1.1 Improving public transport together

The public interest of providing public transport is common sense in current Dutch society. Yet what is gradually but drastically changing is the way the public sector is involved in safeguarding the public interest with respect to new investments in transport plans and mobility programs. It used to be simple. All railways were managed, owned and exploited by the Dutch Railways (in Dutch: Nederlandse Spoorwegen, NS). The company had one shareholder: the Dutch national government. Before liberalization, the public interest was assumed to be taken for granted. The implicit assumption was that public interest was served by public ownership and provision and that separation from private administration would ensure that the public good did not have to compete with economic interest (Charles et al., 2007). In the first decades after the Second World War governments assumed direct providers of public infrastructure particularly to support national development and expansion (Ikenberry, 2000).

In 1980s a significant policy shift emerged in the course of EU-driven liberalization. Keynesian economics was replaced by neo-classical economics, a body of thought which challenged the proportion of national activity undertaken by the public sector versus the private sector (Schipke, 2001; Zohlnhöfer and Obinger, 2006). The result was the large-scale shift of assets and activities from the public sector to the private sector through policy mechanisms such as privatization, contracting out, and public sector downsizing (Megginson and Netter, 2001).

This shift caused two important change of thinking about the public interest. Firstly, the public interest was not taken for granted anymore. The public sector was urged to articulate and regulate what they thought to be the public interest more actively. Secondly, more and more governmental bodies were involved in this process of articulating the public interest.

In 1993, in line with the shift towards an increasing role of the private sector, the liberalization of the Dutch railway sector started. The system with one single company changed into a system with numerous train operators (of which NS is by far still the biggest). Since then, NS encounters many public and semi-public bodies as regulators and stakeholders for their rail services. There is a national transport authority, public shareholders, regional transport authorities at the province level, the municipality level and the in-between city region level, an independent transport regulator, an independent party that develops and maintains the rail infrastructure (ProRail) and a group of institutionalized bodies like ROVER representing customers.

1.2 A common public value?

The public interest of providing public transport is still common sense, but various public bodies tend to articulate slightly different visions on public transport. This can be illustrated by presenting some quotes from involved parties.

‘The quality of public transport should be more adapted to the users. Punctuality numbers in the rail sector should rise to 90% and the connection between several public transport systems should be improved in order to diminish waiting times.’ (Ministry of Transport, 2005)

‘Our goal is to realize attractive and accessible public transport’ (Province of Gelderland, 2010)

‘Ambition: realize a coherent and recognizable public transport’ (City district Arnhem Nijmegen, 2010)

‘Extension and improvement of public transport is the leading theme in our transport policy’ (Municipality of The Hague, 2009)

Our five spearheads are operational excellence, high frequent rail transport, more space for rail freight, custom made service for the regions and serving public interest in a businesslike way.’ (ProRail, 2010)

‘Mission: to transport more passengers safely, on time and comfortably using attractive stations’ (NS, 2010)

All these actors articulate the public interest of train services, but each in slightly different terms. At first sight they essentially share the same mission: improving public transport. But when looking closer, the realization of this value reveals possible conflicts. This is illustrated by a simple example: when the municipality of The Hague wants to improve public transport, they might desire extra stations in and around The Hague. If the Ministry of Transport wants to improve public transport, they might want to improve travelling time between cities, and therefore opening fewer stations. Hence, despite the shared public value, different scope and interests of governments can result in conflicting demands.

1.3 Dealing with the multi-layered public sector

In practice we see that the mutual cooperation of NS and decentral governments is far from a textbook example of allied actors in fruitful cooperation. This is illustrated by the example in the text box below.

Example1: Participating in long term policies beyond the concession term

Regional governments in and around Amsterdam work together to enlarge the international competitiveness of the region in an organization called ‘Metropolis Region Amsterdam’ (MRA). The MRA organization invites NS to participate in a discussion about the accessibility of the MRA region in the future. One of their goals is to create a vision for the mobility of their area for the period 2020-2040. This vision will be created with input of local public transport companies, civil servants of involved decentral governments (municipalities, city districts and provinces), ProRail and hopefully NS.

NS has the exclusive rights to operate the central rail network in the Netherlands up to 2015. Off course NS is keen on operating the main rail network after 2015 as well. The national government tends to extend the concession, however this is not guaranteed yet. So it is difficult for NS to develop a strategy beyond 2015. A NS vision for the period 2020-2040 simply does not exist. The absence of this vision makes it difficult for NS to participate in the discussion about the MRA vision for the period 2020-2040. Another issue is that NS fears that when a new regional tender is organized while they participated in the preparation of this concession, they are not allowed to bid for this concession.

So how should NS participate in this MRA discussion? On the one hand they are very interested in future transport development in this area. On the other hand they do not want to get involved in a way that influences their bidding position for the concession negatively. The central dilemma is that NS and decentral governments are willing to cooperate, but their different time horizons and interests make it very difficult to do this effectively.

In pursuit of more or less the same mission, the Dutch railway sector and governments appear to struggle in complex, multi-layered relationships that are difficult to manage.

1 Based on interviews with NS employees 19

Besides the national government, three levels of decentral governments exist in the Netherlands: provinces, city regions and municipalities. This is visualized in figure 1.

Figure 1: Multi-layered governments (blue layers represent the 3 decentral governments)

The relatively new governmental layer of city regions plays an important role concerning public transport. There is a lot of knowledge on public transport present at the city regions. This is probably the consequence of the new responsibility of city regions to organize the public tendering of bus and certain rail connections. Improving public transport is the core business of the city regions. As a consequence, they gained a central position in developing regional transport.

The complexity of the relations between the railway sector and the different layers of government can be shown by presenting an overview of the relationships. To do this in an orderly way, the Dutch public transport market needs to be described shortly. The relationships between multi-layered governments and market parties are visualized in figure 2.

Figure 2: Relationships between multi-layered governments and market parties (based on scheme ‘samenhang en verantwoordelijkheden’, Roeleveld, 2010)

The railway network is divided into two parts: the central rail network and the regional rail network. The Ministry of Transport awards the concession for the central rail network to one party for several years. At this moment, NS has the exclusive right to exploit train services on the central rail network until 2015. The Ministry of Finance is the only shareholder of NS.

For the regional rail networks, provinces and city regions serve as a transport authority, awarding the specific concession to one of the market parties. Currently, five market parties are active on the regional rail net: NS, Arriva, Veolia, Connexxion2 and Syntus. ProRail is the company responsible for rail construction and maintenance on both rail networks. International rail transport and freight transport fall outside the scope of this research.

Provinces, city regions and municipalities also serve as a transport authority for the bus, tram en metro networks. The majority of the parties that is active on the regional rail networks is also active in this market.

The variety of the projects that are performed within these networks is large. The yearly turnover of the central railway network for example is €1.8 billion. The size of the projects in

2 Veolia and Connexxion will probably merge (ANP, 2010) 21 the regional railway network is significantly smaller. To give an impression: the biggest regional concession (region Groningen-Drenthe) has a turnover of €80 million (NS, 2009). A classification of projects where NS meets multi-layered governments will be discussed in chapter 3.

1.4 Deciding on urban, regional transport plans: from desires to decision

As can be seen in figure 2, NS plays an important role in the public transport sector. NS is the biggest public transport company in the Netherlands. Every year 9 million Dutch citizens use the train to travel from one of the 278 NS exploited stations to another, thereby travelling 16.2 billion passenger kilometers (NS, 2009). NS has to deal with different, sometimes conflicting, interests of multi-layered governments. They face several complexities in managing these relationships.

The influence of the decentral governments has increased over the years. It can be questioned whether NS anticipated this development. In the years between 1990 and 2000, decentral governments had no budget to invest in rail projects as this was part of the responsibility of the Ministry of Transport. Nowadays the majority of the total budget of city regions, for example, is available for public transport. Therefore, the focus of this research will be on the relationship between NS and decentral governments.

Before a project is defined, NS and decentral governments have to decide if and how they want to cooperate. Most of the time the cooperation starts with a desire to realize a certain project. This might be a desire from NS or a desire from a decentral government. Then, negotiations start about a possible cooperation. Eventually a decision can be made about the shape of the cooperation. After this decision, the project has to be executed. However, the execution is not part of the scope of this project. Execution of projects can be very interesting, for example from a project management perspective. But this research tries to answer the question how NS and decentral governments cope with dilemmas to realize public values when setting up a cooperation, and therefore starts with the desires and ends with the decision.

It is chosen to focus on urban regional transport plans, as in urban areas multi-layered governments are present. City regions play an important role in developing transport in urban areas. The more levels of governments are involved, the more complex a projects is. A study of complex projects that are performed can provide an insight in the cooperation between NS and decentral governments.

This research has the opportunity to study NS from within. The principal for this research is the Business Development department. Appendix A shows the organizational structure of

NS. The research fits well within the strategy of NS. One of the three pillars of the strategy of NS is ‘government as a client’ (NS, 2010). An operationalization of this question reveals that the goal is a structural improvement of relationship with decentral governments. This research can contribute to that goal.

1.5 Research question and thesis outline

Following from the previous explorations, we formulate the central research question as follows:

How does NS cooperate with multi-layered governments to reach decisions on new urban regional transport plans and how can this cooperation be improved?

This question will be answered in the following steps. Firstly, based on theories an analytical framework is constructed (chapter 2). This analytical framework consist of three parts: the multiple principals problem, wicked problems and process management. Chapter 3 elaborates on the research approach. In order to get more insight in the real life negotiations that took place two case studies are performed. In total 27 people are interviewed. All interviewees are closely related to the decision making of the cases, working for NS, ProRail or a decentral government. The findings of the case studies are described in chapter 4 and 5 in the format of the analytical framework. In chapter 6 conclusions are drawn, based on a comparison of the two cases. The main conclusion is that NS in the cooperation with multi- layered governments faces the multiple principals problem and should use process management. Currently, the process is under designed resulting in the unsatisfactory situation in which NS is blamed for acting like a monopolist. In chapter 7 recommendations are formulated to improve the relationship with decentral governments. To conclude, a reflection on the research is presented in chapter 8.

Chapter 2. Theory

The goal of this research is to give more insight into the way the complex relationship between NS and multi-layered governments is shaped. A wide variety of scientific disciplines relate to this topic. In this chapter, the most relevant theories relating to the research question are discussed. Therefore, in three steps an analytical framework is developed that serves as a base for the case studies. Firstly, the multiple principals problem is discussed (2.1). Secondly, the term ‘wicked problems’ is introduced to describe the type of problem NS is dealing with (2.2). Thirdly, it is showed that theory about process management can contribute to deal with multiple principals and wicked problems (2.3). After discussing these theoretical notions a summary is presented in an analytical framework in paragraph 2.4. This framework serves as a base for the case studies.

2.1 Multiple principals problem

Firstly, attention is paid to the fact that NS has to deal with different layers of government with conflicting interests. This problem is described by introducing the multiple principals problem.

Conflict between principal and agent In literature interesting notions can be found about cooperation between parties with different interests. One of the most used concepts to illustrate this kind of cooperation is the principal-agent model. Central to principal-agent thinking is a conflict between the agent objective function and the principal objective function (Laffont and Martimort, 2002). When used in this research, the Dutch government serves as a principal, protecting public values. NS represents the agent, trying to optimize their interest and thereby having the incentive to shirk. Literature describes two main categories of shirking: adverse selection and moral hazard. Adverse selection means that the agent seduces the principal to agree on certain incentives while hiding knowledge of the actual effects of these incentives. Moral hazard means that after applying adverse selection, the agent tries to hide the actions and decisions it undertook in response to the incentives (Laffont and Martimort, 2002). Both mechanisms have the consequence that the principal loses control over the agent.

Multi-layered governments As said, until 1993 the organization of the Dutch railway sector was relatively simple. All railways were managed, owned and operated by NS and NS had one shareholder: the Dutch government. The classical principal-agent model can be projected on this situation. However, in the current situation NS no longer has to serve only one principal. The decentral governments gained influence and have more steering possibilities towards NS. Some

24 decentral governments want to have even more decentral influence. This is illustrated by an ongoing discussion about the possible decentralization of regional trains, put forward by decentral governments like the Province of Gelderland (Maarten, 2008). So the situation where NS is steered by the central government changed into a situation where multiple layers of government simultaneously gain more steering possibilities towards NS.

Multi-layered government = classical network? One could question if the multi-layeredness of governments is different from other networks. After all, in networks multiple actors make decisions in a dynamic context in multiple rounds (De Bruijn et al., 2008). However, there is one specialty to the relationship between the governments: despite the dynamic context and changing coalitions, governments have a hierarchic relationship defined by law. The national government is allowed to annul certain decisions of provinces, city regions and municipalities. And the province is allowed to annul certain decisions of municipalities, for example when a municipality does not want to cooperate with the construction of a new highway. That is why figure 1 is displayed in a vertical way. The hierarchic power of higher governments is seldom used, because in a way higher governments are also dependent on lower governments. Lower governments are often responsible for the execution of policy of higher governments. Therefore a good relationship is needed and the use of hierarchic power might harm that relationship. Because of this special relationship of governments, the term multi- layered governments is introduced.

NS serves many principals… Due to the presence of multi-layered governments, NS can be considered an agent not serving one principal, but many principals with conflicting interests. In literature this is called the multiple principals problem. Multiple principals may have overlapping interests but also competing interests in the absence of an imposed hierarchy (Scott, 2000). The outcomes of these interacting incentives may not always be predictable (Miller and Whitford, 2002). A possible strategy to agents is to play principals off against one another (Wood and Waterman 1994). Next, agents might strategically use information asymmetry in a setting of multiple principals (Miller, 2005). Steenhuisen (2010) states that multiple principals are not necessarily losing control to the agent. He concludes that in a multiple principals setting agents face new opportunities to shirk but without the opportunity to refrain from shirking and without the capacity to manage the trade-off. This insight can also be used to further understand and study the problem of the multi-layered government. It is not clear if and how NS prioritizes the competing interests of their principals.

… strategic behavior is likely to occur The principal agent model and the multiple principals problem already addressed the risk of strategic behavior, mentioning adverse selection, moral hazard and information asymmetry. When discussing institutional changes such as privatization and liberalization, strategic

25 behavior is in literature often mentioned as an explanation for disappointing results. According to Ten Heuvelhof c.s. (2009) strategic behavior is reflective, takes shape in interactions, has a time dimension, is aimed at narrow self-interest, is ambiguous and is intentional. Obviously, this definition has a negative connotation. There are three breeding grounds for strategic behavior: - Fewness: a limited number of dominant companies with a disproportionate amount of market power. - Position: an established position, but also acquiring a new position. - Information asymmetry: differences in information provision between actors.

When looking at the Dutch railway market, these breeding grounds seem to apply. There are only a few number of market parties active in this sector, with one dominant player. Secondly, the dominant player NS has an established position because of the historical background where all railways were state owned. Thirdly, because of this position NS has access to unique information. This means that according to this theoretical classification the presence of strategic behavior is present.

2.2 Wicked problems

Deciding on regional transport plans in urban areas is a special type of problem, as there are many different visions about the problem definition. The lack of clarity about the problem definition makes it difficult to structure the problem. In literature these problems are called unstructured or wicked problems.

Different definitions of wicked problems De Bruijn (2008) uses the term unstructured problems. He defines unstructured problems as problems where no objective information is available, there is no consensus about the norms that should be used for the problem solution and the problems and solutions are dynamic. Coyne (2005) uses the term wicked problems and describes them as loosely formulated, subject to redefinition and resolution in different ways over time and not objectively given. The formulation of wicked problems already depends on the viewpoint of those presenting them. There is no ultimate test of the validity of a solution to a wicked problem. Pacanowsky (1995, p.2) adds that wicked problems have an iterative nature, as it is needed to cycle through the phases of problem definition, information gathering, solution and outcome. Wicked problems are not really ‘solved’, rather a more or less effective solution is ‘designed’ based on how the problem is defined. According to Coyle the problem- solving process is characterized substantially by talk, rather than by bodily dissection, surgery and other diagnostic procedures that perhaps promote metaphors of surface and depth.

Networks instead of hierarchies 26

The terms Coyne and Pacanowsky use show some similarities with the vocabulary of De Bruijn when discussing multi actor networks. The essence of multi actor networks is that an actor that wants to change a certain situation, is dependent of other actors. In a hierarchy, it is possible to realize this change with a unilateral action by command and control. But in a multi actor network, actors will not be convinced purely on content arguments mentioned by the initiating actor. Actors need to be involved with the change in order to support it. Decision making in a network is irregular and develops in rounds (De Bruijn et al., 2008).

So wicked problems lack a central, unambiguous problem definition and in multi actor networks there is no central actor that can unilaterally determine a solution. The implication of this notion is that interaction is needed in order to define, or (according to Pacanowsky, 1995) to ‘design’, a solution for a wicked problem.

Expectation: multiple principals, wicked problems need process management It is expected that the empirical study will show that a multiple principals problem can be detected, in which involved actors do not agree about the problem definition. This will probably result in a wicked problem, in which actors will behave strategically. If this is supported by the empirical study is described in chapter 3 and 4. The consequence of this expectation is that it is not possible for NS to unilaterally decide on how an urban regional transport plan should be realized. As NS is acting in a multi actor network, attention has to be paid to the design and management of the process (ten Heuvelhof, 2009). The next section explains what is meant with process management and why it is needed.

2.3 Process management

Now the expected type of problem is defined, it is possible to investigate strategies to deal with these kinds of problems. The choice is made to focus on the theory of process management. Paragraph 2.3.1 motivates the choice to use process management. The four core elements of process management are presented in paragraph 2.3.2. Next, the importance of good timing for process management is demonstrated in paragraph 2.3.3. To conclude, critics to process management are summed up in paragraph 2.3.4.

2.3.1 Why use process management?

According to Ten Heuvelhof (2009) a consequence of acting in a multi actor network is that decision making is capricious and unpredictable. As in practice there is no hierarchically superior actor present, the decision making should be process based. The attention therefore shifts from a project approach to a process of interaction. This implies a less straightforward approach, a less certain and less step by step way of working towards the 27 plan. A more flexible attitude is needed with which it is possible to recognize and deal with dynamics during the realization of a regional transport plan. Assuming the empirical study will support the theoretical analysis that NS and decentral governments are situated in a multi actor network, the theoretical notions of process management are very relevant to describe. De Bruijn et al. (2008, p.21-25) describe six specific arguments to use process management.

1. The reduction of uncertainty about the content Wicked problems have the characteristic that there is no objective information available. Involved actors have access to different sources of information, but there is no consensus about this information. A confrontation of the sources is needed to improve the quality of the information. 2. Enrichment of problem definitions and solutions Wicked problems also have the characteristic that there is no consensus about the problem definition and solution. Interaction between actors is needed to enrich the problem definitions and solutions. 3. Incorporation of dynamics in time New information need to be involved in the process, otherwise complaints may rise about the usability of the ‘old’ information. 4. The creation of (societal) support for a certain solution When actors are not involved in the process of decision making, they might block the decision making in a later phase. 5. Depoliticize decision making The presence of multi-layered governments can result in contradicting demands towards NS. A mayor of a municipality has different political interests from a deputy of a province. Process management can help to depoliticize this situation. 6. Transparency Applying process management, actors can reach consensus about how to follow a certain process, e.g. by fixing go/no go moments or rules of the game. This will make it possible for actors, inside or outside the process, to get insight into the decision making.

These arguments all apply to the problem situation described in chapter 1. In a real life example described below, all six arguments are demonstrated. It is shown that in a multi actor decision making, without using process management, undesirable situations arise.

Example: Lobbying for railway station Maarheeze3

Introduction Maarheeze is a village located in the southern part of the Netherlands, close to the city of Eindhoven. Many people living in Maarheeze want to have a connection to the railway. Therefore, in 2004 a group of citizens raises a foundation that starts to lobby for a railway station in Maarheeze. Soon the municipality of Cranendonck, the city district of the region Eindhoven (Stadsregio Eindhoven, SRE) and the province of North Brabant support the citizens in their goal to open a station. The plan for a station in Maarheeze also fits very well within the accessibility program that is formulated by five cooperating cities in the province of North Brabant. So all levels of decentral government are in favor of constructing this station. Their main arguments are the accessibility of Maarheeze, the possibility to attract car drivers stuck in traffic jams nearby Maarheeze and thereby a reduction of the congestion in the Eindhoven area.

1. Uncertainty about the content NS knows about the plans to construct a station in Maarheeze, but is not very enthusiastic about the idea. Maarheeze is a small village, so the station will only serve stopping trains. NS states that generally operating stopping trains is not a profitable activity and a new station would increase the losses. At that time, the Ministry of Transport uses the standard that every new station should add 1000 new travelers to the existing amount of train travelers. NS calculates that when the Maarheeze station is opened, about 800 new travelers are expected. As this is below the norm of the Ministry, NS is not a proponent of constructing a new station in Maarheeze.

2. Discussion about problem definition and solution But NS is willing to discuss additional measures that can compensate for the financial risk. Decentral governments are willing to pay an operation subsidy to NS. Besides that, SRE is willing to adjust the schedule of busses to the schedule of trains, in order to let busses fulfill a feeder function for the stopping trains. As SRE is the transport authority for all bus concessions in their region, it has the power to realize this. NS and SRE start to negotiate about these issues and in the end an agreement is made. A so called service guarantee contract is signed. This step is crucial for ProRail (responsible party for the basic infrastructure of new stations), because it only starts to construct new stations after such a contract is signed.

3. Dynamics But then unexpected developments take place. The province and SRE are supporting low tariffs for busses in their area. The province even performs experiments with free busses. As a consequence, it is much more attractive for citizens to travel by bus than by train. NS fears that the calculated number of possible train passengers taking the train in Maarheeze is estimated too high, because in the mean time the attractiveness of busses raised. Besides that, people living in Maarheeze are realizing that the construction of a new train station will replace some of the existing bus connections. As train tariffs are not part of the discount actions, people start to fear they have to pay more for public transport than they have to do at the moment.

4. Societal support A new action committee is started in Maarheeze, which is trying to preserve the existing bus lines. They blame NS for acting like a monopolist, trying to replace cheap busses by expensive trains. The action committee effectively organizes protests. A representative of the Socialistic Party in the Second Chamber of the Dutch parliament asks questions about this issue to the minister of Transportation. NS is very unsatisfied with the suggestion that they are trying to compete with the busses, as at first NS was not interested at all in opening a station in Maarheeze.

3 Based on interviews with NS employees 29

5. Politicizing SRE and NS start to negotiate again. SRE is willing to consider extra compensation for the traveler. But they also ask NS to send a request for compensation to the province, as the province also contributes to the low bus tariffs. NS however focuses on the appointments made in the service guarantee contract, which is signed with SRE. After many rounds of negotiations a decision is made. For the next years the busses will still run. SRE will be fully responsible for the financial compensation to NS for the amount of travelers below the standard. In order to decide on the height of this compensation extensive counting will be organized.

January 2010 the construction of the new station Maarheeze started. According to the planning, the first stopping trains will serve station Maarheeze in June 2010. In the mean time the congestion at the roads around Eindhoven is already heavily reduced, due to the construction of new roads.

6. Transparency This example shows that both NS and decentral governments try to serve the passenger. However, the measures they take in order to offer good public transport are interdependent and contradictory. The process and the outcome of the process is not transparent to outsiders. In the end, their efforts to improve public transport do not lead to a desirable outcome for travelers. The central dilemma is that despite the efforts to serve public interests, the process and outcome are unsatisfying to almost all parties involved.

2.3.2 Core elements of a well-designed process: openness, protection core values, speed & content

In the previous paragraph it was demonstrated that process management fits well with the studied problem of deciding on urban, regional transport plans. Six arguments were used to show the appropriateness of process management for wicked problems that occur in multi actor networks . The next step is to investigate if and how NS applied process management as a strategy in practice and what the results are. This will be done by looking at the four core elements of a well-designed process that de Bruijn et al. (2008) describe:

- Openness: this value concerns an open attitude from the initiator towards possible participants and towards the agenda of decision making. This can help to deal with the wickedness of problems, as the different opinions about norms and facts can be discussed with involved actors. - Protection of core values: in order to give participants to a process a feeling that their core values are going to be protected. This means that whatever the outcomes of the process may be, their own core values will not be affected. - Speed: the elements of openness and protection of core values might hamper the speed and progress of a process and therefore need specific attention. - Content: the risk of focusing on the process is that the content of the decision making is neglected.

Based on these four core elements, de Bruijn formulated 12 design principles, which are displayed in table 1.

Core element # Design principle Openness 1 Involve all relevant parties 2 Transform choices related to content into process appointments 3 Transparency of process and process management Protection core values 4 Protect the core values of parties 5 Commitment to the process, not to the result 6 Presence of exit rules Speed and progress 7 Profit as incentive to cooperate 8 Creation of quick wins 9 Appropriate composition of delegation 10 Conflicts are organized as deep as possible in the process Content 11 Roles of experts and concerned are separated and interweaved 12 From variety of content to selection Table 1: 12 design principles for good process management

In the empirical study it will be checked if the four core elements are taken into account and what the effects are. If possible, attention is paid to the specific design principles as described in the table 1. In the empirical study it is also useful to describe how NS prepared their participation in the process. This question will therefore also be addressed in the analytical framework.

2.3.3 Timing of the policy process

No rational decision making An important element of process management is the timing related to the policy process. Several models are developed that describe the process of agenda setting and the policy process. A famous model is the garbage can model (Cohen, March and Olsen, 1972). This model consists of four separate streams: problems, solutions, participants and choice opportunities. The outcomes of decision making are a function of the mix of garbage in the can and how it is processed. It is developed contradicting the rational decision making.

Problems, policies and politics Kingdon (1984) used the garbage model as a basis, but changed it into a model with three streams: problems, policies and politics. Firstly, a stream of various problems captures the attention of people in and around government. Secondly, a stream of policies is produced by a policy community, specialists who concentrate on generating proposals. Thirdly, a stream of politics is composed of public opinion, election results, campaigns of interest groups, etc. Events in the political stream occur independently of the problem and policy streams. Decision making then, is about coupling these three streams. A coupling of the streams

31 results in a so called ‘policy window’: a problem is recognized, a solution is developed and available in the policy community, a political change makes it the right time for policy change. Policy windows generally stay open for only short periods of time. The major strength of Kingdon’s model is that it recognizes that the policy process is fluid and non- linear, and that it involves a vast number of actors and forces. This fits well in the context of multi actor networks where unstructured problems are handled.

Sense of urgency In addition to this policy window De Bruijn et al. (2008) stresses the importance of a sense of urgency. A constraint for a successful process is the need for a sense of urgency consisting of two components: one related to content and one related to the process. Firstly, actors need to be convinced that there is a problem that needs a solution. Secondly, actors need to be convinced that for the realization of this solution cooperation in a process is needed.

The theories about policy windows and sense of urgency are very interesting to NS in their cooperation with decentral governments. When they collectively formulate problems and work on policies, NS can try to influence the political stream and thereby might create a policy window. Creating a sense of urgency can contribute to the realization of a policy window.

2.3.3 Possible critics to process management

There are two main critics to process management. First, it is feared that process management can lead to slow decision-making (De Bruijn et al., 2008). After all, when many parties are involved in decision making, there are also plenty of possibilities to impede the process. Therefore, much attention needs to be paid to the third core element of process management: speed. Besides that, it might be possible that without applying process management the process would not come to a successful end at all. So it might take longer to end up with a solution, but at least a solution supported by a sufficient number of actors is realized.

Another critic concerns the lack of substantive contribution. The term ‘process management’ implies a focus purely on processes, raising the question how the content is safeguarded. In practice, process management tries to add process elements, not to replace the content. As described in paragraph 2.2.4, ‘content’ is even one of the four core elements of a process. In the empirical study, attention is paid to the balance between content and process elements.

2.4 Analytical framework

The theories described result in an analytical framework as displayed in figure 3.

•Consensus about facts? •Consensus about norms? Wicked •How is the problem defintion shaped? problems

•How was the process prepared? •Four core elements taken into account (openness, speed, content & core values)? •Policy window present? Process •Sense of urgency present? management

Figure 3: Analytical framework

Chapter 3. Research approach

The previous chapter serves as a theoretical basis for the research. In this chapter it is emphasized how the formulated research question can be answered.

How does NS cooperate with multi-layered governments to reach decisions on new urban regional transport plans and how can this cooperation be improved?

To answer this research question, several research methods are possible: experiments, historical research, case studies, simulation, etcetera. In this chapter the scientific justification for the chosen research is presented. Firstly, how and why the choice is made to perform case studies will be discussed. Secondly, the process of selecting the case studies is described. Thirdly, the choice for performing interviews as main research technique is discussed with respect to its potential, its limitations and how to maximize its potential.

3.1 Investigate real life cooperation

3.1.1 Looking for the right research strategy

As described in paragraph 1.4 this research focuses on the realization of cooperation with decentral governments, not on the execution of projects. In this phase of decision making NS has to decide if and how they have to deal with decentral governments. The central research question is formulated as a ‘how’ question, because the aim is to get more insight into the way relations currently evolve. According to Yin (2002), experiments, history or case study is one of the preferred strategies to answer such a ‘how’ question. This section will argue why case studies are preferred over experiments and history.

Experiments are performed when an investigator can manipulate behavior directly, precisely and systematically. This usually occurs in a laboratory environment. The cooperation of NS and decentral governments consists of many people doing business at many places. In this setting it is impossible for the researcher to manipulate the behavior of people involved directly, precisely. It concerns real life negotiations, which evidently cannot be transferred to a laboratory environment. Therefore experiments as a research strategy make no sense for this research.

The historical method is dealing with the ‘dead’ past. The main source of evidence consists of studying documents, as no relevant persons are available to report (Yin, 2002). For this research, the researcher has access to many people with a lot of experience in the

34 cooperation between NS and decentral governments. The cooperation is not concluded and cannot be considered as a ‘dead’ past. Therefore, this method does not match this research.

The essence of a case study is that it tries to illuminate a decision or set of decisions: why they were taken, how they were implemented, and with what result (Schramm, 1971). A second important argument to perform case studies is that it studies developments traced over time. Thirdly, case study is the preferred strategy when contextual conditions are important to consider (Yin, 2002). For this research it is very important to get deep understanding of how the real life cooperation between NS and decentral governments takes place. Case study fits this goal very well. By zooming in on a specific case, it is possible to investigate several issues related to the cooperation. The analytical framework formulated in paragraph 2.4 can be leading in performing these case studies. For these reasons the case study is preferred over the experiment and historical research.

3.1.2 Multiple cases, holistic design

Now the choice is made for performing case study, the exact type of case study needs to be determined. According to Yin (2002) it is possible to make a choice on two dimensions. The first is a single-case design or a multiple-case design, the second is a holistic or embedded analysis. For this research the choice is made for a multiple case, holistic design. Firstly, the choice for performing multiple cases is made because the analytic benefits from having two cases are substantial. Analytic conclusions independently arising from two cases will be more powerful than those coming from a single case alone (Yin, 2002). The evidence from multiple cases is often considered more compelling; and the overall study is regarded as being more robust (Herriott & Florence, 1983).

Secondly, the choice is made for a holistic analysis. This means that the case study examines the global nature of an organization, in this case the global nature of NS and decentral governments. If the choice for a (multiple-case) embedded design was made, cases in other contexts would need to be researched, e.g. the cooperation of energy providers with decentral governments. It is chosen explicitly to investigate the specific cooperation of NS with decentral governments and therefore the holistic approach is more suitable. In addition, the holistic approach is advantageous when the relevant theories underlying the case study is itself of holistic nature. As shown in the analytical framework, the main theory used is about process management. This broad theory can be considered holistic and therefore a holistic design is used for the case studies.

The results of the performed case studies will be reported separately, by making use of cross case synthesis. Cross case synthesis is a technique that treats each individual case study as a separate study. In addition to a chapter for both cases (chapter 4 and 5), a chapter is written

35 with the conclusions resulting from a comparison of the cases (chapter 6). The focus is on argumentive interpretation, numeric conclusions are not possible because of limited amount of data available.

3.1.3 Possible critics to case study

When discussing case study as a research method, three common critics are mentioned..

Firstly, a lack of rigor is a major concern. In reaction to this concern, extensive thoughts are given to the research setup before the case studies are performed. The basis of the analytical framework, the choice for a holistic multiple case design and the selection criteria for the cases were all formulated before the interviews took place. This prevents biased views to influence the direction of the findings.

Secondly, it is feared that case studies provide little basis for scientific generalization. However, the choice for multiple cases increases the possibilities for generalizability. The contexts of the two cases may differ to some extent. If under these varied circumstances you still can arrive at common conclusions from both cases, they will have immeasurably expanded the external generalizability of the findings (Yin, 2003, p.53).

Thirdly, performing case studies often takes too long or results in massive, unreadable documents (Yin, 2002). In the preface it was already explained that this research took time under high time pressure, which had a positive influence on the lead time of this research. To prevent the creation of an unreadable document, an analytical framework is developed. This framework can be used for the documentation of the case studies.

A last critic specific for this research could be the risk of ´going native´, being wrapped up in the worldview of the main stakeholder under study (NS). It has to be prevented that the research depends on information from NS only. This independence is safeguarded by involving many people with a different background. Yin (2002) also stresses the importance of multiple sources of evidence. Therefore an extensive study is done in which semi structured interviews are held with people from NS, decentral governments and other organizations (appendix D), internal NS documents and external documents are studied, and observations are performed by visiting conferences. Validation took place by asking feedback on intermediary results to involved actors. Together, these sources of evidence can contribute to an unbiased, well-founded research.

3.2 Case study selection

NS and decentral governments meet in many arenas. Interviews have been held with the four NS regional directors in order to get more insight into the different types of projects where NS cooperates with decentral governments before selecting the cases. During these interviews the projects that might be interesting for the case study were identified. Projects that were performed too long ago are not mentioned in this list, because it would be too difficult to retrieve the process of decision making at that time. The collection of projects during the interviews resulted in a long list of 41 projects, listed in appendix B.

To perform case studies, a selection has to be made out of these 41 projects. The list shows that the diversity of the performed projects is high. In order to structure this list, the scope of the projects is divided into three categories of development:

- Development of infrastructure, e.g. constructing new tracks - Development of transport services, e.g. raising the frequency of trains - Development of station area, e.g. improving the existing station hall

Projects contain one, two or three of these developments. This division is made to give insight in the size and complexity of the projects. A project containing the development of infrastructure, transport services and a station area is thought more complex than a project considering the development of a station area only. The more complex the project, the more clearly the dilemmas between NS and decentral governments will be visible. Based on this categorization, the first selection criterion can be formulated: the complexity of the performed project. The standard for this criterion is that a project for the case study should contain all three categories of development (infrastructure, transport services and station area).

The second selection criterion is based on the multi-layeredness of the involved governments. Figure 1 showed the complexity of four layers of governments, of which three are decentral governments. To get insight in the dilemmas between these layers, it is important that the case study concerns projects where all levels of decentral government are involved: province, city region and municipality.

The third criterion concerns not the selection of the individual cases, but the selection of the set of cases. The characteristics between the different areas in the Netherlands might differ. For example, the political culture differs a lot in every region. Therefore it is chosen to select case studies with a geographical spread. For this classification the division that NS uses is applied. They distinct four regions in the Netherlands: Randstad South, Randstad North, South East en North East.

When these criteria are applied to the projects, a short list can be produced, presented in table 2.

Project Complexity Governments involved NS Region Area development 3 categories of Ministry of Transport North East Arnhem-Nijmegen development Province of Gelderland City region Arnhem Nijmegen Several municipalities Programme Accessibility 3 categories of Ministry of Transport Randstad North North Wing development Provinces of North Holland & Flevoland City region Amsterdam Several municipalities Network study 3 categories of Ministry of Transport, Province of Randstad South Stedenbaan development South Holland City regions Rotterdam and Haaglanden Several municipalities Table 2: Short list of possible case studies

This short list was discussed with the supervisors of NS. Two remarks about the short list were made. The first remark is about the risk of conflicts of interest. During the selection of the projects, it became clear that the writer of this thesis might become vice mayor in the city of Delft. As the city government in Delft is heavily involved in the Stedenbaan project, it was doubtful for NS to use whether this project could be used as a case study. When it became clear that the researcher would become responsible for spatial development in the municipality of Delft, the risk of conflicts of interest were even bigger and therefore Stedenbaan was excluded as case study for this research. It was estimated that for these two cases more than 20 interviews are needed, which fits very well within the scope of a Master thesis. In the next section, the preparation for the interviews will be presented.

3.3 The art of interviewing

In general interviews can be an important information source for case studies. People involved in the cooperation between NS and decentral governments have unique knowledge about this cooperation, which cannot be found in documents. In order to collect as much of this knowledge as possible, the setup of the interviews must be well prepared. This paragraph describes how the interviews are prepared and which practical choices are made to collect as much useful information as possible from the interviews.

Firstly, the type of interview needs to be determined. The main goal of the interviews is to acquire more insight in how the cooperation between NS and decentral governments takes

38 place in practice, without knowing on beforehand what this looks like. Verschuuren (1986) points out that insight in causal relationships without knowing on beforehand how this relationship looks like is an important element of exploratory research. Baarda (et al. 1996) describes that for exploratory research it is advised to make us of semi structured interviews. This type of interview fits very well with the part of the research that is aimed at designing a strategy for NS to improve the stakeholder management with decentral governments.

Secondly, attention is paid to a practical issue: reporting the interviews. It is important to create an atmosphere of trust. Therefore, it is chosen not to record the interviews. Interviewees should be motivated to be as honest as possible. When they know they will be recorded, they will probably hesitate to share confidential information. Therefore only notes are made during the interviews, which are drawn up immediately after the interview.

Thirdly, the time to perform the interviews is relatively short. Most of the interviewees allocate 1 to 1,5 hour in their agenda for the interview. In order to use the scarce time efficiently, no time can be spoiled with questions that do not directly contribute to the research. Therefore three measures have been taken:

- Before the interviews took place, the subjects that need to be discussed were announced by email. Examples of questions were displayed, so interviewees would not be surprised about the kind of questions they got during the interview. - In the introduction of the interview the researcher shortly explained what his level of knowledge was. It was stressed that during the interview the main subject should not be the detailed content of the project, but the way the cooperation evolved. - At the end of the interview people were asked if they would be willing to answer remaining questions on a later moment in time. All interviewees agreed with that extra option to complete the interview.

Fourthly, to structure the interviews a protocol has been developed. This protocol serves as a base for all interviews. The protocol is based on the analytical framework and is displayed in appendix C. The type of interviews used in this research is called focused interviews (Merton, Fiske & Kendall, 1990). This means that interviews are open ended, but are also following a certain set of questions derived from a protocol.

The use of a protocol raises a dilemma. On the one hand it is useful to have a list of questions that can be ‘checked’ during the interviews in order to answer the questions stated in the analytical framework. On the other hand it is important to focus on the global dilemmas that cannot be predicted in advance, in order to discuss the essence of the problem. In practice, both strategies need to be used. The protocol contains a long list of interesting questions, which form the guideline of the interview. However, during the

39 interview this list of questions should not be followed too rigidly. When the point is reached that an interesting dilemma appears, more time needs to be spent to continue asking questions about this dilemma.

Fifthly, the right interviewees need to be selected (displayed in appendix D). For both cases an effort is made to select the persons that were very closely involved in the process of decision making. The closer the involvement, the more insight in the dilemmas of the cooperation they can provide. For the first case this was relatively easy, as there was a regular consultation in which all involved actors send a fixed representative. These representatives were easy to trace and all willing to cooperate with the interviews. For the second case it was more difficult to select the right interviewees as it was not immediately clear who were involved in the decision making. Therefore, first interviews have been conducted with NS representatives who were involved in the process. They provided more insight in the key players that had a role in the process. Based on these insights, the list of relevant interviewees could be completed. By interviewing different people who were participating in the same process, it is possible to explore the different viewing points of actors towards the same problem. It is likely that actors have a different perception of how the process developed. These different perceptions can lead to a rich picture of the dilemmas that occurred in the cooperation. In the case study chapters attention will be paid to the different viewpoints of the actors.

Sixth, a used strategy is to ask external stakeholders to rate NS on five dimensions: consistent vs. inconsistent, decelerating vs. accelerating, formal vs. informal, predictable vs. unpredictable and social vs. commercial. These five dimensions are chosen in cooperation with the regional board, as they expected interviewees to have a strong opinion about these dimensions. The goal of this exercise is not to collect numerical information, but to get more insight in the way stakeholders experience the cooperation with NS. In practice this strategy worked very well, because interviewees came up with interesting examples that formed a good lead of departure to zoom in on the cooperation with NS.

The minutes of the interviews are confidential, but in the case study chapters anonymous quotes are used to give an impression of the interviews.

Chapter 4. Case 1: Programme Accessibility North Wing

This chapter describes the first case, based on the analytical framework introduced in chapter 2. Firstly, an introduction on the problems related to public transport in this area is presented (4.1). Secondly, the presence of the multiple principals problem is discussed (4.2). Paragraph 4.3 shows this case can be considered a wicked problem. Paragraph 4.4 demonstrates which strategy NS used during the process and what the effects are. In the next two chapter quotes collected during the interviews are used. Due to confidentiality reasons, the exact source of the quotes cannot be revealed. However, to place the quotes into context the quotes from NS interviewees are displayed in blue and the quotes of external interviewees are displayed in orange.

4.1 Introduction

The first area that is investigated is the metropolitan region Amsterdam. Both car transport and public transport are major bottle necks for the economic development of this region. Therefore the Minister of Transport in 2001 earmarked money to invest in an improvement of the public transport in and around Amsterdam (de Volkskrant, 2001). The budget for these plans was labeled as Accessibility Offensive Randstad (in Dutch: Bereikbaarheidsoffensief Randstad, BOR). This budget was labeled for investments on a national level, however the focus point was on the urban area around Amsterdam. The working title for the investments in public transport in this region was Programme Accessibility Northwing (in Dutch: Programma Bereikbaarheid Noordvleugel, PBN). As NS and external stakeholders refer to this project as ‘PBN’, from now on this abbreviation will be used for this case description.

The starting situation of this analysis is the moment when the Ministry of Transport announced to provide a budget of 171 million euro to improve public transport in and around Amsterdam, with the constraint that the decentral governments and NS should collectively formulate an investment programme. Decentral governments and NS started to negotiate about this programme in March 2005.

The decentral governments already met each other on regularly basis to discuss which measures should be taken to improve the public transport in their area. Collectively they wrote a vision on public transport called ‘RegioNet’. This vision was written without involvement of the railway partners.

The railway partners followed a different trajectory. At the end of 2001 NS started to write a vision about the future of the railways in the Netherlands, together with ProRail (rail

41 construction and maintenance) and Railion (rail freight). This resulted in the presentation of the vision of the united railway sector with the title ‘Utilization and construction’ (in Dutch: Benutten en Bouwen, 2003) two years later. The conclusion of the report was that the condition of the railways was below an acceptable level. Both the quality and capacity of the system was insufficient. The system was not customer-friendly and robust. In order to improve the railway system three priorities were formulated:

- Reliability: a strong improvement of the reliability of the complete railway system - Better usage: absorb the growth of transport by using the infrastructure more efficient - Investments: invest in expansion of the railway net

These abstract formulated priorities were made specific by presenting a set of measures for the short term and for the long term. For the PBN case, this meant that there were eight investment projects in infrastructure, one of which was the construction of a new station that was also desired by the decentral governments.

4.2 Multiple principals problem

Four layers of governments involved The decentral governments involved in the PBN case consist of the municipality of Amsterdam, the city region of Amsterdam, the province of North-Holland and the province of Flevoland (figure 4). Figure 1 presented an overview of the three levels of decentral governments. This case shows that decentral governments at all levels were involved. The ministry of Transport only announced a budget under certain conditions, and was not part of the process of decision making. Therefore it is displayed grey in figure 4.

Ministry: Transport & Finance

Province: North-Holland & Flevoland

City region: City Region Amsterdam

Municipality: Amsterdam

Figure 4: Involved decentral governments (blue)

Governments with contradictory demands NS faces many principals in this case. Firstly, they have to comply with the general financial standards. The Ministry of Finance as a shareholder sets certain standards, which are translated by NS into a norm that the Return on Investments should be 10% (NS, 2005). Secondly, the Ministry of Transport steers NS with the targets and obligations described in the concession for the central rail network. This means for example that certain punctuality standards must be met. Thirdly, decentral governments only want to cooperate with a joint investment program when certain of their demands are granted. In this case the demands of decentral governments are contradictory with the demands of the central government. Clearly, this is a situation where NS as an agent has to serve more principals with competing interests. This can be illustrated with an example.

Example: Flevolijn

One of the demands of the decentral governments was to run extra trains on the so called ‘Flevolijn’ (Amsterdam – Almere – Lelystad). According to the decentral governments these trains would add value because a growing number of passengers is using this connection. Besides that, it might help to solve the traffic jam on the roads parallel to the railway line. However, NS calculated that running these trains would result in a negative operational result of 6 million Euros per year (NS, 2005). So according to the standards based on the standards of the Ministry of Finance, this investment is not profitable enough.

In this case NS can choose both sides: support the demands of the decentral governments and decide to run extra trains on the Flevolijn or strictly follow the standards of the Ministry of Finance to prevent a negative operational result. The risk of the first solution is that NS will have a negative financial result every year. The risk of the second solution is that NS will not end up with a deal about PBN and thereby miss the opportunity to invest 171 million in the Amsterdam area. Eventually NS found a solution that lies somewhere between these two options. In the final PBN contract the running of extra trains on the Flevolijn is conditionally linked to the growth of passengers. When the market growth on this trajectory is 34% or more, NS will run extra trains. If this growth is not realized, NS will run no extra trains.

No rail competitors The area where the negotiations took place has an interesting characteristic in the perspective of strategic behavior: the absence of rail competitors for NS in this region. In several other areas in the Netherlands, other market parties like Veolia operate regional rail concessions. However, in this area there is only a concession for the central rail network, so until 2015, NS is the only market party.

This position matches the described breeding grounds for strategic behavior: fewness, position and information-asymmetry. NS indeed is blamed by decentral governments to display monopolistic behavior. If this is actually true cannot easily be said. In the beginning, at least the attitude of NS could be described as monopolistic. The central board of NS said that when they would not be able to agree on the deal with the decentral governments, they would arrange the money directly in ‘The Hague’ (where the national governments is seated). However, eventually the NS regional board tried to negotiate a deal in dialogue with

43 the decentral governments. This was commented on positively by the decentral governments.

“The cooperation in this project improved the position of the decentral governments. Based on the law our influence is limited, but due to this cooperation we gained influence.”

Proof of information asymmetry Strategic behavior occurs mostly around the information position of NS. NS was not able or not willing to give insight in their calculations and their tradeoffs to support or reject certain projects. This attitude feeds the perception of NS acting like a monopolist. NS has unique access to certain information. Firstly, they measure all kind of travelers statistics every day, resulting in deep knowledge of the behavior and wishes of train travelers. Secondly, they use advanced models to calculate the impact of different solutions. With a combination of these two information sources NS calculates what the effects of a certain investment in infrastructure is and if it is profitable enough. These calculations are subject to heavy debate. According to the decentral governments NS does not share enough information to get insight in why decisions are made.

“And then NS shows up again with their circle theory4 and calculations that nobody understands.”

NS on the other hand appeals to the confidentiality of the information. This results in a situation where NS has more access to certain information than decentral governments. This makes it difficult to jointly formulate a solution. NS proposes certain solutions, based on their information gathering. But these solutions are not recognized by decentral governments as they do not have insights in the used information. All decentral governments blame NS for the lack of transparent information and calculations.

“NS labeled several projects as unprofitable, but in my belief they just didn’t feel like doing these projects.”

4.3 Wicked problems

‘Increase accessibility’ The two different visions of the decentral governments and the united railway sector led to two different lists of projects that should be performed in the Amsterdam area. Decentral

4 Circle theory (in Dutch: kringentheorie) is used to calculate the potential number of passengers in an area based on certain characteristics of that area. 44 governments identified different bottlenecks in the area from the railway sector. So there was no consensus about the facts. The two parties also differed about the vision how the identified bottlenecks could be solved best, so there was also no consensus about norms. This can be illustrated by zooming in on the different goals actors pursue. In the interviews it was asked which goals actors pursue. Almost every actor replied that the most important goal is the accessibility of the area. However, when asking further, it became clear that it is difficult to explain what exactly is meant with this goal. Does it mean accessibility in their region of interest? Is meant accessibility only by train or also by car or other public transport? Is accessibility always more important than the costs of accessibility?

Zooming in causes confusion When looking from a distance, parties seem to share a common goal, but when zooming in the interests of involved parties differ. For example, when NS has the choice between supporting an investment to open a new station like Amsterdam Science Park and increasing the capacity of one of the main bottlenecks, the Schipholtunnel, they would definitely support the investment in the bottleneck because this increases the robustness of the railway system on a national level. This makes it possible for NS to run extra train services in a wide area. However, for the municipality of Amsterdam it is more attractive to arrange money for opening Amsterdam Science Park, because this will contribute to the economic development of a new area. This example shows that the goals and interests of involved parties differ a lot.

Lack of clarity about decision making at NS internal organization The set of goals becomes even more diffuse when looking at the several visions and documents the actors produced. This can be illustrated by summing up the goals NS describes. The leading NS vision at that time (NS, 2003) describes several goals:

- Customer oriented and robust railway system - Improvement of reliability and realization of transport growth for a price that is acceptable to society - Optimization of the entire railway system within environmental and safety constraints

An internal document about PBN adds three goals, which should serve as decision criteria for new investments: affordable, feasible and attractive to the customer.

In practice, the dilemmas focused on if and how certain specific infrastructure or capacity projects should be financed. Calculations were made, these were discussed and in the end a decision was made whether investments in a specific project were made or not. It seems like eventually the set of investments evolved in the construction of a package deal. There was no integral set of criteria used by the actors that led to a decision. From an internal

45 evaluation about PBN performed at NS it becomes clear that different decision criteria are used even within the company. In an internal memo from the NS regional director to the central board the following quotes were found:

“There is a negative operation result of 10 million Euros, but we expect that this might be compensated by increasing robustness by realizing new infra projects. You win something, you lose something!”

“If there is no negotiation result for the finance of projects, probably the relationship between the railway partners; the relationship between NS and decentral governments and the relationship between NS and the Ministry of Transport will be put under pressure, resulting in possible negative press for NS.”

These quotes contribute to the finding that was also supported in the interviews: the main incentive for the NS regional board to cooperate was the improvement of the relationship with their stakeholders and thereby the image of NS. This incentive does not match the goals the central board prescribed at that time. This internal lack of clarity can help explain why the negotiations took long and there was so much discussion about the mandate.

4.4 Process management

Given the multi-layered government, the presence of multiple principals and the wickedness of the problem, the analytical framework pointed out that process management is needed. If and how process management was applied by NS is described in this section. Firstly, the preparation that actors made is described (4.3.1). Secondly, the four characteristics belonging to a good process design is examined (4.3.2 t/m 4.3.5). To conclude, the timing of the process is described (4.3.6).

4.4.1 Process preparation

Before the first negotiation with the railway sector in March 2005 started, the decentral governments already defined a joint strategy to realize the projects described in their vision. This preparation was organized by one person, the so called Primus Inter Pares (PIP) Railway. The PIP Railway was hired by the decentral governments to prepare the negotiations. As a former employee of NS, the PIP Railway had deep insights in the way NS works. Together the decentral governments made a list of the seven new stations they wanted to open. It was their strategy to stay with these projects, whatever happened. The rationale behind this strategy was to prevent the playing off of decentral governments by the railway sector. The representatives sent by the governments were all experienced senior civil servants,

46 responsible on a managerial level for public transport in their area. The political representatives of the decentral governments were only involved with the signing of the agreement at the end of the negotiations.

“It is good to place politicians at a certain distance, because political interests can pollute the discussion.”

In contrast to the preparation of the decentral governments, the railway sector did not define a strategy for the negotiations beforehand. They already wrote a joint vision, but during the negotiations they did not cooperate closely with each other. For NS it was even difficult to follow a strategy that was shared by the whole company. The negotiations were done by two NS representatives from the regional board, with limited support of the central board. In the opinion of the central board of NS travelers it was better to directly do business with the Ministry of Transport, seated in The Hague.

“The attitude of NS towards decentral governments in the beginning was that if they would not fix the deal with us, they would just go to The Hague and arrange the money over there.”

Therefore, according to the NS negotiator and the other present parties, the mandate to negotiate for NS was absent.

“We completely missed backing from the central management of NS. Because of the lack of support, we could not act proactively.”

To conclude, it is remarkable that the ProRail negotiator was very negative about the role NS played during the negotiations. This confirms that the railway sector did not set a united course during negotiations.

4.4.2 Openness

Firstly, it is important to involve all relevant parties. At first glance all relevant parties are involved: all levels of decentral governments and the leading rail organizations sent a representative to the negotiations. However, the parties that were present were not always speaking on behalf of their organization. All interviewees, including NS representatives, pointed out that the NS negotiators did not have the full mandate of the central organization to negotiate.

Secondly, it is important not to make choices related to the content in advance. Process appointments should concern only the process, not the content. In the first meetings one of

47 the issues was indeed to make process agreements. NS proposed to write down some agreements in a protocol. The concept they wrote is illustrative for the way NS at that time thought about cooperation with decentral governments. NS assumed that the content of the outcome could be unilaterally defined by NS. In the protocol, that should serve as an agreement about the process, they already announce the majority of the content and the desired outcome. Unsurprisingly, the decentral governments were very upset about the tone of voice and the content of the proposed protocol. In their opinion the content should be part of the negotiations between the parties, and therefore cannot be determined without the input of the governments. They rejected the proposed protocol and described what the protocol should look like in their opinion. Eventually, both parties agreed on a completely revised protocol. The protocol contained amongst others appointments about the participants, the planning and the reporting. The difficulty of jointly defining a protocol illustrates the formal atmosphere that was present during the negotiations.

Thirdly, the transparency of the process should be safeguarded. Decentral governments did not evaluate the process as transparent. This is strongly linked to the earlier described supposed confidentiality of the information.

“When NS and ProRail would share more information, it would result in more understanding and respect at the decentral governments.”

In the protocol described above, a planning was presented that should contribute to the transparency. In practice however, this planning was not realistic. Instead of the planned 6 to 9 months, the negotiations lasted for 1 year and 4 months. Next, transparency of the process should lead to trust among the involved actors. The interviews do not give a clear picture about the trust between the partners. However, parties did agree that the trust grew during the process.

“Trust was built up during the negotiations.”

4.4.3 Core values

Firstly, de Bruijn et al. (2008) emphasizes that core values of an organization should be protected. In the interviews it became clear that organizations are not really aware of what their core values are. Interviewees from NS mentioned several values at different levels, for example: ‘keep the central rail network concession’, but also ‘safety for the traveler’. Decentral governments came up with core values as ‘accessibility of the region’. The lack of clarity about the concept of core values makes it difficult to determine if and how core values are protected in the process. As far as interviewees could oversee what the theoretical concept of core values means, they had the impression that these were

48 protected in the process. This can be explained by the presence of exit options, described below.

Secondly, commitment to the process is needed (instead of commitment to the result). In general, all negotiators that were present were committed to the process. However, after a round of negotiations the NS negotiator came back with new information from the NS central board, sometimes contrary with agreements made before. This made it difficult to assess the commitment of NS to the process. This is a direct consequence of the limited mandate the NS negotiator had.

“It takes ages before the central board of NS approves a decision of the regional NS director.”

Thirdly, exit rules need to be formulated. In the protocol there were no explicit exit rules mentioned. But in practice, it was always possible to leave the project without financial damage. However, the risk of reputational damage was high. Everybody knew that the party who stopped the negotiations, would be blamed for breaking the deal, resulting in no budget for any of the plans. This threat worked to continue negotiations. The negotiations were suspended several times. After such a break, NS took the initiative to start negotiating again. Without this intervention it would have been possible the negotiations stranded definitely and the process would have ended. The loss would then consist of the efforts of personnel, shared knowledge, but more important reputational damage towards all partners. So the threat of not getting the budget from the Ministry of Transport took care of enough commitment of the actors to the process.

“It would be very bad for the image of NS when they would fail a project in Amsterdam.”

4.4.4 Speed

Firstly, according to De Bruijn et al. (2008) it is important to organize somewhere in the process profit as reward, as an incentive to cooperate. This is done by the Ministry of Transport who made 171 million Euros available, but only if NS and decentral governments came to a joint investment plan. In addition to such a big profit at the end of a process, it is also good to organize quick wins. These were not available in this process. The negotiations concerned one final deal without any intermediate results.

Secondly, the composition of the delegation must be appropriate. The decentral governments thought about who should join the negotiations and sent experienced people.

“We put a lot of efforts in selecting the right people to participate in this project. People who are able to think strategically and had negotiation experience.”

NS sent the Regional director and the PA manager to the negotiations. Both work for the decentral board of NS. In the beginning an appointment was made that also somebody from the central board would join the process, but in practice he only showed up once. This resulted in a lack of back up and commitment of the NS central board towards the NS negotiators. The NS negotiators needed a lot of extra meetings with the central board.

“The process could have been twice as short when NS organized their mandate and acted more pro active.”

“Sometimes the world inside NS is more complex than the world outside.”

Thirdly, the conflict should be organized as deep as possible in the process. The starting point of this process consisted of two fixed, contradictory visions. This led to an undesired situation in which parties immediately stood opposite each other.

“Everybody had chosen position already. There was not enough room to negotiate.”

Initially, it was planned to negotiate six to nine months about the deal. However, in practice 17 sessions were needed, starting in March 2005 and ending in July 2006. So the negotiations took twice as long as planned initially. Main reasons for this delay were the disagreement about the projects, the belonging financial agreements and the limited mandate for the NS negotiators.

4.4.5 Content

To guarantee a sufficient level of content, the involvement of experts is needed. The risk of focusing on process management is that not enough attention is paid to the content. One of the impressions of the researcher is that in general people working in the railway sector have a lot of technical knowledge about their sector. Employees of NS at a high managerial level still know many details about the rail network and the types of trains. They are very committed to the product and know a lot about the content of their decisions. There are several reasons to explain how this type of organization arose: the history of a state owned enterprise, the high number of engineers that work for NS, the vicious circle that managers with a focus on the content appoint also managers with the same focus, etc. As this is not the topic of this research, this will not be investigated further. However, the consequences are very relevant from a process management perspective, because the risk of involving too

50 many people who focus on the content brings the risk that there is no attention paid to the process.

This seemed to be the case in this project, at least for the internal part. NS put a lot of efforts in making extensive calculations and communication between internal NS departments, but a strategy for the external process was underrated. Despite the presence of the regional director and the PA manager, who are more process oriented than the experts of the NS central division, the process was under designed. In contrast, decentral governments paid a lot of attention to pass through this process successfully. However, they also mobilized experts to support the discussions about the content.

Secondly, a process should start with variety of content, later on working towards a selection. As mentioned before, the starting point of this process consisted of two conflicting visions. This led to an undesired situation in which parties immediately stood against each other.

4.4.6 Timing of the process

Sense of urgency: present As described in the introduction there was a sense of urgency for both the railway sector and the decentral governments to improve the public transport in and around Amsterdam. Besides the sense of urgency related to the content, there was also a sense of urgency related to the process. The Ministry of Transport announced that 171 million euro was available for the improvements, but only if decentral governments and the railway sector would manage to formulate a joint programme of investments. Until that time the decentral governments and the railway sector developed their plans separately, but now there was a necessity to cooperate in order to fulfill the condition of the Ministry.

Policy window: present In terms of Kingdon (1984) it looked like the streams of problems, policies and politics came together, creating a policy window. Firstly, it was recognized that the accessibility in and around Amsterdam by road and public transport was insufficient. Secondly, a stream of policies was formulated by the cooperating decentral governments in this area. Thirdly, public opinion and the attitude of the Minister of Transport resulted in a political stream in favor of investing in public transport in and around Amsterdam. So at the start of the negotiations the possible cooperation looked like a window of opportunity. The process was very complicated, but in the end NS and decentral governments made use of this policy window. They formulated a joint investment program.

Political culture A last remark is made about the political culture in this part of the Netherlands. The western part of the Netherlands has a reputation of what is called ‘governmental multitude’ (in Dutch: bestuurlijke drukte). In a relatively small area there is a huge amount of involved politicians with different responsibilities and agendas. In addition to that, the debate culture is to say things straight forward, which can lead to heavy debates and polarized relations. The saying ‘fortune favors the bold’ is suitable to this area. This is illustrated by some of the quotes that were formulated during the interviews.

“The director of ProRail should be fired immediately because of non-performance, and the director of NS too.”

4.5 Rounding up

The findings of this chapter are summarized using the analytical framework in figure 5.

•Multi-layered, well organized decentral government •Four layers of governments serve as multiple principals for NS with conflicting demands •NS is internally divided: decentral employees pursue different goals than central board •Central board is not interested in involving decentral governments  no mandate for NS negotiators Multiple •High level of market dominance and information asymmetry of NS principals problem •NS was not able or not willing to give insight in their calculations and tradeoffs

•No consensus about the facts and norms that should be used when deciding on investments •Operationalization of ‘accessibility’ causes confusion •No integral set of decision criteria, set of investments evolved in a package deal Wicked •NS is internally divided on how to decide: improvement relationship with stakeholders (NS decentral) vs. ‘affordable, feasible and attractive’ (NS central) problems

•Decentral governments consciously prepared a strategy for the process, NS did not •Formal atmosphere, partly caused by starting point of two fixed, contradictory visions •Developing the protocol demonstrated NS assumed outcome could be unilaterally defined •Threat of not getting budget from the Ministry of Transport took care of commitment to the process Process •Slow process due to disagreement about projects, financing and limited mandate of NS negotiators management •Sense of urgency and policy window seem present

Figure 5: Summary case 1 in analytical framework

Chapter 5. Case 2: City region rail Arnhem-Nijmegen

This chapter describes the second case, again based on the analytical framework introduced in chapter 2. Firstly, the multiple principals problem and the belonging strategic behavior is discussed (5.1). Secondly, it is demonstrated this is a wicked problem as well (5.2). Paragraph 5.3 shows if and how NS uses process management. Again quotes from the interviews are used. Blue represent quotes from NS interviewees, orange quotes from external interviewees.

5.1 Introduction

The second area that is investigated is the city region Arnhem-Nijmegen. With 720.000 habitants on 1000 km2 it is one of the most urbanized regions in the Netherlands (De Stadsregio, 2010). The situation on the rail network is quite complex, because the central rail network is interweaved with three regional rail lines. NS operates the central rail network, but all regional rail lines are exploited by other companies. Figure 6 represents this situation.

Figure 6: overview of market parties and five projects City Region Rail in Arnhem-Nijmegen region (De Stadsregio, 2010)

The biggest project in this region where NS meets decentral governments is the City Region Rail (in Dutch: Stadsregiorail). The main goal of the project is to run 4 stopping trains every hour on all trajectories in the region. This project consists of five investments in infrastructure, three new stations and two capacity related investments:

1. Station Mook-Molenhoek 2. Station Westervoort 3. Station Nijmegen Goffert 4. Turning facility Wijchen 5. Turning facility Elst

These numbers correspond with the numbers in figure 6. The reconstruction and extension of the station in Arnhem is a necessity for the realization of City Region Rail, but is not financed and executed within the scope of this project. The two turning facilities are needed in order to make it possible to run 4 stopping trains every hour. The budget to realize City Region Rail is 55 million Euros. The majority of this budget is made available by the Ministry of Transport. In 2004 the ministry allocated 50 million Euros to the city region Arnhem- Nijmegen, because they appointed City Region Rail as a national pilot. The organizational context of this project is discussed in the next section.

5.2 Multiple principals problem

City region plays central role All layers of decentral government are involved in this project (figure 7). City Regional Rail crosses borders of several municipalities, which are all part of the city region Arnhem- Nijmegen. Municipalities are responsible for the development of the station areas and the activities regulated by public law, like providing zoning plans. For strategic decisions and negotiations about public transport, municipalities give full mandate to the city region Arnhem-Nijmegen.

Ministry: Transport & Finance

Province: Gelderland

City region: City Region Arnhem-Nijmegen

Municipality: Several

Figure 7: Involved decentral governments (blue)

The city region Arnhem-Nijmegen plays a central role in this case, as they are initiator and principal of City Region Rail. Usually, the ministry of Transport is the direct principal for infrastructure developments like these, but in this case it is delegated to the city region.

Central vs. decentral government NS faces many principals in this project. Firstly, again they have to comply with the general financial standards set by the Ministry of Finance and the appointments in the concession for the central rail net formulated by the Ministry of Transport. Secondly, decentral governments have demands that conflict with the demands of the central government. An example is the opening of station Nijmegen Goffert. When the opening of this station was discussed, NS calculated that the norm of 1000 new travelers would never be met. However, several municipalities and the city region were in favor of opening this station. They even wanted to pay an exploitation fee to realize this station. It took a lot of efforts to convince NS, but eventually a deal was made in which the station will be constructed and decentral governments will pay for the exploitation deficit.

Heavy competition As shown in paragraph 5.1 there are multiple rail operators active in this area. Before the regional rail lines were tendered out, they used to be exploited by NS. NS decided not to bid on all regional rail lines in the Netherlands, with the consequence that competitors entered the Dutch railways. The decentral governments have the impression that NS regrets the presence of competitors in their area.

“I think NS regrets they did not go for some regional concessions, now they see the level of success of these concessions exploited by their competitors.”

When Veolia and Syntus started to operate the regional rail lines, they behaved differently than NS used to do. They used a supply oriented approach, in which they experimented with train services. In contrast, NS uses a demand oriented approach, in which they calculate the expected number of travelers. This can be characterized as a risk averse approach. This is even confirmed by NS employees.

“NS calculations are very risk averse. The estimation for the revenues of opening of a Starbucks branch for example, was estimated 3 times too low.”

Mentality change at NS? The new approach of the new entrants does not automatically mean that the level of service for travelers increased. This is still subject to heavy debate. It is not the goal of this research to give a value judgment about the approach new entrants use, but it is important to give attention to the consequences of this development. According to the decentral governments, the presence of other train operators caused a mentality change at NS. NS suddenly shared more information than they used to, as their competitors published information that NS formerly classified as confidential. However, NS representatives confirm that in the new situation they are challenged to show the decentral governments what they have to offer. They do not see it as a negative development. The interaction with governments is indeed growing. This is confirmed by decentral governments.

“Six years ago decentral governments were no serious partners for NS. We only approved the printed timetables, which was a formality. Nowadays there is a lot of interaction, which is a very good development.”

Less dominant position NS In this new situation it seems that the breeding grounds for strategic behavior (fewness, position and information-asymmetry) have become less dominant than in the old situation. However, NS is still blamed by decentral governments to display monopolistic behavior. It is difficult to check whether this is actually true. Strategic behavior mostly occurs around the information position of NS, as they still have a dominant position. As described before, the dispute between the parties focused on using the right numbers. NS was not able or not willing to give insight in their calculations and their tradeoffs to support or reject certain projects.

The information gathering plays a crucial role. In order to determine what the impact of different investments is, NS uses advanced models. These calculations are subject to heavy debate. According to the decentral governments NS does not share enough information to get insight into why decisions are made.

“Decisions are taken based on principles of ‘old fashioned technical calculations’ which leads to inimitable, nebulous decisions.”

5.3 Wicked problems

In this case a clear disagreement about the problem definition is found. According to the decentral governments the major problem is the regional accessibility. The region still grows, while there are no investments made in road infrastructure. Therefore investments in public transport are needed, focused on filling up the gap of travel distances between 10 to 40 kilometers. Several representatives of decentral governments, independently from each other, mention this ‘gap’ as the biggest problem. NS on the other hand, wants to facilitate urban growth. In their vision it is not relevant to classify movements into categories of travel distance. This discussion illustrates there is no consensus about the problem definition. Therefore, this can be labeled an unstructured or wicked problem.

Governmental actors in this area find improving public transport very important. NS also wants to improve public transport in this area. NS considers Arnhem-Nijmegen as a growth region and is willing to facilitate this growth. When looking from a great distance parties seem to share a common goal, but when zooming in, the interests of involved parties differ. The focus of NS lies more on external accessibility, where decentral governments focus on internal accessibility.

This can be illustrated by the extensive discussion about the opening of station Nijmegen Goffert. According to the decentral governments it would attract a big group of new travelers. But according to the calculations of NS the norm of 1000 new entrants would not be met. Decentral governments offered to pay an exploitation subsidy, but NS was still not convinced as they calculated that this station will never be profitable.

5.4 Process

5.4.1 Process preparation

During the interviews it became clear that the city region indeed functions as a central actor. They want to be the spider in the web and they appear to have established that position. The municipalities accept that representatives of the city region coordinate the projects with NS. Municipalities and city region harmoniously communicate about City Regional Rail in flashy folders. This is remarkable, because the two biggest cities in the area, Arnhem and Nijmegen, are famous for their mutual rivalry. For that reason, the chairman of the city

57 region is not a mayor of one of the member cities (as is common in other city regions) but an independent chairman.

“The city region has an independent chairman, because the two biggest cities in the city region (Arnhem and Nijmegen) are like Ajax and Feijenoord.”

The city region is in a heavy competence struggle with the province of Gelderland. According to the province, the governmental body city region is superfluous and should be abolished. The chair of the city region pointed out that it is hard to end this struggle:

“We try to invest in an improvement of the relationship with the province of Gelderland. However, this is quite difficult when the deputy mentions in every interview that the city region should be abolished. We have a budget of 110 million Euros a year and only 50 employees. We are the most efficient governmental level of the Netherlands!”

The province of Gelderland on the other hand is disappointed they are not involved in the formulation of the city region rail project. According to them, they are only allowed to support the execution of the project.

NS meets the decentral governments at several formal and informal platforms. Especially the cooperation between NS and city region improved a lot in the past years. NS and the city region for example collectively promoted the new time table for the Arnhem-Nijmegen area. Key players are the NS regional director and the public affairs manager. Interviewees of the municipalities and the city regions stress their good personal relationships with NS.

“The current relation management of NS is very good. I even got a birthday present.”

In contrast, the relationship between NS and the province of Gelderland is not very good. Politicians and civil servants of the province are unsatisfied with the position of NS.

“It is hard to distinguish who you are negotiating with in the railway sector. There is an inner group consisting of representatives of NS, ProRail and civil servants from different governments. It is a connected group, like a close family, and they lack the ability to think with you.”

5.4.2 Openness

It is difficult to fill in the analytical framework constructed in chapter 2, as a clear structure in which these actors meet each other was absent. The relationship between NS and decentral governments takes place in different formal and informal meetings. One of the findings of the interviews is that a clear trajectory or process of negotiations cannot be distinguished. The global subject of the negotiations is the adaptations and investments that are needed to make the railway system suitable for running 4 stopping trains every hour in the region. This is discussed between NS and the city region. The cooperation is very informal, which makes it difficult to describe how the definition of the City Region Rail project took place. This is illustrated by a quote of one of the involved politicians.

“I do not like to work with protocols. I just invest in good relation management.”

The same politician said that when he started to work at his government, it was not clear what exactly was decided about City Region Rail. This results in a process that is not transparent. Decentral governments also complained about the lack of transparency on the decision making process within NS.

“NS is such a big company that it is an impossible hurdle to find your own way in the company. That is why the public affairs managers are so useful.”

During the interviews it became clear that not all parties are satisfied about their involvement with the project. Especially the province of Gelderland wants to be more involved.

5.4.3 Core values

Interviewees stated that they are not aware of their core own values. As far as interviewees could oversee what the theoretical concept of core values means, they had the impression that these were protected in the process.

Secondly, the actors that are involved in the process seem to be committed to it. Actors are in a constant dialogue with each other and display consistent behavior. However, the province was not involved in the process and therefore also not committed. In several publications they complain about the cooperation in general with NS. It is hard to prove if there is a causal relationship between the lack of involvement of the province and their critical attitude, but it might be possible.

5.4.4 Speed

The incentive to cooperate consisted of the available money for City Regional Rail. In addition to this general incentive, attention was paid to quick wins. The NS regional director organized a survey at different stations in the region to identify small bottlenecks, e.g. the absence of a ticket machine on a platform. Next, the small improvements were solved on a short term. This action created a lot of good will with the governmental parties. Quick wins contribute to the speed of a process, as parties stay motivated to cooperate.

However, decentral governments still complain about the speed at which decision are taken at NS. When at the decentral level a decision is taken, it takes too long before the central board of NS approves this decision.

5.4.5 Content

The content of the projects was discussed in an open, informal way. Parties did not stick to their own formulated visions, but were in constant dialogue to discuss possible projects to improve accessibility and impacts of these projects. Experts of NS performed extensive calculations e.g. to investigate the viability of station Nijmegen Goffert.

Civil servants and politicians of the city region closely cooperated with NS employees responsible for the content (like transport experts) and for the process (like the regional director). This interweaving of experts and stakeholders worked out very well.

5.4.6 Timing of the process

Sense of urgency: present As described in the introduction there was a sense of urgency at both the railway sector and the decentral governments to improve the public transport in the city region. Governmental actors in this area are very keen on improving public transport. Mobility is considered one of the biggest problems in the region. The feeling is that the central government is not willing to invest in improving accessibility. The resources for decentral governments to invest in road infrastructure are limited, but they do have a clear task to develop public transport. Therefore decentral governments feel a sense of urgency to optimize the public transport in their area. The growth potential of travelers in this region is acknowledged by all actors.

Policy window: present In terms of Kingdon (1984) it looked like the problem, policy and politic stream came together, creating a policy window. Firstly, it was recognized that the accessibility in this

60 region by car and public transport was insufficient. Secondly, a stream of policies was formulated by the municipalities and city region in this area. These policies were supported by the fact that there was long time no consensus about which investments should be done to improve the road transport. Thirdly, public opinion and the attitude of the Minister of Transport resulted in a political stream in favor of investing in public transport in this region. The policy window was successfully used and most of the projects are currently executed.

It is remarkable that in this area politicians at all levels are closely associated with the content of the public transport system. The responsibility for the content is not fully delegated to civil servants. This seems to be a unique characteristic of this area.

“Compared to other regions the province of Gelderland and the city region Arnhem Nijmegen are very ambitious concerning public transport. In other regions like Drenthe and Overijssel, politicians pay only limited attention to this subject.”

This is confirmed by one of the politicians:

“Me and my colleagues are heavily involved with public transport in our region. This is needed when one thinks of the situation of the road infrastructure in this area. In 20 years no investment has been made, so there is an urge for smart public transport solutions.”

This involvement contributed to the creation of a policy window, by influencing the ‘politics’ stream.

5.5 Rounding up

The findings of this chapter are summarized using the analytical framework in figure 8.

•Multi-layered, divided decentral government: city region in competence struggle with province •Four layers of governments serve as multiple principals for NS with conflicting demands •NS decentral has freedom to take decisions, relatively loose relationship with central division NS •Confrontation with risk averse strategy NS, as competitors enter market and display different behavior Multiple •NS is challenged to give more insight in their information due to presence competitors principals problem •NS is still blamed for the lack of transparent information and calculations

•Problem definition is permanently subject of debate, no consensus •NS focuses on external accessibility, where decentral governments focus on internal accessibility Wicked •No integral set of decision criteria, set of investments evolved in a package deal problems

•Rivalry among municipalities of Arnhem and Nijmegen brings city region into central position •NS chooses to focus on cooperation with city region: informal and good personal relationships •Relationship between NS and province is other way around: formal and hostile •Clear trajectory or process of negotiations cannot be distinguished Process •Quick wins contributed to good relationship and progress management •Sense of urgency and policy window seem present

Figure 8: Summary case 2 in analytical framework

Chapter 6. Conclusions

Chapter 4 and 5 described the results of the case studies separately. In this chapter the cases are compared and conclusions are drawn. Based on the findings of the case studies the research question can be answered.

How does NS cooperate with multi-layered governments to reach decisions on new urban regional transport plans and how can this cooperation be improved?

In general it can be concluded that NS did not react proactively on the increasing influence of decentral governments. There is no clear strategy about how to cooperate with multi- layered, decentral governments. There are no guidelines that describe how to choose position in the faced tradeoffs, resulting in random choices. The relationship between NS and decentral governments is strongly dependent on the strategy of individual NS employees. This worsens the multiple principals problem and restrains the cooperation to fully exploit the potential of involving all committed interests of decentral governments. As a consequence, it is not clear to external partners if and how NS will cooperate in urban, regional transport plans. Next to that, there is a lot of internal discussion about how to negotiate with decentral governments.

This generic conclusion is specified in paragraph 6.1 till 6.5. The second part of the research question concerns an improvement of the cooperation with decentral governments. This question will be addressed in the next chapter, in which five recommendations are presented, based on the five conclusions.

6.1 Two cases, different environment

The description of the two cases reveal some important differences, summarized in table 3. PBN City Region Rail Relationship between decentral United Divided governments Kind of cooperation Formal Informal Course of negotiations Clear defined process Process without clear beginning or ending Forming of solutions Confrontation of two already Sharing of ideas (dynamic) defined visions (static) Market situation One market party (NS) Three rail companies (NS, Veolia, Syntus) Dominant persons Civil servants are in charge Politicians are in charge Table 3: Summary of main differences between the cases

Firstly, the relationship between the different layers of decentral governments differs. At the PBN case, all three levels of decentral governments worked closely together and formed one alliance towards NS. If the governments had a dispute, it was not revealed to the outside world. They defined a strategy on beforehand, which was not adapted during the process of negotiations. In contrast, the decentral governments at the City Region Rail case were strongly divided. One front was formed by the municipalities and the city region. Another front was formed by the province of Gelderland. Those two blocks constantly struggle with their relationship and even question each other’s right to exist. NS works constructive together with the city region, which is fully supported by the municipalities.

Secondly, the cooperation in the PBN case was very formal. Before the negotiations started, negotiations took place about the protocol that should serve as a guideline during the process. The outcome of the process was written down in many different contracts, in total consisting of hundreds of pages of text. At the City Region Rail case it was hard to find any official documents. The cooperation just emerges, without a clear beginning or end. The atmosphere of the cooperation is very informal and appointments are not formalized.

Thirdly, in order to gain influence, different strategies were used. In the PBN case the strategy consisted of forming an alliance of three layers of decentral governments. This alliance confronted NS with a vision and belonging investments. In the City Region Rail case the strategy consisted of a city region starting a dialogue with NS to jointly determine an investment programme.

Fourthly, the definition of solutions to the experienced problems differs. In the City Region Rail case ideas are shared and constructively is searched for solutions. People did not take position on beforehand, so there is enough space to ‘wheel and deal’. In contrast, before the negotiations about PBN started two (partly) contrasting visions were published with an explicit definition of projects that should be performed. Because actors held on to their position it was difficult to reach consensus.

Fifthly, the market situation in both cases is very different. NS is the only operator in the PBN area, while in the City Region Rail case NS faces two competitors. As a consequence, in the latter case NS is challenged to share more information than they are used to do. In places where competitors are absent, as in the PBN case, there are less incentives for NS to share information with governments.

Lastly, the political culture in both areas is completely different. In the PBN case politicians are only involved at a distance, while in the City Region Rail case politicians are very actively involved with the content. This can be explained by the number of tasks that have to be performed by politicians. For example, in the PBN area the mayor of Amsterdam is the chair

64 of the city region. Besides this tasks the mayor has countless other tasks that makes it almost impossible to be involved with the content. In contrast, the chairman of City Regional Rail is independent and not concerned with other tasks, so he has more time to invest to content related issue. Especially because the majority of the budget of the city region is allocated to invest in public transport. Next to that, the topic of public transport is high on the political agenda in the province of Gelderland, which makes it attractive for politicians to spend time on.

Process management is often seen as an informal political game with an unclear course. This characterization seems to fit with the setting described in the City Region Rail case. However, it would be a misconception to conclude that in the second case process management is applied well. The case studies show that in both cases good process management is absent. In the first case this is more obvious than in the second case. The contrasting cases show a tailor made approach is needed. A strategy that is effective in one environment, can be counterproductive in another environment. For example, when civil servants are ‘in charge’ in a certain area, it is not wise for NS to focus the strategy on politicians only. While in areas where politicians are powerful it is useful to focus on them.

6.2 NS internal organization does not match the increasing influence of decentral governments

The classical way of cooperation in which the Ministry of Transport exclusively performs business with NS belongs to the past. NS is faced with the upcoming power of multi-layered, decentral governments. Unsurprisingly, this is evaluated as a positive development by the concerning decentral governments. The majority of NS representatives also think this is a positive development, as funding and knowledge of decentral governments are used to improve the railway system.

From the case studies the impression arises that NS was surprised by the increasing involvement of the decentral governments. In first instance, NS representatives at the central organization ignored the call for cooperation at a decentral level. Only after the regional NS representatives did several attempts to convince NS central organization to cooperate with decentral governments, negotiations with the decentral governments started.

However, the picture arises that NS has changed a lot over the past years. Around 2001 the company was internally focused and it seems like not one actor (including NS itself) was satisfied with the functioning of the company (Van der Zwan, 2010). Since then, NS paid more attention to their external environment. This change of behavior is confirmed by all interviewed parties. A shift is noticeable from reactive to proactive behavior when it comes to the attitude of NS towards decentral governments. This is amongst others realized by 65 investing in relation management, in which the regional directors and their public affairs managers again play an important role. The organizational choice of NS to work with four regional directors and public affairs managers worked out very well. These employees are externally highly appreciated for their efforts to cooperate constructively.

But the change of the internal organization need to be further developed. It can be concluded that the distance between the central and regional NS division is too big. Especially the PBN case demonstrated that the decision criteria of the regional division are completely different from the central division. The main incentive for the NS regional board to cooperate was the improvement of the relationship with their stakeholders, which does not match the goals the central board (‘affordable, feasible and attractive to the customer’) prescribed at that time. Because of these differences between the central and regional division, it takes very long before decisions made on a regional level are confirmed by the central board. Next to that, for external stakeholders it is often not clear if decisions or plans of the NS regional division will be confirmed by the NS central board. This leads to the conclusion that the current internal organization of NS does not fully match the increasing influence of decentral governments.

6.3 NS is facing the multiple principals problem

Due to the presence of multi-layered governments, NS is facing multiple principals with partly conflicting interests. The cases demonstrated that during negotiations practical solutions are found to deal with the conflicting interests of multiple principals. For NS it is impossible to comply to all principals. In practice, the first reaction of NS is to strictly follow the norms of the Ministry of Finance, which means no investments in projects that are not profitable enough. Then, most of the time decentral governments are disappointed and a discussion is started about the contribution and impact of a certain project. Eventually, a package deal is negotiated on which all actors agree.

In this research no examples are found in which NS plays principals off against each other, as described in the theory of Wood and Waterman (1994). In practice, it seems like the effect of the multiple principals is the other way around, in line with the findings of Steenhuisen (2009). According to NS, the contradicting demands of their principals makes it hard to satisfy decentral governments. NS often makes calculations on request of decentral governments, resulting in a negative exploitation and therefore a negative answer to cooperate. Then, NS is blamed for a lack of cooperative behavior or again, for acting like a monopolist. In the next chapter recommendations are made to better deal with the multiple principals problem.

Decentral governments are very critical about the role of the Ministry of Transport. In their opinion the Ministry omits their role as principal, resulting in a situation where the 66 safeguarding of public values is at risk. As a consequence, NS has too much freedom to operate without improving regional accessibility. This statement of decentral governments is not automatically true. It can be considered strategic behavior to criticize the central government, in order to acquire steering possibilities themselves. A discussion about the current institutional system can contribute to a better institutional arrangement, but up to certain borders. The risk of this discussion is that it results in a less constructive discussion about power, instead of a discussion about goals. As one interviewee stated:

“The focus should not be on who performs projects, but how projects are executed.”

Unmistakably, a system is needed where the local expertise of decentral governments is used. But it should be prevented that the scales are tipped towards a situation in which the national rail network is split up into small, incoherent transport systems. A balance between regional and national accessibility is needed. There is a lot of discussion about who should be responsible for this. The Ministry of Transport should safeguard the accessibility of a national coherent railway system. However, in practice the Ministry is not explicitly fulfilling this role, leading to complaints at decentral governments.

6.4 Process design and process management is needed to deal with wicked problems

Different problem definitions… The case studies revealed that the studied problems can be classified as wicked problems, which means there is no clear problem definition. Decentral governments use more or less the same problem definition. Independently they frame regional accessibility a problem concerning a gap in the transport system for travel distances between 10 to 40 kilometers. It is not clear what the exact origin of this classification is. The definition of 10 to 40 kilometers can be found in several policy documents (Ministry of VROM, 2010; CVOV, 2003). NS does not see the point of using this problem definition. According to NS, rail transport is only a serious alternative for a small percentage of the travelers targeted in this category. Therefore, travel distance is not the right variable to discuss the improvement of the rail network.

… can lead to semantic discussions The risk of framing problems differently, is that a semantic discussion arises, without contributing to solve the experienced problems. In particular the PBN case demonstrated that both decentral governments and NS use their own terminology and stick to their own problem definition and solution. Therefore, situations arise in which parties stand directly against each other. As a consequence, during negotiations only small steps can be made as every party has already chosen position. This situation is undesirable, because it can result in long, slow processes. 67

Cooperation emerges NS has not chosen one strategy to deal with the decentral governments. NS employees do not have clear guidelines about how to set up a cooperation with decentral governments. It seems that cooperation just emerges, and the success of a cooperation is dependent of the people who are involved with the project. Employees who negotiate on behalf of NS do not have a clear mandate nor a clear strategy. In both cases this resulted in a slow process. Due to the unstructured and reactive approach, NS seems to have little influence on the outcome of the process. In both cases eventually a deal was made, but it is unclear if the deal matches the goals of NS, as NS did not make their goals for the individual cases explicit. Summarized, bases on a study of two case studies it can be concluded that both negotiation processes were under designed. NS does not pay enough attention to four important characteristics of process management: speed, protecting core values, content and openness.

The cooperation with multi-layered governments can be improved when process management is consciously used. The next chapter will elaborate on how to organize an improvement of process management.

6.5 Limited information sharing results in a ‘black box’ of decision making

Despite the attempts of NS to deal with decentral governments, NS is still seen as a company acting like a monopolist. This is mainly caused by the lack of transparency when it comes to the explanation of decisions. One of the biggest frustrations of the decentral governments is the limited sharing of two sources of information: - Models that NS uses to calculate the value of certain investments . These models are not comprehended and thereby not recognized by decentral governments. - Statistical information about number of travelers, extra travelers when opening a station, etc. This information is seldom shared, which makes it difficult for decentral governments to understand decisions.

According to the decentral governments NS does not give enough insight into their motivation to take certain decisions. Involved governments experience these calculations as a ‘black box’ and therefore do not trust the outcome. Much effort is currently put in discussing the value and outcomes of the models. This leads to the judgment of NS not being a transparent company.

Chapter 7 Recommendations

Below five recommendations are formulated, based on the five conclusions of the previous chapter. As NS is the central actor with a dominant position in the Dutch railway market, the recommendations are written as an advice towards NS.

7.1 Deal with the diverse environment

In the first conclusion differences between the cases were described. The different settings of the cases showed that process management needs to be adapted to the local situation. In this paragraph recommendations to deal with these differences are presented.

United versus divided A useful tool to design a process that fits the specific situation, is an actor analysis. An example of an actor analysis is presented in appendix E. Such an analysis can reveal e.g. how the relationship between different layers of government looks like. When it becomes clear that decentral governments form an alliance (as in the PBN case) it is recommended to form a strong counter alliance. Therefore at least close cooperation with the other rail parties is needed. It is advised to improve the strategic cooperation with ProRail. NS and ProRail sometimes are played off against each other by decentral governments, while in many cases they have shared interests. Therefore, more insight in the shared interests of NS and ProRail is needed.

If decentral governments are divided, it is important to find out which layer of government can function as the best partner. This decision has to be based on variables described in the actor analysis, like the level of agreement concerning the problem, the resources and the blocking power of the actor. Furthermore it is recommended not to get involved with the conflict between governments, but to make use of the discord. How this can be done in practice is described in the example below.

How an actor analysis could have helped the PBN case If NS had performed an actor analysis at the PBN case, they would have been more prepared for a strong alliance of decentral governments. Then, they at least would have defined a joint strategy together with ProRail to form a counter alliance. An analysis of the alliance of the decentral governments could also have helped to play the decentral governments off against each other. Zooming in on the composition of decentral governments for example could have revealed that the province of Flevoland had complete different stakes than the city region and municipality of Amsterdam. Then, NS could propose for example to invest in new stations around Amsterdam, which mainly the municipality and the city region of Amsterdam would benefit from. The result of this strategy could be that more flexibility was created in the strong alliance of governments with fixed visions.

Formal versus informal Another factor that differs per area is the level of formality. When the atmosphere is very informal, it is not wise to approach actors with protocols and other formal steps. However, it is recommended to formalize at least the negotiated agreements to prevent struggles later on. In the City Region Rail case for example almost none of the agreements are written down. Currently this is not a problem, as actors cooperate well. However, if parties no longer agree on how to execute the project this can become a problem.

In contrast, when the atmosphere is very formal, it is still possible to make use of informal contacts. The PBN negotiations were very formal, but it is still possible to make use of informal contacts when not sitting at the negotiation table. The regional director for example could have called or visited the city region to discuss the upcoming rounds of negotiation.

Monopolist versus competitors In situations where NS faces competitors, it is recommended to pay attention not only to decentral governments, but also to the other market parties. These parties should be involved in the actor analysis as well. Future plans of these companies (e.g. merges, aggressive bids for regional rail concessions, etc.) are very relevant for the strategy of NS. In general, NS benefits from a good relationship with their competitors. After all, when NS and competitors start to quarrel, there is a big chance the concession for the central rail net will be subject to debate again. It is in the interest of NS to prevent discussions on tendering out the central rail net concession, as it their core value to keep this concession.

In areas where NS does not face competition, like in the PBN case, the risk for NS to be blamed for monopolistic behavior is high. Therefore it is recommended to involve decentral governments in the decision making. How this can be done is described in recommendation 7.3 and 7.4.

Dealing with dynamics Decision making in a political arena can always pass of capricious. Relationships that are built up and appointments that are made, can be brusquely disturbed by elections or unexpected political developments. It is therefore important to closely watch the developments in the political arena, so the vision of NS can be framed in a way that it appeals to the ruling political view. For example, when a right wing cabinet is installed it is important for NS to frame their public transport plans as the ultimate solution to the problem of traffic jams. When a left wing cabinet is installed, it would be more wise to focus on e.g. the environmental effects of better public transport.

It is very important to be adaptive to the local political situation. The three streams of Kingdon (1984), problems, policies and politics, can be helpful to determine whether efforts

70 are needed to influence the local agenda. Firstly, it has to be checked whether there is a problem NS is interested in to solve. Secondly, policies can be formulated in cooperation with the local experts. Thirdly, NS can put efforts into influencing the stream of politics. Influencing the stream of politics is crucial if a policy window needs to be created. When public transport is high on the political agenda, probably not many efforts are needed to jointly formulate a sense of urgency. But when accessibility is not high on the list, the first step is to put it on the agenda in the region concerned.

Specific attention needs to be paid to the core values of politicians. In this research politicians did not make their core values explicit. However, it can be assumed that the core value of politicians is their political responsibility, which means that at any time accountability towards democratic organs is needed (Edelenbos, 2000). When participating in a process politicians might be locked up between the participants of the process and the democratic area where they are accountable (De Bruijn et al., 2008). As a consequence, it is not in the interest of politicians to participate in a process.

Therefore, NS needs to make clear, internally and externally, what they have to offer for decentral governments. This does not mean that norms need to be determined and applied rigidly, without being adaptive to the local situation. But it also does not mean that in one region deals can be made that will never be possible in another region. It is important to find a balance between formulating standards and adapt these to the local situation.

7.2 Acknowledge multi-layered decentral governments as influential partners

7.2.1 Three reasons to involve decentral governments

It is recommended to NS to acknowledge the new position of multi-layered decentral governments, because these governments can support NS in three ways. Firstly, decentral governments have budget available to invest in regional public transport. This budget is limited and has to be divided over multiple transport modes. The risk of no or limited participation in cooperating with decentral governments is that they do not invest any of the budget in the railway system. When a cooperation is set up, NS can influence the governments to invest in the railway system. In both cases the involvement of decentral governments led to major investments in the area.

Secondly, decentral governments have extensive knowledge about the regional situation. This knowledge concerns not only public transport, but also future spatial developments. The PBN case showed that a linkage with spatial developments is very important to decide

71 on the right investments in the railway system. A combination of the knowledge of the governments with the knowledge of NS can lead to an enrichment of the solutions.

Thirdly, depending on the level of success of the cooperation, decentral governments can influence the central government and public opinion. The first is important when it comes to the image of NS as a constructive partner. When decentral governments are positive about the cooperation with NS, this will have a positive effect on the opportunities for NS to obtain the central rail network concession after 2015. When decentral governments are not satisfied about the cooperation, they will put the Ministry of Transport under pressure to organize a public tender with more market parties. The same accounts for the public opinion.

7.2.2 Improve internal organization

In order to build up relationships with decentral governments, NS needs to get rid of the image of acting like a monopolist. Because of the historical situation and the unique market position of NS, it is unlikely that the company will ever be considered a ‘regular’ company. However, the case studies reveal that it is possible to boost the current image. Therefore, a change of attitude at all levels of the company is needed. It has to be stressed that arguments as “we do not have to deal with decentral governments because we will arrange the money in The Hague” or “when the proposed connection was really a good idea we would already run it” can no longer be accepted when a serious attempt is made to get rid of the current image.

Case studies show that the NS regional directors are very aware of this change in behavior and they already started the change in the right way. It is recommended to continue the course of building up relationships with decentral governments. Due to the presence of the regional board, decentral governments know who they can address when they want to cooperate with NS. This can also reduce the problem of the ‘many headed monster’. Regional representatives can have a liaison function between the decentral governments and the central organization of NS. However, this strategy can only work when the relationship between the regional board and the central organization of NS is better organized than in the current situation. The case studies reveal that it takes a long time before decisions of the regional board are confirmed by the central board. Next to that, regional NS employees do not have a clear mandate. It is recommended to designate one person at both the regional and central NS organization for every project, who together will be responsible for the internal communication.

Fixing the mandate When this recommendation is applied to the PBN case it means that the position of the NS negotiator (in this case the regional director) needs to be much stronger. In the old situation, the regional director was allowed to attend the negotiations and to negotiate, but the result of the negotiations always needed to be confirmed by the central direction. Many times, the negotiation result was disapproved, resulting in frustration and improbability of NS in the eyes of the negotiation partners. In the new situation, harmonization with the central direction should take place before the negotiations start. The regional director then needs to get a clear mandate. This means that the regional director needs to know how his space to negotiate looks like. This can be done by defining a range for subjects that are likely to be discussed in the negotiations, for example:

- Opening new stations: extra station in Flevoland is a no go, Halfweg Zwanenburg is an option when at least 800 new travelers are expected, Amsterdam Science Park is a desire of NS as well; - Financial consequences: negative results for individual stations are allowed, but the total exploitation must be positive after 5 years; - Opening new stations only takes place when ProRail explicitly supports the value of the new station; - Etc.

After the negotiations take place, the result still need to be double checked with the central division. But the chance of a disapproval is heavily reduced due to the appointments about the mandate. In this way, the process will be faster and the position of the NS negotiator will be more credible. Another advantage is that the NS negotiator can clearly state towards his negotiation partners how far his mandate reaches, and when he is not able to further negotiate.

7.3 Share the multiple principals problem

Due to the presence of multi-layered governments, NS is facing multiple principals with partly conflicting interests. From the case studies it can be concluded that in such a situation NS has no clear strategy to deal with these tradeoffs. If and how demands of decentral governments are taken into account depends on the individual NS employees. There are no guidelines that describe how to choose position in the faced tradeoffs, resulting in random choices. Currently no clear strategy is used to deal with this tradeoff.

It is therefore recommended to share this multiple principals problem with the relevant governments. After all, now the problem is faced at NS while the demands are externally defined. NS can try to make governments sharer of this tradeoff by giving more insights in the difficulty of choices. Sharing the multiple principals problem does not mean to give away influence and say to the environment of an agent, but to gain control and insight over the most desirable trade-off to make a cooperative platform of multiple principals and their agent.

An example could be to yearly publish a list of measures that NS would like to take, but which are not possible due to the demands related to the Return on Investment. Decentral governments can support NS in formulating this list. In this way, the Ministry of Transport

73 and/or Finance gets insight in the consequences of the measures of the multi-layered governments. The tradeoff is then made explicit to the institutions that are responsible for creating this tradeoff, which makes it possible to discuss whether the conflicting demands lead to a desirable outcome. If not, the demands of the governments might be adjusted, leading to a more desirable outcome.

Multiple principals in practice: Flevolijn A perfect example of this strategy is the Flevolijn example that was presented in chapter 4. This example showed that according to the standards of the Ministry of Finance the investment in extra trains between Amsterdam and Lelystad was not profitable enough. However, these trains might contribute to the solving of traffic jams on this trajectory. Therefore, it was a desire of decentral governments to run these extra trains. Eventually NS made appointments about running extra trains when certain growth numbers are met.

In this case, NS tried to solve the problem of the multiple principals with conflicting demands by putting forward extra conditions towards decentral governments. By doing so, NS caused incomprehension at decentral governments. In their opinion it is obvious that extra trains should run on this track to solve the traffic jams. NS employees agree with decentral governments, but due to the standards of the Ministry of Finance they decide to add extra conditions.

It is recommended to move the responsibility for this choice to the central government in future. When NS together with decentral governments or even with the Ministry of Transport agree on the added value of running trains on the Flevolijn, they should collectively address the Ministry of Finance to demonstrate the standards sometimes are not effective. In this way NS could make use of the ‘too big to fail’ principle. This means, the importance of some companies is so big, the government will always support them in the end.

7.4 Be prepared: pay attention to the four core elements of process management

One of the conclusions is that process management can improve the relationship with decentral governments. The four core elements of process management can help to realize good process management.

7.4.1 Content Firstly, it is important to prevent a situation in which a vision is unilaterally determined and imposed to external parties. It is recommended to create more space to collectively determine a vision. This can be done by first defining a global vision internally, without fixing the details. Next, this global vision can be discussed with external parties. The vision can then be enriched by knowledge and opinions of external parties, resulting in an improved vision that automatically is supported by multiple actors. This requires a lot of flexibility of NS. A right balance has to be found between defining a good scope to end with a decision that supports the company and maintain enough flexibility for contribution of external

74 parties. Defining a global vision can contribute to a structured approach to deal with decentral governments.

Secondly, it is important to construct a good composition of the delegation that negotiates. NS has the tendency to focus on the content and to underestimate the characteristics of a good process. This means that in negotiations not only technical expertise, but also process expertise is needed. In practice, the regional directors and public affairs managers already fulfill a process role quite well. However, often they have to communicate messages that are only supported by content related arguments neglecting the consequences for the process. This recommendation is closely related to the recommendation to give more insight into how and why certain decisions are made.

Compose a delegation In the PBN case the NS negotiators consisted of the NS regional director and the PA manager. Both are employees of the decentral division who fulfill a role more focused on process than content. However, they strongly rely on knowledge and input of NS employees from the central division who have a lot of technical knowledge. To bring more balance into the delegation, it would be wise to combine process and content, for example by sending the regional director and a product manager of the central division. In this way regional knowledge about the relationship (process) is combined with technical knowledge form the central division (content).

7.4.2 Openness

If NS participates in a project with multi-layered decentral governments, it is important to make an analysis of the governments involved. The actor analysis mentioned in paragraph 7.1 is a useful tool to identify the position, needs and wishes of actors. This is needed to use their agenda strategically. When the agenda of the actors is known, it is possible to frame the problem in their terms. This can be used in communication with actors, but also with the execution of projects. It is also possible to formulate quick wins when the agenda of other actors is known.

Finding quick wins and package deals Often, decentral governments have many frustrations about relative small issues with a big impact on citizens. An example in the City Region Rail region is the signposting at stations towards busses or taxis or the presence of ticket vending machines at strategic locations. Aldermen can gain a lot of credits from citizens when they improve these kind of little frustrations. NS and ProRail on the other hand can easily improve such things, without big investments. The difficulty of fixing these quick wins is not the realization, but the identification of them. Therefore, it is important for NS to invest in getting insight in the agenda and wishes of stakeholders. This can be realized not only by making actor analyses, but also by intensive communication with the stakeholders. This seems to be a role for the PA managers, but requires a very proactive attitude. Another example based on the City Region Rail case could be to involve running extra trains for the ‘Nijmegen Vierdaagse’ event. From the perspective of NS this is only a minor extra service (compared to the national scale), but from the perspective of the decentral governments it is one of their major events. Involving these kind of issues can contribute to the creation of a package deal from which both parties benefit.

Insight in the agenda of decentral governments can lead to an overview of different issues that need a solution. This insight can help to create a multi issue agenda, which in general reduces conflicts (De Bruijn et al., 2008). When the right fly height is chosen, these multi issues can be discussed resulting in a package deal. It is recommend to pay attention to this topic, as during the case studies no clear examples of package deals were found while the benefits of this strategy can be high.

It can be concluded that it is important to anticipate on the position of decentral governments. Insight in their agenda can lead to more understanding of how to deal with the governments. In order to fulfill this recommendations a lot of knowledge about the functioning of decentral governments is needed. Therefore, an extra recommendation is to hire people that have a lot of experience with decentral governments. It might even be wise to attract people who worked at a city region and have experienced the cooperation with NS. They can give unique insights into the way decentral governments work and how improvements in the relationship can be made.

7.4.3 Speed

In order to speed up processes, several measures can be taken. Firstly, the quick wins that are described above can contribute to a faster process. Secondly, incentives to cooperate need to be constructed. Sometimes these incentives are already present, for example by a shared sense of urgency to solve a certain problem. In that situation it is important to clearly frame the sense of urgency, using terminology that appeals to the involved actors. An example relating to the type of cabinet that governs a country was already presented in paragraph 7.1.

When there is no incentive to cooperate, an incentive needs to be constructed. The most obvious incentive that can be used is to hold out a prospect of money. In both cases the Ministry of Transport used such an incentive effectively. As described in the theoretical chapter, the money should not be paid off too early.

To conclude, it is important that representatives of NS have a mandate when they participate in a process. This means that representatives have certain room to negotiate, without discussing with their supervisors. When the representatives need to discuss every step of the process with their superiors, there is a high risk of other actors becoming frustrated about the speed of the process. Therefore, it is recommended to compose a heavy delegation, consisting of representatives that have experience in processes and are able to stay in close contact with the central management of NS. This also contributes to the authority of the process.

7.4.4 Core values

It is important to define what core values of NS are, for example obtaining the concession for the central rail net after 2015. This research revealed that a lot of public values and core values are written down, but in practice it is not clear if and how these are safeguarded. When participating in a process, NS should on beforehand decide on what will be the bottom line and how tradeoffs between important values will be made.

Introducing exit options is important in order to involve politicians and other actors that need to be convinced to participate in a process. It can be useful to design an exit option that in practice will not be used. This exit option then serves as a ‘valve’. Because the presence of the exit option, parties commit themselves to the process. However, the costs to use the exit option are high, so in practice the risk of actors exiting the process is low.

Setting up a ‘valve’ The City Region Rail case revealed that the province of Gelderland did not participate in the decision making process. As a consequence they constantly struggle with NS about who should be responsible for running local trains. The province gets a lot of media attention, in which NS is displayed as an arrogant monopolist. It would be wise for NS to avoid this discussion and try to involve the Province in the City Region Rail case. As the province will probably hesitate to participate in such a process, exit options can contribute to a safe feeling to join the process. An example of an exit option that serves as a ‘valve’ could be the condition that if after 1 year of negotiations no satisfactory result is obtained, every party is free to leave the process. The construction of the described ‘valve’ might help to convince the province to join the process. Off course, involving the province brings the risk they sabotage the process. However, other levels of government (who are in favor of cooperation with NS), NS and ProRail can jointly put pressure on the province to cooperate.

7.5 Invest in more insight in decision making

When NS wants to continue the course of opening up to the external environment, more insight into the use of their models and information is needed. There are several ways to realize this. One possibility is to externally validate the outcomes of the calculations or even the complete models. Another possibility is by giving more insight in the effects of a decision for the national rail network. Often, NS complained about the lack of understanding of decentral governments when they do not want to acknowledge that the opening of a new station has negative consequences for the national accessibility. Apparently this effect was not convincingly demonstrated. Efforts should be made to make understandable what the effects on the national rail network are, for example by showing simulations. The effects on the national rail network should be made visible instead of just told.

Another possibility is to share more knowledge about travelers. A good example is the research NS performed to the type of travelers. They identified six types of travelers, which can be used for optimizing their products (NS, 2006). Decentral governments will be

77 interested in this kind of information. Such a campaign, in which information is shared in a positive setting, can contribute to a better cooperation.

Benefits of more insight When NS tells an alderman that opening a certain station will only attract 300 extra passengers a day and next to that results in extra travelling time for ten thousands of passengers, the alderman might be convinced. However, the next step is the alderman needs to convince his colleagues and the city council as well. In that case, it would be really helpful when a simple simulation could demonstrate what the effects of opening the station are. This might help for two reasons. First, it is not the argumentation of solely the alderman but it is supported by experts of NS as well. The involvement of NS as an authority might have a positive effect on the discussion. Second, a simulation is far more convincing than a simple story. This strategy will not work for every project. When the disadvantages are not as clear as in this example, extra insight will not help to convince criticasters.

7.6 Make use of the political vacuum

A last recommendation is not based on the case studies, but it is just an idea that came up my mind when recent political developments took place. Recently, the Vice Minister of Transport, Ms. Tineke Huizinga, took the initiative to write a vision about the development of regional public transport. In this process a wide variety of actors participated: decentral governments of all levels, advisory bodies like LOCOV and ROVER, several market parties (bus, metro & railway) including NS and ProRail. But on February 22nd 2010 the Dutch cabinet collapsed. Since then, the outgoing cabinet governs the country. Huizinga is still part of that government, but as an outgoing cabinet is considered not to decide on new or controversial policies, the process will be seriously delayed. If and how this vision will be continued depends on the formation and composition of a new cabinet. It will take months before new developments on this issue are expected, as until now the formation of a new cabinet takes very long.

This situation could serve as an opportunity for NS to take the lead in developing this vision. NS could take the lead in developing a vision collectively with the decentral governments on the development of regional public transport. When process management is applied well, this can contribute to a stronger position for NS. When NS takes the lead the chance of steering the process into the desired direction increases. If decentral governments are involved in a proper way, this can contribute to the image of NS as a constructive, open partner. This image might even contribute to the awarding of the concession of the central rail network after 2015.

7.7 Further research

This research reveals plenty of opportunities for further research.

Firstly, it is interesting to further research the complex relationship with ProRail. One of my first assignments at NS, not specifically linked to this research, was to visualize the relationships between NS and ProRail. The picture I made is displayed in appendix F and shows the complexity. NS and ProRail together represent the Dutch railway sector and partly have shared interests. However, the focus in their relationship now is on their opposing interests. I have the impression that an improvement of this cooperation could lead to an improvement of the position of the railway sector as a whole. However, this needs to be further examined.

Secondly, it is interesting to investigate how recommendation 7.4 can be realized practically. NS can share more information in several ways. When developing a strategy to share more information, the needs of external stakeholders have to be researched. Sharing information indeed only makes sense when it contributes to more understanding at partners of NS. Strategies varying from small measures as externally validating models to bigger measures as developing simulation models need to be investigated. One of the risks is that too much information is shared, so attention has to be paid to the core values of NS.

Chapter 8. Reflection

Rounding up the research, it is useful to reflect on the graduation project. Paragraph 8.1 explains why it is risky to define a roadmap, but eventually motivates why a roadmap is constructed. Paragraph 8.2 demonstrates what I would have done differently when I had to perform this research again. A reflection on theory is presented in paragraph 8.3. To conclude, a short personal reflection on the process of graduation is given (8.4).

8.1 Defining a roadmap

One could question if the conclusions and recommendations of this research can be summarized in a roadmap towards a successful cooperation of NS and decentral governments when deciding on urban, regional public transport plans. In my opinion that is not a realistic option for two reasons.

1. Unstable, diverse environments need a custom made approach. The differences found in the two cases demonstrate that every area has different characteristics and therefore need a different approach. 2. It is hard to build in dynamics into a roadmap or comparable structures. The risk of prescribing a number of steps to follow is that people rigidly stick to these steps, ignoring possible opportunities or threats showing up along the way. As the projects NS performs take place in a very politicized setting, incorporation of dynamics is key to successful cooperation.

Despite these pitfalls I decided to construct such a roadmap. By mentioning the risks of a roadmap, I want to demonstrate that it is not useful to use this roadmap as a blueprint. However, the roadmap presented in appendix G gives the possibility to think of the problem definition, the process design and possible developments during the cooperation. It might be a useful tool to make an analysis before a project is started. This analysis can contribute to a good preparation. It can stimulate thoughts about if and how to participate in a particular project. Currently, no clear assessment about how to cooperate is made. If and how is participated in a project, strongly depends on the persons who are assigned to the project. This roadmap can contribute to a more structural analysis and definition of a strategy in advance.

This roadmap needs to be further developed with internal and external stakeholders. The conclusions and recommendations of this research were already discussed with NS employees (listed in appendix D). By doing so, they validated this research. However, the proposed roadmap need to be validated as well. It is just a first version, based on my

80 research and my personal observations at NS. It needs to be tested in practice. I offered NS to organize a workshop to discuss this roadmap and adjust it to their needs. This workshop will take place in October. The goal of this workshop is to convince NS employees of the need for a better process preparation and to discuss possible strategies to do so. To describe the risks of using a roadmap, a disclaimer is presented in appendix H.

8.2 Learning from the research

If I had to perform this research again, I would have done some things differently. The most important are listed below.

- Spend more time to fully develop the theoretical framework before starting the interviews. Due to time pressure interviews took off a little bit too early. This resulted in insufficient insight in the cases. After the interviews, extra efforts were made to complete the insights. - Involve more people from ProRail. Only one representative of ProRail was interviewed (for the PBN case). This interview demonstrated that the perspective of an ‘outsider’ (not a representative from NS or a decentral government) adds value to the research. - Organize a discussion between representative of NS and decentral governments. I always interviewed people separately. A debate or discussion in which I would ask both parties to react on my findings might have helped to formulate new insights or practical recommendations.

In the theoretical situation I had unlimited time, it would be interesting to interview all stakeholders again to make a more thorough analysis about the current system of responsibilities in the Dutch railway sector. However, this definitely does not fit the scope of a Master thesis, as in total over 50 interviews are needed then.

8.3 Reflection on literature

In the beginning of the research I read a wide variety of literature that might relate to the problem of NS dealing with decentral governments. Some of the literature, I did not use in the final concept of my thesis. For example, literature on public values was studied, used in the interviews for the case studies but in the final version of this thesis theory about the multiple principals problem, wicked problems and process management turned out to be more relevant. The multiple principals problem as described by Steenhuisen (2010) was very useful for this research. His notions fitted this research very well, as one of his case studies was performed at NS as well. From the multiple principals problem it was a small step to the

81 theory of strategic behavior. The behavior Ten Heuvelhof c.s. (2009) describe was found in both case studies.

Another major part of this thesis is based on theory about process management. To my surprise there is relatively few thorough literature on process management. In contrast, literature on project management is ubiquitous. In my opinion literature on project management often neglects the complexity of performing projects in a political area. Rational, successive decision making is not likely to occur in politicized settings as NS is operating in. Process management does pay attention to the capricious and cyclic character of this area. Therefore, process management can contribute to analyze and to optimize the cooperation between NS and decentral governments. However, in literature not many practical examples are found that demonstrate how process management can be applied. I tried to make the recommendations of this research as concrete as possible. By doing so, this thesis might contribute to literature on process management in practice.

8.4 Personal reflection

I started my studies in 2003 and from the very beginning I was very enthusiastic about the typical ‘TB’ courses. During the years I slightly lost my interest in the ‘diehard’ technical courses. It was quite a struggle to fulfill all requirements for technical electives. Therefore I was very glad when I finished all courses successfully and could start my graduation project. I looked forward to design an interesting research that would fit my interests and could contribute to solve a ‘real life company problem’. After all, the past years I solved the problems of MagLev trains, designed a strategy to expand Rotterdam Airport, prevented failures of RandstadRail, designed a strategy for the port of Rotterdam for the next 50 years, etc. But those were only fictive projects and the outcomes were (almost) never used in practice.

Performing my graduation project was by far the biggest challenge of my entire studies. The best part of the research was performing the interviews. I interviewed 27 persons and without any exception those were captivating talks. Not every interview started very smooth. One example was an interview were I arrived 5 minutes late and the interviewee was clearly not really interested in the obvious questions of another student. However, after twenty minutes I managed to steer the interview into a direction where a dilemma was discussed, the interviewee faced in her daily working life. This resulted in an interesting analysis that gave me new insights in the way NS works. Eventually, the interviewee shared a confidential mail correspondence I had been looking for for weeks.

The most difficult part was to write this thesis. Many times I thought of the words Bauke wrote in his dissertation:

“Finishing a PhD is known to be a long and winding, if not depressing, road.”

I realized I was only performing my master thesis, however I did recognize these words. I missed the interaction with people when writing my thesis. The writing challenged my self- discipline to the utmost. Eventually, I managed to write this thesis due to the support of Bauke, Adriaan and my dear family and friends.

References

ANP (2010). Veolia en Connexxion fuseren. Press release 05-05-2010.

Baarda, D.B., Goede, M.P.M. de and Van der Meer-Middelburg, A.G.E. (1996). Open interviewen. Praktische handleiding voor het voorbereiden en afnemen van open interviews. Houten: Educatieve Partners Nederland

Bruijn, J.A. de, Heuvelhof, E.F.T. ten and In ’t Veld, R.J. (2008). Process Management. Why Project Management Fails in Complex Decision Making Processes, Dordrecht: Kluwer Academic Publishers.

Bruijn, J.A. de and Heuvelhof, E.F.T. ten (2007). Management in Netwerken: over veranderen in een multi- actorcontext. Den Haag: Lemma.

Centrum Vernieuwing Openbaar Vervoer (2003). Stedelijk openbaar vervoer in Europa. Rotterdam.

Charles, M.B., Dicke, W., Koppenjan, J. and Ryan, N. (2007). Public Values and Safeguarding Mechanisms in Infrastructure Policies. A conceptual and theoretical exploration. Introductory Paper for the IRSPM Panel on Public Values in Infrastructure, Potsdam April 2-4 2007, Delft/Brisbane.

Cohen, M.D., March, J.G. and Olsen, J.P. (1972). A garbage can model of organizational choice. In: Administrative Science Quarterly (17)1: 1-25.

Coyne, R. (2005). Wicked problems revisited. Design Studies, 26:1, 2005, p. 5-17.

Edelenbos, J. (2000). Proces in vorm: procesbegeleiding van interactieve besluitvorming over lokale ruimtelijke projecten. Utrecht: Lemma.

Herriot, R.E. & Firestone, W.A. (1983). Multisite qualitative policy research: optimizing description and generalizability. In: Educational Researcher, 12, p.14-19.

Heuvelhof, E.F.T. ten, Jong, J. de, Kars, M. and Stout, H. (2009). Strategic Behaviour in Network Industries: A Multidisciplinary Approach. UK: Edward Elgar Publishing.

Ikenberry, L. (2000). After Victory: Institutions, Strategic Restraint and the Rebuilding of Order after Major Wars. Princetown: Princetown University Press.

Kingdon, J.W. (1984). Agendas, alternatives and public policies. New York: HarperCollins.

Laffont, J.J. and Martimort, D. (2002). Coping with ambiguity: reconciling external legitimacy and organizational implementation in performance measurement. In: Public Money and Management, vol. 20, no. 3, p. 13-9.

Maarten, M. (2008). Provinces should steer stopping trains and intercities. In: Public Transport Magazine, nr. 1, 07-02-2008, p.8.

Megginson, W.L. and Netter, J.M. (2001). From State to Market: A Survey of Empirical Studies on Privatization. In: Journal of Economic Literature 39, pp. 321-389.

Merton, R.K., Fiske, M. and Kendall, R.L. (1990). The focused interview. New York: Free Press.

Miller, G.J. and Whitford, J. (2002). Trust and incentives in principal-agent negotiations: the “insurance- incentive trade-off”. In: Journal of Theoretical Politics, vol. 14, no. 2, p. 231-267.

Miller, G.J. (2005). The political evolution of principal-agent models. In: Annual Review of Political Science, vol. 8, June, p. 203-25.

Ministry of VROM (2010). Randstad 2040: Krachtige, duurzame steden en regionale bereikbaarheid. Den Haag.

Nederlandse Spoorwegen (2003). Benutten en Bouwen. Het plan van de spoor sector. Utrecht: NS.

Nederlandse Spoorwegen (2005). Internal memo about progression PBN negotiations.

Nederlandse Spoorwegen (2006). NS Visie op Stations. Van overstapmachine naar dynamisch stadsportaal. Utrecht: NS.

Nederlandse Spoorwegen (2009). NS Reizigers op weg naar 2010. Strategie en Businessplandoelstellingen 2010-2014. Utrecht: NS.

Nederlandse Spoorwegen (2010). Press release annual returns 2009. Utrecht: NS.

Pacanowsky, M. (1995). Team tools for wicked problems. Organizational Dynamics, 23, p. 36-49.

Schipke, A. (2001). Why Do Governments Divest? The Macroeconomics of Privatization. Berlin and New York: Springer.

Schramm, W. (1971). Notes on case studies of instructional media projects. Working paper for the Academy for Educational Development, Washington DC.

Scott, C. (2000). Accountability in the regulatory state. In: Journal of Law and Society, vol. 27, no. 1, p. 38-60.

Steenhuisen, B. (2009). Competing Public Values. Coping strategies in heavily regulated utility industries. Dissertation, NGInfra PhD Thesis Series on Infrastructures, Delft.

Verschuuren (1986). De probleemstelling voor een onderzoek. Utrecht: Het Spectrum.

Waterman, R.W. and Meier, K.J. (1998). Principal-agent models: an expansion? In: Journal of Public Adminstration Research and Theory, vol. 8, no. 2, p. 173-202.

Wood, B.D. and Waterman, R.W. (1994) Bureaucratic Dynamics. The role of bureaucracy in a democracy, Boulder, Colorado: Westview Press.

Yin, R.K. (2003). Case Study Research: design and Methods. Third Edition. Newbury Park, CA: Sage Publications.

Zohlnhöfer, R. and H. Obinger (2006). Selling Off the ‘Family Silver’: The Politics of Privatization. World Political Science Review, 2(1), article 2.

Zwan, A. van der (2010). Het experiment is mislukt. In: PM. Den Haag.

Websites

City region Arhnem-Nijmegen (viewed 03-07-2010) http://www.destadsregio.nl

Ministry of Transport (viewed 23-02-2010) http://www.verkeerenwaterstaat.nl/actueel/nieuws/nieuwsarchief/Steun_in_Tweede_Kamer_voor_Nota_Mo biliteit.aspx

Municipality of The Hague (viewed 15-02-2010). http://www.denhaag.nl/home/bewoners/verkeer-en-vervoer/to/Ontwikkelingen-openbaar-vervoer.htm

NS (viewed 15-02-2010). http://www.ns.nl/cs/Satellite/ns2007/nl/artikel/include/1220863135543/missie?p=1252507138430

ProRail (viewed 23-02-2010) http://www.prorail.nl/Over%20ProRail/watdoetprorail/Pages/Ambities.aspx

Province of Gelderland (viewed 15-02-2010). http://www.gelderland.nl/eCache/DEF/271.html

Appendix A. Organizational structure of NS

NS Poort NS Reizigers Strukton NedTrain

Personnel & Finance Operations Commercie Organisation

Business & Business Systems Sales & service Marketing Customer Service ProductDevelopment

Product development

Market research & advice

Business Development

Transport management

Control & Information

Appendix B. Long list projects NS - decentral governments

# Project description Infra- Transport Station Layers of structure service area government 1 ‘Key project’ Breda X X X 3 2 Area development Eindhoven X X 3 3 Area development Tilburg X X 1 4 Area development Maastricht X X 1 5 Area development Heerlen X 1 6 Railway zone Roosendaal X 1 7 Railway zone Oss X 1 8 Regions of connected knowledge X X 3 9 High speed shuttle Breda X X 3 10 ETMET X X 2 11 Night trains Brabant X 2 12 Intercity stop Best X 2 13 New station Berkel Enschot X X 1 14 Connection Eindhoven Airport X X 3 15 Railway zone Roermond X X 1 16 New station Bleizo X X X 1 17 New station Rotterdam Stadiumpark X X 1 18 New station Schiedam Ketel X X 2 19 New station Dordrecht South X X 2 20 Network study Stedenbaan X X X 4 21 Network study Randstad rail X X X 3 22 Railway zone Delft X X 3 23 Station development Rotterdam X 2 24 Area development Utrecht Central X 2 station 25 Rijn Gouwe Lijn X X 3 26 Area development Amsterdam Lelylaan X 1 27 New station Halfweg-Zwanenburg X X 2 28 New station Almere Poort X X 2 29 New station Lelystad South X X 2 30 New station Dronten X X 2 31 Programme Accessibility North wing X X X 4 32 Area development Amsterdam Central X 2 station 33 Area development Zwolle X X 34 Area development Groningen X X 35 Area development Nijmegen X X 36 New station Nijmegen Goffert X 37 New station Ede X 38 ‘Key project’ Arnhem X X X 3 39 Development agenda Groningen, X X X 3 Leeuwarden, Zwolle

40 Concession Zwolle Emmen X 2 41 Area development Arnhem Nijmegen X X X 4

1 -15: Region South 16-25: Region Randstad South 26-32: Region Randstad North 33-41: Region North East

Appendix C. Interview protocol

Orientation

- What was the motivation to participate in PBN? - Could you describe in which phases the development of PBN took place? - Which dilemmas did you face in this case?

Multiple principals problem

- How did your relationship with other governments look like? - Which goals did your organization pursue? - Who determined these goals? - Are your goals contradictory with the goals of other involved actors? - Did you experience conflicting goals within your organization? - Which goals did NS pursue in your opinion? What is your judgment about this? - How would you describe the market position of NS?

Wicked problems

- Was there consensus about facts and numbers? - Was there consensus about norms? - Was the problem definition subject to debate? If so, what was the consequence for the process of finding a solution? - Did you experience conflicting goals within your organization? If so, how was dealt with this tradeoff? - How would you describe the core values of your organization?

Process management

- How did you prepare for the negotiations? - Did you make an analysis about how other actors would behave during the process? - Would you describe the cooperation as formal or informal? - Were the right parties present during the negotiations? - Are your core values protected during the process? How? - Was there a possibility to exit the process?

- Was the scope of the project changed during the process? - What was the level of trust? - Did the professional background of the involved actors play a role? - What is your opinion about the supply of information? - Was there a fixed planning? Was this planning realized? - Are there moments to pinpoint where package deals were made? - Was the process interrupted? What happened then? - Was there a sense of urgency present to participate in PBN? - Did you experience a sense of urgency at other parties?

To conclude

- What were breakthroughs in the process? Who caused these? How? - How did you experience the role of NS (show 5 dimensions) - How would you describe the political culture in this area? - What are best practices and pitfalls based on this case for future projects?

Appendix D. Lists of interviewed people

Orientating interviews (NS internal)

Function Name Date Regional director Randstad North Joost Ravoo 02-03-2010 Regional director Randstad South Eloy van Raamsdonk 25-02-2010 Regional director South Heleen Herbert 15-02-2010 Regional director North East Arjan Bleeker 24-02-2010 NS public affairs manager Elly van Etten 28-01-2010 Director Business Development Joke van Veen Frequently Business Development Marjanne Poppen Frequently Business Development Adriaan Roeleveld Frequently Product Development Adriaan van Hoogstraten 26-01-2010 NS public affairs manager Wil Joosten 17-02-2010

Interviews case study 1: PBN

Organisation Name Date Province of Flevoland Frans Hasselaar 30-03-2010 Province of North-Holland Douwe Westervaarder 23-03-2010 City region Amsterdam Nico van Paridon 06-04-2010 Municipality of Amsterdam Rene Meijer 23-03-2010 Primus Inter Pares Railways Margreet de Zeeuw 31-03-2010 ProRail Alexandre Vanhoutte 26-03-2010 NS regional director Herman Gelissen 24-03-2010 NS public affairs manager Suzanne Steenbergen 22-03-2010 NS trainee Mariëlle Peeters 15-02-2010 NS product development Carmen Leutscher 12-08-2010

Interviews case study 2: City Region Rail

Organisation Name Date Province of Gelderland Rob Roskes 04-08-2010 City region Arnhem Nijmegen Jaap Modder (chairman) 19-04-2010 City region Arnhem Nijmegen Cor Hartogs 30-03-2010 (sr. civil servant) Municipality of Nijmegen Klaas Jan Gräfe (sr. civil servant) 08-04-2010 Municipality of Elst Jan Walraven (alderman) 31-03-2010 NS regional director Arjan Bleeker 24-02-2010 NS public affairs manager Bert Kluin 15-03-2010

Validation presentations (NS internal)

Function Present Date External supervisor NS Adriaan Roeleveld 22-04-2010 Business Development Marjanne Poppen Transport Management Carmen Leutscher Business Development Eimert van der Beek Director Business & Product Dev. Maurice Oenk 13-07-2010 Director Transport Management Luutzen Stellingwerf Director Business Development Joke van Veen Regional director South Heleen Herbert Regional director Randstad South Eloy van Raamsdonk Regional director North East Arjan Bleeker Business Development Marjanne Poppen

Appendix E. Example of actor analysis

Municipality of City Region Arnhem- Province of Ministry of Nijmegen Nijmegen Gelderland Transport Cooperates closely City region 19 member Ministry City region, with municipalities province Argues with Municipality of Province City region Arnhem Key players + role Jan van der Meer: Jaap Modder: sets the Marijke van Haaren: determines the agenda, strong direct contact with support of Nijmegen network in national Minister of and other small politics Transport and other villages for the City CDA hotshots Region Key issue Desire for station Decentralization of Accessibility of Nijmegen Goffert stopping trains Arhnem and Nijmegen Resources Legal procedures € 200 million for Influencing Ministry Budget accessibility improvements Blocking power Small Big, due to budget Medium, due to Big, due to relationship with budget Ministry Desires Extra bridge Extra bridge Nijmegen

This is an example of an actor analysis. Depending on the complexity of the situation and the need for details, this example can be extended. The table gives the impression of a static analysis, but in practice the content of the table can change rapidly. Therefore it is important to adapt this analysis to the actual developments.

Appendix F. Visualization of the relationship between NS and ProRail

Operations 1. Reizigersinformatie ONDERWERPEN 2. Bijsturing Nummers op deze slide corresponderen met 3. Logistiek organisatie onderdelen op volgende slide 4. Infrastructuur korte termijn 5. Actieplan winterhard spoor

Strategie & Productontwikkeling Regio’s (RN, RZ, NO, Z) 1. Producten lange termijn & infrastructuur Regionale ambities versus. uitbreidingen ambities NS & ProRail 2. Ontwerp dienstregeling & capaciteitsaanvraag

Contracten & concessies Stations • beheerplan 1. Ontwikkeling • vervoerplan • gebruiksvergoeding 2. Beheer • toegangsovereenkomst • netverklaring

Operations Ingrid Thijssen – Pieter Kraaijeveld ORGANISATIE Dir Operations – Dir Vervoer & Dienstregeling 1. Lisette van Herk – Jan Los AFKORTINGEN AH Reisinformatie L&S – Adj Dir Verkeersleiding 2. Bas ter Laak – Hans Burghard Dir = Directeur AH L&STransportbesturing – Dir Verkeersleiding AH = Afdelinghoofd 3. Wim Fabries – Hugo Thomassen Mgr = manager AH Leiding Logistiek – Mgr Capaciteitsverdeling L&S = Leiding & Staf 4. Esmé Kalshoven – Erna Klompers AH Infra – Mgr Vervoersanalyse & Capaciteitsontw. 5. Erik Sigger – Jan Los Operationeel directeur – Adj Dir Verkeersleiding

Regionale (PA) directeuren Strategie & Productontwikkeling Maurice Unck – Pieter Kraaijeveld 1. RN: Joost Ravoo – Hans Lodder Dir Bedrijfs- & Productontw. L&S – Dir Vervoer & Dienstregeling 2. RZ: Eloy van Raamsdonk – vacature 1. Marcel Ingenhoest – Erna Klompers 3. NO: Arjan Bleeker – Cees de Vries Mgr Productontw.– Mgr Vervoersanalyse & Capaciteitsontw. 4. Z: Heleen Herbert – Ine Frings 2. Luutzen Stellingwerff – Hugo Thomassen Mgr Vervoermanagement – Mgr Capaciteitsverdeling

Contracten & concessies Stations Joke van Veen – 1. Barend Kuenen – Kees Jan Dosker Mgr Business Development Dir Asset Development – Mgr Stations Anne-Marie van den Bos 2. Paul Fölsche – Peter Krumm Accountdirecteur Reizigers Dir OVM - Mgr Stations, Programma’s & Projecten

Appendix G. Roadmap for a successful cooperation between NS and decentral governments when deciding on urban, regional public transport plans5

NS internally Together with external partners

1 Select projects Pro-active (desire NS) Re-active (desire DG)

2 Determine priority High: Low: • NS stakes involved • No relevant NS stakes • High impact of DG (budget, involved position) • Low impact of DG

3 Assign employees • Mix of content and process • React on request (calculations) expertise • Give insight in decision making • Central - decentral NS team process (mandate!)

4 Explore problem • Check sense of urgency • List relevant actors • Identify main issues

5 Perform actor scan • Key issues • Resources • Blocking power • Desires (example is presented in appendix G)

6 Check policy window • Problem? • Policy? • Politics?

7 Decide on external support • Core values NS involved? • High budget? • Level of consensus low? • External process architect

8 Define global vision internally • Core values • Rough calculations • Global content of the project

9 Discuss vision with ProRail • Try to gain support • Together define strategy

10 Confront vision with other external stakeholders

11 Formulate dilemmas • Formulate biggest conflicts • Frame them as a dilemma

12 Develop protocol • Entry- and exit rules • Planning • Budget • …

13 Negotiate! • Look for quick wins • Try to make package deals

5 Based on insights of the interviews and the framework of De Bruijn et al. (2008, p.65) 96

Appendix H. Disclaimer for using this roadmap

During my green light meeting it was questioned whether it was a good idea to create a roadmap. The advantages and disadvantages are discussed already in paragraph 8.1. To prevent a wrong use of this roadmap this disclaimer is written.

- The number and order of steps is arbitrary and can differ from case to case. Small cases like placing a ticket vending machine do not need this roadmap. Big cases like discussing a network for high speed trains in the Netherlands need much more preparation than is displayed in this roadmap. - Steps 1 to 8 are recommended to perform internally. However, in the mean time it is always wise to have contact with external actors involved in the project as well. - The ‘real world’ cannot be put into this roadmap, so external developments will change the situation over and over again. Therefore always check if all steps are still relevant or need to be done again. A few examples: o When employees for a reason do not contribute to the project, the delegation must be reconsidered. The functioning of the team is crucial for a successful ending of the project, so step 3 is a recurring step. o The same accounts for the policy window. The problem, policies and politics can differ from week to week (as is seen in the current formation of the Dutch cabinet…) and needs to be monitored constantly. - Step 5, performing an actor scan, needs to be updated from the very beginning till the end of a project. The field of actors changes every day, so attention must be paid to the accuracy of the current analysis. However, it is not useful to extensively report these changes as the project might change into a theoretical, bureaucratic exercise then. The most important thing is that every employee working on the project is aware of the actual position of actors. - Step 13, negotiate, is a topic that needs a lot more explanation than is displayed in this roadmap. Negotiating is a skill and therefore it is recommended to use very experienced people to perform this step. - This roadmap needs to be further developed with internal and external stakeholders. It is just a first version, based on my research and my personal observations at NS. It needs to be tested in practice.