ORGANIZING FOR COLLECTIVE DEVELOPMENT IN PLURALISTIC SETTINGS: THEORY AND EVIDENCE FROM PLANNING THE UK’S RAILWAY

A thesis submitted to the University of Manchester for the degree of Doctor of Philosophy in the Faculty of Humanities

2017

Rehema Sophia Msulwa Alliance Manchester Business School

TABLE OF CONTENTS 1. INTRODUCTION ...... 11 1.0. SUMMARY ...... 20

2. THEORETICAL BACKGROUND ...... 22 2.0. INTRODUCTION ...... 22 2.1. OVERVIEW: THE MEGAPROJECT META-ORGANIZATION ...... 23 2.2. WHAT WE KNOW ABOUT DESIGNING ORGANIZATIONS ...... 25 2.2.1 THE DIVISION OF LABOUR AND INTERDEPENDENCE...... 25 2.2.2. THE INTEGRATION OF EFFORT ...... 28 2.3. DESIGNING SELF-GOVERNING ORGANIZATIONS FOR COLLECTIVE ACTION ...... 38 2.3.1. DESIGNING ROBUST SELF-GOVERNING ORGANIZATIONS ...... 40 2.4. SUMMARY ...... 45

3. RESEARCH SETTING AND METHODS AND DATA ...... 47 3.0. INTRODUCTION ...... 47 3.1. RESEARCH SETTING ...... 47 3.2. AN ABDUCTIVE CASE-BASED RESEARCH STRATEGY ...... 48 3.3. DATA COLLECTION ...... 51 3.4. DATA ANALYSIS ...... 55 3.5. SUMMARY ...... 57

4. EMPIRICAL SETTING ...... 58 4.0. INTRODUCTION ...... 58 4.1. CONTEXT OVERVIEW ...... 58 4.2. SUMMARY ...... 61

5. ANALYSIS ...... 63 5.0. INTRODUCTION ...... 63 5.1. RESOLVING COORDINATION PROBLEMS ...... 63 5.1.1. COORDINATION PROCESS OVERVIEW ...... 64 5.1.2. INTEGRATIVE DEVICES: COORDINATION ...... 72 5.2. RESOLVING COOPERATION PROBLEMS ...... 86 5.2.1. COOPERATION PROCESS OVERVIEW ...... 86 5.2.2. INTEGRATIVE COOPERATION MECHANISMS ...... 96 5.3. SUMMARY ...... 123

6. DISCUSSION ...... 125 6.0. INTRODUCTION ...... 125 6.1. THE SOCIAL AND TECHNICAL ASPECTS OF COLLECTIVE DEVELOPMENT ...... 125 6.2. ORGANIZING COLLECTIVE DEVELOPMENT IN PLURALISTIC SETTINGS ...... 128 6.3. MULTI-LEVEL GOVERNANCE STRUCTURES ...... 133 6.4. RESOLVING THE COORDINATION PROBLEM ...... 138 6.4.1. TECHNICAL WORKING GROUPS ...... 142 6.4.2. BOUNDARY OBJECTS ...... 145 6.5. RESOLVING THE COOPERATION PROBLEM ...... 149 6.5.1. ROBUST META-ORGANIZATION BOUNDARIES ...... 153 6.5.2. NEGOTIABLE RESOURCE BOUNDARIES ...... 158 6.5.3. APPEALING TO THE PROPORTIONALITY COSTS AND BENEFITS ...... 162 6.6. SUMMARY ...... 166

7. CONCLUSION ...... 168 REFERENCES ...... 190 APPENDICES ...... 206 WORD COUNT: 54230

2

LIST OF TABLES Table 2.1 Design Principles (Ostrom et al., 1994) ...... 47 Table 5.1 Overview of the Coordination Process Across Organizational Boundaries ...... 71 Table 5.2 Overview of the Cooperation Process Across Organizational Boundaries...... 94

LIST OF FIGURES Figure 5.1 The Composition of Inter-Organizational Technical Working Groups ...... 73 Figure 5.2 The Different Hierarchical Levels of Project Decision-making ...... 103 Figure 5.3 The Evolution of Local Station Decisions ...... 110 Figure 5.4 The Evolution of HS2 project cost over time (2009-2017) ...... 120 Figure 6.1 Framework for Organizing Collective Development in Pluralistic Settings ...137

3

APPENDICES

Appendix A Interview Details...... 206 Appendix A1 Interviewees, job roles and employers ...... 206 Appendix A2 Number of Interviews and Job Roles by Unit of Analysis ...... 207 Appendix B Exploratory Questions for HS2 Ltd. Informant Interviews ...... 208 Appendix C Teaching Case Studies ...... 209 Appendix C 1 Britain’s High Speed 2: All Aboard? (A) ...... 209 Appendix C 2 Britain’s High Speed 2: All Aboard? (B) ...... 247 Appendix D HS2 Ltd’s Wider Route Network Options ...... 258 Appendix E HS2 Route ...... 259 Appendix E 1 Proposed High Speed 2 Route Map, 2013 ...... 259 Appendix E 2 Route Map, 2017 (includes changes in Sheffield) ...... 260 Appendix F Longitudinal Chronology of the HS2 Planning Process ...... 261 Appendix G Birmingham ...... 265 Appendix G 1 West Midlands Sifting Process ...... 265 Appendix G2 Possible Stations Locations in Birmingham – Stage Two ...... 266 Appendix G3 Birmingham Curzon Street Connections from High Speed Two ...... 267 Appendix G4 Virtual Image Showing of Birmingham Curzon Street HS2 Station ...... 268 Appendix H Leeds...... 269 Appendix H 1 Leeds Sifting Process ...... 269 Appendix H 2 Proposed Leeds City Centre Station Final Options ...... 270 Appendix H 3 Proposed New Lane Station Layout...... 271 Appendix H 4 New Lane Indicative Station Illustration ...... 272 Appendix H 5 Leeds Station North Layout ...... 273 Appendix H 6 Leeds Station North Indicative Illustration ...... 274 Appendix H 7 Options for Leeds Station ...... 275 Appendix H 8 Integrated Leeds Station ...... 276 Appendix I London ...... 277 Appendix I 1 London Sifting Process ...... 277 Appendix I 2 Proposed Footprint for a Completely Rebuilt Euston Station Footprint ... 278 Appendix I 3 London Rail and Underground Connections for High Speed 2 ...... 279 Appendix I 4 Proposed Plan for a Scaled Down (Redeveloped) Euston Station ...... 280 Appendix J Sheffield...... 281 Appendix J 1 South Yorkshire Sifting Process ...... 281 Appendix J 2 Sheffield Victoria Station Connectivity ...... 282 Appendix J 3 Proposed Sheffield Victoria Station Layout ...... 283 Appendix J 4 Sheffield Victoria Station Indicative Illustration ...... 284 Appendix J 5 Sheffield Meadowhall Connectivity Map ...... 285 Appendix J 6 Proposed Meadowhall Station Layout ...... 286 Appendix J 7 Meadowhall Station Indicative Illustration ...... 287 Appendix K Manchester ...... 288 Appendix K 1 Manchester City Centre Station Long List to Final Options 288 Appendix K 2 Manchester Connectivity Map ...... 289 Appendix K 3 Manchester Piccadilly Station - Platform Level Plan ...... 290 Appendix K 4 Manchester Piccadilly Indicative Station Illustration ...... 291 Appendix K 5 Architects Rendition of HS2 Piccadilly Station ...... 292 Appendix K 6 Proposed Salford Central Middlewood Station Layout ...... 293 Appendix K 7 Salford Central Middlewood Indicative Station Illustration ...... 294 Appendix K 8 Proposed Salford Central combined station layout ...... 295 Appendix K 9 Salford Central combined indicative station illustration ...... 296

4

ABBREVIATIONS

AGMA Association of Greater Manchester Authorities

CPR Common Pool Resource

DfT Department for Transport

ES Environmental Statement

GLA Greater London Authority

GMCA Greater Manchester Combined Authority

GMCC Greater Manchester County Council

HSR High-Speed Rail Programme (Board)

HS2 High Speed 2

HS2 Ltd. High Speed 2 Limited

IAD Institutional Analysis and Development Framework

MP Member of Parliament

NAO National Audit Office

PAC Public Accounts Committee

TCM Tacit Coordination Mechanism

TFGM Transport for Greater Manchester

TfL Transport for London

UK

5

ABSTRACT

In capital-intensive organizations formed to plan new infrastructure development projects, the promoter of the project (as a single organization or as part of a coalition) rarely controls all of the critical resources required to achieve the system-level goal. Instead, the direct control of interdependent resources is diffused across multiple legally independent stakeholders (Lundrigan, Gil and Puranam, 2015). As such, the core structure in these so-called ‘megaproject’ meta- organizations is a classic empirical instantiation of a pluralistic setting (Denis, Langley and Rouleau, 2007). In pluralistic settings, the authority to make strategic decisions is diffused across actors with heterogeneous objectives, interests, values and expertise. Hence, to achieve the goal, the promoter needs to cooperate with multiple stakeholders.

Since some critical resources are not transactional or measurable, the cooperation problem is not a ‘buy’ problem. Instead, resolving the cooperation problem necessitates a search for mutually consensual solutions that reconcile conflicting interests. Moreover, this search unfolds without recourse to top-down authority characteristic of unitary organizations. Therefore, the promoter has to play a coordinating role that traverses organizational boundaries to coalesce competing preferences into a one-off plan.

Against this backdrop, this doctoral research investigates how designed rules and structures influence consensus-building during the collective development process. We conduct the research by drawing on two cognitive lenses consolidated in two vast bodies of literature that have remained largely disparate: organization design (Puranam, Alexy and Reitzig, 2014; Burton & Obel, 1984; Lawrence & Lorsch, 1967; March & Simon, 1958; Mintzberg, 1979) and collective action (Ostrom 1990, 2005). Combining these two research streams allows us to investigate how to resolve the coordination and cooperation problems inherent in pluralistic settings.

Our research method is a single case study with embedded units of analysis. This method allows us to probe deeply into operational details while maintaining the holistic features of the focal phenomena (Yin, 2009; Yin, 2013; Siggelkow, 2007; Eisenhardt & Graebner, 2007). Our focal case is the planning stage of High Speed 2 (HS2), a new multi-billion-pound cross-country railway project in the UK. The scheme is promoted by the UK Government. However, the planning effort has required that the Government share local decision rights for planning choices related to the stations along the route with multiple local authorities. These local authorities are independent, resource-rich stakeholders who are impacted by local choices, and they have deep knowledge of local needs and constraints. Thus, in the HS2 case, organizing for collective action is a prerequisite for achieving the system-goal.

Our research presents two major theoretical contributions. First, we contribute to organizational design literature by advancing our knowledge of how organizations can be designed to achieve system-level goals when decision-making authority is diffused across multiple organizational boundaries. Specifically, we advance our conceptual understanding of polycentric systems—a form of organizing that distributes decision-making authority across multiple local groups of independent stakeholders. As such, we illuminate the designed processes and structures that enable the core actors in a polycentric system to integrate effort and reconcile their differences over time. Organization design choices are about designing governance structures that enable and constrain collective action. Hence, we also contribute to the project management literature with insights on the governance of the planning stage of megaprojects. Specifically, we offer a deeper understanding of how to organize an inter-organizational setting to make planning decisions and manage interdependencies with the environment. Furthermore, we reveal that ambiguous evaluations of megaproject performance are rooted in collective efforts to resolve coordination and cooperation problems. Our research is grounded in the planning effort for the HS2 project and thus embedded in the UK context. We, therefore, encourage future studies to investigate the generalizability of our claims on organizing for collective action in other institutional contexts.

6

I declare that no portion of the work referred to in the thesis has been submitted in support of an application for another degree or qualification of this or any other university or other institute of learning.

7

COPYRIGHT STATEMENT i. The author of this thesis (including any appendices and/or schedules to this thesis) owns certain copyright or related rights in it (the “Copyright”) and s/he has given The University of Manchester certain rights to use such Copyright, including for administrative purposes. ii. Copies of this thesis, either in full or in extracts and whether in hard or electronic copy, may be made only in accordance with the Copyright, Designs and Patents Act 1988 (as amended) and regulations issued under it or, where appropriate, in accordance with licensing agreements which the University has from time to time. This page must form part of any such copies made. iii. The ownership of certain Copyright, patents, designs, trademarks and other intellectual property (the “Intellectual Property”) and any reproductions of copyright works in the thesis, for example graphs and tables (“Reproductions”), which may be described in this thesis, may not be owned by the author and may be owned by third parties. Such Intellectual Property and Reproductions cannot and must not be made available for use without the prior written permission of the owner(s) of the relevant Intellectual Property and/or Reproductions. iv. Further information on the conditions under which disclosure, publication and commercialization of this thesis, the Copyright and any Intellectual Property and/or Reproductions described in it may take place is available in the University IP Policy (see http://documents.manchester.ac.uk/DocuInfo.aspx?DocID=2442 0),in any relevant Thesis restriction declarations deposited in the University Library, The University Library’s regulations (see http://www.library.manchester.ac.uk/about/regulations/) and in The University’s policy on Presentation of Theses

8

For Those Who Dare Greatly

Our actions may be impeded But there can be no impeding our intentions or our dispositions. Because we can accommodate and adapt. The mind adapts and converts to its own purposes the obstacle to our acting. The impediment to action advances action. What stands in the way becomes the way.

― Marcus Aurelius

9

ACKNOWLEDGEMENTS

First and foremost, I would like to express my sincere gratitude to my supervisor Professor Nuno Gil. It has been an invaluable experience to be his PhD student. Nuno has taught me, both consciously and unconsciously, how good research is done. I appreciate all his contributions of time, ideas, funding and field work opportunities, which have made my PhD experience both intellectually stimulating and productive. His dedication to producing novel multi-disciplinary research has been an excellent example and one from which I draw immense motivation.

I am also grateful to my co-supervisor Professor Bruce Tether for his continued support and encouragement. I am especially grateful for his insightful comments and sometimes- challenging questions, which encouraged me to interrogate the research from various perspectives.

Besides my supervisory team, I would like to thank some of the academics who have (sometimes unwittingly) guided me during my PhD. Many thanks to Dr Luciana Nicollier, Dr Laszlo Czaban and Professor Graham Winch for their feedback during my first, second and third-year reviews, respectively. Thank you also to Professor Maria Nedeva for a stimulating course on Epistemology, which provided a wonderful introduction to the world of academic research. To the members of the Organization Design Community whose research I hold in the highest regard, thank you for producing work that inspires. Particular thanks to Professor Carliss Baldwin and Professor Phanish Puranam whose work has been a guiding light. A special thanks to my examination committee, including Professor Ray Levitt and Professor Graham Winch, for their insightful comments, questions and feedback during my final examination. It was an honour to have Professor Levitt as my external examiner given his stellar research record and expertise in organization theory, project organising, and in the projects domain more generally.

I gratefully acknowledge the funding sources and support in kind that have made my PhD work possible. I am sincerely grateful to the Karimjee Jivanjee Foundation for giving me a scholarship for the duration of my PhD. I appreciate this funding greatly and do not for a moment take it for granted. I am also grateful to the African Development Bank, and particularly the Operations Secretariat Committee, for the opportunity to contextualize my research in practice.

A special thanks to all who have offered words of encouragement, a smile or a laugh on cloudier days of the PhD. In Manchester, I am eternally grateful to Tinashe, Franziska and Mercy; and in Abidjan to the Barnabas family: Mama Hadija, Dada Aisha, Kaka Riyak, Areng and Adol. Thank you all for being a silver lining.

Last but certainly not least, I would like to thank my family: my parents, Dr Khambi Msulwa and Mrs Rosemary Msulwa, for their steadfast love and support. My sister, Subira Msulwa-Gwanoya, for always having the best advice and for her refreshing sense of humour. My brother, Baraka, for being a constant sounding board and my precious niece, Tinotenda, for her warmth and sense of wonder. Thank you also to my extended family, both directly related or otherwise, for being a motivating force. In the words of Maya Angelou, “I come as one, but I stand as 10, 000.”

10

1. INTRODUCTION

Developing large-scale infrastructure projects, or so-called megaprojects is a process of collective development conducted over relatively long periods of time (Miller &

Lessard, 2000). Megaprojects are typically partly or wholly publically financed because they are capital-intensive and strategically important both within and across national borders (Altshuler & Luberoff, 2004; Levitt & Scott, 2017). Furthermore, given their sizeable physical scale and broad scope, these undertakings require multiple resources, in addition to finance, in order to come to fruition. Other resources include land, planning consent, political support, and specialized skills and expertise. Importantly, a single organization rarely controls all of these resources. Instead, direct control is distributed across a diverse array of legally autonomous organizations spanning the public and private sector (Lundrigan, Gil & Puranam, 2015; Levitt & Scott, 2017; Kivleniece & Quelin,

2012).

Megaprojects are therefore carried out by a ‘meta-organization’ (Gulati, Puranam &

Tushman, 2012), a particular inter-organizational form made up of multiple heterogeneous actors unified by a superordinate goal: develop and deliver a single functional asset

(Lundrigan et al., 2015). Initially, this goal is advanced by the promoter, which comprises a small group of actors that form a coalition to promote the megaproject. This coalition exists at the organization’s core and includes, for example, a national government and a local government who together promote an urban transport system. Over time, the membership of this core grows as more resource-rich actors, in support of the goal, join the megaproject organization. These actors include public agencies, funding bodies, future operators, legislators, and private firms who collectively control the crucial resources required to forge ahead with the development.

Given their resource contributions towards the system-level goal, the actors in the core attempt to directly influence the design concept and project narrative. Taken together,

11 these set out the project performance targets, such as future costs and benefits, the scope regarding design architecture, and key functional components. The project concept and narrative also capture the scheme’s budget, timescale, and a delivery plan (Gil & Beckman,

2009; Gil & Tether, 2011; Miller and Hobbs, 2009). Thus, actors in the core influence these performance targets by proposing, and at times even asserting, their legitimate preferences and claims for the strategic design choices that contribute towards the final form of the infrastructure system. Their legitimate claims notwithstanding, actors within the core have to make trade-offs and compromises because of the various political, financial and technical constraints on the project. As such, the coordination of actors’ heterogeneous preferences and interest-based negotiations ensue simultaneously. To that end, the process of collectively developing large-scale infrastructural artefacts unfolds in a contested environment.

Additional tensions can exist between project supporters and opponents with conflicting views on a project’s material claims, value propositions or evaluative criteria, and a scheme’s symbolic significance (van der Westhuizen, 2017). To complicate matters further, core actors assess the design concept on an on-going basis and can defect from the megaproject organization at any time. Such an occurrence puts the sustainability of a megaproject organization at risk, especially if the resources that actors contribute are non- substitutable. At the limit, defections can result in a scheme being interrupted temporarily, or in the complete abandonment of the system-level goal.

Overall, the core structure of megaproject organizations is a classic empirical instantiation of a pluralistic setting. Pluralistic organizations are characterised by the diffusion of authority to make strategic choices across multiple autonomous and heterogeneous actors who hold diverse objectives, interests, values and expertise (Denis,

Langley & Rouleau, 2007; Denis, Langley & Rouleau, 2010). Other known pluralistic

12 organizations include large educational organizations such as universities, as well as regulated firms, and healthcare organizations.

In the pluralistic megaproject setting, strategic choices are the result of negotiations undertaken to shape the overall intent of a scheme (Bonke & Winch, 2002). Ultimately, the aim is to facilitate concerted action between organizations and to subvert the legitimate forces of competition within the core and in the broader project environment (Astley and

Fombrun, 1983; Astley, 1984). However, as in other pluralistic organizations, the process can result in repetitive cycles of decision with no action (Denis, Dompierre & Langley,

2011). Thus, to avoid this outcome, the project promoter has to play a coordinating role that traverses the boundaries of a single organization and coalesce conflicting design preferences into a one-off design definition.

Against this backdrop of megaprojects as pluralistic meta-organizations, this doctoral research examines how the actors in the meta-organization core manage to agree on the design definition of a future infrastructure system that they collectively develop. The research focus is on strategic design choices because these choices substantially define the scope of the project (Samset & Volden, 2016). Strategic design choices also influence the cost-benefit analysis that partially informs decisions to make the capital investment and the budget and timescale that inform decisions to allocate resources to the scheme.

The objective of this research is to further our understanding of how to govern the pluralistic core of megaproject organizations. Multiple empirical accounts suggest that consensus-building on megaproject organizations is challenging (Ross & Staw, 1993;

Shapiro & Lorenz, 2000; Morris, 1994; Miller & Lessard, 2001; Wachs, 1989). Sometimes these organizations get stuck in escalating indecision where strategic decisions are made and remade continuously, but nothing gets done (Denis et al., 2011). Under these circumstances, the project fails to move into implementation. Other times, pluralistic megaproject organizations succeed to move from planning into implementation, but they

13 encounter significant slippages in the original performance targets (time, cost, and scope) throughout the process of development (Flvbjerg, Bruzelius & Rothengatter, 2003). This state of affairs raises the core research question that motivates this study: How do we design organizations for collective development in pluralistic settings? Our core research question triggers some subsidiary questions: Which mechanisms can be used to enable collective action in pluralistic meta-organizations? To what extent is consensus a prerequisite for strategic decision-making? How does the design structure of the infrastructural artefact influence the collective’s ability to converge? How is the authority to make decisions distributed? How are inter-organizational disputes resolved? Why do some actors in the core struggle to converge their interests more than others? Is there a link between governance structures and the performance of the collectives regarding their capacity to resolve differences in cost and schedule targets?

We understand governance in terms of the structures that enable and constrain collective action even in situations of high interdependence and complexity (Gulati,

Lawrence & Puranam, 2005; Puranam & Vanneste, 2009; March & Simon, 1958; Simon,

1957; Lawrence & Lorsch, 1967; Thompson, 1967; Ostrom, 1990). In unitary organizations, such structures comprise authority hierarchies, employer-employee contracts and pecuniary incentives, specified by those with authority to do so (Puranam &

Vanneste, 2009; Clément & Puranam, 2017; March & Simon, 1993, Blau, 1964; Gulati &

Sytch, 2007; Raymond, 2008). However, similar mechanisms are not available in more pluralistic settings because authority is diffused. Thus, in these settings, cooperation cannot be ‘bought’ through market transactions and other legal contracts (Williamson,

1975). Instead, legally independent organizational actors tend to engage in processes of interest-based negotiation and mutual gains bargaining (Lawrence, Hardy & Phillips,

2002). We, therefore, seek to understand the rules and structures governing these

14 pluralistic organizations, and how they enable actors to reach mutually acceptable solutions. In addition, we seek to understand how these rules and structures emerge.

To address our research questions, we will combine two previously unrelated research streams. The first is organization design literature in management studies (Gulati,

Puranam & Tushman, 2012; Lee and Puranam, 2015; Burton & Obel, 1984; Lawrence &

Lorsch, 1967; March & Simon, 1958; Mintzberg, 1979). The focus of this vast body of literature is design mechanisms that encourage desired patterns of interaction between individuals given an organization’s goals. These include reporting lines, incentive systems and arrangements of physical space (Gulati, Lawrence & Puranam, 2005; Clément &

Puranam, 2017; Nadler & Tushman, 1997; March & Simon, 1958; Simon, 1957). Directly relevant to the focal problems in this research are the insights from this literature stream into how organizational structures are essential for consolidating divisions and integrating effort (Srikanth & Puranam, 2014; Knudsen & Srikanth, 2013; Puranam, Raveendran &

Knudsen, 2012; Tushman & Nadler, 1978).

Organization design scholars argue that the integration of effort within an organization requires the resolution of coordination and cooperation problems (Gulati,

Lawrence, & Puranam, 2005; Puranam, Alexy & Reitzig, 2014; Lawrence & Lorsch, 1967).

Coordination is accomplished through information sharing and information provision, both of which allow interdependent actors to act as though they can predict each other’s actions

(Puranam et al., 2012). When such predictability does not exist, interacting individuals are unable to anticipate the actions of others and adjust their own accordingly, which gives rise to coordination failures (Puranam et al., 2012; Schelling, 1960; Srikanth & Puranam, 2014).

Coordination failures, in turn, can take the form of ineffective communication, delays in knowledge transfer, misunderstandings, and the poor synchronization of activity, all of which undermine the collective effort (Srikanth & Puranam, 2014; Gulati et al., 2012).

Thus, organization design scholars stress the importance of developing ‘intelligent-filtering

15 systems’ during decision-making processes so that actors “will get only the information that is most important and relevant to the decisions they will make” (Simon, 1996: 145).

In the organization design literature, which is mostly empirically informed by the firm, cooperation is achieved through the use of employment contracts, incentives and inducements, which align the interests that multiple interdependent actors face (Puranam et al., 2012, Srikanth & Puranam, 2014). When conflicting interests prevail, however, cooperation failure takes the form of holdup, shirking and other manifestations of opportunistic behaviour, which undermine the collective effort (Srikanth & Puranam, 2014;

Alchian & Demsetz, 1972; Klein et al., 1978; Williamson, 1979). Under these circumstances, the organization design literature assumes the availability of hierarchical authority to resolve ensuing problems (Simon, 1962; March & Simon, 1993).

Recent contributions have started to explore how the literature on organization design can extend beyond the boundaries of the firm (Gulati et al., 2012; Gulati & Singh,

1998; Gulati, Lawrence, and Puranam, 2005; Reuer, 2004). Gulati et al. (2012), for example, propose the meta-organization construct to capture new forms of organizing that do not rely on authority hierarchies and employer-employee relationships to achieve system-level goals. However, this emerging literature is still incipient and tells us little about organizational design in inter-organizational pluralistic settings such as the megaproject meta-organization core.

In marked contrast to organization design literature, collective action literature comprises extensive theorizing about rules and structures that can be designed to govern collectives of independent actors. (Olson, 1965; Hardin, 1968; Ostrom 1990; Ostrom, 2005;

Dietz, Ostrom & Stern, 2003; Ostrom, 2009). Elinor Ostrom’s work, in particular, is empirically grounded in collectives of individuals and organizations formed to ‘self-govern’ both human-made and natural resources such as fisheries, forestries, and groundwater basins. The focus of this literature has been to propose principles for designing rules and

16 structures to govern collective action arenas and to ensure the sustainability of both the shared resources and the arenas within which they are embedded (Ostrom 1990; Ostrom,

2005; Dietz, Ostrom & Stern, 2003). Ostrom’s (1990, 2009) work is therefore directly relevant for addressing our core research question: How do we design organizations for collective development in pluralistic settings?

In Ostrom’s work, robust rules and structures are developed to foster cooperative behaviour and dissuade actors from acting opportunistically (Ostrom, 1990; 2009). Such arrangements create the right conditions for collective strategies to emerge, and for higher

(joint) returns compared to those gained from strictly narrow, short-term pursuits of self- interest (Ostrom, 1990; Walker & Ostrom, 2009, Hyden, 1990; Oakerson, 1993). In other words, if some claimants to a shared resource appropriate from it beyond a socially optimal point, they will deprive other claimants of similar benefits. Furthermore, in the long-term, the resource risks falling prey to a “tragedy of the commons” (Olson, 1965; Hardin, 1968).

To avert such a tragedy, Ostrom (1990) proposes the use of robust governance structures that can adapt to the many social, economic and ecological disturbances that occur over time. In turn, such rules and structures can significantly contribute to the sustainability of shared resources by minimizing free-riding and other manifestations of opportunistic behaviour among actors in a collective.

For at least two reasons, Ostrom’s (1990, 2009) work is unlikely to extend directly to the more pluralistic megaproject core. Firstly, in Ostrom’s robust governance model, the

Government does not directly interfere in local governance, and the claimants to the shared resources are well known. On megaprojects, however, the Government actively participates in decision-making, and the organizational boundaries between the project and the broader environment are blurred. Secondly, Ostrom’s work is rooted in settings where actors organize around pre-existing resources. For actors in the megaproject core, however, the infrastructural design, the resource that is the focus of collective action, is yet to be

17 created. Instead, this resource emerges over time. Hence, actors in the core are concerned with developing a shared resource rather than sustaining a pre-existing one. In this regard, the organization design literature complements Ostrom’s work seeing as the former is concerned with the governance structures in settings where human-made artefacts, such as new products and services, are created.

Organization design literature and collective action literature, therefore, suggest a complementarity that can be articulated to further our understanding of how to organize for collective development in pluralistic settings. Furthermore, the two streams of literature share assumptions that support this complementarity. Both organization design and collective action literature assume that actors are boundedly rational, fallible learners who possess imperfect information and are undertaking a costly information search (Simon,

1996; Ostrom, 1990). Crucially, our aim is not to contribute to a vast body of literature on political science and governance, but rather to extend organization design literature outside the boundaries of the firm.

In conducting the research, we adopted an abductive case-based research strategy.

Abduction involves a systematic combination of theoretical and empirical findings, and it is useful as it benefits from the dual impact of both inductive and deductive research approaches (Klag and Langley, 2013). A case-based research strategy is useful for supporting inquiries into areas in which little previous research exists (Eisenhardt, 1989).

While organization design research has started to extend into new forms of organizing; we still know very little about how to govern pluralistic settings from an organization design perspective (Gulati, Puranam, Tushman, 2012; Puranam, Alexy & Reitzig, 2014). Thus, an abductive case-based research strategy is in line with our motivation to contribute to the extension of this research stream.

We chose to conduct a single case study on High Speed 2 (HS2) a cross-country railway project unfolding in the UK because it allowed us to trace the planning process in

18 its natural context. In turn, this presented an opportunity to develop a theory by

“describing and explaining the temporal sequence of events” (Van de Ven & Huber, 1990:

213). A revelatory case study on the planning process of such a project is unusual because similar projects are infrequently undertaken, and they typically give only limited research access (Yin 1994).

Consistent with an embedded single case study approach, we embedded multiple units of analysis into the case (Yin, 2003). These units of analysis comprise local groups formed to agree on station locations in five cities along the route. Specifically, we studied the rules and structures formed to govern the decision-making processes in groups in

Manchester, Leeds, Sheffield, London, and Birmingham. Embedded case studies are advantageous because they allow for a more focused study enquiry, add a significant opportunity for extensive analysis, and they can enhance the insights from a single case

(Yin, 1994).

In this research, we aimed to uncover relationships between emerging governance structures and the outcomes of consensus-building efforts that we could logically explain.

To that end, two cognitive lenses guided our data collection process: the coordination- cooperation dichotomy from the organization design literature and Ostrom’s (1990) design principles for robust and sustainable common pool resource management. We also drew on Ostrom, Gardner and Walker’s (1994) Institutional and Development (IAD) framework, which outlines a systematic approach for analyzing institutions that govern the actions and outcomes within collective action arrangements (Ostrom, 2007; Ostrom et al., 1994). As

Hess and Ostrom (2005: 3) argue, the framework contains universal sets of intellectual building blocks that can be used to ask a fundamental question: “How do fallible humans come together, create communities and organizations, and make decisions and rules in order to sustain a resource or achieve a desired outcome?”

19

Our empirical findings have led to two main insights. Firstly, we show how organizations can be designed across multiple firm boundaries to integrate effort and achieve system-level goals when decision-making authority is diffused. To that end, we outline coordination and cooperation mechanisms for organizing collective development in pluralistic settings. Moreover, we advance our conceptual understanding of polycentric structures, a form of organizing that distributes decision-making authority across multiple local groups of independent stakeholders (Ostrom, Tiebout and Warren, 1961; Ostrom,

1990). Through this exercise, we are able to illuminate the polycentric rules and structures that enable core actors to reconcile their differences over time. We also offer a deeper understanding of designing an (infrastructural) artefact in an inter-organizational context, and the process of making strategic design choices while managing environmental interdependence. Furthermore, we reveal how efforts to achieve common ends by resolving coordination and cooperation problems contribute to ambiguous evaluations of megaproject performance.

1.0. Summary

This thesis investigates how to organize collective development in pluralistic settings, which are characterized by the diffusion of decision-making authority across multiple autonomous and heterogeneous actors who hold diverse interests, objectives, values and expertise (Denis et al., 2007; Denis et al., 2010). Theoretically, we position our research at the intersection of organization design and collective action literature. Through this lens, we can examine how, in the absence of unitary top-down authority, legally autonomous but interdependent actors achieve collective ends. Methodologically, we adopt an abductive case-based research strategy, which allows for the fruitful “cross-fertilization” of theoretical insights and empirical findings (Dubois and Gadde, 1999). Developing a case study also allows us to provide a rich empirical instance of designing meta-organizations under pluralism. This phenomenon is an underexplored topic in organization design and

20 one where little previous research exists. Thus, a case-based research strategy is well- suited for the study (Yin, 1994; Eisenhardt, 1989). Empirically, our insights are grounded in the development of High Speed 2 (HS2), the new high-speed railway in the UK. By juxtaposing theory and HS2 related data, we uncover organizational mechanisms for resolving the coordination and cooperation problems inherent in pluralistic settings.

Moreover, through this process, we contribute to the literature on organization design, project management and stakeholder management.

We organize the remainder of this study as follows. In Chapter 2 that follows we combine the organization design and collective action literature streams to illuminate what we know and do not know about the governance of pluralistic meta-organizations. In

Chapter 3, we discuss our research methods and the research site. In Chapter 4, we provide some context for the research with a description of the empirical setting. In Chapter 5, we present our analysis and examine the rules and structures governing the different decision- making groups within the broader meta-organization formed to develop HS2. We pay particular attention to the relationship between these structures and the outcomes of consensus-building efforts in the local groups formed in five cities. Subsequently, in

Chapter 6, we discuss our findings as they relate to existing theory and present a conceptual framework that outlines the process of organizing collective development in pluralistic settings. In Chapter 7, we then conclude with a discussion about the implications of our research for theory and practice, following which, we outline the limitations of our study.

21

2. THEORETICAL BACKGROUND

2.0. Introduction

In this chapter, we review the organization design and collective action literature.

These two research streams are previously unrelated, and their integration allows us to deepen our understanding of organizing collective development in pluralistic settings. To begin, we frame the megaproject organization as a so-called meta-organization (Lundrigan et al., 2015; Gulati, Puranam & Tushman, 2012). Actors in the meta-organization core are legally autonomous and operate in the absence of an overarching hierarchical authority or contracts governing their interactions. As a collective, the core actors determine the project strategy, which outlines the scheme’s budget, timescale, and a delivery plan (Lundrigan,

Gil, and Puranam, 2015; Gil & Beckman, 2009; Gil & Tether, 2011; Miller and Hobbs,

2009). Thus, they require alternative governance mechanisms if they are to achieve their system-level goal. To that end, we ask: How do we design organizations for collective development in pluralistic settings? Why do some collectives appear to converge more rapidly than others? How is the authority to make decisions distributed within a pluralistic system? Moreover, how are inter-organizational conflicts resolved? To address these questions, we proceed by outlining what we know about organizing from the mainstream organization design perspective. Our objective here is to illuminate what we know and do not know about the governance of pluralistic meta-organizations. We then consider robust self-governing organizations in the context of managing common pool resources. Here our objective is to determine whether the principles used to manage common pool resources can be applied to meta-organizations in pluralistic settings. Overall, the aim of this review is to identify organizational mechanisms that can facilitate collective development in pluralistic settings.

22

2.1. Overview: The Megaproject Meta-organization

The planning stage of large infrastructural artefacts as an empirical phenomenon has received attention across vast, disparate bodies of literature including planning studies

(Flyvbjerg, Bruzelius & Rothengatter, 2003), project management studies (Morris, 1994), and organization and management studies (Pitsis et al., 2003). In the latter body of literature, scholars have adopted different cognitive lenses to make sense of planning activities. These include institutional theory (Mahalingam & Levitt, 2007), sociological theories on the escalation of commitment (Ross & Staw, 1993), and networks of indecision

(Denis et al., 2011). Except for Lundrigan, Gil, and Puranam (2015), organization design, which is our cognitive lens, has yet to extend into these pluralistic settings.

Lundrigan et al. (2015) conceptualize the megaproject organization as a ‘meta- organization,’ i.e., an organization comprising actors, including organizations, who are themselves legally autonomous and not linked through employment relationships (Gulati et al., 2012). In this conceptualization, the megaproject meta-organization consists of a core and a periphery, each comprising a distinct set of actors. The core includes actors with varying professional roles such as public agencies, funding bodies, future operators, and legislators. These actors possess critical resources for developing the infrastructural artefact such as land, planning consent, political support, and fine-grained localized information. As such, core actors claim the rights to influence strategic decisions directly in exchange for contributing their resources.

Notwithstanding the influence core actors have on strategic design choices, they lack the expertise and knowledge to carry out the design and construction works. Strategic design choices are therefore operationalized by the meta-organization periphery

(Lundrigan et al., 2015). This periphery includes a wide array of specialized firms who provide planning, management, and design capabilities to the core, and who are governed

23 by contractual relationships. Typically, the core actors appoint an agent who contracts out the necessary services to a vast supply chain (Lundrigan et al., 2015).

Inter-organizational disputes are endemic among core actors because while they share authority and influence over strategic choices, they also have multiple diverse objectives (Denis et al., 2011). Disputes also arise because core actors have rivalrous preferences for one-off design choices (Lundrigan et al. 2015). In addition, it is challenging (though not impossible) to exclude core actors from the strategic decision- making process because they are resource rich. Juxtaposing the low-excludability of actors in the core with their rivalrous preferences, Gil and Baldwin (2013) framed strategic design choices in one-off infrastructure developments as ‘common-pool resources’. Thus, in order to achieve the system-level goal, core actors need to manage these common resources well (Ostrom, 1990). Furthermore, core actors need to overcome the agency problem of free riding and subvert issues of misappropriation and hold-up (Alchian and

Demsetz, 1972; Olson, 1965; Fjeldstad et al., 2012; Gulati et al., 2012).

If the meta-organization core is pluralistic, then no single entity has the authority to design organizational rules and structures and thus determine the terms of interaction.

Instead, given the higher level of parity that actors in the core share, patterns of interaction emerge through interest-based negotiations and mutual gains bargaining (Lawrence, Hardy

& Phillips, 2002; Aggarwal, 1998, Gil et al., 2017). Core actors, therefore, need to solve the problems of organizing without explicitly relying on formal authority as outlined in employment contracts (Gulati, Puranam and Tushman, 2012).

Consequently, an opportunity arises to merge the organization design literature with research on collective action because Ostrom (1990, 2005) theorized extensively about organizing in the absence of an authority hierarchy and employer-employee relationships.

Even so, Ostrom’s (1990, 2009) ideas are grounded in communities comprising relatively homogenous actors. In contrast, the pluralistic megaproject core comprises heterogeneous

24 actors with diverse objectives, interests, values and expertise (Denis et al., 2011).

Therefore, while Ostrom’s work is useful for giving us pointers on how to organize in the absence of authority hierarchies and contracts, it does not automatically extend to more complicated pluralistic settings such as the organizational core of a megaproject. We will expand on these ideas further, but first, we discuss organizing from the mainstream organization design perspective.

2.2. What We Know About Designing Organizations

Organization design can be viewed as the pursuit of integration for a given allocation of tasks to agents (Puranam et al., 2012; Camerer & Knez, 1996, 1997; Heath &

Staudenmayer, 2000). This conceptualization captures the essence of the two fundamental and interlinked problems, which lie at the heart of designing organizations: the division of labour and the integration of effort (Burton & Obel, 1984; Lawrence & Lorsch, 1967;

March & Simon, 1958; Mintzberg, 1979). Decomposing these problems further, Puranam et al. (2014) argue that the division of labour comprises task division and task allocation, which refer to the breakdown of an organization’s goals into contributory tasks, and the allocation of those tasks to individual members within the organization. Complementing this division, the integration of effort comprises information provision and reward

(motivation) provision, which are necessary for resolving coordination and cooperation problems, respectively (Puranam et al., 2014; Gulati et al., 2005; Lawrence & Lorsch,

1967). Within firms, such integration is aimed at improving the “quality of the state of collaboration that exists among departments that are required to achieve unity of effort by the demands of the environment" (Lawrence and Lorsch, 1967: 11).

2.2.1 The Division of Labour and Interdependence In organization design, the division of labour requires that decisions to be made within a system are considered and then divided into relatively independent subsystems.

This division is important because organizational actors (and computers) are boundedly 25 rational and therefore limited in their capacity to process alternatives or intricate chains of consequences (Simon, 1947). Naturally, dividing tasks into parts creates interdependence in the system (Simon, 1945; March and Simon, 1958; Lawrence and Lorsch, 1967;

Thompson, 1967; Galbraith, 1977; Tushman and Nadler, 1978; Burton and Obel, 1984).

Such interdependence arises at the task, resource or agent level among those actors who carry out the divided labour (March and Simon, 1958; Thompson, 1967; Puranam et al.,

2012; Pfeffer & Nowak, 1976; Pfeffer & Salancik, 1978). Thus, it follows that all approaches to organizational coordination are interested in the governance of interdependence as a matter of joining resources and collaborating (Grandori, 1997).

Organizational scholars propose that alternative forms of organizing correspond to different levels of interdependence. Thompson (1967), for example, presents three progressively complex levels of interdependence coupled with effective organizational mechanisms for each (Thompson, 1967). Firstly, there is “pooled interdependence”, when subtasks can be carried out relatively independently and concurrently. Achieving coordination in the context of pooled interdependence requires the specification and enforcement of rules and standards, which can be realized through detailed work processes.

Secondly, there is “sequential interdependence”, when subtasks cannot be carried out until one or more prerequisite has been met. Achieving coordination in this context requires hierarchical coordination, which can be achieved through scheduling tasks, and rescheduling them if changes arise. Lastly, there is “reciprocal interdependence”, when subtasks need close coordination as they proceed. Achieving coordination in this context requires “mutual adjustment”, which can be realized through face-to-face meetings

(Thompson, 1967).

In an extension of Thomspon’s (1967) conceptualization of mutual adjustment,

Levitt (2015) introduces more specific forms of organizing, which depend on whether the reciprocal interdependence among actors is compatible or contentious. Compatible

26 interdependence arises when tasks allocated to interdependent actors do not have conflicting or inconsistent sub-goals. In this case, mutual adjustment can be achieved through information sharing, self-organized relationships and networks (Levitt, 2015). On the other hand, contentious interdependence arises when task allocation gives rise to goal misalignment, and actors have sub-goals that conflict and that are progressively more inconsistent. In this case, mutual adjustment requires further effort and can be achieved through negotiation processes and the escalation of decisions in the case of an impasse, to deliver outcomes at the system level (Levitt, 2015).

In the megaproject meta-organization core, we can expect that both compatible and contentious interdependence arise. However, we conjecture that the nature of interdependence will tend towards the latter because core actors have heterogeneous interests, objectives and values (Denis et al., 2011). Furthermore, each core actor controls specific resources or possesses specific knowledge and capabilities, which equip them to carry out particular tasks better than others (Gil 2009, Gil and Tether 2011; Maurer, 2010).

Such know-how would, therefore, have to be represented and exchanged across organizational boundaries, a process that is not necessarily politically neutral process (Star

& Greisemer, 1989; Carlile, 2002; Carlile, 2004; Srikanth & Puranam, 2011, Teurtscher et al., 2014).

In a scheme to promote a national railway line, for example, the Government and its promoting agent will be the sole actors with the architectural knowledge for the totality of the scheme. To that end, they will know how the different components of the system work together and how they are related to each other (Baldwin & Clark, 2000; Henderson &

Clark, 1990; von Hippel, 1990). With this higher-order knowledge, they are thus better equipped to forecast global costs and timescales. However, local actors will have local knowledge and other resources that will be a critical input for making robust strategic choices such as locating an urban train station. As the decision-making process unfolds,

27 these powerful actors will negotiate and bargain in a bid to influence the technical outcome of the scheme (Gil 2009, Gil and Tether 2011; Maurer, 2010; Levitt, 2015). In the context of such divisions, an additional integration burden arises to bring the distinct (organization or technical) components together into an integrated whole.

2.2.2. The Integration of Effort

The integration of effort is central to organization design, and it is both necessary and paramount in the context of unitary organizations (Puranam & Goetting, 2011;

Puranam et al., 2014). Integrating effort is about promoting actions that maximize the value of the system as a whole, for a given division of labour. Similarly, in pluralistic settings, integrating the effort of autonomous actors is critical for achieving the system- level goal (Denis et al., 2001,2007, 2011; Jarzabkowski & Fenton, 2006). For the megaproject meta-organization, for example, it is only through integrating effort and aggregating core actors’ resources and efforts that contributions will translate to the development and delivery of a functional infrastructural asset (Garud & Karnoe, 2003).

When organizing in pluralistic settings, and high dispute contexts more generally, the integration of effort is a particularly challenging task (Bechky, 2003; Knorr-Cetina, 1999).

The challenge arises in part because multiple actors take decisions in a more decentralized manner, which naturally brings issues of group problem formulation to the fore (Lee and

Puranam, 2015; Baer, Dirks & Nickerson, 2013). Furthermore, actors share authority, yet have rivalrous preferences, and cannot be excluded from the strategic decision-making process (Gil & Baldwin, 2014).

Despite the abovementioned challenges, integrating effort is a prerequisite for achieving the system-level goal, as is the voluntary contribution of critical resources by self-interested actors. Thus, when disagreements arise in these settings, coalitions can rarely impose their strategic preferences on the collective (Pettigrew, 1990). Attempts to

28 do so can result in actors defecting, which ultimately contributes to the disintegration of the organization, and failure to achieve the system-level goal.

From organization design literature, we know that integrating effort requires the resolution of both coordination and cooperation problems among interdependent actors

(Gulati et al., 2005; Lawrence & Lorsch, 1967). Therefore, in the following section, we review what we know about each in the context of unitary organizations. We then introduce Ostrom’s ideas of organizing collective action arenas in the context of diffuse power and authority. We carry out this exercise to gain insight into organizing collective development in pluralistic settings.

2.2.1.1. Resolving Coordination Problems

As previously discussed, interdependence at the task, agent or resource level requires some degree of coordination (Thompson, 1967; Levitt, 2015). Typically, this coordination takes the form of information processing, whereby flows of information allow interacting and independent actors to accurately anticipate each other’s actions (Puranam et al., 2012;

Tushman and Nadler, 1978; Schelling, 1960). During this process, information that supports coordinated action is gathered, interpreted and synthesized for decision-making and communication (Tushman & Nadler, 1978; Clark, 1996; Schelling, 1960; Knudsen &

Srikanth, 2014; Okhuysen & Bechky, 2009). The dynamic creation of ‘common ground’, i.e., knowledge that is shared and known to be shared (Srikanth & Puranam, 2011), then gives rise to reciprocal predictability in that interacting and interdependent actors can accurately anticipate each other’s actions (Puranam et al., 2012).

Information processing allows interacting actors to ‘speak the same language’, at least from a technical perspective (Colfer & Baldwin, 2010: 20). This coordination is especially important in the context of contentious interdependence (Levitt, 2015), or, more generally, when interdependent actors interact under conditions of high uncertainty

(Lawrence & Lorsch, 1967; Tushman, 1978; Tushman & Nadler, 1978, Galbraith, 1973; 29

Arrow, 1974). Information processing is, however, complicated by bounded rationality or an individual’s limits of knowledge and computational capabilities (Simon, 1957; Simon,

1955; Knudsen & Srikanth, 2014; Puranam, Stieglitz & Osman, 2015). Information processing is further complicated by task and environmental uncertainties (Tushman and

Nadler, 1978).

Surely, we can expect that information processing needs are salient in the early stages of organizing under pluralism, when actors are searching for ways to reconcile their heterogeneous preferences (Knudsen & Srikanth, 2014; Gavetti, 2012; Puranam et al.,

2015; Greve, 1998). Indeed, information processing is critical in the early stages of large- scale infrastructure development projects when actors are deciding on the project mission, concept and narrative (Winch, 2015). However, the large physical scale of these projects makes it challenging to process information and thus resolve the coordination problem.

Because a megaproject impacts many independent actors or so-called stakeholders, we can expect that the necessary information for consensus-building and for resolving emerging disputes will be dispersed. Thus, the promoter has to undertake the challenging task of processing vast amounts of information and coordinating work across multiple, legally independent actors with heterogeneous preferences.

Complicating matters further, the assets developed on large-scale infrastructure projects have long operating lives. Long timescales constitute a significant cause of uncertainty, and thus increase information processing requirements, as do technological complexity and innovation (Tushman & Nadler, 1978). Issues, therefore, arise when infrastructure developments are technologically complex, or if they involve untested technologies (Gann & Salter, 2000). However, as Simon (1996) argues when discussing design in general, in many cases, “the inner environment, the hardware, is simple.

Complexity emerges from the richness of the outer environment” (Simon, 1996: 110).

Therefore, taken together, the complexity of the inner and outer environments contributes

30 to increasing information processing requirements during the design decision-making process.

Notwithstanding the challenge of processing information, organizations need to match task information requirements with their information processing capacity to achieve collective action (Galbraith, 1973; March and Simon, 1958; Tushman & Nadler, 1978).

Organization design literature suggests three groups of mechanisms that can be used to achieve this matching: i) hierarchy ii) targets and goals and iii) coordination by rules and programs (Galbraith, 1974). In pluralistic organizations, however, these mechanisms are not necessarily available. By definition, a hierarchy is absent in a meta-organization

(Gulati, Tushman and Puranam, 2012). Plus, if an organization is pluralistic, the goals and targets of the participating actors are often divergent (Denis et al., 2011). There is no reason to expect that rules and programs cannot be used to coordinate work under pluralism. However, this requires collective problem formulation as well as agreement on the rules and programs that will guide the search for solutions. Consensus-building presents a challenge in itself given that negotiations among actors with different objectives can lead to political manoeuvring. In turn, this consumes scarce resources such as attention, memory capacity, and time; and distracts actors away from problem formulation, which ultimately undermines the collective effort (Baer, Dirks, and Nickerson, 2013).

Organizational scholars further suggest that coordination challenges can be overcome through mechanisms that enhance the formation of predictive knowledge

(Thompson, 1967; Puranam et al., 2012). These mechanisms include communication, mutual observation, learning and (joint) decision-making (e.g. Galbraith, 1974; March and

Simon, 1958; Tushman & Nadler, 1978, Puranam et al., 2012). For Galbraith (1974) such mechanisms, reflected in vertical and lateral structures, increase an organization’s capacity to handle information. In addition to hierarchy, these mechanisms include teams, task forces, committees and project managers (Galbraith, 1974; Levitt, 1984).

31

Srikanth and Puranam (2014) describe coordination mechanisms more generically as ongoing communication, tacit communication and modularity. Explicit communication occurs through direct interactions such as face-to-face meeting. On the other hand, tacit communication occurs through common procedures, documents, goals, and authority, which serve to enhance the observability of context, actions and outcomes (Srikanth &

Puranam, 2014; Clark, 1996). These mechanisms allow for the emergence of shared knowledge, and reciprocal predictability, which neutralize attempts to subvert coordination through political manoeuvring (March & Simon, 1958; Puranam et al., 2012; Srikanth &

Puranam, 2014). Decomposing problems into modular task architecture gives rise to pooled interdependence (Thompson, 1967). As such, different actors can carry out their work in parallel with limited coordination needs because the design interfaces between modules ensure that actions are coordinated at a higher level (Baldwin and Clark 2000;

Galbraith, 1974; Tushman and Nadler, 1978; Simon, 1962; Puranam et al., 2014: 173).

Together, various coordination mechanisms help actors to leverage what Puranam et al. (2012: 6) call epistemic interdependence, “a situation in which one agent’s optimal choices depend upon a prediction of another agent’s actions.” In the absence of such mechanisms, coordination failures can occur such that actors lack predictive knowledge and “thus cannot predict the actions of another actor in situations where such a prediction is essential for optimum outcomes” (Puranam et al., 2010: 10).

Face-to-face communication is a coordination mechanism that extends easily into pluralistic organizations (Srikanth & Puranam, 2014; Clark, 1996; Okhuysen & Becky,

2009). However, we can expect the costs of face-to-face communication to increase substantially in large-scale projects where actors are not co-located and don’t have a history of working together (Srikanth & Puranam, 2014). Thus, we should not take it for granted that face-to-face communication happens extensively.

32

Alternatively, tacit communication mechanisms can be used to create and leverage common ground (Clark, 1996; Cramton, 2001; Gutwin et al., 2004). These mechanisms include common procedures, or common goals, which enhance the observability of the context, actions and outcomes (Srikanth & Puranam, 2014, 2011; Schelling, 1960). They also include shared documents that enable multiple actors to exchange knowledge, assess each other’s knowledge, and transform knowledge to better address the design problem

(Star, 1989; Carlile, 2002). Relying on tacit coordination mechanisms can be difficult, however, because specialization within boundaries impedes communication and hinders the free flow of information (Dearborn and Simon, 1958). Thus, the transfer of tacit knowledge can be slow, costly and uncertain (Kogut and Zander, 1992; Zander and Kogut

1995).

Pluralistic organizations can also try to decouple tasks through investment in nearly decomposable, or modular structures (Simon, 1962; Baldwin & Clark, 2000; Eppinger &

Ulrich, 2016). However, we cannot assume that these mechanisms are readily available in megaproject organizations. Also, modularity is expensive (Henkel & Baldwin, 2009), and it is difficult to decompose strategy formulation in pluralistic organizations ex-ante because in these settings strategies tend to be emergent (Denis et al., 2007).

Coordination failures occur when interdependence is unknown or unstable because actors are undertaking nonroutine tasks with a high frequency of exceptions. Such circumstances stress the information processing capabilities of an organization (Galbraith,

1974, Thompson, 1967; Scott, 2003; Jarzabkowski et al., 2010). We can also expect this to be the case in one-off project-based pluralistic organizations such as those created to develop large infrastructure. The higher the degree of uncertainty surrounding the task or the task environment, the higher the amount of information that decision-makers have to process (Galbraith, 1974, Arrow, 1974). Furthermore, in the context of high

33 interdependence and high uncertainty, the functional capabilities of the system can become overloaded (Galbraith, 1974).

Galbraith (1974) proposes that, in addition to adopting mechanisms that increase their information processing capabilities, organizations can achieve coordination by reducing their information processing needs from nonroutine tasks. To that end, organizations can create slack resources (e.g. by extending planning targets or increasing budget contingencies) to decouple tasks. Slack resources contribute to what Cyert and

March (1963) call the ‘quasi-resolution of conflict’ because they allow actors to reconcile divergent goals without resolving conflict when they fail to meet their collective aspirations. However, we cannot assume that these mechanisms are readily available in pluralistic organizations. Slack is likely to be finite and using side payments to resolve disputes can create issues of precedence and equity. Furthermore, these organizations unfold under the scrutiny of multiple third parties, which makes it difficult to use slack to resolve conflict, particularly when resources are constrained (Gil and Tether, 2011; Gil and

Pinto, 2016).

Crucially, overcoming coordination challenges does not guarantee a desirable outcome. Highly interdependent work needs to be organized such that there is a high level of communication between actors (Galbraith, 1977; Tushman and Nadler, 1978), which allows them to develop common ground (Simon, 1947; Schelling, 1960; Srikanth and

Puranam, 2011). However, it is possible for actors to coordinate successfully on an adverse outcome, such as when employees collude to shirk on the job (Puranam and Goetting,

2011). Hence, cooperation problems, which are discussed next, also need to be resolved.

2.2.1.2. Resolving Cooperation Problems

While coordination problems arise when actors struggle to predict each other’s actions, cooperation problems are rooted in diverging or misaligned interests among interacting and interdependent actors (Gulati et al., 2012). Furthermore, cooperation 34 problems are more pronounced when actors perceive that the costs of cooperating outweigh the benefits (Gavetti et al., 2007). In other words, actors cooperate if they feel it is worth their effort to carry out the tasks allocated to them and adhere to professional norms (Lee and Puranam, 2015).

For organization scholars, the notion of cooperation has informed work on encouraging cooperative behaviour using inducements and rewards, such as monetary incentives, power, and reputation (Gulati, 2007; Gavetti et al., 2007; Simon, 1965; March and Simon, 1958; Cyert and March, 1963). Employer-employee relationships are another way to encourage cooperation in that employees cooperate in exchange for a guaranteed salary (Coase, 1937). Organization design theorists also argue that cooperation challenges can be overcome through legal, private and social sanctions, which prevent opportunistic behaviour while sustaining commitment to specific investments, trust and contractual safeguards (Gulati et al., 2012).

Research extending organization design into inter-organizational settings recognizes that, in these settings, problems of cooperation arise due to conflicting interests among more symmetrically dependent actors. In dyads or triads formed to negotiate strategic alliances between firms, for example, cooperation problems can arise due to an organization’s loss of interest in the partnership. Moreover, they can arise due to a firm’s diminished commitment that gradually weakens a relationship initially underpinned by a collective profit-seeking goal (Gulati et al., 2012; Doz, 1996). These problems are exacerbated when coupled with opportunism, i.e., the pursuit of self-interest with no regard for unenforceable commitments or moral obligations (Williamson, 1985). To that end, in the absence of norms of cooperation, actors can renege on their agreements by contributing less to a common pool of resource, or by trying to claim more benefits than agreed.

Uncooperative actors can also misappropriate their partner’s resources, exploit a superior

35 bargaining position or hold up the process to negotiate more favourable terms (Gulati et al.,

2012).

To minimize adverse occurrences, organizations entering into strategic alliances agree on the scope of their relationships upfront. They negotiate the inputs they are willing to contribute (time, resources, market access, for example) and outline what they expect from the collaboration (Gulati & Singh, 1998). Agreements are then captured in a contract, which governs the strategic alliance. Thus, despite the challenges associated with forming inter-organizational relationships, they are still seen as worthwhile for sharing risk or realizing benefits (operational, commercial, technological, or reputational) that are challenging to get through market transactions (Gulati et al., 2012).

Pluralistic settings are however fundamentally different from unitary organizations, strategic alliances, and open innovation settings, which predominantly inform the study of cooperation in the organization design literature. In pluralistic settings, power and influence are shared by multiple organizations (Denis et al., 2011). Thus, unlike in unitary organizations, there is no authority hierarchy, an organizational mechanism that is useful for applying sanctions and providing inducements and rewards (Simon, 1962; March &

Simon, 1993). Furthermore, sanctions and rewards to encourage cooperation are not readily available and must first be agreed to by actors in the pluralistic setting.

Pluralistic settings are similar to strategic alliances because of the resource interdependence among actors in both contexts (Pfeffer & Nowak, 1976; Pfeffer &

Salancik, 1978). However, strategic alliances are governed by a formal contractual structure used by partners to formalize the alliance ex-ante. Contrastingly, in pluralistic settings, it is not always possible to reach upfront agreements on input contributions and expected outputs. The notion that heterogeneous actors with conflicting goals share decision-making authority in pluralistic settings is also an obstacle for developing meritocracy-based forms of authority. This form of authority is useful for encouraging

36 cooperation and resolving disputes in less heterogeneous organizations such as communities of scientists (Teurtscher, Garud & Kumaraswamy, 2014) and open source communities (O’Mahony & Ferraro, 2007). Furthermore, pluralistic settings are very different from buyer-supplier relationships (where the buyer can buy cooperation from the supplier) given that pluralism denotes the diffusion of decision-making authority and goal heterogeneity.

All in all, organization design literature suggests that overcoming problems of cooperation in pluralistic meta-organizations is complicated. Complications arise because authority is diffused in these settings and actors are heterogeneous in terms of planning horizons, goals, capabilities, knowledge bases, values and interests (Gulati et al., 2012;

Denis et al., 2011). Nevertheless, this heterogeneity is important as it indicates that different actors can potentially contribute very different resources, which can be captured in an advantageous division of labour.

In contrast to unitary organizations, pluralistic meta-organizations are also without a hierarchal authority structure that designs and applies inducements and sanctions that encourage cooperation. Hence, pluralistic organizations need to achieve their system-level goals while mired in disputation. From management studies, we know that decision- making processes in pluralistic settings are endemically political (Gray, 1989; Gray &

Clyman, 2003; Puranam, Raveendran, & Knudsen, 2012; Langley, 1995). Furthermore, in these settings, there is also a high risk that cooperation failure becomes the norm, and actors become trapped in networks of indecision (Denis et al., 2011). Notwithstanding the high level of complexity and uncertainty in pluralistic settings, we know from organization design scholars that organizational differentiation among actors requires the integration of effort to achieve the system-level goal (Lawrence & Lorsch, 1967; Tushman, 1978;

Tushman & Nadler, 1978, Galbraith, 1973; Grandori, 1997; Dougherty 1992, Heath and

Staudenmayer 2000, Lawrence and Lorsch 1967, Sosa et al. 2012).

37

Outside of management literature, Elinor Ostrom has theorized extensively about collectives that similarly operate without an authority hierarchy and employer-employee relationships. Her research recognizes that pluralistic arenas are endemically political and that they resemble bargaining structures. However, the focus of her work has not been on negotiation processes. Instead, her work has focused on the design of rules and structures that support negotiation processes, and make self-governing organizations sustainable

(Ostrom, 1990; 2009). We now turn to discuss the relevance of her ideas in our effort to extend organization design literature into pluralistic settings.

2.3. Designing Self-Governing Organizations for Collective Action

Collective action occurs when otherwise self-interested, individual or organizational, actors come together to further a common interest (Ostrom, 1990). As is the case in megaproject meta-organizations, these actors are interdependent because individually they cannot realize the shared purpose, or system-level goal, that unifies them and brings mutual benefits. In the collective action literature, this goal is the sustainability of a

‘common-pool resource’ such as a fishery, groundwater basin, or irrigation system that exists under the shared ownership of a community (Ostrom et al., 1994; Frischman, 2005).

The defining characteristics of common-pool resources are low excludability and high subtractability (Ostrom, 1990). Low excludability implies that it is not trivial to exclude claimants from using the shared resource. High subtractability implies that the use of the shared resource is prone to rivalry. i.e. the benefit that one user derives from the resource detracts from another user benefitting simultaneously. For example, the water extracted from an irrigation system by one user cannot simultaneously be extracted by another user, and if the amount of water available is limited, the more one user extracts, the less is available for all others.

38

The notion of a common-pool resource is fundamental to Ostrom’s work on organizing self-governing arenas of collective action (Ostrom, 1990; Ostrom, 2005). This notion also suggests important points of similarity and difference between Ostrom’s settings and the pluralistic settings at the heart of our research. An important point of similarity is that in both settings, collective action is underpinned by resources shared by non-excludable actors with rivalrous preferences. To that end, the strategic design choices that need to be made when developing large-scale infrastructure, conflate the properties of low excludability and high rivalry (Gil and Baldwin, 2013). These strategic choices exhibit low excludability because they directly impact multiple independent actors, and some actors will even share the future asset in use. These actors contribute critical inputs to the

‘commons’ such as technical knowledge, finance, planning powers, and political support.

Thus, in exchange for their resource contributions, they claim rights to influence strategic design choices directly, and they are therefore difficult to exclude from the strategic decision-making process. In a regime of strong property rights, laws in the environment will also ensure that legitimate claimants cannot be excluded from the decision-making process. Such actors include those with land at risk of being expropriated, future users, or actors who may suffer blight to their properties.

The emergent strategic design choices also exhibit high subtractability because claimants to the design have heterogeneous preferences; however, technical and financial constraints dictate that they have to make trade-offs and compromises. Thus, we can expect rivalry in regards to their preferred design choices. This rivalry would not exist if the design could be broken apart into individually-owned modular components, each one meeting the needs of a particular actor (Simon, 1962; Baldwin & Clark, 2000; Eppinger &

Ulrich, 2016). However, rivalry is generally high if a design is not decomposable (Levitt,

1984). Under these circumstances, actors have to negotiate their preferences for one-off strategic design choices until mutually consensual solutions emerge. To that end, Ostrom

39

(1990) presents organizational design principles, which can be used to design robust rules and structures that facilitate collective action. Next, we briefly review these principles before discussing their applicability in pluralistic settings.

2.3.1. Designing Robust Self-Governing Organizations

Revised theories of collective action, formalized by Ostrom (1990), emerged in response to traditional theories, which argue that failures of collective action are rooted in a misalignment between actors’ self-interest and the common interest. In traditional theories, the former is, in fact, incompatible with a collectively beneficial outcome (Olson,

1965; Hardin, 1968, Ostrom, 1990). Taking this view, pessimistic scholars of collective action argued that self-governing organizations were unsustainable because self-interested actors would act against the optimum outcomes for the group.

In other words, traditional theories of collective action argued that self-interested actors in self-governing collective action arenas would be faced with a conflict between

“individual rationality” and “group rationality”, which would give rise to a social dilemma

(Olson, 1965; Hardin, 1968). Although actors would try to bring about desirable outcomes, they would ultimately advance their self-interest at the expense of the common interest.

Furthermore, the larger the collective action arena, the more likely it would be that actors’ advancing their self-interest would trigger its collapse (Olson, 1965). As Hardin (1968:

1244) put it: “Freedom in a commons [organization] brings ruin to all.” The notion of a

‘tragedy of the commons’ was captured in canonical models such as the prisoners’ dilemma. The same idea was also used to explain empirical phenomena such as the underprovision of local public goods, overharvesting, and overfishing. Furthermore, the research went on to conclude that cooperation could only be achieved through government regulation or privatization (Olson, 1965; Hardin, 1968).

40

In light of empirical evidence to the contrary, Ostrom (1990, 2008, 2009) and her colleagues revised theories of collective action (Walker & Ostrom, 2009, Hyden, 1990;

Oakerson, 1993, Dietz et al., 2002). Through the study of robust self-governing organizations, they found that communication, and the shadow of future interactions, encouraged the development of norms of cooperation. This finding suggested that actors within a collective are not locked into a narrow, short-term pursuit of self-interest.

Revised theories of collective action also challenge the in-built assumptions of perfect information and utility maximization previously used when predicting a tragedy of the commons. Instead, and in line with organization design theorists, these theories assume that actors are boundedly rational and possess imperfect information (Ostrom, et al., 1994; Simon 1955, 1965, 1972; Williamson, 1985). Along with imperfect information- processing capabilities and incomplete information, Ostrom (1986) also recognized that any individual could make a mistake when choosing a strategy designed to realize a collective goal. Thus, the assumptions of imperfect and incomplete information are complemented with the assumption that actors are fallible learners (Ostrom, 1986; Ostrom

& Ostrom, 1977).

Related to our setting, Ostrom (1990, 2005) argues that effective rules and structures can enable and constrain collective action. Effective as used in this regard refers to rules and structures that are perceived as legitimate and enforceable, and that are understood by actors within the collective in a similar manner (Ostrom, 1990, 2008; Dietz et al., 2002:

20). Effective rules and structures also attend to the biophysical structure of the shared resource, and they exist at multiple levels of governance. Thus, either implicitly or explicitly, they account for the structures at different levels of governance (Ostrom, 1990).

Designing robust and effective self-governing organizations is, however, not a trivial problem and it can be a slow process to overcome emerging disputes (Ostrom, 1990). The process is especially challenging for emerging organizations that need to adapt to

41 technological and economic change (Ostrom, 1990). Thus, both efforts to create and change governance structures already in use are equally disputational (Ostrom, 1990;

2005).

As with scholars in organization design, commons theorists recognize that there is no such thing as an optimal set of organizational structures because optimality is contingent on the context (Ostrom, 1990, 2005). That notwithstanding, Ostrom (1990) proposes eight principles that can avert a tragedy of the commons when incorporated into the design of self-governing organizations. The design principles, listed in Table 2.1, capture the structural similarities among sustainable self-organized systems.

Table 2.1 Design Principles (Ostrom et al., 1994)

Principle Explanation

Clearly defined The boundaries of the resource and its rightful participants are clearly boundaries defined Rules affecting appropriation and provision are congruent. In other words, Proportionality of the distribution of costs is proportional to the distribution of benefits, and benefits and costs those who pay the most, receive the most. Collective Choice Individuals affected by rules can participate in modifying them and rules are arrangements congruent with local social and environmental conditions

Monitors who can audit the condition of the resource and participant Monitoring behaviour are responsive and accountable to the participants

Participants who violate rules are likely to be assessed graduated Graduated sanctions sanctions. Small and first offences are identified, but not severely punished Participants have rapid access to low-cost local arenas to resolve conflicts Conflict resolution among users or with officials. Ideally, conflict resolution should be face-to- mechanisms face Recognition and The rights of participants to manage the commons, change local rules, non-interference by monitor and sanction peers and resolve their own conflicts are not authorities challenged by government authorities Nested “polycentric” When a common-pool resource is closely connected to a larger system, enterprises governance activities are organized in multiple nested layers

Each principle listed in Table 2.1 contributes to encouraging actors to behave cooperatively (Ostrom, 1990: 415). Taken together the principles suggest that robust structures comprise small closed groups of actors, who share in resources with well- defined boundaries. Actors in robust structures can also easily monitor each other, enforce sanctions for rule infractions; and they have access to low-cost dispute resolution mechanisms when necessary (Wade, 1987; Baland & Platteau, 1996; 1999; Ostrom, 1990). 42

The idea is that governance structures underpinned by these principles can better address commitment problems, monitor individual compliance with existing structures, and cope with free riding. All of these problems are endemic to self-governing organizations.

Overall, the design principles aim to be adaptive to heterogeneous local conditions as opposed to providing a universal blueprint for imposing on self-governing collective arenas (Ostrom, 2008). In addition, the design principles do not assume that an omnipotent external force, e.g. the Government or a systems architect, operating in a vacuum, designs emerging organizational structures (Ostrom, 1990). Instead, robust rules and structures are emergent, and the outcome of negotiated processes among the actors in the self-governing organization in a given context (Ostrom, 1990; 2008).

The final design principle, nested “polycentric” enterprises, denotes “many centres of decision-making that are formally independent of each other” (V. Ostrom, Tiebout &

Warren, 1961:831; V. Ostrom, 1972; V. Ostrom, 1999a, 1999b). This principle suggests that the governance of common pool resources can span multiple levels of decision- making. In the collective action literature, these include the operational, collective choice and constitutional levels of governance (Ostrom, 1990; 2005; 2008). Operational structures set out the rules for everyday activities that directly involve the use of various resources.

Collective choice structures set out rules about how to change the operational rules, and constitutional structures set out rules about how to change the collective choice structures, and about actors’ roles at a collective choice level. The rules and structures at all three levels are shaped by the social norms and traditions of a particular context, which influences their perceived acceptability and legitimacy (Ostrom, 1990, 2005).

The notion of polycentricity is underexplored in the collective action literature.

However, it is this idea that most approximates commons theory to the pluralistic organizations at the heart of our research. In polycentric structures, as in pluralistic organizations, multiple actors with conflicting goals can influence the design of

43 organizational rules and structures (Ostrom, 2012). However, a crucial difference arises between polycentric self-governing organizations as conceptualised in the Ostrom’s work and pluralistic organizations such as those formed to develop large infrastructure. In the former, government authorities are expected to set high-level collective choice rules but do not actively participate in the operational management of a shared resource. In contrast, in our pluralistic meta-organizations, the Government and other public and private agencies actively participate in and are direct claimants to the strategic decision-making process.

Central authorities also design the rules of the game within which actors form coalitions, bargain, and contribute and withhold resources. Moreover, they resolve disputes, as necessary. Thus, in pluralistic organizations formed to develop large infrastructure, different levels of decision-making overlap and sometimes interpenetrate.

Ostrom’s (1990) design principle regarding clearly delineated organizational boundaries and roles seems difficult to enact in pluralistic settings where authority is diffused, boundaries are fluid and contested, and actors have multiple objectives (Denis et al., 2011). Interestingly, ensuring the proportionality of costs and benefits is necessary for maintaining a sense of fairness and legitimacy in a common pool resource system, however, in pluralistic settings measuring costs and benefits is itself a subject of contestation (Langley, 1991). Nevertheless, the ideas of polycentricity, sanctions, monitors and low-cost conflict resolution mechanisms can potentially extend to pluralistic organizations. These differences and similarities suggest that there are opportunities to combine Ostrom’s ideas (1990, 2005) with organizational design literature. In so doing, we can build a cognitive lens to interrogate empirical data on organizing in pluralistic settings.

We can also develop new theory about why some pluralistic organizations struggle to build consensus more than others.

Trying to understand how organizational structures encourage the integration of effort in a pluralistic organization is akin to asking how multiple organizational actors

44 coordinate information and reconcile conflicting interests to agree on mutually accepted strategic choices. In asking this question, we are close to the core motivation of Ostrom’s

(1990, 2008) research to further our understanding of how organizational structures can foster independent self-interested actors to collaborate. In Ostrom’s (1990: 25) words, this requires actors to switch from “simultaneous, non-contingent, and frequency-independent actions” to “sequential, contingent, and frequency-dependent decisions.” It is our task here to investigate which organizational rules and structures can facilitate this switch.

We carry out this study with the appreciation that the design of organizational rules and structures is the outcome of negotiated processes in a given context. From organization theorists, we understand that context matters because different organizational structures can have varying effectiveness in different contexts (Burns and Stalker, 1961; Chandler,

1962; Galbraith, 1973). In line with this assertion, commons theorists do not postulate a causal relationship between a specific organizational structure and optimal performance.

Instead, they recognize that the design principles contribute to creating robust self- governing organizations across diverse common pool resource settings. Moreover, in agreement with Ostrom (2005: 275), we also understand that there is no “holy grail” for structuring a pluralistic organization. Nevertheless, as both organization design literature and collective action literature suggest, some organizational structures are more amenable to integrating effort than others.

2.4. Summary

In this chapter, we reviewed the organization design and collective action literature to deepen our understanding of organizing in pluralistic settings. We began by outlining the two fundamental problems of organizing from an organization design perspective: the division of labour and the integration of effort. With a particular focus on the latter, we considered whether coordination and cooperation mechanisms used to govern unitary organizations can be applied to pluralistic meta-organizations. By taking the similarities

45 and differences between the two settings into account, we determined that, except for hierarchy, the coordination mechanisms used in unitary organizations can be extended to meta-organizations in pluralistic settings. Contrastingly, the cooperation mechanisms used in unitary organizations would not extend as easily given the absence of hierarchical structures and contractual arrangements in the meta-organization core. To that end, we introduced the collective action literature (Ostrom’s, 1990, 2009). Although this literature is not framed in terms of the division of labour and the integration of effort, it provides a useful starting point for investigating structures that can encourage cooperation in pluralistic settings. In the next chapter, we outline the research setting, methods and data.

We then go on to use the theoretical framework outlined in this chapter to interrogate empirical evidence in order to deepen our understanding of organizing collective development in pluralistic settings.

46

3. RESEARCH SETTING, METHODS AND DATA

3.0. Introduction

In this chapter, we outline the abductive case-based research strategy that we adopted to explore organizing in pluralistic settings empirically. An abductive research approach benefits from the dual impact of both inductive and deductive approaches, and thus from the constant interplay of theory and data (Klag and Langley, 2013). Moreover, case studies support inquiries into areas with little previous research, as is the case with conceptualizing governance in pluralistic settings from an organization design perspective (Eisenhardt,

1989; Gulati, Puranam, Tushamn, 2012; Puranam, Alexy & Reitzig, 2014). To that end, we present a single case study of High Speed 2, the new railway line in the UK. Furthermore, we embed five units of analysis into the HS2 case (Yin, 2013). Our units of analysis are the inter-organizational groups formed to decide on station locations in the following cities along the proposed route: Birmingham, London, Manchester, Leeds and Sheffield. To develop the case, we drew on the theoretical framework we developed in Chapter 1. We also triangulated primary data from semi-structured interviews with secondary data from archival documents and other HS2 publications (Yin, 1994). We then recursively cycled between the data, extant literature and emerging theory in an attempt to construct logical explanations about organizing collective development in pluralistic settings (Yin, 1994; van de Ven, 2007; Klag and Langley, 2013).

3.1. Research Setting

The research setting is High Speed 2: a national high-speed railway network promoted by HS2 Ltd., a public agency wholly owned by the Department for Transport of the UK national government. The HS2 infrastructure project was proposed to build a high- speed rail line from London, in the south of England, to Manchester and Leeds in the north, via Birmingham, the East Midlands, Sheffield and Crewe. The project was proposed to be built in two stages: a first stage connecting London to Birmingham by 2026 (later extended 47 to Crewe, 48 miles north of Birmingham); and a second stage connecting Birmingham

(later replaced by Crewe) to Manchester and Leeds by 2032. Since 2009, HS2 has been supported by all three of England’s major political parties - Conservatives, Labour and the

Liberal Democrats - in government and opposition. Thus, HS2 enjoys widespread cross- party support in Parliament. Nevertheless, the project remains controversial outside of

Parliament with disagreements around the strategic case of the scheme, for example.

3.2. An Abductive Case-Based Research Strategy

The research was undertaken abductively through an ongoing process of exploring theoretical and empirical findings permitted to suggest a theory (Pierce, 1958). This process was triggered by a ‘breakdown in understanding’ when, initially, the topic of organizing collective development was interrogated using collective action literature and design theory (Van de Ven, 2007; Gil and Baldwin, 2014; Msulwa, 2013). Certain aspects of the phenomena were left unexplained through this lens, which resulted in ‘genuine doubt’ that generated a search for alternative explanations and the production of conjectures about how organizing influenced the collective development process (Klag and

Langley, 2013, Locke, 2007). What followed was a systematic combination of theoretical and empirical findings in search of the best fitting ideas to construct logical explanations about our empirical observations (van de Ven, 2007; Klag and Langley, 2013).

The process of abduction involved an extensive iterative processing using literature, data and emergent theory. Multiple lines of research were considered during this exercise including open source communities (O’Mahony and Ferraro, 2007; Lakhani, Lifshitz-

Assaf & Tushman, 2013; von Hippel and von Krogh, 2003); co-development of infrastructure assets (Gil and Baldwin, 2013; Lundrigan et al., 2014), global communities of scientists (Teurtscher, Garud and Kumaraswamy, 2014), and international development

(Levy, 2014). Complementing these research efforts, in both 2015 and 2017, an opportunity arose to attend the Academy of Management (AoM) and Organization Design

48

(ODC) Conferences. Attending both of these conferences, but especially the latter, was pivotal for the research. The social embeddedness provided an avenue for discussing, expanding and generating ideas (Klag and Langley, 2013). Furthermore, attending presentations by scholars using related literature streams helped us to understand nuances we would have otherwise missed.

This abductive approach differs from a purely inductive approach, which requires that empirical observations be interpreted without any preconceptions or theoretical foundations (Glaser and Strauss, 1967). An abductive approach also differs from a purely deductive approach, where theory is drawn from outside of the data (Mounce, 1997;

Langley, 1999). Instead, the abductive research approach benefits from the dual impact of both inductive and deductive approaches. On the one hand, the theoretical literature offers a conceptual framework from which to consider the phenomena. It also helps to delineate important variables, suggest relationships among them, and direct the interpretation of findings. On the other hand, the data or field situation allows us to look for evidence to support, refute, or modify our emerging hypotheses (Strauss and Corbin, 1990: 148).

A case-based research strategy is particularly suitable for providing a rich empirical instance of phenomena for which little previous research exists (Yin, 1994; Eisenhardt,

1989). This is true of designing pluralistic meta-organizations, which is a topic underexplored from an organizational design perspective. Another key strength of a case- based research strategy is that case studies can account for the real-world context in which phenomena occur (Eisenhardt & Graebner, 2007; Eisenhardt, 1989). This strategy is particularly useful when the boundaries between the phenomenon and the context are not clearly evident (Yin, 2009).

To that end, megaproject organizations are enmeshed in the environment because their decisions are subject to numerous regulatory frameworks. For example, the core will have to formally consult multiple actors in the environment irrespective of whether they

49 possess resources that are critical for the development or not. Failure to do this puts the core at risk of having their decisions challenged legally, and ultimately not being upheld by the court system. Megaproject organizations are also enmeshed in the environment because multiple environmental actors categorically oppose the development. While not part of the pluralistic organization core, these actors can legally challenge the decisions of the core, and thus cannot be ignored. Developing a single case was, therefore, useful for capturing these nuances. It was also revelatory as we were able to richly describe the phenomena while retaining the holistic features of planning large-scale infrastructure projects (Yin, 2009; Yin, 2013; Siggelkow, 2007; Eisenhardt & Graebner, 2007).

Consistent with an embedded case study approach, multiple units of analysis were embedded into the case study (Yin, 2013). Embedding units of analysis allows for a more focused study enquiry, offers a significant opportunity for extensive analysis, and it enhances the insights from a single case (Yin, 2013). Single case studies with embedded units of analysis also permit replication (or corroboration) and an extension of the insights from each unit. This, in turn, enables the development of more elaborate theory and a more complete theoretical picture (Yin, 2013; Eisenhardt & Graebner, 2007). The embedded units of analysis are the inter-organizational groups formed to decide on an HS2 station location in cities along the route. Specifically, we studied local groups in Birmingham,

London, Manchester, Leeds and Sheffield.

Our sample was diversified along two dimensions: the urgency to agree on strategic choices and goal congruence. Regarding urgency, we included two units of analysis

(London Euston and Birmingham) that faced greater urgency as they were planned for

Phase One of the project. Both London and Birmingham faced high urgency to build consensus because the Government was keen to get the HS2 bill approved before May

2015, when national elections would be held. The other three units (Leeds, Sheffield and

Manchester) faced less urgency as they were scheduled to be delivered during Phase Two

50 of the project. Although subject to change, the Government intends to get the second bill approved around 2020. Varying our units of analysis with regard to urgency allowed us to interrogate how time, or the lack thereof, affects decision-making in pluralistic settings.

Regarding goal congruence, we included two units (London Euston and Sheffield) where actors had very heterogeneous interests. In London, local actors demonstrated varying degrees of support and opposition to the goal. On the one hand, regional actors in the Mayor's office including the Greater London Authority (GLA) and Transport for

London (TfL) were in support of HS2 subject to certain conditions. On the other hand,

Camden Council, the local authorities governing the locality where the proposed station would be located, expressed outright opposition to the project (though some actors within

HS2 Ltd. viewed this as a ploy to gain political support). Thus, as will be discussed in the following section, the two groups in London made demands in line with their respective interests. Goal congruence in Sheffield was also low, and in this case divergence was rooted in local actors advancing competing interests, which eventually led to an impasse in the design decision-making process. In the other three cases (Leeds, Manchester and

Birmingham) local actors were united in their support of the goal, and differences arose between local actors and the promoter to varying degrees. We were therefore interested in understanding what implications such heterogeneity has on collective development efforts.

Ultimately, a diversified sample permits the replication and extension of emerging logics, which enables the development of more elaborated theory (Yin 1984, Eisenhardt and

Graebner 2007, Siggelkow 2007).

3.3. Data Collection

The primary method of data collection was semi-structured interviews. We interviewed HS2 Ltd. staff members and multiple organizational actors that had direct interactions with HS2 Ltd. including local government officials, private consultants, transport authority officials, and other third parties. Appendix A includes two detailed lists

51 of all the interviewees, outlining their job roles and employers, structured by unit of analysis. We adopted a snowball sampling technique (Biernacki & Waldorf, 1981) to choose our interviewees. Hence for each unit of analysis, we asked the respondents to suggest other people from their organizations and across organizational boundaries who could provide a complementary point of view.

Interviews allowed for a targeted and direct focus on case topics and also granted insightful explanations about the infrastructure planning process (Yin, 1994). Furthermore, interviewing a range of informants with different perspectives was important for limiting individual bias in the accounts of the interactions, and for triangulating interview data

(Miles and Huberman, 1984; Eisenhardt & Graebner, 2007). More generally, interviewing multiple informants was also important for correcting any inaccuracies on our part due to poor recall or reflexivity when asking questions (Yin, 1994).

We used an interview protocol that outlined the key research questions and shared this protocol during electronic exchanges when arranging the interviews. Appendix B presents the exploratory questions we used during the initial stages of data collection.

During interviews, we adopted a semi-structured interview style, which gave interviewees the opportunity to shape the course of the interview (Cassell et al., 2004). A semi- structured interview style also gave interviewees an opportunity to volunteer relevant knowledge that we were not necessarily aware of (Yin, 1994).

During data collection, interviews were recorded and transcribed in their entire length. Initially, transcription was carried out by the author and then by a professional transcriber, in which case important conceptual details were shared to guide the interpretation of recordings (Miles and Huberman, 1994). These details included the project aims, the setting, interview participants and interview topics. Reviewing transcriptions in parallel to data collection allowed us to refine our research questions gradually. It also allowed us to build on insights gathered from previous interviews and

52 from the analysis of secondary data sources. In doing so, and following recommendations from scholars of qualitative research methods, we foregrounded topics that proved increasingly relevant and discarded areas that we initially thought could be relevant but eventually proved otherwise (Cassell et al., 2004; Yin, 1994).

As we refined our interview protocol, we became increasingly focused on investigating the rules and structures supporting the decision-making process to locate local railway stations. We were particularly interested in how these governance structures were helpful for developing common knowledge and overcoming disputes between the promoter and local actors, as well as amongst local actors themselves. We started the interview process in 2013 when local planning talks were underway and our data collection period lasted three and a half years. Following up on our units of analysis for several years was important for investigating where disputes came from, and how they were resolved in the end. During this period, we also invited four interviewees to talk to Masters students at the University of Manchester. Following their presentations, respondents were invited to stay for lunch. Extending these invitations created opportunities to meet with respondents in a more informal setting, and to take detailed notes of both their presentations and of the conversations at lunchtime. We stopped collecting data in early 2017 when we reached theoretical saturation (Glaser and Strauss 1967).

We triangulated interview findings with secondary data to improve the accuracy of our findings and conceptual insights. Building on Yin’s (1994) recommendations, we examined archival documents both internal and external to the pluralistic organization formed to develop HS2. Internal documents included meeting minutes, cost-benefit analyses, feasibility studies, project reports, design documents, budgets, timescales, and other documents uploaded online. We were also given some documents by interviewees during interviews and found much more by spending time on the websites of various organizations including HS2 Ltd., the Department for Transport, local governments and

53 public agencies. Across our units of analysis, we examined 83 local and strategy planning documents and 38 other documents prepared by local authorities. We also examined 30 documents prepared by HS2 and 30 parliamentary documents. Over and above these locally focused documents, we examined route-wide documents primarily presented by the promoter. Other relevant internal documents included HS2 Ltd.’s responses to requests for data made under the Freedom of Information Act, and also the documents submitted by multiple public institutions during parliamentary proceedings. Many documents relating to

HS2 are publically available because the planning process is highly regulated.

We triangulated internal documents produced by the project promoter and other resource-rich actors with external reports produced by third parties and official watchdogs in the project context. These included reports published by the Major Projects Authority, the National Audit Office and the Public Accounts Committee. We also triangulated the project organization’s documents with news articles encountered in the specialized press

(New Civil Engineering, Construction News), the national press (Financial Times,

Guardian), and with media interviews given by elected leaders and HS2 Ltd. top officials.

Documentation was important for the research as it was stable and could be reviewed repeatedly in an unobtrusive fashion. It also gave us access to highly specific information such as exact names, references and details of events while simultaneously giving us broad coverage of how the events were unfolding in each city. Archival documents were especially useful for precise mapping and quantitative data (Yin, 1994).

Using multiple methods for data collection was useful for triangulation, complementarity and expanding the data (Yin, 1994; Maxwell, 1996). With regards to triangulation, in particular, Yin (1994) suggests that a very important advantage when using multiple sources of evidence is the development of converging lines of inquiry. We, therefore, played interview data against data from archival documents and developed a convergent pool of evidence with the aim of strengthening the construct validity and

54 reliability of our conceptual findings (Yin, 1994; Maxwell, 1996). We also collected both real-time and retrospective data, which is a useful practice for mitigating retrospective bias and developing rich and reliable theory (Langley, 1999).

3.4. Data Analysis

To guide the data collection and analysis process, we made sure to condense and analyse the data as each wave of interviews was transcribed. We did this by coding the data or using other forms of iterative reflection (Miles and Huberman, 1994). One such form of iterative reflection, which we used to organize our initial data set, was a detailed chronological and factual account of the planning effort (Msulwa and Gil, 2014 a,b). This account was presented in a teaching case format and is included in Appendix C. The emerging preliminary constructs and frames were helpful for focusing data collection, and they provided a starting point for theory development (Eisenhardt, 1989; Miles and

Huberman, 1984).

A crucial aspect of our analysis was coding the data in line with concepts from both organization design and collective action literature, as well as emergent concepts from our empirical observations. We coded transcripts manually, by putting labels in the transcription margins. During this process, we flagged codes dealing with similar topics and further clustered these into categories (Saldana, 2013; Locke, 2007). We then used this categorization to populate multiple excel worksheets, i.e. one for each unit of analysis, and one for the project as a whole. Presenting the data in this way allowed us to evaluate each category both within and across embedded units of analysis. This exercise, in turn, was also helpful to determine whether categories should be put in the foreground or background at different levels of analysis (Agar, 1980; Miles and Huberman, 1994). The process was carried out iteratively and, with each iteration, we weighed the appropriateness of codes and categories and revised them as necessary (Miles and

Huberman, 1994; Strauss and Corbin, 1998; Locke, 2007). The purpose of this exercise

55 was to ensure that the data could be organized and related back to our core research question: How do we organize collective development in pluralistic settings? (Locke,

2007).

From Ostrom’s research, we became sensitized to the importance of examining interactions at multiple levels of governance (Ostrom et al., 1994). Thus, as the analysis unfolded, we drew on Ostrom’s (2005) Institutional Analysis and Development (IAD) framework, which is useful as it outlines a systemic approach for analyzing institutions that govern the fluid and dynamic actions and outcomes within collective action settings.

The IAD framework is also useful as it allows for the detailed analysis of specific resources and situations while being general enough to apply to multiple types of inquiries

(Oakerson 1993; Hess and Ostrom, 2005).

We also followed recommendations to produce tabular displays and diagrams to reveal patterns (Miles & Huberman 1994; Langley, 1999). Our focus was on the inter- organizational groups formed in cities along the proposed route, and specifically, how emergent rules and structures influenced the design decision-making process in each city.

As we iterated between data analysis and theory development an idea started to emerge that pre-existing governance structures at the local level directly influenced the decision- making process. After multiple iterations, the juxtaposition of private and collective interests also became clearer. Ideas, therefore, emerged around the interplay between pre- existing governance structures, newly created structures, an d the array of interests within the core during the collective development process. Hence, we explored the relationship between structures and outcomes and were interested to see how these were influenced by either homogenous or heterogeneous interests.

To develop theory, we followed recommendations from qualitative scholars.

Eisenhardt & Graebner (2007), for example, suggest a recursive cycling between the case data, emerging theory, and extant literature. Likewise, Yin (2013) recommends sifting

56 collected data to build inferences about what has happened, why and in what circumstances. Hence, we iteratively played empirical data against the extant theory to further our understanding of the questions central to this research. These iterations helped to verify that the constructs developed fit with the evidence, and they enabled the research questions to evolve over the research and writing period.

3.5. Summary

In this chapter, we outlined the abductive case-based research strategy that we adopted to explore the notion of organizing collective development in pluralistic settings.

Abduction involves a systematic combination of theoretical insights and empirical findings

(Klag and Langley, 2013; Van de Ven, 2007). Thus, we benefitted from the dual impact of both inductive and deductive research approaches by adopting this approach (Klag and

Langley, 2013). Additionally, a case-based research strategy is useful for supporting inquiries into areas where little previous research exists (Eisenhardt, 1989). Our research strategy is, therefore, suitable because while organization design research has started to extend into new forms of organizing; we still know very little about how to govern pluralistic settings from an organization design perspective (Gulati, Puranam, Tushamn,

2012; Puranam, Alexy & Reitzig, 2014). To that end, we present a single case study on planning High Speed 2 (HS2), a cross-country railway project unfolding in the UK.

Moreover, we trace the collective development process in five units of analysis embedded in the case. Structuring the research in this way is beneficial because it adds a significant opportunity for extensive analysis and enhances the insights from a single case (Yin, 2013).

In the chapter that follows, we outline the empirical setting within which we undertake our research following which we present our empirical analysis.

57

4. EMPIRICAL SETTING 4.0. Introduction

In this chapter, we provide an overview of the empirical setting within which we develop the HS2 case study. HS2 is a new multi-billion-pound cross-country rail project in the UK. The scheme is promoted by the UK Government or, more specifically, by a limited coalition of actors including HS2 Ltd., Network Rail and the Department for

Transport (DfT). Of these actors, HS2 Ltd. is responsible for the working-level integration, which is required for the DfT to be able to deposit a Hybrid Bill in Parliament. Getting a

Bill passed is a significant hurdle because the Government will only receive authorization to construct and subsequently operate the rail line once this is achieved. To that end, HS2

Ltd created a so-called meta-organization to advance the system-level goal of developing the design for a single indivisible functional asset (Lundrigan et al., 2015). In doing so,

HS2 Ltd collaborated with local authorities and had to integrate the effort of a diverse set of actors across multiple organizational boundaries. Below we provide a brief overview of the empirical setting before we proceed with our investigation into the mechanics of organizing collective development in pluralistic settings.

4.1. Context Overview

In January 2009, the UK Government incorporated HS2 Ltd. to serve as an arm’s length delivery body for the Department for Transport (DfT). Upon incorporation, HS2 Ltd. was tasked with i) examining the case for a high-speed rail service in the UK, ii) exploring route-wide and local alternatives by developing route proposals for a rail line linking

London (in the south) to the West Midlands (in the north), and iii) examining the prospect of eventually extending the line further afield to Scotland. Therefore, HS2 Ltd. looked at different configurations for joining up the various cities in the UK, and considered various route options, before shortlisting the three route options presented in Appendix D.

58

Route options were assessed against various technical, financial and environmental criteria including cost estimates, journey time savings and possible service patterns.

Options were also assessed against broader demand and economic analysis. The view was to develop a high-capacity rail system linking and serving the high capacity population centres across the UK. The choice of station locations was market driven, and internal assessments suggested that smaller markets would not fare well with a high-cost high capacity system. Thus, the UK’s main markets would be served.

The ultimate decision-making power in HS2 Ltd. rested with Cabinet Ministers

(‘Ministers’) notably the Prime Minister, the Secretary of State for Transport, and the

Chancellor. Hence, in 2012, HS2 Ltd. presented its proposals to the Government for consideration. The proposals were subjected to a wide-ranging consultation, and then subsequently endorsed. In line with HS2 Ltd.’s proposal, the Government’s preferred route option was a Y-shaped network with stations in London, Birmingham, Leeds, Manchester,

Sheffield and the Midlands as presented in Appendix E.

Before the project could go ahead, HS2 would have to be legislated by a Hybrid Bill.

This legislation is required because, in the UK, the Government cannot give HS2 Ltd. the authority to purchase the land necessary to build large-scale infrastructure compulsorily.

Instead, this power rests with Parliament. To that end, the Hybrid Bill is an Act of

Parliament that gives individuals, organizations or groups especially affected by the proposed rail line an opportunity to petition against the scheme. In turn, this provision adds significant time to the parliamentary process. Once the Bill is passed, however, it acts as a formal contract, and all resources (capital, planning power, political support) are effectively secured to forge ahead with the project. The Bill also outlines the scope of the project in a design concept, and whilst not part of the enacted legislation, it is presented with a budget and a timescale associated with this scope.

59

Prior to Hybrid Bill proceedings, the laws in the environment obliged the

Government to formally consult over 4,000 stakeholders that would be materially impacted by HS2 plans. The promoter also had to show what alternatives had been considered before committing to a final plan. Thus, upon endorsing the scheme, the Government tasked HS2

Ltd. with managing the planning effort and translating its proposal into a more tangible strategy that would enable the DfT to deposit a Hybrid Bill in Parliament towards the end of 2013. As per legislative requirements, HS2 Ltd. would, therefore, have to i) deliver a safe and affordable route design, ii) assess the environmental impacts of the design, iii) produce an environmental statement with propositions for dealing with the impact of the project on the environment and iv) consult with relevant bodies and parties on the environmental statement and other aspects of the proposals. A significant portion of this was the process of nailing down a design concept for each station along the route.

Although HS2 Ltd. was legislatively required to consult with materially affected actors, the agent went beyond legislative requirements by creating working groups with key local actors in the respective cities along the route. By engaging local actors in this way, the promoter aimed to minimize the time Select Committees 1 spent settling unresolved disputes in Parliament between the promoter and those organizational actors materially affected by the scheme. The Hybrid Bill would provide powers for the construction and operation of HS2. However, certain matters and details, such as plans, specifications, site restoration schemes and construction arrangements, would be reserved for subsequent local planning authority approval. Thus, at a high level, it was also in HS2

Ltd.’s interests to engage with local authorities extensively. Reaching agreements early on would avoid unnecessary delays in obtaining the necessary approvals, which could further delay the construction process.

1 A committee made up of a few Members of Parliament who are appointed to examine the work of government departments, which in this case includes the Hybrid Bill deposited by the DfT.

60

In addition to facilitating the design decision-making process, HS2 Ltd. also had to gather and distribute information relating to amendments, modifications and mitigation measures to be adopted with regards to the rail line. By failing to do so, the promoter would risk missing its cost and schedule targets in part due to a protracted parliamentary process. Therefore, HS2 core actors undertook a process of collective development, which allowed them to share their preferences for a local station and helped them to figure out mutually consensual design solutions. Ideally, such design solutions would include i) a station location and ii) a station design and configuration including route alignments, route passes and cost contributions where applicable.

Between 2009 and 2017, the physical element of the system-level goal of the scheme remained stable - to develop a new national railway network linking London to the

Northern regions. However, partly in response to emerging controversies, the rhetoric around promoting this system-level goal in the wider project environment evolved. These changes in rhetoric are endemic to pluralistic settings (Denis et al., 2011). However, they fall outside the scope of our research. Rather our focus is on the strategic choices made by inter-organizational groups in cities along the route in order to define the form of the future scheme. Appendix F provides a longitudinal chronology of the HS2 planning process and summarizes major events between 2009 and 2017. Over the 7-year period, the HS2 scope, estimated cost forecast and corresponding budget (cost forecast plus contingency) evolved significantly. In part, these changes are attributable to emerging strategic design choices undertaken during the collective development process.

4.2. Summary

Up until this point, we have argued for the integration of two research streams that are previously unrelated: organization design and collective action. This integration gives us a cognitive lens through which we can interrogate empirical evidence about organizing collective development during the HS2 planning process. In the previous chapter, we

61 detailed the abductive case-based research strategy that we used to conduct our empirical analysis. And in this chapter, we presented an overview of the HS2 empirical setting to provide a better understanding of the broader context within which the collective development process unfolded. In the next chapter, we present our analysis, which highlights the organizational rules and structures that emerged during the design decision- making process. We also show how these rules and structures influenced local strategic choices. Studying these choices and the governance structures guiding them is important because, as we will later argue, both technical and organizational dimensions have an impact on the collective development effort. By implication, both dimensions also have an impact on project performance indicators, including cost and schedule targets.

62

5. ANALYSIS 5.0. Introduction

In this chapter, we detail the process of collective development undertaken by the meta-organization formed to develop HS2, the new high-speed railway in the UK. HS2 is a single, very large phased programme. However, this development is pluralistic in that the promoter, a limited coalition of actors including HS2 Ltd., Network Rail and the DfT, needs to negotiate with each of the cities along the proposed rail route. To deepen our understanding of organizing collective development in pluralistic settings, we investigate how this process unfolds in five of these cities: London, Birmingham, Manchester, Leeds and Sheffield. To that end, we examine how local actors collaborate or not, with the aim of highlighting the mechanisms used by local groups to resolve the coordination and cooperation problems inherent in pluralistic settings.

5.1. Resolving Coordination Problems

From organization design scholars, we know that resolving coordination problems is important for the development of shared problem definitions and agreed-upon facts

(Srikanth & Puranam, 2014; Clark, 1996; Puranam et al., 2012). Indeed, during the HS2 planning process, coordination was carried out through the acquisition and dissemination of information both within and across multiple organizational boundaries (Tushman, 1978;

Tushman & Nadler, 1978, Galbraith, 1973). This information processing, in turn, reduced the level of uncertainty on the project by delineating a solution space within which decisions would be made.

Coordination efforts were also important because they gave organizational actors an opportunity to develop predictive knowledge of each other’s actions (Puranam et al., 2010) through ongoing communication, mutual observation, learning and joint decision-making

(Galbraith, 1973; March & Simon, 1958; Tushman & Nadler, 1978). By processing information in this way, inter-organizational groups were able to agree on multiple feasible

63 station options for the various cities along the route. Crucially, the resulting station shortlists provided a common basis for the inter-organizational negotiations that would take place during the ensuing cooperation process. During this process, the leaders representing organizations within the HS2 meta-organization core would negotiate on the parameters of the final stations.

5.1.1. Coordination Process Overview

The scale, geographical scope and complexity of HS2 necessitated coordination across multiple organizational boundaries. As such, coordination involved information sharing among actors in the project core. These actors included HS2 Ltd., Network Rail, the owner and operator of the UK rail network, the Department for Transport (DfT)), HM

Treasury, local government authorities and regional transport bodies. Given that no single actor possessed the capacity to deliver the project independently, the core’s collective capability to exchange and process information formed the basis of the planning process.

While the DfT maintained ownership and control over high-level decision-making, working-level integration was largely assigned to HS2 Ltd. In the role of a “systems integrator,” HS2 Ltd. was responsible for bringing the different project components together and making sure that they worked together in a synchronized fashion (Prencipe et al., 2005). This synchronization entailed developing a common understanding of the project constraints and communicating the broader implications of local decisions. The

Head of Route Engineering at HS2 Ltd. explains this below,

You’ve got to be able to manage [the groups of people you engage with], keep them focused, keep them directed, make sure they understand the objectives of what you are trying to deliver…[this is about] setting the assumptions in the early stages of the process and continually reviewing them and talking through them and making sure people understand that…they are moving for a reason and everyone understands what that reason is. [Head of Route Engineering, HS2 Ltd]

64

The starting point for coordination across the full range of actors was the Y-shaped route proposed by HS2 Ltd. in 2012 and subsequently endorsed by the Government in

2013. In deciding on the route, HS2 Ltd. did much of the work internally and only engaged with local and regional authorities on a limited basis. The rationale for this was that engaging more extensively on the route, and publishing multiple route options, posed a significant land blight risk. Indeed, the promoter on a previous project, , published multiple routes resulting in many people being unable to sell their homes. Thus, the primary criteria used to select the route were developed by internal working groups within HS2 Ltd. As explained below, this criteria revolved around the demand forecast for the high-capacity rail system, which was envisioned to link and serve Britain’s main markets and centres of population,

The entry point was: where is the demand? Well, the demand is the biggest city in Europe, London; second biggest in the country, Birmingham; Britain’s third city is Manchester, and the fourth city is Leeds. It so happens that Liverpool is about to be sixth and happens to be very close to Manchester. East Midlands is about ninth, and South Yorkshire is about tenth…so you are connecting up to eight of the ten top regions of the country [Technical Director, HS2 Ltd]

The Y-shaped route identified by HS2 Ltd. formed the core component of the scheme’s design architecture (Baldwin, 2015; Tushman & Murmann, 1998; Tushman &

Rosenkopf, 1992). This pre-specified route brought a sense of order to the planning process by constraining the potential options for subsequent decisions. It also contributed to the development of reciprocal predictability (Puranam et al. 2012) among the array of project participants during the design decision-making process. Specifically, core actors could predict that strategic choices would be made within, and in relation to, this predefined solution space. To that end, and with land blight restricted to a single corridor,

HS2 Ltd. began to work extensively with local authorities to identify and select station locations.

65

Local authorities worked with HS2 Ltd. as so-called ‘delivery partners’ helping to refine the route and ensure effective and efficient connectivity between the proposed rail network and existing local services. Crucially, the local authorities also contributed to narrowing down a long list of potential station options into a shorter list of more feasible options. Through this ‘sifting process’, HS2 Ltd. together with local authorities narrowed down long lists of as many as 30 options to short lists of between two and four viable options for each chosen city. The shortlists for each station are listed in Table 5.1.

In discussing how the process was initiated in Manchester specifically, and across the cases more generally, the Head of Route Engineering at HS2 Ltd. explained that the company, “gave [local actors] a map…and some magnetic rectangle blocks the size of a station, and said ‘right, where do you want to put the station?’…[and] very quickly, things started to drop off.” These maps and the station shortlists (once developed) therefore served as boundary objects (Carlile, 2002, 2004) providing common ground (Srikanth and

Puranam, 2014, 2011) that would inform further investigation and deliberation by both

HS2 Ltd. and local leaders.

From a technical perspective, it is fair to say that the promoter kept unilateral control over the system architecture but shared decision rights with the local authorities for their respective stations. The promoter did, however, restrict those shared decision rights to local planning choices. From an organizational perspective, the promoter created a meta- organization by including local authorities in the project organization. Furthermore, by sharing decision rights over local strategic choices, the promoter made the local authorities part of the meta-organization core (Lundrigan et al., 2015). Making local authorities part of the core went beyond the legislative requirement to consult them. Thus, in effect, at the cost of some decision-making autonomy, the promoter reduced uncertainty on the project and increased the likelihood of achieving a collectively beneficial outcome.

66

Over multiple stages, core actors used a set of principles and procedures, initially developed by HS2 Ltd, to narrow down long lists of potential stations. The principles included maximizing the economic benefits from HS2, increasing capacity on the existing rail network (and especially on overcrowded services), integrating with other transport networks, and designing a sustainable railway. Based on these principles, design criteria were further developed to guide option development. Criteria included cost, construction timescales, and operational feasibility of the stations along with economic and environmental appraisals. The criteria became progressively more detailed in each subsequent stage of option development, and in response to feedback from inter- organizational discussions.

The key integrative devices used for processing information were inter- organizational technical working groups in each city, and strategic documents produced by these local working groups. Inter-organizational technical working groups were important as they enabled face-to-face interaction throughout the coordination effort. This direct communication allowed for the establishment of common ground (Srikanth & Puranam,

2011, 2014), and a shared problem representation in the form of station shortlists. In turn, these shortlists were fundamental for the final decision-making undertaken by the leaders of the various organizations in the core. An HS2 official explains this importance below,

When you look at stations, you can’t look at them just as a separate piece of work...you also have to take very much into cognizance the effect of growth within the city and consider how the city partners, the city Chamber of Commerce, and all the people who make the city what it is… want the city to grow and to work and to develop in the future…Projects like these are fundamentally projects that can trigger a significant change in a city and how a city is growing, and it can actually change in some cases the direction, which that city is taking itself [Head of Route Engineering, HS2 Ltd]

As a complement to the inter-organizational technical working groups, strategic documents were published to support the decision-making process. Documents captured fine-grained and localized information, which was used as evidence to support, oppose or defend both decisions and proposals. As a statutory requirement, documents were also

67 made publically available for comment, giving all actors in the project environment access to the published information. Although both strategic documents and working groups contributed to coordinating core actors’ needs and interests, the former was seen to support the latter. Indeed, some work in organization decision-making suggests that the production of meanings through face-to-face interactions is better for ambiguity reduction than the imposition of rational criteria or the distancing involved in setting up number systems

(Denis et al., 2006; Thomas & Trevino, 1993). Below an HS2 Ltd. official details the relative importance of direct and indirect forms of communication in the project context,

There are fundamentals to it, emails and letters and paper reports are all good as backing evidence, but ultimately what you need is a group of people around a table, and it needs to be a managed group… where everyone accepts everyone else’s aims. So HS2 has its aim to do something and deliver something, and within that HS2 has its realm of assumptions that it’s doing it against and everyone knows why it’s making certain decisions. [Head of Route Engineering, HS2 Ltd]

Coordination during the planning process was important for developing a common problem definition across organizational boundaries (Srikanth & Puranam, 2014). By implication, it was also an important antecedent for collective problem-solving activities undertaken by core actors at the leadership level. Through information processing, actors were able to make their tacit knowledge explicit and transfer complex knowledge and information. Furthermore, information flows made it possible to publically present a unified frame of the options and trade-offs despite conflicting sub-goals and contentious interdependence (Eriksson et al., 2017; Levitt, 2015).

Crucially, the process of coordination gave HS2 Ltd. access to local information, which made it possible to adapt the station designs. The process of coordination also gave local authorities an opportunity to express their interests and priorities for the scheme and thereby directly influence the final design choices. Sharing and accessing fine-grained local information also made it possible to adapt the collective organizational rules and structures accordingly (Ostrom, 1990). Structuring the process in this way reduced the

68 scope for discretionary decision making thus guarding core actors against ad-hoc responses to emerging problems. In turn, it reduced the level of uncertainty on the project, which could otherwise complicate the planning process. Having outlined the coordination process, we now turn to discuss inter-organizational working groups and strategic documents, and how these integrative devices were used to facilitate coordination across organizational boundaries in the pluralistic meta-organization core.

69

Table 5.1 Overview of the Coordination Process Across Organizational Boundaries

Unit Integrative Coordination Devices Sifting Process Outcomes Illustrative quote

Inter-organizational technical working group: Information processing takes from 15-20 options 1 Local Council, HS2 Ltd., Regional Development (Long List) down to 8 options (Intermediate List) Extending New Street would offer the best solution for passengers, Agency, Public Transport Delivery Agency, Regional and then to 4 options (Short List) and then trade-off but unfortunately, it is not feasible.” [HS2 Ltd] Government Office, Highwaysemerges Agency, Network Rail Because of its orientation, Moor Street is not compatible with a route Documents Short List: via the Water Orton corridor. [HS2 Ltd] Local: The Big City Plan, City Centre Masterplan, City Centre Transport Strategy, Birmingham Core Strategy 1. Curzon Street “It was concluded that a new station would be needed as all the other

Birmingham Draft Promoter: Birmingham HS2 masterplan Regional 2. Moor Street options had significant engineering, capacity and/or land use Rail Development plan The Big City Plan, 2010; 3. New Street constraints. Curzon Street was selected as the best location as it had Birmingham Development Plan; Birmingham City 4. Warwick Wharf fewer local impacts.” [HS2 Ltd] Council Response to Phase 2; Centro Response to Phase 2 consultation Inter-organizational technical working group: Information processing takes from 27 options (Long List) In construction terms, the King’s Cross site presents several 1 Local Council (included after station selection), HS2 down to 7 options (Intermediate List) to 2 options (Short difficulties. The location would not easily accommodate a station Ltd., Crossrail, Network Rail, London Underground Ltd., List) before settling on Euston box… The Thameslink tunnels at the northern edge of the

Greater London Authority, Transport for London site…would force the alignment deeper below the ground and

Short List: preclude a station on the surface. The Regent’s Canal would require Documents: either to be permanently diverted…and the sewer would also require

London 2014, Euston area plan (local actors’ initiative) 1. Euston Station diversion.[HS2 Ltd] 2012-15, 5 versions of Euston HS2 strategic plans 2. King’s Cross (promoter) The Camden Plan 2012-2017; Camden’s After examining 27 potential locations in 2009, HS2 Ltd. concluded Transport Strategy 2011; Mayors Economic that Euston is the only viable option for the London terminus. [HS2 Development Strategy; Euston Area Plan, 2014 Ltd] Inter-organizational technical working group: Leeds New Lane would benefit from a faster overall journey time, and Information processing takes from 30 (long list) to 8 1 City Council; HS2 Ltd; Metro (West Yorkshire as a result of the lower-cost approach would be less expensive (Intermediate list) to 3 (short list)(final) before trade Passenger Transport Executive); Network Rail; overall. The station would be well located to support and enhance the offs emerge.

Highways Agency; Government Office for Yorkshire and city’s future development to the south.[HS2 Ltd]

Humber*; Yorkshire Forward* Short List: Whilst the Leeds station North site provided the easiest possible Leeds Documents interchange with services at the existing station and would also be 1. Leeds New Lane 2015 Rethinking Leeds report; Leeds masterplan; Leeds closer to the existing city centre, the site was constrained and would 2. Leeds Station North: transport for West Yorkshire; Local Transport Plan (Rail have involved a longer connection on to the main line that would 3. Sovereign Street South: Plan 7) 201 generate highest impacts on local communities [Leeds City Council].

70

Unit Integrative Coordination Devices Sifting Process Outcomes Illustrative Quotes

Inter-organizational technical working group: Information processing takes from 33 (long list) to 15 Salford Central Middlewood station would be located to the west of 3 City Councils (2 in all meetings and 1 only in certain (Intermediate list) and then to 3 (shortlist) before trade- the existing Salford Central station. The site benefits from good meetings); HS2 Ltd; Transport for Greater Manchester; offs emerge. connections to major highways but suffers from poor connections to Highways Agency, Network Rail, Government Office for existing Metrolink services[HS2 Ltd] North West*, North West Development Agency* Short List: The Piccadilly proposition is more expensive than the Salford

Documents 1. Manchester Piccadilly alternatives, but it has far more benefits…So it’s actually about value Local: Climate Change Call to Action; Manchester City 2. Salford Central Middle Wood for money, not just cost. The challenge was to get them [local actors] Plan; Salford City Plan; Transport and Economic Model; 3. Salford Combined Station to come up with some answers [HS2 official] ‘Manchester Family‘ Business Plan, TfGM Transport Strategy, 2014; Greater Manchester’s third Local Plan We [locals] didn’t have a preconceived notion as to what the answer Manchester 2011/12 – 2015/16; Greater Manchester Strategy – might be… It had to follow that consensus around being able to Better Together, 2013; Greater Manchester Growth and demonstrate to ourselves the economic value [Rail Programme Reform Plan, 2014; HS2 Consultation Response, 2014; Director, TfGM] HS2 Manchester Piccadilly Strategic Regeneration Framework, 2013; GMCA Response to Consultation, We [local transport body] did quite a lot of modelling work… we were 2014; Manchester City Council Report for Resolution, seeking to influence the outcome …..I also think it helped to bring 2014 Salford on our side [Rail Programme Director, TfGM]

Inter-organizational technical working group: Information processing takes from 30 (long list) to 7 A new station at Sheffield Meadowhall would be situated alongside 2 City Councils; South Yorkshire Passenger Transport (intermediate list) to 2 (short list) before trade-off the M1, providing convenient access by road serving the South Executive; Highways Agency; Network Rail; Government emerges. Yorkshire region. The station is well placed for support growth in Office for Yorkshire and Humber*; Yorkshire Forward* Sheffield and already has excellent public transport networks.[HS2 Short List: Ltd] Documents: Sheffield City Region Transport Strategy 2011-2016; Sheffield City Council: An Assessment of 1. Victoria Station Sheffield is in grave danger of committing economic suicide…were the Case for a city centre high speed rail station in the 2. Meadowhall Station we to stand idly by while London decides on our behalf to develop

Sheffield Sheffield City Region; Sheffield City Council Response this city’s biggest ever economic opportunity four miles away from the to High Speed Rail Consultation, 2014; Rotherham heart of the action [Sheffield City Council] Response to Phase 2; Sheffield City Centre Masterplan, 2013; Sheffield City Centre Masterplan, 2013; Rotherham Growth Plan 2015 – 2025; The Economic Case for High Speed 2: Leeds City Region and Sheffield,

71

5.1.2. Integrative Devices: Coordination

5.1.2.1. Inter-Organizational Technical Working Groups

Face-to-face interaction formed the bedrock of coordination during the planning process as inter-organizational actors looked to agree on the location and form of local

HS2 stations. Initially, core actors came together in inter-organizational technical working groups and were tasked with identifying potential station options and the trade-offs associated with each of these. They also examined the assumptions, data and analysis presented by the promoter. Because discussions were primarily technical, the groups primarily consisted of “the sorts of people who knew what the transport system was doing,” as the Head of Route Engineering at HS2 Ltd. explained.

In a bid to avoid unnecessary land blight, HS2 Ltd. was selective about who joined the technical working groups. The same official, therefore, explained that “you have to be very careful whom you bring into this little room to help us guide and understand how the city works.” Preference for group membership was, therefore, given to senior planners and heads of transport strategy. Participants representing local leaders were also invited to join the working groups so that the leaders were kept informed and their views were represented as the process of station-shortlisting unfolded.

As shown in Figure 5.1, distinct technical working groups were created for each city represented on the route, and groups included a diverse array of actors. The project promoter (at the centre of the figure) comprised a coalition of actors included HS2 Ltd.,

DfT and Network Rail, and was represented in all of the groups. Other actors in the respective working groups included local councils, regional development agencies, regional transport agencies and regional highways agencies.

As the figure shows, in all of the groups, except London, local councils served as the primary point of interaction for the promoter. Interactions were either with a single council,

72 as was the case in Birmingham and Leeds; or with multiple councils, independently as in

Sheffield, or as a unit as in Manchester. In London, however, the primary point of interaction was with the Greater London Authority and Transport for London, the regional development and transport agencies, respectively. Similarly to the interactions with the councils in Manchester, the promoter interacted with these bodies as a unit. As will be discussed in greater detail ahead, the exclusion of Camden Council in London during the initial stages of the decision-making process gave rise to tensions within the local working group.

Figure 5.1 The Composition of Inter-Organizational Technical Working Groups

The solid lines in Figure 5.1 represent direct communication lines shared by core actors within the technical working groups while the dotted lines represent indirect communication lines. Core actors primarily engaged in direct communication, during face- to-face meetings, for example. However, they also communicated indirectly through published strategic documents, discussions with the promoter in one city that were later shared in another, or media reports. Ultimately, both forms of communication were

73 important for creating common ground (Srikanth & Puranam, 2014) and for the development of predictive knowledge (Puranam et al., 2012).

The number of local councils in each technical working group provided the greatest source of size variation across the groups. This varied from a single council in London,

Birmingham and Leeds, to two councils in Sheffield and three councils in Manchester.

Generally, limiting the size of the technical working groups reduced the amount of information that would otherwise have to be processed upfront. Conversely, the higher the number of local councils in the technical working groups, the greater potential there was for information processing to become complicated. Nevertheless, on occasions when sensitive information was shared, it was done on a confidential basis to minimize the risk of it being shared publically. On such occasions, HS2 Ltd. asked the actors in all groups to sign confidentiality agreements and prohibited them from taking documents away from the meetings. Next, we consider the process of coordination in the single and multiple council groups, respectively, in order to highlight further similarities and differences.

Single-Council Technical Working Groups

Information sharing was relatively easy in Birmingham and Leeds, where each technical working group comprised a single local council. Through a shortlisting process in Birmingham, presented in Appendix G1, the local technical working group developed a shortlist consisting of three stations all within an arc of a central point: New Street, Moor

Street and Curzon Street (Appendix G2). The proximity of the three sites was seen as a key advantage. However, HS2 Ltd. concluded that a station at Curzon Street (Appendix G3) was the best choice.

According to HS2 Ltd., Curzon Street was ideally located next to Birmingham’s economic centre, and its construction would be less disruptive than the other two options.

Furthermore, the site identified for the station was a wasteland that would have a significantly lower environmental impact than the other options. Nevertheless, this 74 proposition clashed with a masterplan that had previously been agreed to for the area, and the proposed site was committed for private development. Final decision-making was, therefore, deferred to the local inter-organizational strategic board, a point we return to when we discuss the problem of cooperation.

The shortlisting process was similarly straight-forward in the case of Leeds

(Appendix H1) where the technical working group arrived at a shortlist of two station options: a Leeds City Centre Station and one at New Lane, located to the South of the city centre (Appendix H2). While HS2 Ltd. preferred the station at New Lane (Appendix H3,

H4), local actors were unified in their preference for a station in the City Centre (Appendix

H5, H6). As these differences emerged, it was important for HS2 Ltd. that, ahead of selecting a final station location, local leaders were kept informed that there were two options to be considered. As the HS2 Ltd. CEO stated in the press, “HS2 Ltd., Network

Rail and the Council need to continue working together in detail…taking into account the need to ensure the best possible synergy with local services.4”

The technical working group in London also comprised a single local council.

However, in contrast to both Leeds and Birmingham, the Council only joined the group once a station location was identified. Up until this point, local preferences were represented by the GLA and TfL. According to HS2 Ltd. officials, the Council was excluded from the short-listing process because all 32 London local authorities were initially considered as potential station locations. A longlist of 27 station options

(Appendix I1) was subsequently developed before being narrowed down to a station at

Euston (Appendix I2, I3). Foreseeing the challenges associated with having to process information from such a large number of independent actors, HS2 Ltd. opted to exclude all local authorities from the decision-making process instead.

4 Source: High Speed Two Limited. (2014). Rebalancing Britain: From HS2 Towards a National Transport Strategy. London: High Speed Two Limited.

75

By choosing to exclude the local councils from the initial shortlisting process in

London, HS2 Ltd. reduced the level of uncertainty and interdependence among actors within the meta-organization core. Restricting decision-making participation also reduced the level of interdependence between the actors in the meta-organization core and those in the broader project environment. In other words, if all of the long-listed local councils were included in the decision-making process, there would be greater potential for speculation about the final location of the London terminus. In turn, the uncertainty across the city would trigger unnecessary land blight and potentially widespread opposition in the project environment. Ultimately, such an outcome could undermine the collective effort and threaten the project organization’s survival. An HS2 Ltd. official explains this below,

It’s a challenge, and it’s a physical management challenge that says once you get a group of people in a room, how do you manage them, how do you control their outputs and how do you ensure that they come to the table with an open mind rather than a closed mind…that’s the most important thing [Head of Route Engineering, HS2 Ltd]

By deliberately excluding the Council during the short-listing process, the promoter circumvented the coordination problem, at least temporally. However, the promoter reduced the level of uncertainty within the core at the expense of a larger pool of information. Although much effort was made to engage the Council later on, their early exclusion negatively impacted the ensuing consensus-building efforts. While we discuss cooperation in a subsequent section, it is fitting to say here that the Council was combative towards HS2 Ltd. Their combative nature was partly attributed to the fact that they were principally against the scheme. However, it was also seen as a tactic to extract more concessions. An HS2 official explains the challenges associated with Camden Council’s combative nature below,

There’s been more consultation at Euston than you can dream of, and it’s been very hostile…very hostile. It’s a wonder my staff tolerate it at times; it’s been abusive. You’ve got politics driving it, you’ve got people with genuine concerns that whip up the ante, and then consultation becomes very difficult. Ultimately there’s some genuinely opposed; and there’s some trying to capitalise [Chairman, HS2 Ltd]

76

The Birmingham and Leeds cases show that, in the absence of antagonistic behaviour, information processing in smaller groups was relatively straightforward, even though actors exhibited differing sub-goals or interests. Contrastingly, the London case shows that when actors are antagonistic, smaller groups can have high information processing requirements. Such behaviour, on the part of Camden Council, complicated the process of coordination, and made it challenging for other actors in the HS2 meta- organization core to acquire and disseminate information.

Multiple-Council Technical Working Groups

Across the cases, the technical working groups comprising multiple local councils had to acquire, collate and disseminate a greater load of information. The greater load of information that these groups had to process added to the complexity of the process.

Furthermore, information processing was more complicated in those groups where the councils operated independently and were advancing competing interests as compared to groups where the councils acted in unison. As we will show, Sheffield offers an example of how such fragmentation at the local level complicated matters because it meant that

HS2 Ltd. had to engage with the competing councils independently in attempts to create common ground. This pattern of interaction contrasts the Manchester case where councils were engaged as a unit.

The local technical working group in Sheffield shortlisted two stations (Appendix J1), which offered very different spatial opportunities in the region: one at Victoria, a city centre location (Appendix J2 – J4), and one at Meadowhall, a regeneration project and the site of a large shopping mall which opened in 1990, and had since drawn business out of the city centre (Appendix J5 – J7). The process of information sharing in the region was complicated by the fact that there were competing groups of local authorities lobbying for a station. As a result, communication among local actors regarding HS2 ceased, and there was a breakdown in the coordination efforts. In explaining this further, the Executive 77

Director of Places in Sheffield said, “we meet once a month…and discuss other things actually. But when it comes to HS2, we’re still at loggerheads.” Coordination efforts also broke down because, in view of the deep-seated local rivalry, HS2 Ltd. opted not to engage with either group of councils. This particular breakdown in communication wasn’t well received by local actors as detailed in the quote below,

We are not meeting [with HS2] now. Just because we have different opinions, they haven’t spoken to us in months. The consultation period ended, and we heard nothing from them…they didn’t speak to us at all. There was no information coming back…because we have a difference of opinion [Director of Places, Sheffield City Council]

In sharp contrast to the Sheffield case, the Manchester case demonstrates how an alignment in interests attenuated group size related challenges. In this case, HS2 Ltd. did not have to create a shared local context because local authorities in the region had a history of working together. This history dates as far back as 1974 when the Greater

Manchester County Council (GMCC) was created to undertake some strategic functions alongside the ten districts in the region. Although the GMCC was later abolished in 1986, the districts maintained their relationship through the Association of Greater Manchester

Authorities (AGMA), a voluntary organization representing the ten district councils in

Greater Manchester. AGMA serves to have the councils work together and develop policies on issues affecting the region as a whole like transport, regeneration and attracting investment. In addition, AGMA lobbies the UK Government and others actors.

Seeing the opportunity for even greater collaboration AGMA advocated for the

Greater Manchester Combined Authority (GMCA), which was created in 2011 as a corporate legal body with powers of its own. Since 2011, the GMCA has reached a number of devolution agreements with the UK Government and continues to explore ways to bring more powers and responsibilities to Greater Manchester. The ten councils on the GMCA have also collaboratively built consensus around Greater Manchester’s priorities in a range of sectors such as transport and housing. Furthermore, the Combined Authority promotes

78 playing to a ’Manchester advantage’ and putting the collective regional good ahead of the attainment of local council benefits. Thus, although the HS2 technical working group comprised three local councils, information sharing and station short-listing were carried out within a unit. Contrastingly, information sharing in both London and Sheffield was carried out with technical working groups comprising multiple actors advancing different agendas. To that end, the Head of Route Engineering at HS2 Ltd. dubbed Manchester’s way of working “the epitome of how you should work as a city.”

The technical working group in Manchester arrived at a short list of three stations during the coordination process (Appendix L1): one station at Manchester Piccadilly, close to the centre of the region (Appendix L2-L5) and two stations in Salford (Appendix L6 –

L9). Discussing the three options the Director of Program Cost and Commercial

Arrangements at HS2 Ltd. explained that, “all three locations are acceptable, so they’re feasible, and the costs are credible though there are still cost swings between £50 and

£100 million.” From the onset, the Manchester authorities preferred a station at Piccadilly, the most important station in the city centre. Contrastingly, HS2 Ltd. and Salford Council preferred a Salford station. However, upon further information processing, Piccadilly emerged the economically superior option, and HS2 Ltd. conceded. Furthermore, studies showed that a station in Salford would require a substantial demolition of public housing, which divided Salford Council. Through additional dialogue, and demonstrating a willingness to compromise, the local actors agreed on a station at Piccadilly. Thus, despite initial differences, the actors in the Manchester working group avoided an impasse.

Arguably, information sharing and shortlisting were relatively easier in Manchester because the three councils undertook systematic evidence-based consensus-building rather than political infighting as displayed in Sheffield. This approach is detailed below,

At a very, very early stage, we agreed that our objective was to maximize the economic benefits of the site, and that was agreed at a Greater Manchester level…We didn’t have a preconceived notion as to what the answer might be…It had to follow that consensus

79 around being able to demonstrate to ourselves that it had that economic value. And I think that’s really important [Rail Programme Director, Transport for Greater Manchester]

Through the creation of inter-organizational working groups, the promoter addressed both a technical and organizational problem. From a technical perspective, working groups provided actors with an opportunity to short-list technically feasible design options, and to ensure the integration of potential stations with existing proximate transport networks.

Groups were also able to establish the potential environmental impact and broader growth and regeneration implications of the proposed station options. From an organizational perspective, the technical working groups allowed actors to develop both common ground

(Srikanth & Puranam, 2014) and predictive knowledge (Puranam et al., 2012).

As our findings show, depending on the composition of the technical working groups, creating common ground took more effort in some cities than in others. Broadly, groups with more heterogeneous interests and goals required greater information processing.

These groups also required additional coordination devices to reduce the uncertainty among the interdependent actors. To that end, both within and beyond technical working groups, actors used strategic documents as an integrative device. We discuss these next.

5.1.2.2. Strategic Documents

During the HS2 planning process, and as organization scholars suggest, the codification of information and communication via documents was effective and efficient for coordination as the number of decision-making participants, and matters grew, even in the absence of diverging interests (Galbraith, 1977; Stinchcombe, 1990). Documents were easily retrievable as both physical and digital documents and could, therefore, inform and support decision-making processes across organizational boundaries.

Both the promoter and local actors periodically published strategic documents. The promoter’s documents articulated the case for high-speed rail, the progress of the scheme, and the approaches taken to identify station options and engage with stakeholders. The 80 promoter also published technical specifications complete with visualizations of route passes and alignments, as well as station locations, footprints and configurations of emerging proposals. Concurrently, local authorities published a range of documents detailing their local strategy and planning aspirations as outlined in Table 5.1.

In 2013, the Government created an HS2 Growth Taskforce to offer advice on how to maximise the return on investment from HS2. One of the Taskforce’s recommendations was that local authorities publish growth strategy reports related to how high-speed rail would generate local jobs, growth and regeneration. For the promoter, these subsequently published documents were important for understanding city plans and local outlooks. As the Head of Route Engineering at HS2 Ltd. stated, ‘we need to…ensure we understand where their plans are taking us.’ The HS2 Growth Taskforce also recommended that the

Government support local authorities to deliver their HS2 growth strategies and realize the local regeneration potential associated with the scheme. As explained below, this was especially challenging for HS2 when local plans were insufficiently developed,

One of the things that is most challenging…is getting people to look, and actually take, that long-term set of objectives. People do, as it were, descend down into the detail all too quickly, and they fret about all manner of things about the here and now rather than actually thinking where they expect to be in 30 years’ time…we are just now starting to work with the Councils to articulate their development aspirations in more detail than they have to date [HS2 Ltd. Leeds, Manchester and Heathrow Project Director]

Over the course of the planning process, local authorities within the core also published a variety of documents detailing their positions on HS2 Ltd.’s proposals. In 2011, for example, Manchester published a local transport plan declaring its support for the scheme. The plan stated, “High-speed rail has the potential to significantly accelerate the economic development of the cities of the north, and Greater Manchester is committed to working with government, HS2 Ltd. and other stakeholders 5.” Contrastingly, Camden

Council published multiple documents stating that it was against the scheme. In one such

5 Source: Manchester City Council Executive (2016). Manchester City Council Report for Resolution. Manchester: Manchester City Council.

81 document, Camden Council declared, “The Council is opposed to HS2 in its existing proposed form because of its major adverse impacts on the borough of Camden6.” These documents were important as they gave the promoter an indication of local concerns and potential petitions that could arise during the Hybrid Bill proceedings in Parliament.

When local authorities felt that the promoter’s proposals did not align sufficiently with local plans or that the promoter’s analysis and forecasts were based on incomplete assumptions, they published documents detailing potential amendments. In Sheffield, for example, the city council published documents advocating a city centre station. In one such report, the Council stated that it “undertook a quantitative assessment of the relative economic impacts of the alternative HS2 stations in Sheffield… refining HS2 Ltd.’s own analysis of development and employment capacity 7 .” In their analysis, local officials argued that a station located in the city centre would generate £2-£5 billion additional growth over 25 years, over and above a station located at Meadowhall, an area on the periphery and HS2 Ltd.’s preferred option. Furthermore, the Council found that a city centre station would create 9500 net new jobs compared to 3000 at Meadowhall. A similar analysis was carried out in Manchester where local officials proposed their own procedure for capturing regeneration benefits. To that end, local authorities in Manchester stated the following,

Building upon the work undertaken by HS2 Ltd., the Council and local partners are undertaking more detailed work to evaluate the local economic impacts of HS2. Work continues to refine [the existing] analysis…and will be developed to provide the detailed analytical framework required to underpin business cases for station development, regeneration, planning, and crucially, funding and delivery. The intention is that this

6 Source: Camden Council (2014). HS2 Phase 2 Consultation, Comments of London Borough of Camden. London: Camden Council. 7 Sheffield City Council and South Yorkshire Passenger Transport Executive. (2012). Maximising the Economic Impact of HS2 Investment in Sheffeld. Sheffield: Sheffield City Council.

.

82 further work will be undertaken with local partners and Network Rail, HS2 and DfT. [Chief Executive, Manchester City Council8]

Given the potential impact of a station on the growth and development trajectory of each city, HS2 Ltd. welcomed local documents and analysis. However, propositions were not immediately incorporated into the scheme. Instead, local propositions were crosschecked against the promoter’s criteria. Cross-checking was necessary because local authorities’ visions of the future were based on a mixture of forecasts and conjectures, a point we return to in more detail later. Therefore, strategic documents were not an end in themselves but were rather a means by which to establish common ground (Srikanth &

Puranam, 2014). In turn, this shared knowledge would inform the high-level decision- making process undertaken by the organizational leaders within the core. Explaining the importance of documentation in this regard the Senior Manager of Planning Decisions at the Greater London Authority stated that “if we can…give them a plan saying ‘this is how we think it should be planned, and this is how it should set itself up spatially’, they can all understand that. They might not agree with it…but at least we can all understand where we’re trying to get to.”

Both the promoter and local actors continually updated their strategic documents over time. Updated documents reflected the assimilation of information and a refined basis from which mutually consensual solutions could be identified when high-level decision- making efforts ensued. Documents were therefore living and emergent artefacts that would morph into the most important planning documents: the project bill which would be presented to Parliament for regulatory approval, and the associated drawings formalizing the final planning choices. The HS2 bill would supersede all previously developed documents and once submitted it would be scrutinized, debated, and amended if

Parliament saw fit. The promoter was therefore set on carrying out robust information

8 Manchester City Council. (2014). Report on HS2 Consultation – A City Council Response. Manchester: Manchester City Council.

83 processing and developing the best evidence to make its case given this high level of scrutiny ahead. An HS2 Ltd. official explains this below,

Our job is to develop it, design it, manage it, process it, substantiate it, consult with communities at the same time and produce the necessary documentation for a hybrid bill…[because] until we have the hybrid bill and royal assent, we do not have a full project. [Phase 1 Programme Director, HS2 Ltd]

It is important to note here that, unlike typical inter-organizational coordination problems, the resolution of the coordination problem for HS2 was much more enmeshed with the environment. That is to say that agreements reached by local actors and the promoter still needed to be sanctioned by Parliament. Thus, given the inevitability of the parliamentary process, and cognizant of the time it could take, the promoter prioritized the issues that could significantly stall the proceedings rather than attempting to resolve every single problem ahead of time. In some ways, Parliament acted like an external referee (Gil and Pinto 2017). However, given that HS2 had cross-party political support, Parliament was not necessarily as impartial as a referee in the traditional sense.

Importantly, the exchange of strategic documents made the decision-making process more transparent. It also made it possible for core actors to recognize each other’s stake in the proposed design choices. Furthermore, documents served as a basis for both predictive knowledge (Puranam et al., 2012) and the architectural knowledge of the designed artefact

(Baldwin, 2015). A high level of transparency arguably also made it easier for local actors to remain supportive of the project despite conflicting sub-goals and interests (Eriksson et al., 2017). Moreover, disseminating documents widely made it possible to keep actors in the wider environment informed about the progress of the scheme.

While outside the scope of this study, we conjecture that the transparency with which the design process was carried out attenuated the level of project opposition within the broader environment. This conjecture is in line with the argument that, especially in the public sector, plans are important for the legitimation of an organization and its strategy

84 with external stakeholders (Abdallah and Langley, 2013; Langley, 1988; Stone and Brush,

1996). An immediate difficulty which we will return to later is that sometimes documents presented competing analyses based on similar methodologies. Unsurprisingly, this occurrence complicated the process of establishing common ground.

The Hybrid Bill and the accompanying Environmental Statement (ES), together running to some 50,000 pages in total, were published and submitted in November 2013.

These documents were largely informed by both the coordination efforts outlined here and the cooperation efforts outlined in the next section. The latter were aimed at building consensus and intersecting the diverse array of interests through high-level decision- making. As the Head of Bill Preparation at HS2 Ltd. explained while preparing the documents, “we’re starting the preparation of these parliamentary documents on the basis of a draft design; when the design changes then we have to go back and change our parliamentary documents.” In February 2017 after three years of parliamentary scrutiny, the project received Royal Assent granting the Government the powers to build Phase One of the scheme between London and Birmingham.

Our findings demonstrate that coordination mechanisms, such as technical working groups and strategic documents, were important for the creation of common ground in the

HS2 megaproject meta-organization. As in unitary organizations, coordination in this pluralistic setting typically happened through the exchange of information, albeit across organizational boundaries (Srikanth & Puranam, 2014). However, although information processing was important for reaching agreed upon facts, it was insufficient for agreeing on mutually beneficial design choices during the HS2 planning process. Instead, cooperation mechanisms were a necessary complement for moving from collectively formulated problems towards collectively endorsed solutions. In the next section, we, therefore, discuss how cooperation mechanisms were used to reconcile the array of interests held by actors within the core.

85

5.2. Resolving Cooperation Problems

In theory, we know that cooperation problems such as shirking, misappropriation, hold up and similarly opportunistic behaviour are rooted in diverging or misaligned interests among interdependent actors (Gulati et al., 2012). In unitary organizations, such problems are averted through inducements and incentives that serve to align interests

(Puranam, 2015). For the HS2 megaproject organization, inducements took the form of concessions awarded during the planning process. However, such concessions were given sparingly and only after other avenues had been explored. Thus, inducements and incentives, in the form of ‘rewards’, do not adequately capture the cooperation mechanisms adopted during the HS2 planning process. Instead, we show how cooperation was primarily fostered through rules and structures created to incentivize core actors to comply with the design decision-making process and to assure them that others were doing the same (Ostrom, 1990).

5.2.1. Cooperation Process Overview

Local strategic boards created by HS2 Ltd. formed the linchpin for cooperation efforts in the respective cities along the route. These strategic boards differed from the technical working groups in that actors on the former had the authority to sign off on final decisions whereas actors on the latter were primarily tasked with exchanging and processing information. The strategic boards generally consisted of the top executive leadership of HS2, top DfT officials, local and elected political leaders, and agencies operating interdependent assets such as Network Rail or local highways authorities. As the

Area Manager for Euston at HS2 Ltd. put it, “we implemented a separate, distinct mechanism [that] had the most senior people you could get for those companies in the room when we were reviewing the options.”

86

Strategic boards gave the promoter an opportunity to encourage cooperative behaviour among core actors. Boards also gave core actors an opportunity to resolve emerging disputes rooted in divergent interests. Achieving cooperation was important because inter-organizational tensions could adversely affect the passage of the Bill through

Parliament. Given this risk, the promoter made an effort to prioritize collective solutions that wouldn’t alienate local actors whose preferences were not selected, especially if alienation would arise because it seemed difficult to reconcile local preferences with those of the promoter. The promoter was also careful not to alienate resource-rich core actors because doing so would reduce their incentives to contribute towards the project.

Ultimately, if local actors defected and refused to attend meetings, it would be more difficult to develop the predictive knowledge necessary for identifying mutually consensual solutions. Furthermore, this would increase the risk that core actors would petition against the scheme in Parliament. While crucial for resolving conflicts, deferring decisions to Parliament created more uncertainty during the planning process. Thus, it was important for the promoter to pre-empt the alienation of actors in the core so as to minimize conflict and reduce the risk of defection ahead of parliamentary proceedings.

Below an HS2 official explains how they did this,

We’re trying to minimize the conflict as much as possible rather than necessarily reach a resolution. So, it’s not like a contractual relationship where you have to escalate to a point; it’s more…how can we respond to your concerns; how do we make things better; how do we get you to a place that you’re happier [Commercial Director, HS2 Ltd]

Face-to-face interaction in the monthly strategic board meetings consolidated inter- organizational relationships between local actors and the promoter. Furthermore, the meetings created networks of communication across organizational boundaries, which facilitated the emergence of norms of cooperation, such as the willingness of participants to compromise. This was most apparent in Manchester (“I think [HS2] took our involvement very seriously”, said the Rail Programme Director at TfGM), where the

87 strategic board was able to reach agreements relatively quickly. To that end, our findings also suggest that the effective coordination process in the city contributed to solving the cooperation problem more easily than in cities where the coordination process was less effective. In the Sheffield case, for example, norms of cooperation were less apparent (“we are not meeting with HS2 now. Just because we have different opinions they haven’t spoken to us in months”, said the Executive Director for Places in Sheffield) as was the case in London (“They’ve always treated us like a stakeholder rather than a partner…they need to work with us as a partner because we’re part of the solution”, said the Assistant

Director for Regeneration and Planning for Camden Council). In both cities, strategic boards were fragmented and therefore took relatively more time and effort to forge agreements and act in a concerted manner. Furthermore, challenges with the coordination processes in these cities made the process of achieving cooperation more complicated.

Our findings suggest that encouraging cooperative dynamics across all groups was important for the promoter. Generally, a willingness to compromise, as was demonstrated by local actors in Manchester, made it easier to agree on the design choices to be advanced in Parliament. In contrast, unresolved conflicts as in London (and potentially Sheffield), led to an impasse, which was an unfavourable outcome for HS2 Ltd. as it created more uncertainty in the decision-making process. Uncertainty arose because it would be difficult to predict the time it would take to resolve differences in Parliament. It would also be difficult to predict the amendments and scope changes that would be enacted by the legislative body and, by implication, the ensuing costs that would be borne by the promoter.

One HS2 Ltd. official explains this below,

Hugely complex projects such as HS2 inevitably take time. This is not just because of the technical issues involved, but also because of the legislative process. At the heart, however, there is one simple truth: the linkage between time, certainty and cost. Put simply, the shorter the timescale, and the more certainty about the timescale, the lower the costs will be. [Chairman, HS2 Ltd9]

9Higgins, D. (2014). HS2 Plus. London: High Speed 2 Limited.

88

HS2 Ltd., therefore, aimed to resolve the cooperation problem by addressing concerns and providing assurances ahead of the parliamentary process. However, the process of reaching consensus was more complicated for some local groups than for others, and this had implications for the time it took to build consensus. Complications arose as a result of disagreements between locals and the promoter, disagreements among locals themselves, or both. In regions where HS2 met with a single local council, as in

Birmingham, Leeds and London, disagreements were primarily between local actors and the promoter rather than among local actors themselves.

As detailed in Table 5.2, the strategic board in Birmingham converged after five months whereas those in Leeds and London took four and six years, respectively. The core actors on the strategic board in Leeds had sufficient time to resolve their differences because the local station would only be delivered during Phase 2 of the project (beginning in 2033). Contrastingly, the strategic boards in Birmingham and London had added time pressure because their respective stations were due to be delivered during Phase 1 of the project (beginning in 2026). Greater urgency notwithstanding, in London, the process of convergence was further complicated by the fact that Camden Council, which had been excluded from the initial discussions to locate the HS2 terminus at Euston, was conflicted about the project. On the one hand, the Council had chosen to oppose the scheme politically and to legally challenge the promoter’s plans through the public court system.

On the other hand, the Council continued to engage with HS2 because, if they lost the legal battle, they were still interested in influencing the final design choices. An HS2 Ltd. official describes their view of the Council’s position below,

Camden don’t like HS2; they don’t want HS2, but equally, they do want regeneration and redevelopment of Euston, and to be honest, the catalyst for that is HS2 is coming and bringing inward investment. So, in a way they are a bit conflicted, because they don’t want it, but in some ways, if it doesn’t happen, who else will come along and kick-start regeneration in the area [Euston Area Manager, HS2 Ltd]

89

While the local strategic board in Manchester consisted of multiple local authorities, they demonstrated homogeneity in interests and were, therefore, able to engage with HS2

Ltd. as a united front. Thus, as detailed in Table 5.2, they converged after 24 months even though the station in Manchester would only be delivered during Phase 2 of the project. In marked contrast, local actors within Sheffield held divergent views and were each competing for a station in their locality rather than cooperating with each other. Referring to the divergences between the two groups of local actors, the Executive Director for

Places in Sheffield stated, “We’ve countryside in between, which is very nice. But it also creates a bit of a problem because they think they are distant, which in reality they are not.”

In the absence of unilateral authority to impose a local preference and with no immediate urgency given that the local station would only be delivered in Phase 2, the strategic board iterated between additional information processing and negotiations over the course of 5 years.

Strategic board meetings were important as forums for conflict resolution and mutual monitoring. During the process of collective action, both of these design principals are crucial for facilitating the discussion and resolution of any issues or perceived infractions that emerge (Ostrom, 1990). Conflict resolution and monitoring mechanisms are also important as they allow participants to actively audit the technical and organizational temperature of the collective endeavour (Ostrom, 1990, 2009). Beyond the strategic boards, conflict resolution and monitoring mechanisms were also available within the wider project environment. For example, local actors could approach the courts if they were unhappy with the process through which decisions were made, and especially if they deemed HS2 Ltd.’s approach unlawful.

To that end, Camden Council and other local authorities challenged HS2 Ltd. in a judicial review submitted with the aim of having the project stopped. The group argued that in deciding to proceed with the scheme HS2 Ltd. and the Secretary of State had

90 followed a legally flawed process. According to the group, the process was legally flawed on ten counts. However, during court proceedings, nine out of the ten accusations were dismissed. Moreover, on appeal, the courts ruled in the Government’s favour. For

Camden, this was an unfavourable outcome that made them question the independence of the judicial process itself (“it was helpful in confirming to me that perhaps the judicial review process isn’t the most independent one, as you might expect”, said the Assistant

Director for Regeneration and Planning for Camden Council). For HS2 Ltd., however, the judicial review was a success as it upheld what the promoter saw as a tremendous amount of effort put into stakeholder management. As the HS2 Ltd. Chairman put forward, “They

[Opposers] will return, but they will return knowing that they’ve already been rebuffed once, and the guys here are in a position to say ‘they’ve been actively consulted, and we’ve made a lot of effort to work with them.’’

In addition to public courts, Parliament was an additional forum for conflict resolution. During parliamentary proceedings, all individuals, organizations or groups materially affected by the proposed rail line would have an opportunity to petition against the scheme. It was therefore in HS2 Ltd.’s interests to resolve disputes ahead of Parliament, especially those disputes arising among actors in the project core, our focus here.

Otherwise, petitions could result in amendments to the scope of the scheme with unfavourable implications for the project costs and schedule.

HS2 Ltd., therefore, took proactive measures to minimize the risk of local authorities in the core petitioning. As the HS2 Ltd. CEO explained, “it’s a long process of trying to understand where the complaints are going to come from and then how many of those we could actually address before we get to the bill.” It follows that despite the promoter’s efforts to encourage cooperation, local actors could still end up petitioning against the scheme in Parliament. In this case, it was important for the promoter to engage with local

91 actors to be able to demonstrate that alternative proposals had been thoroughly considered ahead of parliamentary proceedings. An HS2 Ltd. official explains this below,

Once it goes into the parliamentary process, when people are petitioning, obviously it’s not in our control but we need to present the best evidence we can as to why we’ve made a certain proposal and why we don’t think that the proposal that the petitioner is making is the one that the committee should accept so we still have a chance to influence. But the decision-maker at that point is Parliament, it’s not even our Secretary of State at that point, it’s Parliament [CEO, HS2 Ltd]

Strategic decisions about where to locate the local stations were, therefore, made through consensus-building instead of through a top-down decision-making process evident in settings with unitary authority. Given that design choices were not imposed on local groups, collective arenas gave actors an opportunity to influence these choices ahead of the parliamentary proceedings. Design choices were, however, typically within the solution space delineated during the coordination process, and subject to the technical and financial constraints presented by core actors.

Crucially, our findings reveal that HS2 Ltd. encouraged cooperation through the use of three key integrative devices: (i) flexible organizational boundaries for different levels of participation in decision-making; (ii) negotiable technical and financial resource boundaries to help incorporate local interests and make concessions, either as a form of inducement or to attenuate conflict; and (iii) maintaining the proportionality of costs and benefits when considering concessions, in order to maintain the legitimacy of the scheme in the eyes of core actors and third parties. We now turn to discuss why each of these devices was important both for encouraging cooperation and for maintaining the collective development effort.

92

Table 5.2 Overview of the Cooperation Process Across Organizational Boundaries

Integrative High-level goal nit Salient Issues Outcomes Illustrative Quotes

U Cooperation Devices congruence/ Urgency

High Goal Congruence/ Promoter’s Rapid consensus 2010, With a little bit of imagination we can still make this work.. this is the Face-to-face High Urgency preferred location building right location, and we believe we can rejig our masterplan …. We are going interaction requires back ~5 months to upset a few people because we have to change their plans, but actually we Strategic board (every pedalling on plans can sell it on the basis that we are going to get something better here. [Local 6-8 weeks) committed for the High goal leaders] Arriving by high-speed area congruence made 2010, the University is demanding a £30 million refund from taxpayers, after Negotiable Resource train in the centre will be it easy to forge an learning its long-awaited new campus will have to be scrapped because of a

Boundaries an uplifting experience agreement proposed new high-speed rail station. [Local media] Birmingham (Compromise) [City Plan] 2011,The council has reached an agreement with the university for the suitable location [Local media]

Meeting forums Low Goal Congruence/ London-wide Consensus only 2010, We can’t see how not going to Euston would work .[HS2 official] HS2 Strategic board High Urgency agencies support reached after the 2012, The aspirations that they [Camden] have really require the whole Euston (every 6-8 weeks) the promoter’s regulatory station to be obliterated and start again.. that [fairy-godmother scheme] is just Not going to Euston is the preference process starts not possible [HS2 official] Euston area planning Ryanair solution vs. (Euston), but not ~6 years 2013, this entire scheme is ill-conceived and poorly planned it amounts to a group (monthly, just the local shed being bolted on to an existing lean-to. [Camden leader] local actors) Going to Euston is open government Low goal 2014, maybe we should go for a really big re-development of Euston. Let’s heart surgery on a (Camden Council) congruence face it [elected leader] Interest-based conscious patient resulted in 2014 How can Camden properly consider? It is not acceptable to present only negotiations protracted interest- one side of the story [Euston OAPF Strategic Board] based negotiations 2015, Our focus now will be on working to ensure that HS2 deliver on these

Negotiable Resource promises, highlighting the unfair compensation deal given to Camden on London Boundaries Select Committee [Regulator] (Concessions) 2016, if Camden drops opposition, it will be included on a panel of bodies that would oversee strategy and integration for the Euston project (HS2) Flexible participation 2016, These hard-fought agreements mark a long-overdue recognition of boundaries HS2’s impact on Camden …. A major step forward to get the best deal [Council Leader] Escalation of Decision-making

93

Integrative High-level goal Cooperation Salient Issues Outcomes Illustrative Quotes congruence/ Urgency Devices HS2 Steering group High Goal Congruence/ Preferred choice Rapid consensus 2011: there was a bit of a tension between the Salford and the Manchester Low Urgency by local actors building reps…The Salford reps kind of wanted it to be in Salford, and the Manchester Multiple forums: (Piccadilly) costs ~24months reps kind of wanted it to be in Manchester [HS2 Official] “The Manchester more than the 2012: it was really a compromise. Salford was very excited to get something Family of Once-in-a-generation promoter’s High goal … [but] there would be a huge amount of demolition.… the Salford head organizations.” opportunity to unlock the preferred choice congruence made concluded Piccadilly was the right location, even before Manchester said ‘is Manchester potential of the North (Salford) it easy to forge an Piccadilly viable?’ It just by nature worked that way[HS2 Official] Combined Authority; West [City Leader agreement 2013: This is all about getting ahead of the game. Instead of waiting until the Association of the building process, we are starting now – if we start now, we can’t lose. [Local Greater Manchester leader] Authorities; Local 2014, it will be important that an open and close working relationship is

Manchester Enterprise maintained with HS2 Ltd. and other stakeholders so that a number of detailed Partnership (LEP) technical issues are resolved [Manchester City Council Chief Executive]

what was good about it was that HS2 recognized that we [local transport Negotiable Resource body] had a better grip on modelling our own city then they did…HS2 was Boundaries persuaded by the Manchester family that Piccadilly was the best location (Concessions) [Rail Programme Director, TfGM]

Interest-based High Goal Congruence/ Promoter prefers Slow Consensus negotiations Low Urgency location 5 minutes Building 2013, The Secretary of State has had closed-door discussions with the Face-to-face away from central ~ 4 years Leeds leaders and said ‘OK, I am going to make this proposition work for interaction This is a once in a 100- station, whereas you.[HS2 Ltd Official] HS2 Strategic board year opportunity to … local actors ask 2013, Leeds sent to her a quite strong letter explaining their concern that Regional Economic create an unparalleled for co-locating the they didn’t feel the HS2 recommendation was the right answer. [HS2 Ltd Board location for major two stations High goal Official] Negotiable Resource sustainable development congruence made

Leeds Boundaries it easy to forge an 2014 The original proposals fulfilled HS2’s brief, but did not sufficiently take (Concessions) agreement albeit a into account the changing nature of the wider factors… The sum would have Flexible participation higher cost been lesser than the parts [HS2 Plus] boundaries Escalation of 2015, This is transformational and a real chance for us to create something Decision-making special in the centre of Leeds [Local Elected Leader]

94

Integrative High-level goal Cooperation Salient Issues Outcomes Illustrative Quotes congruence/ Urgency Devices Interest-based Low Goal Congruence/ Promoter prefers No consensus negotiations Low Urgency location outside after 2013: No one option is a perfect fit, so, inevitably, I recognize that this has city centre; some ~5 years to be a judgement call… [HS2 Ltd Chairman] local actors Negotiable Resource Sheffield is in grave concur, but other Low goal 2013: Rotherham City Council fully endorses the responses by SYTIA and Boundaries danger of committing local actors congruence made Sheffield City Region…An HS2 station at Meadowhall will provide a (Concessions) economic suicide [if the vehemently it difficult to forge catalyst for economic growth [Rotherham City Council] station is not in the city oppose an agreement Flexible participation centre] vs evidence 2013: The location of the HS2 station is crucial to the future of Sheffield. boundaries suggests Meadowhall is The Victoria option would reinforce the existing economic strategy focus on the best way the city centre, stimulating new quality development opportunities and

Sheffield Escalation of investment in line with the City Region’s economic [Genecon on behalf of Decision-making Sheffield City Council]

2016: We are delighted the Government has concluded that a city centre station location is the best option[Sheffield City Centre Council Leader]

2016: I am shocked and surprised by this significant last-minute change…They have never been suggested during the last four years’[Doncaster Mayor]

95

5.2.2. Integrative Cooperation Mechanisms

5.2.2.1. Flexible Polycentric Participation Boundaries

The organizational boundaries delineating direct participation in the decision-making process were malleable from the outset of the planning process. This adaptability was reflected in the changing composition of the groups of actors at different stages of the planning talks. Such flexibility meant that organizational boundaries were constantly adapted to suit decision-making requirements, which was crucial for deferring decision- making in order to avoid an impasse. Thus, multiple actors participated at different stages of the HS2 planning process. For example, HS2 Ltd.’s internal working groups were tasked, by the Government, with route selection during the early stages of the decision- making process. The process then opened up to local actors. At this time, inter- organizational technical working groups were formed to short-list station options and collate information; and strategic boards were formed to select the final station locations based on the shortlists. In addition, public courts were involved when decisions were challenged legally, and ultimately Parliament would decide on which design choices would be enacted legislatively.

Ahead of the Parliamentary proceedings, strategic boards were crucial for fostering agreement among the leaders of the various organizations. Board meetings were formally held every 6-8 weeks for organizational leaders to discuss high-level design choices and to select the final station locations. In addition, ad hoc meetings were convened as and when issues arose. Strategic boards worked well where coordination efforts had been effective in creating common ground. However, the strategic boards worked poorly where there was little underlying agreement. Thus, the decision-making process was more or less complicated depending on the level of goal congruence on the board. In both Sheffield and London, for example, it was a challenge to create common ground through coordination efforts, and the strategic boards in both cities were fragmented. This

96 fragmentation made the process of arriving at final design choices more difficult. In both cities, local actors questioned the accuracy of the promoter’s analysis, the distribution of payoffs, and the promoter’s sincerity when it came to sharing decision rights. As a result, meetings were held inconsistently in both London and Sheffield, and face-to-face interaction was characterised by alternating periods of joint activity and communication breakdowns. This further contributed to a sense in the local actors that the promoter was not cooperating with them. One London official expresses this sentiment below,

We’ve sometimes been concerned that the way the process has worked has been somewhat like tennis. So, they’ll send us a proposal drawing, we review it, we say that’s alright, or we don’t like that, and we don’t like that then we pass it back to them, they do a few weeks they look at it and pass it back. What it really should be is that the two of us are in the same room discussing the design. And we have done that, but not always and that’s been a worry [HS2 Lead Sponsor, TfL]

The London case is telling about how the promoter changed organizational boundaries and created forums to encourage participation and facilitate the search for mutually consensual solutions. First, HS2 Ltd. created higher-level structures that brought together the Chairman of HS2 Ltd. and leaders in the other promoter organizations, namely the Director General of the DfT and the Chief Executive of Network Rail. This high-level board, internal to the promoter, helped to develop a strategy for dealing with local actors’ concerns. Ahead of the high-level meetings, the Chief Executive of HS2 Ltd. would meet up with the Chief Executive of Camden in bilateral meetings or meetings opened up to other actors. HS2 Ltd. also set up a planning group for Euston to further facilitate the search for consensual solutions. This group included the HS2 Ltd. Chairman, the Political

Leader of Camden Council and the Deputy Mayor of the GLA. Moreover, HS2 Ltd. set up mitigation groups which saw lower level HS2 officials interacting with Camden Council officials to determine how best to deal with project related demolition. All of these forums and different memberships were important for getting agreement on the more difficult issues, and for giving local actors assurance that their concerns were heard,

97

For a long time, we didn’t have interaction [with Camden], but now we’ve moved past that, so that’s good. I think a number of things helped. One is that we started to demonstrate that we had taken some of their issues seriously, so they had started to see some progress in some of the things that we’d done [Euston Area Manager, HS2 Ltd]

Local actors also created their own local boards, which complemented the structures created by HS2 Ltd. In turn, HS2 Ltd. was invited to sit on these boards and always accepted. These forums provided additional arenas for devising solutions to difficult local problems. The forums also created further opportunities for encouraging cooperative behaviour, and in this regard, HS2 Ltd.’s attendance showed goodwill and a willingness to cooperate. From a theoretical perspective, locally created forums correspond with

Ostrom’s (1990) design principle that in robust organizations, structures and rules are congruent with local social and environmental conditions, and local actors can participate in modifying these rules and structures.

Sometimes, the organizations represented on the different forums were identical.

However, the membership differed as participants were from different hierarchical levels of their respective organizations. This was the case in London, for example, where

Camden Council created a mechanism for escalating dialogue because, as the Assistant

Director for Regeneration and Planning for Camden Council put it, “we were just not getting traction from HS2 Ltd. at the officer-to-officer level particularly on some of the blight and other complex issues.” Other times, the representative organizations would overlap only partially. Local authorities in Camden, for example, also created a political oversight group, which was a mix of the London Mayor’s office, local authorities and an open invitation to HS2 Ltd. This oversight group was particularly important for getting local politicians to buy into the proposed design choices as one local leader explains,

Within that structure…what we try and do is avoid , which can happen on these project, is a situation where you have a project where officers are developing something and politicians don’t know what’s going on, and when they find out they don’t like it, and then the whole thing’s up in the air; so it’s trying to keep the thing in a controlled, managed direction, where there’s always political buy-in to what you’re doing, and there are

98 continuous links from the political decision-makers to the actual people who are doing the project [Senior Manager Planning Decisions, GLA]

The Birmingham and Manchester cases offer other examples of how local actors created boards to devise consensual solutions. In Manchester, locals created a board focused on the regeneration around the station area. The Chief Executive of Manchester

City Council, the CEO of HS2 Ltd. and the Director for HS2 from the DfT were all invited to sit on this board, and they accepted this invitation. Of this structure, a local official stated that there were “two, quite high, influential people…sitting on the Manchester

Board.” Likewise, Birmingham created a scrutiny board to examine what would need to be done to maximize the local benefits of HS2 for the city. The board was open to all key local leaders as well as HS2 Ltd., Network Rail and DfT.

As is depicted in Figure 5.2, our findings suggest that the participation boundaries during the decision-making process also switched back-and-forth between different inter- organizational hierarchical levels. In the figure, we describe these as the operational level, the collective-choice level, and the constitutional level. These three levels mirror those available to appropriators of common pool resources (CPRs) as a key approach for solving problems (Ostrom, 1990, 2005). Similarly, the design choices made during the planning process can be viewed as CPRs because it was hard for the promoter to exclude local actors from the decision-making process, but often choices were subtractable because they were either/or due to technical and financial constraints (Gil and Baldwin 2013).

Furthermore, in this setting, as in the context of CPRs, decision-making traversed multiple levels, and governance rarely involved a single set of rules.

In the context of CPRs, actors at the operational level make day-to-day rules and decisions regarding the appropriation and provision of the resource. At this level actors also decide on what information they will exchange and withhold, on the necessary arrangements for monitoring the CPR including who the monitors will be, and on how

99 rules will be enforced (Ostrom, 2005). As represented in Figure 5.2, the operational level of the HS2 planning process saw an interplay between internal HS2 Ltd working groups, represented by a hierarchy in the figure, and the inter-organizational technical working groups and strategic boards, both represented by inter-organizational governance structures.

Information flowed between these different groups during the collective development process.

At the collective choice level, actors take decisions relating to policy-making, management, and the adjudication of policy decisions. These decisions are, therefore, taken by government officials in bureaucratic structures, by elected representatives in local and national legislatures and by judges in judicial arenas (Ostrom, 1990, 2005). Similarly, during the HS2 planning process, the decisions taken at the collective choice level involved amendments proposed by local actors, which would have financial implications beyond HS2 Ltd.’s mandate. Moreover, a so-called HSR Board including DfT, HS2 Ltd., the Treasury and Infrastructure UK (a division of the Treasury focused on the UK’s long- term infrastructure needs) held delegated authority to make decisions which were expected to have substantive cost implications and, therefore, required approval from multiple departments within the Government. Nevertheless, the Secretary of State would have to sign-off on the decisions taken at this level.

Thus, as depicted in Figure 5.2, issues deferred from the operational level would first be considered by a High-Speed Rail (HSR) Board, represented by an inter-organizational governance structure in the figure. Following the HSR Board’s consideration, issues would then be escalated to the Secretary of State, at the top of the DfT hierarchy, for a final decision. In describing this process of escalation the HS2 Ltd. Commercial Director explained that it involved “holding meetings, writing lots of papers, and sending submissions to the Secretary of State who then gives a review.”

100

At the constitutional choice level, rules are set regarding how to change collective choice structures, and about actors’ roles at the collective choice level (Ostrom, 1990;

2005). In the HS2 setting, the Hybrid Bill proceedings represented this level of decision- making. The Government (at the collective choice level) would only be granted the powers to construct and operate HS2 once the Bill had been through all the necessary stages in the two houses of Parliament, and subsequently once the Monarch had agreed to make the Bill an Act of Parliament. This process is shown at the top of Figure 6.1.

The HS2 project organization was therefore polycentric with “many officials and decision structures assigned limited and relatively autonomous prerogatives to determine, enforce and alter legal relationships” (V. Ostrom, 1972; V. Ostrom, Tiebout & Warren,

1961 McGinnis, 1999: 55). For example, issues were escalated beyond the operational level to the Secretary of State when local actors and HS2 Ltd. reached an impasse in decision-making. As the head of the DfT, the Secretary of State also monitored interactions between HS2 Ltd. and local actors and intervened in decision-making where necessary.

Crucially, despite exercising control over HS2 Ltd., local actors generally saw the

Secretary of State as a legitimate arbitrator. Our findings suggest that this was because both the UK Treasury, at the collective choice level, and Parliament, at the constitutional level, would ultimately audit the DfT’s decisions. Thus, the Secretary of State had an incentive to act fairly during the consensus-building process. In turn, this was essential for maintaining a sense of fairness within the collective action arena. The Cost and Efficiency

Manager at HS2 Ltd. (seconded from the consultancy, CH2MHill) explained this when stating that “the First Point of Scrutiny is DfT, and of course the DfT will be quite aggressive scrutinizers of us because they, in turn, will then be scrutinized by the Treasury.”

101

Figure 5.2 The Different Hierarchical Levels of Project Decision-making

Unsurprisingly, Leeds, Sheffield and London were among the most contentious locations and were also the locations where the Secretary of State became most involved.

In Leeds, this involvement was spurred by what the Head of Route Engineering at HS2 Ltd. described as a “strongly worded letter that went straight to the Secretary of State [showing] very clearly that Leeds weren’t 100 % convinced with the proposition on the table.” In

Sheffield, the Secretary of State was called to intervene because local tensions led to an impasse in decision-making; and in London, Camden Council approached the Secretary of

102

State for a commitment that assurance and mitigation measures would be put in place to address the local demolition of social housing as a result of HS2. The latter was eventually committed to, and the Government agreed to pay for tenants to be rehoused. In all three cases, the Secretary of State gently guided negotiations and steered conflicts toward mutually acceptable solutions.

The inclusion of different actors during the various stages of the decision-making process, and for different levels of decision-making, suggests that organization boundaries were both polycentric and adaptable over the course of the planning process. Such flexibility was instrumental in advancing collective action, avoiding communication breakdowns and thus for preserving the survival of the project organization. In the next chapter, we will discuss how the notion of adaptable participation boundaries differs from

Ostrom’s design principle that fixed organization boundaries are important to design robust consensus-oriented organizations.

Up until this point, the analysis has focused on the organizational architecture adopted during the HS2 planning process. We have shown how decision structures, comprising both technical and strategic actors, were important for both sharing information and maintaining the voluntary engagement of actors in the core. We also showed that flexible and polycentric participation boundaries were necessary to foster cooperation. In this regard, having multiple decision structures guarded against an impasse in decision- making. In the sections that follow, we will outline technical and financial aspects of the scheme and examine the extent to which related mechanisms also facilitated cooperation.

5.2.2.2. Negotiable Resource Boundaries

During the HS2 planning process, changes to the design choices were both important and inevitable, and refinements and extensions were a crucial part of the technical advance of the scheme (Clark, 1985). Design choices continually evolved as new information

103 emerged or when core actors challenged design proposals. Constant change in this regard is in line with studies of decision-making in pluralistic settings, which attribute changes to the fact that collective searches for single solutions are difficult (Denis et al. 2011, Simon,

1969). Thus, negotiable technical and financial boundaries delineating the resource pool were another important cooperation mechanism. Flexibility was important in this regard as it made provision for the priorities and interests expressed by local leaders to be incorporated into the final design choices. As such, negotiable boundaries served as a means of resolving conflict and reducing the risk that local participants would petition against the scheme.

Design decision-making involved a continuous process of defining, redefining, exploring and testing design choices in light of new information. Across the cases, the promoter’s preferred station location served as an initial reference point against which the technical feasibility and socio-economic value of proposed changes would be assessed.

HS2 Ltd. incorporated proposed alternatives if they were considered relative improvements that would refine design choices. By implication, HS2 Ltd. did not incorporate proposed alternatives if they were not considered relative improvements. Similarly, proposed alternatives were not included if they were not seen to represent value-for-money, or if cooperation efforts to build consensus failed. Instead, HS2 Ltd. communicated this decision and escalated their recommendation to the Government. If the Government upheld the decision, HS2 Ltd. would then prepare to defend its position in Parliament. An

HS2 Ltd. official explains this approach below,

If someone says could you move it there we look at it, and we use a sift criteria of why we should and what benefits there are…it isn’t that we’re going to move the thing completely, it’s just we have to understand, appreciate and take due regard for what people are looking for [Cost and Efficiency Manager, CH2M Hill]

As shown in Figure 5.3, the design decision-making process often involved multiple iterations before consensus was reached. The number of iterations depended in part on the

104 extent to which core actor’s technical preferences diverged; on the extent to which preferences were financially justifiable from HS2 Ltd.’s perspective; and also on the extent to which additional financial resources could be made available to reconcile and accommodate multiple interests. As depicted in the figure, actors in Birmingham agreed to the initial station proposal, albeit with a compromise from the local council given that the identified land was previously committed for private development. In London, a final decision was reached after two iterations and much disputation. For each of the Phase 2 stations–Manchester, Leeds and Sheffield–a decision was reached after a single iteration, albeit with varying levels of difficulty.

The technical working group in London identified Euston as the ideal station location for the city. When this was put forward to the local council, HS2 Ltd.’s initial proposal was for a complete overhaul of the existing Euston station and for the installation of a new double-decker station with 24 platforms, estimated to cost £2.1bn. Publically, the Council was opposed to the scheme because 400 residents would be displaced by the HS2 proposal.

Thus, the Council called for HS2 Ltd. to consider other options for locating the HS2 terminus in the city. Behind closed doors, however, the Council engaged with HS2 Ltd. through the technical working groups to establish regeneration opportunities in case the scheme went ahead.

Upon further review, HS2 Ltd. found that completely overhauling Euston station would be more complex and costly than had been initially understood. Significant parts of the station would have to be demolished resulting in 10 years of construction disruption for the local area and existing railway operators. In turn, this would significantly raise the costs of the scheme. Finding this so-called “fairy godmother station” to be unrealistic, HS2

Ltd. took a unilateral decision to refine the initial station option. To that end, HS2 Ltd. proposed that instead of a complete overhaul only parts of the existing station would be renovated. This change in the technical boundaries lowered the station costs from £2.1bn

105 to £1.8bn and then to £1.6bn when HS2 Ltd. dropped plans for an over ground railway line around Euston. An HS2 Official explained:

Originally the scheme that was developed and went through a consultation process was a bit like a fairy-godmother scheme where we came along with our magic wand and we sort of obliterated everything and started again with everything new and shiny. What we actually found as we developed that concept and really started to understand in much more detail what it would mean to deliver that work whilst trying to keep a station running...was that the complexity of what would be involved and the cost and the time was more than originally had been understood with the level of development that we’d been able to do in the early days [Euston Area Manager, HS2 Ltd]

HS2 Ltd.’s new preferred design choice would retain 13 platforms from the existing station and construct new HS2 platforms where the former platforms had been (Appendix

J4). These changes would result in savings of approximately £400 million relative to the initial cost forecast. Still unsatisfied, Camden council strongly opposed this proposal and the suggestion of a more modest intervention. This time their opposition led to stalled efforts to forge consensus and a breakdown in decision-making. By 2013, HS2 Ltd. chose to forge ahead with its preferred design choices and presented these in the Hybrid Bill submitted to Parliament. This decision suggests that, at this point, the promoter had determined that given time constraints, any negotiations were best continued once parliamentary proceedings had begun. Alternatively, it also suggests that the promoter determined that they had sufficient evidence to make a strong case for a scaled down

Euston Station in Parliament.

Once in Parliament, a cross-party select committee of Members of Parliament (MPs) would listen to petitions and study all relevant evidence before settling issues that were raised. Hence, the more issues the core actors left unresolved, the longer the Parliamentary process would take. Taking this possibility and the associated cost and schedule implications into account, in 2014 the CEO of HS2 Ltd. conceded that the plans for Euston needed to be reconsidered because they lacked ambition. Implicitly this meant that more resources would be mobilized to facilitate the search for a consensual solution. Following

106 this, HS2 Ltd. revised the plans for Euston further and proposed constructing 11 new platforms over two stages. Camden Council again vehemently opposed these revised plans and continued to argue that the original plan was superior. At this stage, the promoter refused to concede, and negotiations continued while both parties concurrently presented their arguments to the parliamentary select committee.

By the end of 2016, the promoter maintained that only 11 new platforms would be built in Euston. However, the promoter also agreed to revamp the station over the two stages of construction. This proposal was likely to cost over £4.5bn, more than double the cost of the initial proposal. Because the increase in costs left a significant shortfall relative to the funds HS2 Ltd. initially committed to Euston, the promoter also committed to working with local actors to find ways to make up for the shortfall. Furthermore, the promoter gave Camden council a seat on a governing body for developing the area around

Euston. In return, the local council agreed to withdraw its petition against the promoter’s plans.

Tensions in London were primarily rooted in differences between local actors and the promoter. As the case demonstrates, technical and financial resource boundaries were changed to encourage cooperation and to resolve differences that emerged. Camden

Council’s first position was that they did not want HS2 at all and their next position was that if they could not stop it, they wanted to get the most out of it. Their stance suggested a limited willingness to cooperate. This stance, in turn, was partly in response to the fact that they were excluded from earlier decision-making. As the promoter grew increasingly concerned about the complexity and cost implications of a grand station, integrating effort became more complicated. Ultimately, resolving local issues required that the promoter relax the boundaries around the resource pool.

The London case contrasts the Sheffield case because in the former, primary tensions were between local actors and the promoter whereas in the latter primary tensions were

107 among local actors themselves. In this case, boundaries around the resource pool evolved in response to the promoter’s efforts to encourage cooperation and minimize conflict at the local level. Of the two station options shortlisted by the technical working group in

Sheffield, a station at Victoria in the city centre and a station at Meadowhall, HS2 Ltd.’s analysis favoured the latter. For HS2 Ltd., this option most clearly satisfied the balance of design criteria including demand, connectivity, cost and environmental impacts. Local councils around Meadowhall supported this proposal, seeing it as an opportunity to consolidate growth in the area. However, Sheffield City Council, the largest council in the city region, felt strongly that the station should be at Victoria. The Council argued that locating a station at Meadowhall was inconsistent with the city centre station locations in

Leeds, Manchester and Birmingham. The Council also argued that locating the station at

Meadowhall would reinforce Meadowhall’s emergence as a business and transport hub in the Yorkshire region and accelerate the decline of the city centre. As a local official explains, the City Council, therefore, launched a strong lobby effort against the station at

Meadowhall,

We’re lobbying big stuff for Victoria station, because something like this gets built every 100 years in this country. This will be the first major rail expansion since Queen Victoria was on the throne. It’s way overdue. The importance of getting that location as the station in the city centre can’t be overstressed, for the economy of Sheffield. That’s absolutely nothing against all the other towns. [Executive Director of Places, Sheffield City Council]

Sheffield City Council was willing to borrow money to help finance the extra costs of moving the station into the city centre. However, the promoter remained reluctant to change the station location regardless of the fact that a local contribution would increase financial resources on the project. For some years, HS2 Ltd. held off making a final station recommendation to the Government. However, in 2016, after further analysis, HS2 Ltd. argued against both a station at Meadowhall and a major station in the Sheffield city centre.

Instead, HS2 Ltd. announced that it would propose a more modest rail interchange that would connect the HS2 services to the City Centre station via a ‘spur’ or a short side track 108 that would connect to the main HS2 track. In making this recommendation, HS2 Ltd. argued that the station at Meadowhall, which was encountering strong opposition by key local actors, wouldn’t stand the test of time. Likewise, the promoter argued that locating a station in the city centre would violate fundamental design criteria. Furthermore, HS2 Ltd. was not in a position to mobilize the significant financial resources that this option would require. Moreover, a proposal for a city centre station ran the risk of being overturned by

Parliament if local actors petitioned against it in Parliament.

109

Figure 5.3 The Evolution of Local Station Decisions 110

Responding to HS2 Ltd.’s recommendation, some local actors welcomed the change

(“this is fantastic news for Sheffield”, said the local council leader10) while others were left disappointed (“I am dismayed to see these surprise new plans to drop Meadowhall as South

Yorkshire’s HS2 stop”, said an MP in the region11). For HS2 Ltd., however, the outcome was well justified albeit the process of arriving at this decision was amongst the most time and energy consuming. An HS2 official explains why below.

The decision as to where to locate the HS2 station in South Yorkshire has proved one of the most difficult we have faced… it is, in part , due to topography…as well as the industrial legacy…On top of that are the different political and economic perspectives of the city and the region…Each of those perspectives is entirely valid from the viewpoint of those who hold them, but they have tended to be incompatible [CEO, HS2 Ltd]

Compared to the changes in Manchester and London, the Sheffield case is interesting because in this case, the proposed design solution was a shrinkage of the technical and financial resource pool rather than its augmentation. By opting for an HS2 interchange instead of a full station, the promoter estimated that it would make major savings of approximately £1 billion. This suggests that while a case could be made for spending more on the project (to a point), a case could also be made for reducing the project scope.

For the promoter, this was especially welcome as the decision was made at a time when project costs had escalated. Not oblivious to the cost savings, local actors opposing the decision argued that Sheffield was being sacrificed because concessions that the promoter had made elsewhere had depleted the contingency funds and put the promoter under pressure to find savings. One local leader, for example, went as far as to warn HS2 Ltd. not to sell the region “down the river12.”

10Source: BBC. (2016). Council claim victory over HS2 Station Move. 28 June 2016. 11 Source: BBC. (2016). HS2 Route Change : Updates and Reaction. 7 July 2016. 12Source: Langston, 2017. Transport Ministers Confirm Meadowhall for HS2 Still an Option. Yorkshire Post. 7 February 2017.

111

Unlike the Euston case, the final outcome of the Sheffield case is not known at the time of writing and could, therefore, unfold in a number of ways. For example, local actors who had supported HS2 Ltd.’s initial plans could petition against the promoter’s decision to drop Meadowhall station. In this case, Parliament could side with the promoter and uphold the decision to drop the station and the opportunity for major savings, or Parliament could side with those advocating for a station at Meadowhall. Alternatively, Parliament could also side with those proposing a station in the city centre, which would mean increasing the local financial resource pool.

Regardless of the outcomes in Sheffield, our analysis suggests that changing resource boundaries to accommodate the concerns of local actors was necessary for encouraging cooperation. Carrying out this process collaboratively was also far less complicated and reduced the need for parliamentary arbitration. Manchester is a good case in point. Of the three Manchester stations shortlisted during the coordination process, two stations in

Salford and one at Manchester Piccadilly, the centre of the region, HS2 Ltd.’s initial preference was for a station in Salford. The route approach into Salford would be cheaper and would result in less environmental impact than the alternative options. However, constructing the station in Salford would disrupt more environmental actors as it would involve significantly more demolitions.

Local actors in Manchester, representing both Salford and Piccadilly, challenged the promoter’s proposal based on locally commissioned analysis. This analysis showed that a station at Piccadilly, although more expensive, was the more compelling option from a value creation perspective. After further deliberation, HS2 Ltd. supported this alternative proposal, and the Government subsequently endorsed a station at Piccadilly. Thus, taking a more collaborative approach to adapting the resource boundaries created more certainty around design choices and allowed for more accurate cost forecasts to be made. It also meant that HS2 Ltd. could predict that local actors would not petition against the

112 promoter’s plans in Parliament. It is worth noting that Piccadilly was the more expensive option and thus local cost forecasts increased, which in turn decreased the contingency funds available to finance proposals from other disputes likely to emerge over time. Cost was, therefore, only one of the factors influencing the decision-making process.

The process of agreeing to change resource boundaries in Leeds was also collaborative. However, it proved more complicated than the relatively straightforward process in Manchester. Part of what complicated the local process was that the evidence backing the case for an alternative proposal in Leeds was weaker than for HS2 Ltd.’s preferred station. In addition,the cost implications of the proposed change were greater. Of the two options shortlisted by the technical working group in Leeds, the promoter preferred a station at New Lane, 500m south of the existing station (Appendix H7, Option 3). The local council, however, expressed dissatisfaction with this proposal arguing that the distance between the two stations would be a barrier to achieving a seamless interchange with HS2, and it would undermine city region connectivity more generally. As expressed below, local actors, therefore, proposed that HS2 Ltd. consider a station that was sufficiently integrated with existing rail lines,

The Leeds HS2 station should be sited in the city centre in such a position that it can be fully integrated with the classic rail network and other transport networks. A station at Leeds New Lane as announced by the Secretary of State in January 2013 allows for a city centre HS2 station location in an area of the South Bank, which is considered to have the potential for regeneration, and integration both within the centre of Leeds and with the existing station. However, the distance between the two stations has been highlighted as a potential barrier to achieving seamless interchange and City Region connectivity to High- Speed Rail [Leeds City Council, in press13]

Although HS2 Ltd. acknowledged the value of Leeds’ preference, they were reluctant to incorporate the proposed change because doing so would require a substantive relaxation of the cost forecast target. The major issue was not the change in the station location per se but the impact of such a change on the route coming into Leeds. The

13 Source : RTM. (2015). Higgins to Look at Other Sites for Leeds HS2 Station. Rail Technology Magazine. 6 February 2015.

113 physical environment was more challenging and the integrated station would have a greater negative physical impact on surrounding communities. Co-locating the stations as local actors in Leeds were proposing would, therefore, require a significant increase in costs. Insistent on their proposal, local authorities voiced their concerns to the Secretary of

State in what HS2 Ltd. saw as ‘a very strong letter.’ Encouraging cooperation, the

Secretary of State tasked HS2 Ltd. with holding further discussions to explain their proposition to local authorities more clearly. HS2 Ltd. and local actors then engaged in a new round of talks and exchanged more evidence. At some point during the discussions,

HS2 Ltd. officials thought they had convinced Leeds, with one official saying,

We explained to them why their route into Leeds was more challenging, more detrimental, had more impact, and was more harmful to Leeds than what we put on the table and they understood and said I wish we had known that beforehand. They then bought into our proposition of a New Lane Station [Head of Route Engineering, HS2 Ltd]

However, Leeds continued to insist upon a more integrated station configuration, as local officials expressed that HS2 Ltd.’s proposed station was “too detached from the existing station and too isolated from the city centre.” Conceding to local demands, in

2015, HS2 Ltd. proposed that Leeds Station be remodelled into an integrated ‘Yorkshire

Hub’ (Appendix I8). This proposal would see an HS2 station adjacent to the existing station and a concourse connecting the two. This integrated proposal is illustrated in

Appendix I7 (Option 2). Incorporating this change meant that the local cost forecast would have to increase significantly. Nevertheless, the change appeased local participants:

This is very welcome news…to offer certainty in that everyone can now know exactly where and how it will come into Leeds. That is hugely important as now planning at all levels can be carried out in an informed way around HS2, what it means and where it will go….It also confirms the remodelling of Leeds Station as a single integrated transport hub, which is very pleasing as it is a key element of the South Bank Leeds development which aims to provide more than 35,000 jobs, over 4,000 homes and educational facilities for over 10,000 students [Leeds City Council Leader, in print14]

14 Source : Leeds City Council. (2016). Council Leader Welcomes HS2 Preferred Route and Revamped Leeds Station Confirmation. Leeds: Leeds City Council.

114

Across the cases, resource boundaries were adapted to accommodate emerging information and interests. In Manchester, this was the case given that the promoter accepted the station location that local actors preferred, even though this would increase local costs. Through discussions, the promoter recognized that greater benefits were associated with the local preference and deemed it the most satisfactory choice. The case of Sheffield is interesting because our findings show a mutual willingness to cooperate.

However, Sheffield’s inability to make a persuasive case for a city centre station, coupled with local disagreements, made the promoter reluctant to relax the budget. Instead, the promoter chose the option of constructing an interchange rather than a new station as being the most satisfactory, creating an opportunity in this case for significant savings. Since the

HS2 budget is subtractable, we conjecture that such savings were convenient at a time when the HS2 Ltd. had already made significant concessions in Manchester and London.

The Birmingham case is consistent with these findings. In Birmingham, local actors contributed to the consensus-building efforts by agreeing on a compromise because the land identified for a station by HS2 had already been committed to local development plans. In light of this clash, a meeting was called between the leaders of HS2, the senior leadership in Birmingham, and Centro, the local transport agency. Together these parties examined the different approaches for getting the route into the city and weighing the evidence presented. Following these meetings, Birmingham decided to change their local plans to accommodate HS2 Ltd.’s preference even if that meant the city would have to compensate the private developer for changing their plans. HS2 Ltd. was impressed by this decision and felt it was a strong statement of support.

They had it committed with the University, they had the funds of the University, they had other people who were either looking to commit or build or were in the process of, and us coming along who were effectively going to blight that land for a period of time…I was impressed with them because they said, “No, we are looking beyond the blinkers of a five year period here, we are looking at where Birmingham can be in ten to fifteen years’ time and we think with a little bit of imagination we can still do the development we want and we can protect that piece of land for the station…[Head of Route Engineering, HS2 Ltd]

115

Our findings suggest that having negotiable technical and financial boundaries encouraged cooperation. Boundary changes reflected a willingness to incorporate new information into the design decision-making process and signalled the acknowledgement of differing interests on the project. Most cases suggest that the promoter financed the proposed concessions that were necessary to reach mutually consensual solutions.

However, the Birmingham case shows that local actors could also mobilize resources (with

Sheffield council also open to doing the same). Thus, crucially, the willingness to relax boundaries was not one-sided on the part of the promoter.

Our findings also show that the boundaries delineating the resource pool could not be relaxed infinitely. Instead, the decision to relax resource boundaries involved a certain level of discretion. This discretion was exercised in London, for example, when the mayor asked for an additional railway line, so-called Crossrail 2, to help transport HS2 commuters across the city once they arrived at Euston. While not contesting the case for

Crossrail 2, the HS2 Ltd. Chairman insisted that the line should not be financed from the

HS2 budget “[Crossrail 2] needn’t be part of this bill – there’s no reason why they can’t have their own act. They’re just trying to get somebody else to pay for it.”

Taken together flexible participation and resource boundaries illustrate the correspondence between emergent social and technical architectures during the HS2 planning process. This correspondence was necessary for maintaining cooperation in the pluralistic meta-organization core. However, it also contributed to ambiguous performance measurements at the system level given that corresponding performance targets were also continually changing (Lundrigan et al., 2015). To that end, we now turn to analyze how the costs and benefits of the scheme were managed, and how they were used to encourage cooperation.

116

5.2.2.3. Proportionality of costs and benefits

Both collective action literature and organization design literature propose that the proportionality of costs and benefits matters for consensus-oriented collective action

(Ostrom, 1990; Gulati et al., 2012). Indeed, this proportionality is one of the principles underpinning the design of robust arenas for governing collective action in the context of natural common-pool resources (Ostrom, 1990). In this context, consistently relating costs to benefits is a crucial element for establishing a fair system. Doing so also helps to build trust in relationships, and it encourages actors to keep a resource well-maintained and sustainable. Our analysis shows that the same idea extends to our setting where heterogeneous interests inform single design choices.

As discussed in the previous section, multiple technically feasible design choices were proposed over the course of HS2 planning discussions. However, the promoter was careful to only agree to choices that represented “value-for-money.” As such, our findings show that core actors systematically made an effort to demonstrate the commensurability of the costs and benefits underlying their respective design choices. For the promoter, emphasizing value-for-money was important because design choices were being scrutinized by third parties to the scheme, and these choices could be legally challenged if they were found to be imprudent. Maintaining a sense of fairness in the use of public funds was also important to uphold the legitimacy of the scheme in the eyes of core actors and other actors in the wider project environment. Hence, the promoter was selective about incorporating changes to maintain a sense of fairness and to avoid setting the unsustainable precedent of conceding to all local demands. To that end, an HS2 Ltd. official said,

Precedent is a big issue here...if we change something fundamentally on Phase Two that could have an adverse effect on Phase One, previous towns will say, well you could do it there why can’t you do it on phase one, so you have to be really careful now that whatever we do isn’t setting precedent. We have to look at everything transparently and say that is the proposition for this station wherever it is, we can change it for these reasons, we can do whatever it is we need to do but we also need to make sure that when we change stuff

117 that is what we do, we don’t do it and set a precedent that might underpin or undermine the current bill that is going through Parliament [Head of Route Engineering, HS2 Ltd].

Given the importance of value-for-money, coupled with HS2 Ltd.’s concerns about precedence when granting concessions, alternative design choices were assessed against a number of different criteria in the search for mutually consensual solutions. As the Head of

Route Engineering at HS2 Ltd. explained, “at the end of the day we’re not an open chequebook, we can’t turn around and say well I’ll put the station here and I will also rebuild that stuff – no.” Although HS2 Ltd. did not explicitly announce the financial constraints for local stations, the notion that demands had to be well justified was understood by local actors. Thus, in the same vein, the Leeds, Manchester and Heathrow

Project Director at HS2 Ltd. explained that “Government will not willingly go for a more expensive option unless they can be satisfied that they’re actually getting value for the additional cost.” For local actors, the challenge was, therefore, to convince HS2 Ltd. or the

Government that additional costs were justified.

Our findings show that the promoter was also invested in developing a common basis through which both the costs and benefits of alternative design choices could be assessed. However, measuring the project benefits proved challenging because, at least initially, the Government was not equipped to do so for a project of HS2’s scale. The HS2

Ltd. Chairman explained this stating, “we haven’t got a model that relates to a brand-new railway that connects cities because we haven’t built one since the Victorian era… we’ve already included some wider economic developments [in our modelling], but I’m certain there are much much more.” Therefore, as planning unfolded, senior HS2 Ltd. officials, a panel of experts, along with in-house and central government economists created a working group to find new ways to measure and capture the economic impact of the scheme.

118

To that end, our findings suggest that developing consensual measurements for value-for-money involved reconciling the promoter’s project-wide models with the more local models developed by actors at the regional and local levels. The promoter recognized that local authorities possessed more fine-grained information regarding demand, population data, growth forecasts, jobs and job creation opportunities. This information was coupled with a better understanding of how their respective cities worked.

Furthermore, local actors also expressed more confidence in their own models. The Rail

Programme Director for TfGM, for example, stated, “our modelling of the city centre was better than HS2’s, so we persuaded them that they should use our transport models rather than their own to actually evaluate the benefits case as well as the cost case for the

Piccadilly area.” Therefore, local level models, which emphasized the benefits of urban regeneration, were merged with the promoter’s models, which emphasized technical aspects, capital cost, environmental impact and the economic benefits of increased transport connectivity. An HS2 official described this process of integrating different perspectives as follows,

Our model needs to allow us to look at long-distance and national journeys, whereas they’re focused very much on what happens very much within the region…the stakeholders didn’t always necessarily agree with the approach that we were taking, but eventually, we got to a position where they understand our approach…and they have their approach, and they’ll present the information from their point of view on a more detailed level as well [Senior Route Engineer (North West), HS2 Ltd]

Local Searches, Global Performance Targets

At this point, it is worth stepping back from the dynamics of searching for consensus at the local level to examine the broader implications that local searches had on global performance targets. As depicted in Figure 5.4, between 2009 and 2016, the project cost forecast doubled from an initial estimate of less than £15 billion (in 2011 prices) for Phase

One to the region of £40 billion (inclusive of construction risk allowance) but excluding contingency funds. This evolution of the cost forecast was largely determined by the

119 evolution of the design over time. The first significant cost escalation on the project happened in 2010 after the Government significantly increased the project scope from a

London-Birmingham line to the cross-country Y-layout. Subsequently, increases in the cost forecast partially mirrored local searches for stations along the route. This evolution of the cost forecast is in agreement with the idea that relaxing the boundaries of the resource pool was a prerequisite for achieving cooperation and finding mutually consensual solutions.

Figure 5.4 The Evolution of the HS2 Project Cost Over Time (2009-2017) Source: Gil, Msulwa and Pinto, 2017

In an effort to manage the project costs and moderate the opposition stemming from project cost escalation, the Government invited local actors to finance the shortfall associated with local concessions directly. Local authorities, however, responded to this call for contributions with mixed reactions. Taking a less cooperative position, local actors in Leeds and London, argued that the financing for track and station costs was the

Government’s responsibility. To that end, the TfL/HS2 Lead Sponsor stated, “It’s not our project, and we believe, and I think government agrees, that this has to be funded primarily by the HS2 project rather than the mayor.”

120

In marked contrast, local authorities in both Manchester and Sheffield were open to making contributions. In Manchester, particularly, the Government asked the Council to finance the costs of a second station near Manchester Airport. The promoter felt this station had a weak business case, but local actors championed the proposed station stating that it represented long-term economic value. The Secretary of State subsequently endorsed the proposal partly because it was strongly backed by local organizations.

However, the endorsement was conditional on a local contribution towards the station. In turn, Manchester officials assured the promoter that they had the financial capacity to contribute at some future point arguably because the ten Manchester councils own the local airport, which is highly profitable. Although the size of the local contribution would be determined at a later date, local authorities agreed to the Government’s terms. On the other hand, as expanded on below, Sheffield City Council tabled a range of figures that they were willing to contribute if the promoter agreed to move the station to the city centre,

We’ve tabled a range of figures, between £100 to £400 million, depending on how it’s structured. So, that’s pretty sizable for a city of this size to be coughing up, because we’re confident, so confident of that. I have looked at most of the studies, KPMG is working on something at the moment that is basically proving that no matter what the government says, we’re going to have to have some sort of public sector money to pull prime station development, level of connectivity because the numbers are so big. Now in our case, we’re saying we’ll put our hands in the pocket because we think it is so crucial to our city [Director of Places, Sheffield City Council]

Despite Sheffield City Council’s strong lobby and willingness to make a financial contribution towards a city centre station, HS2 Ltd. remained adamant that such an option did not represent value-for-money. For the promoter, such considerations were important given that the project’s value propositions were monitored by third parties within the project environment, some of whom reported to Parliament.

Third Party Monitoring

Given the public nature of the HS2 project, value-for-money propositions were scrutinized by multiple actors in the project environment. The so-called Major Projects

121

Review Board within the Treasury, for example, cross-examined HS2 officials quarterly on the economic case of the scheme. To that end, the Project Director for Infrastructure UK,

HM Treasury (seconded to HS2 Ltd.) explained that “Treasury isn’t here to be visionary, necessarily, that’s the role of the ministers.” Other scrutinizers included public spending watchdogs such as the National Audit Office (NAO) and the Public Accounts Committee

(PAC). Both bodies are responsible for holding the UK Government to account by scrutinizing the value-for-money of public spending for the delivery of public services.

The benefit of having such monitors in the environment is that they act as a deterrent for politicizing design choices or currying favour from particular local actors.

In their assessments of HS2, both the NAO and the PAC raised concerns about HS2

Ltd.’s value propositions (NAO, 2013, 2016; PAC, 2013,2016). The NAO, in particular, criticized HS2’s cost-benefit analysis and argued that the strategic case for the project was not well presented (NAO, 2013). At the same time, both bodies cautioned that failure to sequence strategic choices correctly could result in a reduction in future benefits and an increase in costs (NAO, 2016; PAC, 2013). With regards to the capital costs of the scheme, in particular, both the NAO and PAC were sceptical about the cost forecasts presented by

HS2. They also questioned the legitimacy of the underlying assumptions used. In this regard, the Chair of the Public Accounts Committee stated explicitly that HS2 Ltd. had used “fragile numbers, out-of-date data and assumptions which do not reflect real life”

(PAC, 2013: 5). Furthermore, the Chair concluded that the costs for Phase 2 of the scheme were “still volatile and needed to be firmed up urgently” (PAC, 2016: 5).

Both bodies also questioned how the project promoter would realize its ambitions of rebalancing economic growth and integrating HS2 with the wider UK rail system. They argued that the promoter would not to be able to realize the potential urban regeneration and growth benefits unless the Government committed funding for those complementary

122 activities. They were especially concerned about this since, from the onset, the HS2 project budget had not included provision for regeneration around HS2 stations.

For the promoter, being seen to address the concerns relating to the legitimacy of costs and benefits was important for sustaining cooperation with local actors. It was also important for fostering support for the scheme and neutralizing opposition within the project environment. Managing the cost forecast during the planning process, and especially using the contingency funds sparingly ahead of the parliamentary process, was therefore imperative. This was because once the Hybrid Bill was before Parliament, the parameters defined in the Bill would be fixed, which would limit the solution space from which the promoter could make decisions. Parliament’s intervention would, therefore, constrain the promoter’s ability to influence the search for consensual solutions.

Furthermore, any delays associated with the parliamentary debates around design choices would also translate into extra uncertainty about the construction timescale and, therefore, about its cost. In a sense, the contingency funds created a buffer for the promoter ahead of tough negotiations around contentious issues.

The promoter’s approach was to integrate effort, as far as possible, ahead of parliamentary proceedings by reconciling its more utilitarian vision for the scheme with the grand visions expressed at the local level. Doing so effectively would minimize the level of intervention from Parliament over the final design choices, and it would therefore also reduce the uncertainty and costs associated with parliamentary interventions. In this way, the promoter could foster cooperation and minimize defections, reduce uncertainty, and ultimately advance a collective interest in Parliament.

5.3. Summary

In this chapter, we empirically explored the notion of organizing collective development in pluralistic settings. We did this by using the intersection of organization design and collective action literature as our cognitive lens. We focused our analysis on 123

HS2, the UK’s new high-speed railway, and the strategic decision-making process in five cities along the route: London, Birmingham, Manchester, Leeds and Sheffield. As such, we highlighted the centrality of the integration of effort in each city, and we also showed how the different groups went about resolving the coordination and cooperation problems inherent in pluralistic settings. On the one hand, our findings demonstrate how the rules and structures in some cities (Manchester and Birmingham) were amenable to collective development, which saw these groups largely achieving their local aims. On the other hand, our findings demonstrate how in other cities (London, and especially Sheffield) this was not the case.

Collectively developing HS2 was a complicated process as can be expected in the context of collective strategizing (Astley, 1984). Our empirical findings suggest that complications were rooted in problems of “inflationary consensus” (Denis et al., 2011), collective “paralysis by analysis” (Langley, 1995), or the dilution of strategic intent (Denis et al., 2011; Jarzabkowksi & Fenton, 2006). These observations are wholly consistent with management theory in settings where authority is diffused. However, in the discussion chapter that follows, we focus our attention on the organization design perspective. To that end, we further interrogate our findings using key concepts from our theoretical framework, to develop a coherent conceptual understanding of organizing collective development in pluralistic settings.

124

6. DISCUSSION

6.0. Introduction

In this chapter, we use the previously outlined theoretical insights and empirical findings to identify some important mechanisms for organizing collective development in pluralistic settings. More particularly, we identify coordination and cooperation mechanisms that can be used to resolve collective action problems inherent in pluralistic settings. The coordination mechanisms we identify include technical working groups and boundary objects, which are important for processing information within the meta- organization core. The cooperation mechanisms we identify include adaptable polycentric structures, negotiable resource boundaries, and appeals to proportionality in costs and benefits, which are important for aligning the interests of the different core actors. To that end, we propose that the coordination mechanisms used for collective development in pluralistic settings are similar to those mechanisms used in unitary organizations, as outlined in the organization design literature. Furthermore, the cooperation mechanisms used for collective development in pluralistic settings are similar to those outlined in the collective action literature. Moreover, we uncover mechanisms, which support coordination and cooperation in pluralistic settings that go beyond each of these literature streams. Overall, we discuss how organizational choices can help collectives of heterogeneous actors achieve their system-level goals.

6.1. The Social and Technical Aspects of Collective Development

As is characteristic of meta-organizations, while planning HS2, resource-rich actors in the core remained legally independent but were nonetheless unified by a ‘superordinate goal’ (Gulati et al., 2012; Lundrigan et al., 2015; Gil et al., 2017). The goal was to develop the design for a single indivisible functional asset in the form of a cross-country railway line. Initially, this goal was advanced by a limited coalition of actors including HS2 Ltd.,

125

Network Rail and the Department for Transport (DfT). Of these actors, HS2 Ltd. was responsible for working-level integration, which entailed undertaking the necessary work to enable the DfT to deposit a Hybrid Bill in Parliament towards the end of 2013.

For large infrastructure projects, advancing both the technical and social (or organizational) aspects of a scheme is crucial for collective development (Lessard et al.,

2014). Similarly, in so-called business ecosystems innovation is distributed, and recombining the component parts into whole integrated goods requires that diverse entities be integrated into a coherent network (Baldwin, 2012; Adner & Kapoor, 2010; Iansiti &

Levien, 2004; von Hippel, 1988). In the HS2 context, achieving social and technical coherence was complicated given that the meta-organization was enmeshed in the environment. Nevertheless, achieving an appropriate level of coherence was paramount because it would determine whether or not the Government received powers for the construction and operation of the proposed scheme during parliamentary proceedings.

Thus, in line with legislative requirements HS2 Ltd.’s primary objectives were to i) deliver a safe and affordable route design to Parliament, ii) assess the environmental impacts of the design, iii) produce an environmental statement complete with propositions for dealing with the impact of the project on the environment and iv) consult with relevant stakeholders on the environmental statement and other aspects of the proposals. More broadly, HS2 Ltd. had to integrate effort across organizational boundaries to facilitate the development of a collective strategy that could withstand parliamentary proceedings. This exercise involved coalescing (sometimes conflicting) design preferences into a single design for the project.

The parliamentary process posed a significant hurdle, and so the promoter deemed it imperative that agreements were reached ahead of time. Failure to do so would contribute to a protracted parliamentary process with potentially negative implications for the cost and schedule of the scheme. In the worst-case scenario, it would see the project halted

126 altogether because unresolved issues would undermine the super-ordinate goal and the development of a collective strategy. Beyond the parliamentary process, if HS2 were legislated, certain matters and details would be reserved for local planning authority approval, including plans, specifications, site restoration schemes and construction arrangements. Engaging with local authorities early on would, therefore, also prevent unnecessary delays when obtaining these approvals. Thus, HS2 Ltd had every incentive to engage with local authorities and resolve differences as they emerged. Hence, the promoter invited local authorities to join the megaproject organization and proceeded to make an effort to encourage norms of cooperation (e.g. compromise, reciprocation and mutual trust).

Against this backdrop, the primary aim of this doctoral research has been to further our understanding of how to organize for the collective search, identification and development of mutually acceptable solutions in pluralistic settings. To do this, we examined how legally autonomous inter-organizational actors with heterogeneous interests designed and developed an indivisible artefact in the form of a cross-country railway line.

Through this exercise, we gained insight into how governance structures influence the collective development process, the quality of inter-organizational conversations, and the performance of the meta-organization more generally.

Going forward, we draw on insights from organization design and collective action literature and use this conceptual lens to interpret our empirical observations.

Independently, each stream of literature is insufficient for illuminating the dynamics of collective organizing for planning HS2. However, taken together they provide a solid conceptual foundation on which we can develop a theoretical understanding about how to govern collective development processes in pluralistic settings.

127

6.2. Organizing Collective Development in Pluralistic Settings

In organization design studies of unitary organizations, a basic mission or strategy typically takes the form of objectives, policies or plans articulated by organizational leaders. These leaders set out the overall direction the organization is to take and a predetermined course of action for organizational activities (Mintzberg, 2003; Mintzberg,

Ahlstrand & Lampel, 2001). Following this articulation, the intended strategy is broken down into goals and quantifiable objectives captured in the tasks allocated to actors within the organization (Puranam, Alexy, Reitzig, 2014). As these tasks are carried out to advance an organization’s strategic objectives, internal decisions combine with external events, and so the realized strategy emerges (Mintzberg & Waters, 1985; Gavetti et al.,

2012; Lee & Puranam, 2015; Puranam et al., 2015). Therefore, while strategic choices are guided by top-down hierarchical structures, they are realized over time (Posen and

Levinthal, 2012, Eisenhardt and Brown, 1999; Levinthal and March, 1993; March, 1991,

Cyert and March, 1963).

To advance the strategic objectives in unitary organizations, action is also guided by purposefully designed coordination and cooperation mechanisms that serve to integrate effort among the different organizational actors (Lawrence and Lorsch, 1967; Mintzberg,

1979; Puranam & Goetting, 2011; Puranam, Goetting & Knudsen, 2011). Coordination mechanisms take the form of information channels that support decision-making. These channels include hierarchical lines of communication and formal procedural specifications, which serve to communicate the content of an organizations activities (Galbraith, 1973;

March & Simon, 1958; Tushman &Nadler, 1978; Srikanth & Puranam, 2014; Knudsen &

Srikanth, 2014). Through these channels, actors are made aware of the planned behaviour of others (Puranam, Goetting & Knudsen, 2011, Srikanth & Puranam, 2011). Cooperation mechanisms are incentives or inducements that reward effective action. Through these mechanisms cooperation is ‘bought’ and rewards are given with a related expectation that

128 commands within the hierarchy will be accepted. They include employment contracts and other pecuniary and non-pecuniary mechanisms such as promotions, reputation and budgets (Lee and Puranam, 2015; Burton & Obel, 1984; Lawrence & Lorsch, 1967; March

& Simon, 1958; Mintzberg, 1979).

Consistent with contemporaneous conceptions of the strategy process, the process of collective development on HS2 was a coalescence of a deliberate and emergent strategy process because an intended outline was developed, and then details emerged over time

(Mintzberg et al., 2001; Mintzberg and Waters, 1985). In the HS2 case, the deliberate aspect was reflected in the promoter’s initial definition of the parameters of the scheme as a Y-shaped route linking England’s major cities. Following this, emergent details were the result of consensus-oriented discussions. Through the delineation of an intended outline, the promoter reduced the amount of uncertainty surrounding the megaproject organization.

The promoter was also able to identify those local authorities that were crucial for the collective development of the design concept. Therefore, once the Government endorsed the Y-shaped route, the promoter was able to extend the boundaries of the project organization and create a so-called meta-organization (Puranam et al., 2012; Lundrigan et al., 2015).

Local working groups were formed to agree on the location and form of each station along the route. Including local authorities in these groups gave them an opportunity to refine the plan for the Y-shaped route that HS2 Ltd. had developed in isolation. The promoter also shared local decision rights with local authorities, which afforded them an opportunity to influence the local station location and configuration directly. They were also able to assess the mitigation measures for the proposed rail line and propose possible amendments. Ultimately, creating a project meta-organization gave the promoter and all the actors in the core a platform to communicate, gather and distribute information, and work towards agreeing on consensual designs.

129

In the absence of centralized authority over the local decision-making process, the promoter had to guide each local working group towards a common end. As in other meta- organizations, consensus-building had to be carried out with an appreciation for the symbiotic and reciprocal relationship that the different organizations shared (Gulati,

Puranam & Tushman, 2012). Thus, in our focal setting, rules and structures emerged over time to guide multiple legally autonomous yet interdependent actors in the elaboration of a collective strategy (Astley, 1984). In turn, this exercise involved the iterative combination and recombination of multiple organization-level strategies of the project participants

(Mintzberg, 1985; Posen and Levinthal, 2012). Furthermore, actors worked at pooling their individually controlled resources to jointly develop a capital-intensive infrastructural artefact, which none of them could develop on their own.

With no recourse to command-and-control authority characteristic of unitary organizations, collective development in this pluralistic setting went beyond top-down processes that support traditional strategy-making. To that end, as we examined the planning process, we found that each actor maintained a certain level of decision-making authority. On aggregate, the emerging meta-organization, therefore, exhibited a polycentric architecture in that various decision-making forums were nested in multiple layers (Ostrom,

1990).

Within the emergent polycentric structure, coordination mechanisms, akin to those used in unitary organizations, were important for the transmission and exchange of information among core actors. This similarity is in line with Srikanth and Puranam’s

(2011) findings that inter- and intra-organizational coordination mechanisms are similar when communication is not constrained by factors such as geographic distance and differences in time zones. In line with the authors, we also found that tacit coordination mechanisms including boundary objects (Carlile, 2002, 2004) were used to both leverage pre-existing common ground and build new common ground by making actors, actions and

130 outcomes more observable. As in unitary organizations, these coordination mechanisms were important for reducing uncertainty on the scheme (Srikanth & Puranam, 2014,

Lawrence & Lorsch, 1967; Tushman & Nadler, 1978; Galbraith, 1973; Arrow, 1974). In contrast to unitary organizations, however, information channels were, more richly horizontal than centrally directed.

Further contrasting unitary organizations, though not unexpectedly, we found that given diffuse authority, and multiple conflicting objectives among actors in the megaproject meta-organization core, cooperation mechanisms akin to those used in firms were both inconspicuous and inapplicable. In the absence of a command and control decision structure typical of unitary organizations, the project promoter had no recourse to reward actors in the meta-organization core with employment contracts, extrinsic compensation or other inducements and incentives, such as salaries, promotions, reputation and budget. Thus, alternative means had to be adopted to secure their adherence to the collective effort. We, therefore, drew on revised theories of collective action to make sense of these. Elinor Ostrom’s (1990, 2009) work was crucial in this regard given that it conceptually frames the problem of cooperation as one of organizing and searching for appropriate means to solve related problems (Ostrom, 2009).

In the context of shared influence and authority, collective action scholars argue that cooperation can be achieved through the creation of robust institutions and structures that are perceived to be legitimate (Ostrom, 1990). Robustness in this context refers to the maintenance of a system’s performance even when it is subject to unpredictable disturbances in the behaviour of its component parts or its environment (Carlson & Doyle,

2002; Anderies, Janssen, and Ostrom, 2004; Ostrom, 1990, Ostrom, 2009). Regarding organizing for collective action, robust structures avert opportunistic behaviour in which

“individually rational but socially defecting choices may lead to a higher payoff for an

131 individual partner, but where once all partners adopt such a strategy, the alliance will fail”

(Zeng & Cheng, 2003: 587).

According to Ostrom (1990), robust governance structures are typically characterised by eight design principles (Ostrom, 1990). These principles capture the rules that emerge as a result of implicit or explicit efforts. They also promote order and predictability among groups of actors despite unpredictable environmental disturbances (Ostrom, 1990; Ostrom,

1986). The principles, previously summarized in Table 2.1, define explicit rules about what shared resources participants in the system can use, what the responsibilities of each participant are, and how the participants will be punished if they break the commonly understood rules (Ostrom, 1990). Hess and Ostrom (2005: 7) argue that these principles are “one of the truly most important findings in the traditional commons research.” It is our job here to leverage our empirical findings to illuminate why some of these principles remain valid to organize collective development in pluralistic settings, while others do not.

In the remainder of this chapter, we propose that robust governance structures can be created to guide collective development in pluralistic settings. However, given the added complexity rooted in the plurality of interests and payoff functions, we also argue that these governance structures will not always mirror the principles offered by Ostrom (1990).

Most importantly, we argue that designing a robust meta-organization to forge ahead with a collective development requires that the boundaries around both the membership of the organization and the pool of shared (technical and financial) resources be relaxed over time.

This contention is in contrast with Ostrom (1990)’s principle that a fixed shared resource, well-defined boundaries delineating access to the shared resource, and well-defined participant roles are crucial for designing robust collective action systems that are sustainable over time.

Overall, we argue that for collective development in pluralistic settings, robust governance structures reflect coordination mechanisms similar to those outlined in the

132 organization design literature for unitary organizations; and cooperation mechanisms similar to those outlined in the collective action literature for the management of common pool resources. Moreover, we uncover mechanisms to support coordination and cooperation in pluralistic settings that go beyond each of these literature streams, such as flexible boundaries around both the membership of the organization and around the pool of shared (technical and financial) resources.

6.3. Multi-level Governance Structures

In agreement with the collective literature (Ostrom 1990, 2009), we find that multi- level or polycentric governance structures27 support collective development in pluralistic settings. The basic idea behind a polycentric structure is that it has “many centres of decision-making that are formally independent of each other” (V. Ostrom, Tiebout &

Warren, 1961:831; V. Ostrom, 1972; V. Ostrom, 1999a, 1999b). In Figure 6.1, we present a conceptual framework that shows how such a polycentric structure looks in the context of developing large-scale infrastructure.

In developing the framework, we drew on certain aspects of the Institutional

Analysis and Development (IAD) framework in collective action literature (Ostrom et al.,

1994). This framework captures how humans repeatedly interact within rules and norms that guide their choice of strategies and behaviours. Furthermore, the IAD framework suggests that collective action arenas cannot be understood unless we examine the participants and their norms of behaviour, the level of common understanding, the degree of homogeneity in preferences and interests, and the distribution of authority and resources among them (Ostrom, 1990: 45).

27 In line with Ostrom and Janssen (2005), we employ the term “multi-level” and polycentric as having similar meanings. Using the term in this way is supported by Hooghe and Marks (2001) who consider polycentric governance as one of two major forms of multi-level governance.

133

Likewise, we cannot understand the organization created to develop HS2 unless we examine the locus of power and influence among actors, the range of their incentives and their varying roles. For example, HS2 Ltd.’s mandate is to produce a cross-country railway line that meets technical and budgetary specifications; while local authorities included in the project core are primarily concerned with the local growth and regeneration implications of the scheme. Furthermore, the UK Government is invested in increasing the capacity of the transport network (DfT) provided that the proposed scheme represents value-for-money (HM Treasury). Moreover, Parliament is concerned with whether private and public interests are sufficiently accounted for during project planning, given the nature of the project and the legislative context within which the project is embedded. There are also over 4000 stakeholders that are impacted by the development choices, but that remain external to the meta-organization.

Consistent with the notion of polycentricity, a network of higher-order actors spanned a network of lower-order actors during the HS2 planning process. This polycentric structure is represented by the dotted rectangles nested within one another in

Figure 6.1. As shown in the figure, the HS2 meta-organization core was situated in the broader meta-organization (including the meta-organization periphery), which was itself situated in the wider environment. During the HS2 planning process, technical officials and organizational leaders, within the meta-organization core, were involved in coordination or cooperation efforts, respectively. However, the specific actors involved in each depended on whether the emerging decisions and actions to be taken were technical, strategic or a combination of both.

As Figure 6.1 shows, patterns of interaction arise as a by-product of the coordination and cooperation efforts undertaken by actors in the meta-organization core. Furthermore, these emerging patterns influence how the development problem is defined and what collective solutions are available to address the problem. In the HS2 case, for example,

134 core actors in Manchester and Birmingham were unified and thereby largely achieved their local aims while others in London, Leeds and especially Sheffield struggled in this regard.

To that end, when core actors are unified, the collective development process is relatively straightforward. Primarily, it involves technical artefact development and iteration between a shared problem representation, sifting through alternatives and selecting from those alternatives. However, when core actors are fragmented additional coordination or cooperation mechanisms are required to better inform artefact development. This additional requirement is represented by the bold arrow circling back to the organizational or social aspects of the scheme following an unsuccessful iteration of technical artefact development. From the HS2 case, we, therefore, note that collective development is an iterative problem-solving process (Simon, 1996).

During the iterative process of developing the HS2 concept, core actors drew on a range of evaluative criteria to help them assess alternative technical choices as well as the suitability of the emerging organizational rules and structures. Some criteria were rooted in deep-seated norms of project performance (e.g. the project management triangle of on time, on budget and meeting pre-specified quality); while other criteria considered more fundamental problems relating to the projects strategic success (Samset, 2009). Some examples of strategic considerations are presented in the conceptual framework and include relevance, which asks ‘are we doing the right thing?’ and sustainability, which asks

‘are the positive effects or impacts of the project sustainable?’

While we appreciate the view that the strategic failure of a project is more detrimental than its tactical failure (Samset, 2009), we also recognize that performance measurements can be ambiguous (Lundrigan et al., 2015). This is particularly the case in pluralistic settings where “the conflation of committals to different baselines, differing perspectives for efficiency and effectiveness, and rivalry in high-level choices gives rise to competing performance narratives which cannot be reconciled” (Lundrigan et al., 2015:

135

33). With this in mind, the notion of evaluative criteria in Figure 6.1 has been kept deliberately broad. We propose, however, that it is important to understand that the collective development process unfolds within a polycentric architecture, in which different patterns of interaction can influence the performance indicators that actors draw on, and therefore the inferences made about the meta-organization’s performance.

It is worth noting here that because the HS2 promoter had pre-selected the route (a higher-level choice), emerging decisions made by actors in the core (lower-level choices) would have to align with the route selection. Ultimately, synchronization between the initial higher-level choices made by the promoter and local choices made by inter- organizational groups at the local level was necessary for achieving the system-level goal.

Furthermore, the HS2 meta-organization was nested within the wider project environment, comprising constitutional and collective choice level rules and structures, third parties monitoring the collective development process and resolving deflected conflicts, and other actors both for and against the development. Thus, the HS2 meta-organization also had to take these dynamics into account.

136

Figure 6.1 Framework for Organizing Collective Development in Pluralistic Settings 137

As we recognize the many actors spanning the multiple levels of governance during the HS2 planning process, it is easier to understand how coordination and cooperation mechanisms were required to consolidate the various divisions in interests, preferences and values (Ostrom, 2007, Srikanth & Puranam, 2011; Clark, 1996). Next, we discuss each of these in more detail.

6.4. Resolving the Coordination Problem

Compared to a product development process, which unfolds in a well-structured context with unified authority, collective development, as with planning for a society or an organized body of members, is associated with greater complexity (Simon, 1962). This added complexity, i.e. a large number of parts interacting in a nonsimple way, requires that design calculations and criteria reflect an array of interests and therefore the possibility of larger and broader positive and negative externalities (Simon 1996). Coordinating structures are therefore important for addressing uncertainty, an inherent feature of collective action (Lawrence and Lorsch, 1967; Tushman, 1978; Tushman & Nadler, 1978;

Galbraith, 1973; Arrow, 1974: Srikanth & Puranam, 2014; Knudsen & Srikanth, 2014).

Without uncertainty, the collective development process would otherwise be carried out mechanistically through an evolutionary process involving a trivial selection of a known design to meet a known set of needs (Nelson and Winter, 2009; Williams & Samset, 2010).

Given the complexity and uncertainty in the HS2 case, such a mechanistic approach did not unfold. Instead, the promoter had to address the needs of multiple actors with heterogeneous interests in the meta-organization core through the creation of rules and structures.

During the collective development process, uncertainty can be attributed to a wide range of factors. In agreement with the project management literature, specific factors relevant to the HS2 planning process included what Samset (2010) refers to as task or

138

technological uncertainty, project mission uncertainty and environmental uncertainty.

Project mission uncertainty is related to task uncertainty and stems from a lack of shared objectives and agreement about the strategic intent of the project (Winch, 2015; Youker,

1999). During the HS2 planning process, project mission uncertainty arose because the project did not begin with a defined collective strategy that could be executed linearly.

Instead, coordinated collective action emerged over time through the combination and recombination of participant specific strategies (Denis et al., 2011; Lee & Puranam, 2015).

Hence, rather than a purely deliberate rational strategic planning process, the collective strategic process involved consensus building from which a collective strategy emerged

(Mintzberg, 1985). This process of strategy formation included correcting assumptions, which proved untenable as new information became available (Winch, 2002). The way the project mission evolved is telling of the uncertainty surrounding the development effort.

Task uncertainty stems from the uniqueness of outputs produced, which makes it difficult to interpret, define and identify critical tasks and problems (Samset, 2010). During the HS2 planning process, a major source of task uncertainty was that the proposed cross- country rail line was the first of its kind since Victorian times. This uncertainty meant that the promoter had to unpack what it meant to “move from a rail network designed for

Victorian needs, and Victorian concepts of time, to one that matches [today’s] different needs28.” In addressing task uncertainty project participants, therefore, used coordination mechanisms to consider several alternative viable design concepts, including a zero option when executing the task (Williams & Samset, 2010). This process was initially done through the shortlisting of feasible options against a progressively more detailed list of criteria in technical working groups, and then through incorporating new knowledge and information into the decision-making process as it emerged. Adopting an iterative process

28 Source: High Speed 2. (2014). Rebalancing Britain: From HS2 towards a national transport Strategy. London: High Speed 2 Limited.

139

when executing the task allowed project participants to coordinate around shared problem definitions, which in turn reduced task uncertainty on the project (Simon, 1996; Greve,

1998; Gavetti et al., 2012; Puranam et al., 2015)

Environmental uncertainty, meanwhile, stems from the complex and ever-changing dynamics in the project setting (Samset, 2010; Brix & Peters, 2015). The uncertainty and its associated complexity stemming from the environment arises because, as with social planning, in the HS2 setting “those for whom plans were intended became designers themselves who [sought] to use the system to further their own agendas” (Simon, 1996:

153). Coordination mechanisms were, therefore, essential to address the environmental uncertainty from the various forces within the environment (Miller and Hobbs, 2009; Scott,

2012). In part, this meant periodically publishing documents to keep actors in the broader project environment informed. In addition, actors in the meta-organization core had to develop the capability to cope with a turbulent environment (Williams & Samset, 2010).

Within the HS2 meta-organization, this meant developing collective forums, procedures and rules for joint decision-making.

Coordination mechanisms reduce uncertainty through information processing, which fosters common knowledge and alignment of understanding (Tushman & Nadler, 1978;

Galbraith 1973, Srikanth & Puranam 2014; Knudsen & Srikanth, 2014; Clark, 1996). By exchanging information, decision-making participants are also able to assess each other’s domain or location specific knowledge (Srikanth & Puranam, 2014; Star, 1989). In other words, coordination mechanisms aim to establish reciprocal predictability of actions among interdependent actors (Puranam et al., 2010). On HS2 this was especially important because information was localized and embedded in each specific station location. Therefore, through exchanging information, the promoter gained access to first- hand low-cost knowledge about multiple local situations. This predictive knowledge

140

(Puranam et al., 2012) was, in turn, crucial so that the promoter could pre-empt core actors’ complaints and concerns and address them ahead of parliamentary proceedings.

Developing common ground across interdependent local groups reduces their temptation to freeride, skirt commitments, and violate rules (Ostrom, 1990; Srikanth &

Puranam, 2014; Knudsen & Srikanth, 2014; Clark, 1996; Schelling, 1960). Reaching a common understanding also reduces the investment necessary for enforcing and monitoring actions, and imposing sanctions on those who take unauthorized actions

(Ostrom, 1990). Taking such action minimized the number of petitions against the scheme and in turn the likelihood of a protracted parliamentary process. We conjecture that without informational exchange across organizational boundaries the planning process would have been incomplete at best and impeded or stopped at worst.

Thus, in agreement with ideas in organization design (Knudsen & Srikanth, 2014;

Srikanth & Puranam, 2014; Schelling, 1960; Simon, 1945) our findings suggest that resolving the coordination problem revolved around creating mechanisms which, firstly, allowed for the exchange of information across organizational boundaries; and secondly, allowed for interacting actors to pre-empt each other’s actions and calibrate their own when necessary. In the HS2 context, this was done through the creation of technical working groups and the use of strategic documents, which incorporated both project and local level knowledge and information. Both mechanisms permitted coordinated action across a large number of interdependent actors.

Working groups and strategic documents made provision for the acceptance and assimilation of information for creating shared solutions (Simon, 1996: 144). By addressing information processing requirements as they arose, the two mechanisms also reduced the uncertainty on the scheme (Tushman & Nadler, 1978; Lawrence &Lorsch,

141

1967; Gulati Lawrence & Puranam, 2005; Heath & Staudenmayer, 2000). Next, each of these mechanisms is considered in more detail.

6.4.1. Technical Working Groups

Our findings suggest that inter-organizational technical working groups are a crucial coordination mechanism for reducing uncertainty during the collective development process. These mechanisms allow core actors to establish a shared conception of the technical problem at hand. They are also important as they set the foundation for the norms of engagement in ensuing interactions. During the project planning process, these structures provided settings for communication, mutual observation, learning and (joint) decision-making as outlined in the organization design literature (Galbraith, 1973; March and Simon, 1958; Tushman & Nadler, 1978).

Inter-organizational structural linkages were especially important for exchanging information across organizational boundaries at the outset of the project planning process.

Without these channels of communication, the relevant information would otherwise have been unavailable to each of the core actors, including the promoter. Collective coordination structures, therefore, permitted access to a broader array of knowledge.

Furthermore, through face-to-face interactions, technical actors were able to develop mutually agreed upon station shortlists that would support the final decisions taken by strategic leaders. They were also able to determine the trade-offs associated with each of the options and challenge each other’s assumptions, data and analysis. Ultimately, coordination efforts allowed actors to refine and extend the evidence base that would support decisions, in turn reducing task, project mission and environmental uncertainty.

Interestingly, our analysis suggests that overcoming coordination problems during the collective development process does not necessarily require more information, but

142

rather that people receive the most important and relevant information for the decisions they will make. For HS2 Ltd., for example, relevant information would typically relate to the overall fit of proposed choices with the organization’s remit, operational and engineering feasibility as well as demand. This finding is wholly consistent with Simon’s claim that “the task is not to design information-distributing systems but intelligent information-filtering systems” (Simon, 1996: 144). To that end, our findings also suggest that another aspect of the coordination process was vetting the quality and level of information exchanged. During the HS2 planning process, the promoter did this by being selective about who joined the technical working groups and limiting participation to senior officials, senior planners and senior heads of transport, or other actors with direct access to local leaders.

Through face-to-face interactions, the well-informed core actors were able to reduce task uncertainty by narrowing the parameters of the solution space. They were also able to reduce environmental uncertainty through contributing diverse perspectives about how to navigate the environment. An example of the latter is the creation of the scrutiny board for the HS2 station in Birmingham, a structure through which local actors, who were accountable to the local community, were able to interrogate the promoter on the merits of

HS2. This suggests that the membership of collective development groups is crucial because actors who can constructively examine and challenge each other’s assumptions, data and analysis ultimately contribute to the collective interest.

While we would logically expect that a higher number of actors in the technical working groups would result in a greater information processing load, this was not necessarily the case. Working groups with single local authorities did not consistently exhibit effective information processing nor did those with multiple local authorities consistently exhibit ineffective information processing. This paradox is particularly clear

143

when we consider that although the local group in Manchester comprised multiple councils, information processing in the group was also the most effective. To that end, we recognize that additional factors can complicate information processing during the collective development process. Some such factors include a lack of prior collaboration experience on which actors can ground their expectations and predictions (Gulati, 1995); or limited time to forge relationships (Nordqvist et al., 2004). Both factors invariably complicated the coordination effort between the promoter and the respective local authorities and overcoming the ensuing challenges required continued interaction.

Technical working groups can, therefore, be seen as forums with the specific purpose of creating common ground by accessing the dispersed knowledge and expertise that lie outside of the boundaries of a single organization’s boundaries (Knudsen & Srikanth, 2014;

Srikanth & Puranam, 2014). In ideal working groups, there is a sense that sharing and free revealing of information are mutually beneficial. However, interdependence and the pressure toward joint-decision making can also introduce the possibility of intergroup disagreement and conflict related to a misalignment of interests (Pettigrew, 1973,

Ericksson et al., 2017; Levitt, 2015).

Ostrom (1990, 2005) suggests that to resolve such conflict, actors need to holistically examine both the conflicts and the context within which they are embedded. In so doing, they should take into account knowledge and information about participants, strategies, technological changes, economic relationships and (in the context of natural resources) the underlying ecological conditions. Taking these factors into consideration increases the chances that a collective situation can be changed more effectively from one in which appropriators act independently to one in which they adopt coordinated strategies to obtain higher joint benefits (Ostrom, 1990, 2009). Our findings are in concurrence with these ideas. Overcoming the challenges in both Sheffield and London, for example, required an

144

understanding of the social context and of the various local actors’ motivations, including, that the issues in Sheffield were deep-seated in a pre-existing local rivalry.

Representing knowledge and information across both functional and organizational boundaries is both a challenging and complex undertaking (Srikanth & Puranam, 2014;

Carlile 2004: Dougherty 1992). Such an endeavour can be both difficult and inefficient for a least two reasons. Firstly, the increased specialization within boundaries impedes communication and hinders the free flow of information (Lawrence & Lorsch, 1967;

Dearborn and Simon, 1958; Wilensky, 1967; Allen and Cohen, 1969, Rogers and

Shoemaker, 1971). Secondly, the particular perspectives that different actors generate can act as barriers to the integration of ideas (Carlile 2004, Dougherty 1992). Furthermore, making consequential architectural dependencies visible is one of the biggest challenges in complex technology development settings. Hence, our findings show that addressing information processing challenge through the creation of coordinating structures takes varying degrees of time and effort (Henderson and Clark, 1990, Carlile, 2002; Srikanth &

Puranam, 2014). In addition, we find that multiple coordination mechanisms are required in addition to inter-organizational structures. To that end, we discuss boundary objects next.

6.4.2. Boundary Objects

Our findings show that strategic documents were a critical coordination mechanism.

These documents are an example of “boundary objects”: objects that help to create a shared context that “sits in the middle” (Star & Griesemer, 1989: 451; Carlile, 2002;

Carlile, 2004; Teurtscher, Garud & Kumaraswamy, 2014). In the HS2 case, our findings suggest that as boundary objects, strategic documents enabled actors to represent tacit knowledge or knowledge that was yet to be codified. Such knowledge is by definition

145

more difficult, slow and costly to transfer (Srikanth & Puranam, 2011; Carlile, 2004;

Kogut and Zander, 1992, Zander and Kogut 1995).

The planning process was therefore characterized by the continual emergence of new documents, which contributed to the technical advancement of the scheme and the advancement of collective action. Strategic documents were crucial for outlining the knowledge that informed these advancements. These documents codified the most important information from different actors’ points of view and were therefore important for reducing uncertainty on the project as matters grew and interests diverged (Galbraith,

1977; Stinchcombe, 1990). Of the many documents published by the promoter, the most important were arguably those articulating the case for HS2 and detailing the progress of the scheme. These documents established a common orientation, which was crucial for mobilizing collective action, particularly in the absence of unified authority (Srikanth &

Puranam, 2011). In turn, documents published by local authorities were also important for detailing local strategy and planning aspirations as they related to HS2.

It is thus reasonable to say that during the collective development process strategic documents served as boundary objects for the representation and exchange of knowledge across organizational boundaries (Star & Greisemer, 1989; Carlile, 2002; Carlile, 2004;

Srikanth & Puranam, 2011, Teurtscher et al., 2014). The promoter’s documents, in particular, contributed to the articulation of agreed-upon facts and shared problem definitions, which provided a common basis for discussion and negotiation (Srikanth &

Puranam, 2011). Developing this common knowledge would have been otherwise challenging given that the necessary project-wide and local knowledge was dispersed across multiple independent actors (Carlile, 2002). Technical drawings, such as the route outline, also served as a “social glue” around which core actors and actors in the wider environment could discuss the station design and the design decision-making process

146

(Henderson, 1991). Strategic documents, therefore, made it possible to exchange knowledge, assess each other’s knowledge and transform knowledge to better address the design problem (Star & Greisemer, 1989; Carlile, 2002).

Up until this point, we have established that boundary objects create ‘mutual ground’ and facilitate information sharing. To that end, artefacts such as strategic documents as well as maps and technical drawings establish a shared context across different problem- solving contexts (Star, 1989; Carlile, 2002; Bechky, 2003). However, extant literature depicts these objects as concrete, stable and unchanging although subject to “interpretive flexibility” (Star, 1989). Once an object is created it is said to be “both plastic enough to adapt to local needs and the constraints of several parties employing it, yet robust enough to maintain a common identity across sites (Star and Griesmer, 1989: 393). However, our findings suggest that boundary objects can also evolve and take character over time. We attribute this evolution to actors’ bounded rationality and thus the need to adapt the form of the boundary object as new information emerges (Simon, 1996; Gavetti et al., 2012;

Puranam et al., 2015; Ostrom, 1990).

Over the course of the planning process, published strategic documents challenged the data, analysis and assumptions used by core actors. In Sheffield, for example, local actors published a document with the purpose of “refining HS2 Ltd.’s analysis” (Sheffield

City Council, 2012). Strategic documents, therefore, contributed to the richness of the collective development process as they provided information that could be negotiated and transformed in order to reach new agreements. Yet, the publication of strategic documents by multiple actors raised the risk of information overload, ineffective communications and undermined creativity, a phenomenon Langley (1995) calls “paralysis by analysis” to acknowledge situations where more information than necessary is collected and processed

(Denis et al., 2011; Feldman & March, 1981; Langley, 1995; Brunsson, 2000). To address

147

this risk, we propose that nominating a third party who requests the required information, such as the HS2 Growth Taskforce, could contribute to advancing collective action. Such a body could also monitor and assess this information for quality, relevance and strategic alignment, which would serve to strengthen the collective development process.

Among core actors, strategic documents, therefore, played a crucial role in coordinating diverse knowledge, conveying information between groups and mobilizing action to resolve the consequences that exist at inter-organizational boundaries (Star and

Griesemer 1989; Carlile 2002; Bechky, 2003, Henderson 1999). Beyond the core actors, however, strategic documents also contributed to the reduction of environmental uncertainty by keeping external stakeholders in the environment informed about the project’s progress. For HS2, this was done through the online publication of multiple reports documenting the route choices and how those choices were made, as well as the publication of business case updates, compensation plans, the technical specifications of the proposed stations and so on. Adopting a similarly transparent approach to collective development processes, arguably, contributes to legitimating the process in the eyes of both core actors and third parties.

Our findings suggest that the capacity of strategic documents during the collective development process is two-fold: to inform and to persuade (Carlile, 2002). Thus, while strategic documents are used to establish a shared means for representing and specifying differences, there is also a political aspect to them such that “creating and reshaping boundary objects is an exercise of power that can be collaborative or unilateral” (Boland and Tenkasi, 1995: 362). Strategic documents can, therefore, be used for lobbying purposes, as was the case in Sheffield with Sheffield City Council using documents to advocate for a station in the city centre. Documents can also be used for expressing and legitimizing a position on a contentious issue, as was the in the case of London where the

148

Council published documents opposing the scheme, and both London wide authorities and the Council published consultation responses, petitions and masterplans. It is therefore imperative to reconcile the informative and political dimensions of such documents and to filter out the information that leads to unproductive social processes.

For the promoter, strategic documents were especially important for consolidating divisions and misunderstandings while developing the Hybrid Bill that would ultimately be deposited in Parliament. However, while necessary, reconciling and integrating information and ideas is not sufficient for heterogeneous actors to build consensus. In addition to this, interests and action must be aligned, which can be fostered through the use of cooperation mechanisms. We discuss these next.

6.5. Resolving the Cooperation Problem

During the process of collective development, coordination induces but does not guarantee that a group of interdependent actors operating under a shared goal will invest time, effort and knowledge in order to reconcile their differences (Gulati et al., 2012).

Inertia in this regard can arise if participants opt to advance their private interests at the expense of the collective interest. This misalignment of private and collective interests gives rise to a social dilemma, which can escalate to a “tragedy of the commons” if left unaddressed (Olson, 1965; Hardin, 1968). In an inter-organizational context, such a tragedy can be triggered by cooperation failures wherein “organizations…shirk by contributing less than agreed or try to claim more benefits than agreed via misappropriation of partner resources or [collective] outcomes or via holdup, the exploitation of a superior bargaining position to negotiate more favourable terms” (Gulati et al., 2012: 7). Aligning interests and actions during collective development is, therefore, imperative for averting the deterioration of the collective effort or its ultimate dissolution.

149

During the HS2 planning process, the tension between private and collective interests arose because station designs would have to address two higher-order goals i) resolve a transport capacity problem on the wider rail network, and ii) catalyze economic growth and regeneration within regions. Importantly, the two goals did not coexist from the outset. Instead, the second goal emerged over the course of the collective development process. The promoter was fundamentally invested in advancing the former; however, through advocacy and lobbying, local authorities were able to persuade the promoter of the importance of the latter over time. Reconciling these two goals, and the ensuing conflicting interests, was seldom a trivial undertaking. The process of aligning interests was further complicated by the fact that the local solution space was constrained by more global performance targets and public announcements. Nevertheless, at best, left unaddressed misaligned interests could lead to severe conflicts between core actors, weakening commitment and overall collaborative relations. While at worst, a “tragedy of the commons” could occur wherein the cost of the scheme, borne to the taxpayer, reached exorbitant heights or local authorities defected and formed a coalition to oppose HS2 and ultimately lobby Parliament to halt the project.

Our findings suggest that HS2 Ltd. played the role of a systems integrator: an actor with an explicit interest in integrating effort during the development process (Brusoni, et al., 2001; Prencipe, et al., 2003; Hobday, et al., 2005). Much like the lead firm in business ecosystems (Baldwin, 2012), a systems integrator is an “architect, catalyst and guide” that simultaneously directs participants towards the overarching system-level goal and shapes the environment around it (Williamson & De Meyer, 2012). In the absence of command- and-control authority, HS2 Ltd. used a range of mechanisms to encourage decision-making participants to cooperate and collectively search for consensual solutions. These mechanisms differed somewhat from those adopted by unitary organizations when

150

awarding inducements and incentives such as promotions, reputation and budgets (Simon,

1951; Gibbons, 1998; Puranam, Alexy & Reitzig, 2014). Instead, we contend that they revolved around (even if not restricted to) the creation of shared rules and structures legitimated over time (Ostrom, 1990, 2005).

During the HS2 planning process, and as a last resort, the promoter also encouraged cooperation through the use of significant concessions. In Euston, for example, the promoter decided to redevelop the station area. This redevelopment would likely cost over

£4.5bn, more than double the cost of the initial proposal, however, it led to Camden

Council dropping their petition against the scheme. While such concessions can be seen as a form of inducement, the strategy of awarding them was not uniformly adopted. Instead, concessions were awarded cautiously given the financial and scheduling constraints associated with the scheme. Moreover, these constraints were monitored by independent watchdogs at the national level. Thus, HS2 Ltd. was careful not to give out too many concessions early on to avoid setting an unsustainable precedent and to reduce the risk of prematurely running out of slack resources, notably contingency funds. Therefore, while in unitary organizations inducements and incentives are defined a priori with the expectation that employees will cooperate (Puranam, Alexy & Reitzig, 2014), HS2 Ltd. only gave out concessions when all other avenues for fostering cooperation had been exhausted. This approach is in agreement with Gil and Pinto’s (2015) suggestion that slack resources in meta-organizations are useful for bridging differing goals without resolving conflict and thus facilitating a “quasi-resolution” of conflict (Cyert and March, 1963).

Our findings, therefore, suggest that HS2 Ltd.’s main approach for fostering cooperation was through the creation of governance structures, and through collectively delineating a sufficiently inclusive and negotiable solution space from which final decisions could be made. Cooperation efforts involved substantial amounts of dialogue and

151

communication aimed at finding ways to address concerns and provide assurances ahead of parliamentary proceedings. For example, to resolve problems at Euston, multiple strategic boards with overlapping purposes and missions were created.

The use of governance structures to encourage cooperation is in line with the collective action literature, which argues that robust governance structures facilitate cooperation. Robustness is reflected in rules and structures that incentivize actors to cooperate and assures them that others are doing the same (Ostrom, 1990). Governance structures are also essential for creating multiple opportunities for face-to-face interaction, which generates social norms, rules of governance, goodwill, civility and mutual respect.

In turn, these contribute to increasing joint returns (Ostrom, 2009; Ostrom et al., 1994;

Ostrom, 1998; Ostrom et al., 1994; Sally 1995; Balliet 2010).

Through the use of sufficiently inclusive governance structures and a constrained but negotiable solution space, actors in the HS2 meta-organization found ways to accommodate the array of interests in each station working group. More specifically, the creation of multiple forums complementing the technical working groups helped to reconcile actors’ inconsistent and incompatible interests. For example, strategic boards gave leaders from the various independent organizations an opportunity to resolve differences collectively and to determine future action. These boards were therefore reasonably advantageous for all the actors in the meta-organization core.

During the collective development process, fostering cooperation among interdependent actors is crucial for creating political capital that can be wielded for mutual gain (Ghemawat & Khanna, 1998). Without adequate support from the local authorities, for example, there was the risk that HS2 would not progress through parliamentary proceedings in a timely manner, with significant implications for both the project costs and schedule. At the extreme, there was the risk that the planning process could unravel. Thus,

152

encouraging norms of cooperation was imperative. Next, we describe three mechanisms used to foster cooperation i) a polycentric and adaptable organizational form, ii) negotiable technical and financial resource boundaries, and iii) appeals to the proportionality of costs and benefits.

6.5.1. Robust Meta-Organization Boundaries 6.5.1.1. Polycentric Structures

The project organization created by HS2 provides a strong example of a polycentric structure. As previously discussed, polycentricity, a key principle in Ostrom (1990)’s theory of collective action, denotes a multi-level governance structure comprising different decision centres that have specified areas of authority in regard to which they need not defer to others (V. Ostrom, Tiebout & Warren, 1961; V. Ostrom, 1972; V. Ostrom, 1999a,

1999b). A polycentric structure differs from a monocentric one, as in unitary organizations, in that the former has multiple decision centres that are not formally linked, whereas the latter is governed by hierarchical authority. During the HS2 planning process, the multiple decision centres were evident in both the technical working groups and strategic boards across the different cities, and at different inter-organizational hierarchical levels.

In line with Ostrom (2009, 1999: 39), we propose that in collective development settings such a polycentric structure is desirable for at least three reasons: First, with different decision centres, mistakes are less likely to have systematic effects compared to the context of a single rule-making body. Hence in London, for example, clashes between the promoter and the local council were confined to local decision-making bodies and did not impede decision-making progress in Manchester. In this regard, polycentricity resembles the idea in organization design of finding ways to decompose complex systems to reduce management complexity (Simon, 1962, Baldwin & Clark, 2000; Levitt, 1984).

Second, a polycentric structure is desirable because adaptation can be speedier given that 153

information about what has worked well in one setting can be transmitted to others that may try it in their settings without waiting for approval from an overarching authority or majority. A case in point was the way cities like Leeds and Sheffield followed

Manchester’s example by also creating local governance structures that complemented those created by the promoter. Third, when small systems fail, there are larger systems to call upon, and vice versa. Arguably, the latter works particularly well in a pluralistic setting where there are multiple interests, values and perspectives to draw from (Denis et al., 2011).

The analysis suggests, however, that the local structures and arrangements created within the polycentric system had varying levels of effectiveness when it came to resolving conflicts and encouraging cooperation. For example, local structures created in

Manchester, Leeds, and Birmingham were more effective than those created in London and Sheffield, which were subject to intermittent breakdowns in communication. With this in mind, we conjecture that the observed polycentric system could have been more robust were impartial, low-cost conflict resolution mechanisms and monitors available across all local structures. These are principles espoused by collective action scholars.

To that end, Ostrom (1990) argues that having direct recourse to an appointed official on the ground for the sake of conflict resolution would improve the capacity to organize and ultimately the capacity to achieve collective action. Plausibly, adding local level conflict resolution mechanisms and third-party monitors at the local level could have helped to resolve more conflicts before they escalated, which would, in turn, limit the extent of so-called escalating indecision (Denis et al., 2010). Although the Secretary of

State was able to advise actors when they reached an impasse, their impartiality was debatable given that Cabinet Ministers were championing HS2. Furthermore, while the

National Audit Office (NAO) and Public Accounts Committee (PAC) were monitoring the

154

project, and core actors could approach the Courts for legal recourse, these bodies were third party to both the disputes and the meta-organization as a whole.

Resolving the cooperation problem during the collective development process in pluralistic settings thus requires that structures and rules be created to steer large numbers of autonomous actors in a common direction. Effective governance structures, in turn, attenuate cooperation failure as a result of shirking, misappropriation, hold up and similarly opportunistic behaviour (Gulati et al., 2012). In the HS2 case, we found that averting such problems required a range of different actors. Therefore, next, we explain why adaptive participation boundaries were important.

6.5.1.2. Adaptive Participation Boundaries

Contingency theory posits that different approaches to structuring organizations can have different levels of effectiveness under varying conditions (Burns and Stalker, 1961).

As such, organizations should be designed to fit the complexity and dynamism of both the organizational and external environment (Burns and Stalker, 1961; Chandler, 1962;

Galbraith, 1973; Miller and Hobbs, 2009). Thus, organizations should be adaptive or otherwise flexible, responsive and resilient (Doz and Kosonen, 2008, Martinsuo, Hensman,

Artto, Kujala, & Jaafari, 2006, Morris et al., 2012). Such characteristics improve the organization’s ability to dynamically adapt as necessary. Organization design scholars also posit that an adaptable organization is one that adopts the appropriate designs at any given point in time for a given mission, task, or set of circumstances (Henderson and

Newell, 2011; Puranam & Swamy, 2016, 2016; Knudsen & Srikanth, 2014; Ethiraj &

Levinthal, 2004). As long as the performance of a given structure does not match up with pre-defined aspirations, an organization adapts through exploration, or a problemistic search (Knudsen & Srikanth, 2014, Clement & Puranam, 2017; Gavetti et al., 2012;

Puranam et al., 2015; Greve, 2008). 155

In the collective action literature, adaptive governance is similarly important given that on the one hand, the efficacy of a specific set of rules and structure can erode over time, and on the other hand, rules and structures may need to be reinforced because

“humans devise ways of evading governance rules” (Dietz, Ostrom & Stern, 2003: 1908).

Adaptability is also important because trial-and-error adaptation increases the chances of discovering beneficial organizational forms (Ostrom, 1990, 2009). As knowledge of different kinds of rules, institutions and technologies is discovered, it can be combined in order to internalize externalities through a dynamic evolutionary process (Ostrom, 2005).

Thus, both bodies of literature deem adaptability desirable even if they implicitly suggest that it occurs through different processes: from one organizational form to the next in response to changes in the environment (organization design literature) and through a dynamic evolutionary process as the organization takes form (collective action literature).

Wholly consistent with both the organization design and collective action literature, our findings suggest that adaptability is similarly important for collective development in pluralistic settings. As such, during the HS2 planning process, the promoter adopted a trial- and-error approach when creating inter-organizational linkages and was also sensitive to the environment, changing the structure of the project organization accordingly once it took form.

Collective action literature suggests that robust structures adapt a fixed set of participants to a dynamic environment (Ostrom, 1990). In the literature, fixed participation boundaries are seen to contribute to the creation of robust rules and structures (Ostrom,

1990). Contrastingly, our analysis suggests that changes in participation boundaries contribute to the robustness of the collective development process in a pluralistic setting, and, thus, that there is scope to relax these assumptions. One explanation for this is that introducing different participants into the process can contribute to a ‘pragmatic’ problem-

156

solving approach: as issues arise or circumstances change, decision-making can be deferred to the actors identified as most suitable for making said decisions or providing the resources that can avert impasse. Issues can also be escalated to higher order actors with the mandate to relax constraints when necessary (particularly when lower order actors lack similar administrative authority). Furthermore, different actors may be more suited to facilitate the resolution of emerging disputes and monitor the collective development effort.

Such was indeed the case during the HS2 planning process where conflicts were heard by higher-order actors such as the HS2 Rail Board or the Secretary of State, and so meta- organization actors did not always have to resort to the court system.

Our findings suggest that dispute resolution hinged on switching back-and-forth between the different actors in the meta-organization. This switching was important for fostering cooperation as it allowed multiple project participants in overlapping structures to obtain and integrate reliable information (Sanchez & Mahoney; 1996; Clark & Fujimoto,

1991; Clark & Wheelwright, 1993). Switching between actors also allowed them to monitor outcomes of the planning process, where monitoring is another of Ostrom’s (1990) principles. Initially, the switching took place between the technical working groups and the strategic boards, for example, and then hierarchically between the strategic boards, the

High-Speed Rail Board and the Secretary of State. Moving between the multiple levels of governance was particularly instrumental in avoiding communication breakdowns and thus advancing collective action. This finding is consistent with the power of polycentric structures to resolve disputes (Ostrom, 2010).

Crucially, governance structures were not solely changed by HS2 Ltd. Instead other core actors, namely local authorities, were also able to create structures and to invite actors into these structures as they saw fit. This mutual effort to organize is in line with the assertion that in robust collective action arenas, those affected by the rules can participate

157

in modifying them (Ostrom, 1990). In Birmingham, for example, local actors created a scrutiny board to assess HS2’s growth and regeneration potential. When considered together, all the structures and arrangements contributed to cooperation as they served as forums where core actors were able to monitor each other’s behaviour and consolidate divisions in interest, values and preferences (Ostrom, 1990; Denis, Langley & Rouleau,

2007).

The HS2 case demonstrates that an adaptable polycentric structure is instrumental in encouraging mutually beneficial patterns of interaction and outcomes. Nevertheless, we appreciate that “no humanly designed, complex system can be fully planned to achieve optimal performance” (Simon, 1990 paraphrased by Ostrom, 2005: 2007) and that “there is no guarantee that we can take the steps, including exercises in political design and management, needed to alter the operation of prevailing arrangements in a well-planned fashion” (Ostrom & Dietz, 2001: 286). To that end, in addition to creating structures, our findings show that other mechanisms were important for fostering cooperation, and ultimately the integration of effort towards collective ends. These mechanisms included negotiable resource boundaries and appealing to the proportionality of costs and benefits.

We discuss both mechanisms next.

6.5.2. Negotiable Resource Boundaries

During the collective development process, multiple actors come together to develop a designed artefact. Generally, designing artefacts requires complex information- processing and that the basic function of the artefact be reconciled with its desirable qualities (Simon, 1996; Baldwin and Clark, 2000; Ulrich, 2011; Crilly 2010). This process involves building associations between the technical ‘internal environment’ of the artefact and the ‘external environment’ within which it is embedded (Simon, 1996). In so doing, it

158

is invariable that design choices will evolve and will only tend toward completeness as new knowledge is combined and recombined. Changes will result from feedback mechanisms, through a process of calibration between the systems actual and desired states, and through a process of calibration between the system and the environment (Simon,

1996). Thus, during collective development in pluralistic settings, augmenting the design scope is important for technical advancement.

Based on our findings, we propose that augmenting the design scope can also foster cooperation if it reduces conflict among actors. Indeed, during the HS2 planning process, relaxing the resources boundaries was important as it made provision for the incorporation of the local actors’ priorities and interests into the final design choices. In London, for example, the scope of the station oscillated between small and large because of the cost implications of a complete overhaul. In the end, the core actors settled on a costly redevelopment of the station site as part of an agreement with Camden Council. In exchange, the Council agreed to drop its petition against the scheme.

This notion of flexible resource boundaries is in contrast with Ostrom’s (1990) claim that fixed resource boundaries are important for sustaining cooperation. However, Ostrom

(1990) makes this claim in the context of natural resources that already exist. Contrastingly, in the context of collective development, an artificial resource or artefact is being created over time (Simon, 1996; Baldwin & Clark, 2000). Thus, our findings suggest that during the process of creating an artefact, negotiable boundaries are desirable as they allow the collective to embrace both exogenous and endogenous changes. In pluralistic settings where multiple actors have inconsistent beliefs and heterogeneous or conflicting interests and needs, negotiable resource boundaries are also desirable as they help actors reconcile these differences, attenuate conflict, and in turn, they contribute to the robustness of the meta-organization.

159

While negotiating the resource boundaries for HS2, the promoter’s preferences consistently served as the initial reference point against which other changes were assessed

(Kahneman, 1992). As contributions from other core actors were incorporated, the reference point changed. Challenges arose in this regard if the resulting changes were perceived as relative losses by some actors in the core. One explanation for this is loss aversion, i.e. people’s tendency to weigh losses more heavily than foregone gains of equal magnitude (Kahneman & Tversky, 1979; Hardin, 1982). In the extreme, as in London and

Sheffield, the result was a breakdown in communication and delayed progress in reaching a collective end.

Nevertheless, our findings show that local groups typically undertook multiple design iterations before a final design choice was reached. Thus, in line with Simon (1996:

143) we argue that at the start of collective development processes, “what is needed is not a “correct” conceptualization so much as one that [can] be understood by all participants and that [will] facilitate action rather than paralyze it.” We further argue that the number of iterations necessary before a final design choice is reached depends in part on the level of information within the group and on the extent to which actors’ technical preferences converge. Manchester, for example, was able to agree on a collective design choice in fewer iterations than London given that local project participants acted in collaboration.

They were partly able to do this because they could produce information that inspired a sense of certainty in the promoter about the efficacy of their proposals and the accuracy of locally produced cost-benefit forecasts.

Unsurprisingly, our findings show that a higher number of actors laying claim to single, indivisible design choices complicates the collective development process (Gil &

Baldwin, 2013). Complicating matters even further, and also making boundary relaxation a necessity, is the fact that changes in technical resource boundaries are frequently reflected

160

in financial resource boundaries. Hence, a decision to relax technical resource boundaries to resolve local disputes can lead to substantive cost overruns with negative implications for the performance of the meta-organization in the eyes of third parties. The promoter pre- empted this in the Sheffield case, for example, and opted not to build a new station in the city centre because doing so would require a significant increase in costs. Thus, to sustain the collective development effort, consistently weighing additional costs against potential benefits is necessary as are trade-offs and compromises.

Our findings thus suggest that flexible resource boundaries are important for fostering cooperation. Negotiable resource boundaries, as a rule, reflect the willingness to search for a concept that satisfies an array of core actors (Simon, 1996; Ulrich, 2003). By incorporating actor’s interests into design choices, conflict is attenuated, and bottlenecks during the collective development process are removed. Removing bottlenecks, in turn, opens up the opportunity to capture and create value (Baldwin, 2015). However, as

Baldwin (2015) argues, successfully managing bottlenecks requires an understanding of both the technical architecture of the system and the rich network of interdependent relationships within which it is embedded.

For large-scale infrastructure projects, strategic design choices influence the cost- benefit analysis that partially informs decisions to make the capital investment, as well as the budget and timescale that inform decisions to allocate resources to the scheme. Hence, it is important to balance the benefits associated with achieving collective action with the costs thereof. Achieving this balance or at least a perception of this balance is crucial for legitimatizing the process and ultimately for maintaining the collective development effort

(Ostrom, 1990). In the next section, we discuss this topic in more detail.

161

6.5.3. Appealing to the Proportionality Costs and Benefits

Our final proposition is that during the collective development process in pluralistic settings, it is important to maintain a shared perception that the benefits that actors appropriate are congruent with the costs that they accrue. We make the distinction between the perception of congruence between costs and benefits and actual congruence because, as the HS2 case demonstrates, determining the exact costs and benefits of a future-oriented endeavour can be challenging. Nevertheless, when there is congruence between the rules that assign costs and the rules that assign benefits, participants view the undertaking as fair and legitimate (McKean, 1992). In turn, this perception of legitimacy assures actors that they will benefit from cooperation and, therefore, from making decisions in the collective interest (Ostrom, 1990; Ostrom, 2009; Gibson, 2001). In contrast, if participants do not perceive the situation as fair and legitimate, they will be uncertain about the level of opportunistic behaviour that they can expect from other participants. The perceived inequity threatens the collective endeavour as it can frustrate actors to the point that they refuse to voluntarily contribute resources given that they believe the payoff structure is unfair (Ostrom, 1990; Williamson, 1975).

Without a fair, orderly, and efficient method of allocating collective resources, participants have little motivation to contribute to the continued provision of the resource system (Ostrom, 1990, 2009). Thus, encouraging proportionality can be valuable even when measurements of proportionality are debatable. During the HS2 planning process, one way through which HS2 Ltd. sought to maintain the perception of congruence between costs and benefit was by insisting that all proposals made by project participants meet value-for-money requirements. Although most proposals were technically feasible, HS2

Ltd. was cautious about granting concessions, which helped to maintain a sense of fairness on the project.

162

By not easily acquiescing to demands as they arose, the promoter was also able to avoid setting an unsustainable precedent. However, maintaining congruence during the planning process still proved challenging for at least two reasons. Firstly, as already mentioned, HS2 is a future-oriented project that will take decades to implement and that will ultimately be long-lived, it was, therefore, challenging to establish exact costs and benefits because it is hard to predict the future state of the world. The long planning horizon added a layer of complexity especially given that benefits will only be appropriated in the distant future. Secondly, the evolving technical resource boundaries meant that knowledge about the structure of the resource system was itself evolving, and by implication so were the associated costs and benefits. This dynamism, in turn, added another layer of complexity to the problem.

As outlined in the discussion about coordination, uncertainty and complexity are reduced over time as information is exchanged and processed (Galbraith, 1973; March and

Simon, 1958; Tushman & Nadler, 1978, Puranam et al., 2012). This was also the case with regards to the HS2 costs and benefits. Even so, and as expected in collective action settings, the completeness, accuracy and symmetry of information available to participants varied across the local groups (Ostrom, 1990). In the collective action setting this variance is attributed to, the costs of acquiring information, the number of participants involved, the complexity of the situation, and the stability of factors affecting both the participants and the resource system (Ostrom, 1990, 2005). These factors are similarly applicable in the context of collective development, although to varying degrees. Our findings suggest, for example, that of all the local groups, Manchester had more accurate information because actors in the city had a history of collective decision-making.

Furthermore, the GMCA had the corresponding structures in place to determine the boundaries of the local ‘resource system.’ Local actors in Manchester could also

163

convincingly estimate the costs and benefits associated with their proposed design changes.

It is thus not by chance that Manchester managed to produce a range of analyses aimed at demonstrating the legitimacy of their proposals and influencing design choices in their favour. HS2 Ltd., in turn, viewed this work as legitimate enough to inform changes to the concept. The outcome was different in Sheffield, for example, where local analysis was viewed less credibly.

Thus, as in other pluralistic settings, apart from being used informationally and rationally, analysis was also used as a tool of persuasion (Langley, 1991; Tushman &

Rosenkopf, 1992). In their case, Manchester was persuasive. The use of analysis as a tool for influence is explained by March and Simon who predicted “that almost all disputes in organizations will be defined as problems in analysis, and bargaining (when it occurs) will frequently be concealed within the analytic framework” (March & Simon, 1958: 131).

Formal analysis is similarly important in pluralistic settings in which people cannot quite trust one another (Langley, 1989:609). With the weight given to both perceived costs and benefits, it is, therefore, unsurprising that cost-benefit analyses are contested through counter-arguments and other kinds of analysis from those with competing interests

(Feldman & March, 1981; Brunsson, 1982, 2000, Langley, 1995).

When actors exhibit divergent value systems, there is a considerable amount of trial- and-error learning (Langley, 1991) in order to agree on a value conceptualization. Hence, during the collective development process, many decisions are taken without full knowledge of their consequences given the level of uncertainty about the costs and benefits.

Left unmonitored, this can lead to consequential errors. We, therefore, argue that in the absence of unitary authority, governance structures include impartial third parties that monitor the propositions, data, analysis and assumptions presented in collective development settings. This proposition is in line with Ostrom’s (1990) claim that robust

164

collective action arenas require impartial third-party monitors. In the HS2 setting, the

NAO and PAC adopted this role. By challenging the promoter when necessary, both bodies were important for maintaining the transparency and legitimacy of the system. We, therefore, conjecture that having a similar mechanism at the local level would contribute to the robustness of the collective development effort. We foresee that such a body could force out ill-conceived proposals and analyses that would otherwise go unopposed because errors and biases were both unchallenged and undetected.

Our findings further suggest that while maintaining the perception that costs and benefits are proportional, managing costs is especially important. We argue that in the collective development setting one way to do this is to convince actors that the collective resource they share is limited and that all resource-related decisions are interdependent.

This approach serves to have actors perceive the vulnerability of the resource, the interconnectedness of their fates, and to implicate themselves as part of the problem if things go wrong. The argument in natural resource settings is that when participants are aware of their interdependence, they will constrain the use of the resource in a manner that they agree is workable and fair instead of acting independently (Gibson, 2001; Ostrom,

1990). During the HS2 planning process, the participants seemed to be aware of their interdependence or at least that financial resources were limited. Hence, they invested in making their case for alternative proposals. Nevertheless, getting core actors to act on this awareness was not always easy.

By inviting local authorities to make a financial contribution towards the project,

HS2 arguably sought to offset the high costs of concessions and maintain the congruence between costs and benefits. However, maintaining this proportionality was sometimes seen as the Government’s responsibility. To that end, actors on the project were faced with

“the paradox of perverse incentives” because although they had a strong influence on the

165

total cost, they had no direct financial obligation (Welde & Odeck, 2017: 627). Indeed, local actors lacked incentives to make concessions given that the Government was financing the whole project. We, therefore, propose that maintaining the perception that costs and benefits are proportional in the context of collective development, and maintaining collective action more broadly may, in fact, require disproportional resource contributions from individual actors (in this case the Government) to keep the system in equilibrium.

Maintaining the perception that costs and benefits incurred by each core actor are proportional is important so that other core actors in the meta-organization view the collective development process as fair and legitimate (Ostrom, 1990, 2005). However, doing so can be challenging in the context of future-oriented human-made artefacts where the measurement of costs and benefits is disputable, and related analysis is used to defend and bargain for private interests and to exert influence as much as it is used to resolve conflict. We, therefore, propose that third parties who monitor the data, analysis and assumptions at all hierarchical levels of the development process are helpful for legitimizing decisions and strengthening consensus-building efforts. Alternatively, the incentive structure of the development process can be altered such that core actors are encouraged to contribute towards both the provision and appropriation of collective resources. However, defining the specific means by which this can be achieved is a task for future research.

6.6. Summary

In this chapter, we conceptualized a polycentric organizational structure for the governance of collective development in pluralistic settings. We positioned this conceptualization between the mainstream perspective on organization design and views on collective action in the context of managing common pool resources. By bringing

166

together the propositions from these two previously unrelated research streams, we were able to identify coordination and cooperation mechanisms for the integration of effort in pluralistic settings. With regards to coordination mechanisms, we argued that similar mechanisms to those outlined for unitary organizations be employed for meta- organizations in pluralistic settings. More specifically, we identified technical working groups and boundary objects as key integrative devices during the collective development process.

With regards to cooperation mechanisms, we argued that similar mechanisms to those used in unitary organizations are inadequate. Thus, we proposed that some of the mechanisms outlined in the collective action literature be employed instead. However, there is added complexity under pluralism, as compared to the management of common pool resources, because of the plurality of interests and payoff functions. Therefore, we also uncovered cooperation mechanisms that are unique to our setting. To that end, we identified polycentric structures, negotiable resource boundaries, and appeals to proportionality in costs and benefits as key integrative devices during the collective development process. In the next chapter, we present our conclusions before outlining the theoretical and practical contributions of this study.

167

7. CONCLUSION

In theory, as in practice, organizations and organizing can be distinguished from one another (Weick 1969, 1974; Ostrom, 1990). On the one hand, organizations are social or human-made artefacts (Simon, 1996; Baldwin, 2015) that need to be designed by people to help them achieve pre-specified goals. On the other hand, organizing can be regarded as a perpetually fluid problem-solving process concerned with “resolving...equivocality in an enacted environment by means of interlocked behaviours embedded in conditionally related process” (Weick, 1969: 11; 1974). In other words, organizing is a process that reduces uncertainty (equivocality) through recursive routines, and repetitive, reciprocal, and contingent behaviours among actors (Weick, 1969, 1974). When the process of organizing is carried out successfully in a collective action setting, it is institutionalized in sets of rules and structures that have legitimacy in the eyes of all participants (Ostrom

1990). It is through this legitimization that the organization is “enacted” (Weick, 1969).

Given that it provides a scaffolding within which actors can take targeted strategic action and make progress on other technologies (while concurrently being influenced by them), organizing has been described as “the mother of all general-purpose technologies”

(Puranam, 2012: 19; Bresnahan & Trajtenberg, 1995).

Against this backdrop, our aim with this research has been to gain a deeper understanding of organizing the collective development of designed artefacts that the members of a collective will share in use. The collective development of designed artefacts is central to the megaproject meta-organizations formed to develop new, long-lived infrastructure assets (Lundrigan et al., 2015). These assets have a high degree of design- choice interdependence, and they cannot be developed without the resource contributions of multiple legally independent actors. Moreover, in exchange for their resource contributions, these actors seek to directly influence design choices, which are ultimately

168

the outcome of the collective development process. Crucially, this process is carried out in pluralistic settings in which actors have differing interests, objectives and values. Given the resultant rivalry, it is necessary to create the relevant governance structures to govern the collective development process. Hence, our investigation into the nature of appropriate rules and structures.

The following research question has guided our research: How do we design organizations for collective development in pluralistic settings? Other questions we consider in this research include: Why do some local groups appear to converge more rapidly than others? How is the authority to make decisions distributed within the broad system? And how are inter-organizational conflicts resolved?

To address the above research questions, we adopted an abductive case-based research strategy, which benefits from the dual impact of both inductive and deductive approaches (Klag and Langley, 2013). On the one hand, we can draw on a theoretical framework to consider a phenomenon; while on the other hand, we can draw on the data or field situation to look for evidence that supports, refutes, or modifies our emerging hypotheses (Strauss and Corbin, 1990: 148). The case-study method was most suitable for this study because it is useful for supporting inquiries into areas in which little previous research exists (Eisenhardt, 1989). This is true of governing pluralistic settings from an organization design perspective (Gulati, Puranam, Tushamn, 2012; Puranam, Alexy &

Reitzig, 2014). Thus, an abductive case-based research strategy was in line with our motivation to contribute to the extension of this research stream.

Theoretically, we positioned our work between mainstream organization design and the contributions of Elinor Ostrom and her colleagues on collective action and the associated challenges when managing common pool resources. Both research streams are concerned with reducing uncertainty through interlocked behaviours in an effort to achieve

169

collective ends, albeit in different contexts. Furthermore, both research streams assume that the actors in each respective context are fallible learners with imperfect information, and a limited capacity to access and process information (boundedly rational) who, therefore, partake in a costly search for information (Simon, 1972; Ostrom, 1990).

Empirically, we focused on the rules and structures created to support the design decision-making process during the planning phase of long-lived infrastructure. More specifically, we studied the planning of HS2, a new high-speed rail network in the UK.

This project-based endeavour requires resources from multiple legally autonomous actors exhibiting heterogeneous interests. Actors include national and local governments, and other resource-rich, materially impacted stakeholders who together form a part of the HS2 meta-organization core.

Our analysis showed that at times actors in the core had rivalrous preferences for the one-off design choices to outline the local stations. However, the promoter could not simply exclude disagreeing local actors from the decision-making process. This low excludability was reinforced by the overall institutional framework surrounding the scheme, which made it hard to exclude local actors once they had been invited to participate in the design decision-making process. Furthermore, if the promoter excluded local actors, or prompted their defection, the promoter would lose access to their valuable resources, notably, fine-grained local information and local planning powers. In other words, excluding resource-rich local actors could potentially complicate project implementation, or worse still completely halt the planning process it gave rise to successful legislative challenge or destabilizing opposition. This conflation of low excludability and high rivalry allows us to argue that local design choices qualify as common-pool resources (Ostrom, 1990, Gil and Baldwin, 2013).

170

Given their heterogeneity, and sometimes conflicting preferences, core actors needed to coordinate their actions to arrive at mutually consensual choices for the final design.

Crucially, as with other common-pool resources, this ‘community of design development’ was not governed by formal contracts, by an authority hierarchy (firm or government), or by a meritocracy-based hierarchy. Rather, these known governance mechanisms were absent during the planning process. Hence, we aimed to investigate how this collective fostered collective action. We were also interested in the collective’s overall performance relative to expectations created at the onset of the development process.

From the organization design literature, we know that all organizing has to resolve two fundamental problems: the division of labour and the integration of effort (Burton &

Obel, 1984; Lawrence & Lorsch, 1967; March & Simon, 1958; Mintzberg, 1979). By implication, this requires that the four universal sub-problems of organizing be resolved: task division and task allocation, which make up the division of labour; and information provision and reward provision, which, in turn, make up the integration of effort (Puranam et al., 2014; Galbraith, 1973: 3; March & Simon, 1993: 2; Porter, Lawler, & Hackman,

1975). The integration of effort and its related sub-problems have been the focus of this research. Furthermore, we explored the integration of effort in the absence of command- and-control authority characteristic of unitary organizations. We chose this particular focus given that the HS2 project promoter was primarily concerned with reconciling multiple streams of information, divergent interests and a range of resources held by actors in the meta-organization core. Doing so was necessary to advance the system-level goal.

Organization design scholars argue that for unitary organizations integrating effort is about addressing the coordination and cooperation requirements within the system (Gulati,

Lawrence, & Puranam, 2005; Puranam, Alexy & Reitzig, 2014; Lawrence & Lorsch, 1967).

Coordination is undertaken through information processing, which serves to inform actors

171

of the planned behaviour of others as well as communicate the content of the organization's activities. Information processing is reflected in formal procedural specifications and hierarchical lines of communication (Galbraith, 1973; March and Simon, 1958; Tushman

& Nadler, 1978; Gulati, Lawrence & Puranam, 2005; Puranam et al., 2012; Camerer, 2003;

March & Simon, 1958). Cooperation, on the other hand, is ‘bought’ through employment contracts and inducements, such as promotions. These are given with a related expectation that commands within the hierarchy will be accepted (Ostrom, 1990; Gulati et al., 2012).

Our research shows that coordination mechanisms similar to those used in unitary organizations are applicable when organizing for collective development in pluralistic settings. Conversely, we find that there are limitations to ‘buying’ cooperation in these settings. Thus, alternative cooperation mechanisms to those used in unitary organizations need to be considered. To that end, we build on collective action literature and propose that robust governance structures are essential for this purpose. Ultimately, we identify mechanisms to support coordination and cooperation in pluralistic settings that are a combination of the mechanisms from the organization design and collective action literature streams, respectively. Furthermore, we also identify mechanisms, such as flexible resource and participation boundaries, that go beyond each of these two streams of literature.

Given that collective development is undertaken by multiple actors, large amounts of information relevant to the project were found beyond any one organization’s boundaries.

Hence, organizations had to interact with each other to facilitate both the transfer of information and its timely assimilation and transformation into collective problem definitions and solution sets (Srikanth & Puranam, 2014). The creation of a meta- organization with technical working groups and strategic boards facilitated this exchange of information. With both structures, face-to-face meetings allowed for the provision of

172

knowledge and information, across organizational boundaries (Srikanth and Puranam,

2014; Schelling, 1960; Clark, 1996).

On the one hand, direct communication through face-to-face meetings made for a more open, distributed and democratic decision-making process. While on the other hand, it increased the level of uncertainty within the meta-organization as refinements and extensions were continually proposed in light of new information (Tushman & Nadler,

1978). In light of this uncertainty, we found that the use of strategic documents as boundary objects contributed to achieving concerted action by creating a shared context across different problem-solving contexts (Star 1989, Carlile 2002). Together, these communication and decision-making channels created linkages among actors, a high degree of transparency, and predictability of particular actors’ activities and decisions

(Gulati and Singh, 1998; Gulati and Sytch, 2007; Puranam et al., 2012; Srikanth and

Puranam, 2014).

Along with creating a shared informational context, actors partaking in collective development processes have to coalesce their interests in order to forge consensus if they are to achieve the system level goal. For example, during the HS2 planning process, integrating effort meant merging the promoter’s vision for a functional cross-country railway network, with local authorities’ visions for local economic development and regeneration. While necessary, aligning interests can be especially challenging when actors are laying claim to indivisible design choices. Contrastingly, with a modular design structure, functional components can be individually owned and designed (Gil and

Baldwin, 2014).

Given that during the HS2 planning process many local design choices were indivisible and the solution space was constrained, core actors had to make trade-offs and compromises. As this contentious process unfolded, rules and structures were, therefore,

173

imperative for sustaining repeated voluntary face-to-face interaction and ultimately, for encouraging cooperation. Rules and structure were particularly important because employment contracts, and other such inducements and incentives that foster compliance in unitary organizations, were not feasible for the megaproject meta-organization.

A central insight of this research was that collective development took place within a polycentric structure with multiple decision centres. System-level decisions centralized by the project promoter or systems architect, included the choice of route, budget-related decisions, and the development schedule. A level below, more local actors along with HS2

Ltd. and the UK Government needed to agree on the location and configuration of local stations. At this level, local design choices were a common-pool resource shared among multiple actors. Thus, while actors exhibited rivalrous preferences, it was difficult to exclude them from the collective development process (Ostrom, 1990; Gil and Baldwin,

2014).

Rivalry among core actors during the HS2 planning process is also to be expected in pluralistic settings characterised by multiple objectives, diffuse power and knowledge- based work processes (Denis, Langley & Rouleau, 2007). Nevertheless, in the spirit of polycentricity, failure to resolve emerging conflicts would see them deferred to the UK

Parliament, an independent authority that was a third party to the conflict and the meta- organization. For the promoter, however, resolving conflicts in Parliament was the last resort because parliamentary proceedings could be very protracted. During these proceedings, all petitioners materially impacted by the scheme would have to be heard, which included over 4000 stakeholders kept external to the HS2 meta-organization. It was therefore in the promoter’s interest to encourage collective action and minimize conflict with local authorities ahead of time.

174

Polycentric systems accommodate multiple nested decision-making groups (Ostrom,

1990). Consistent with this, information exchange during the HS2 planning process was encouraged in multiple joint decision-making forums at multiple levels of governance.

Thus, ahead of parliamentary proceedings, when conflicts arose, this nested structure made provisions for the escalation of decision-making to resolve them. Escalating decision- making was especially useful when there was an impasse at lower levels of governance.

Furthermore, issues could be discussed simultaneously in different forums with different organizational actors, or with different representatives of the same organizations, which created an opportunity for the solutions reached in one forum to be applied in other forums.

Importantly, emerging governance structures also served as mutual monitoring and low-cost conflict resolution arenas during the HS2 planning process. That said, and as suggested by Ostrom (1990), we propose that incorporating impartial third parties into the pluralistic meta-organization would help collectives resolve disputes and make the meta- organization more robust. Furthermore, we suggest that these actors be third party to the disputes, but not to the designed structures, and thus internal to the megaproject meta- organization, unlike Parliament. In our setting, local conflict resolution was partly undertaken by the Secretary of State, who had the administrative authority to relax resource boundaries to foster cooperation. However, as a member of the project promoter, we cannot say that the Secretary of State was fully independent. This lack of an impartial, affordable conflict resolution mechanism is a fragility in the HS2 setting that can potentially undermine the collective development process.

The goal of cooperation efforts during the HS2 planning process was to minimize the risk that local actors would petition against local choices (the location of the station and the station configuration) in Parliament. In addition to polycentric structures, cooperation was also fostered through the provision of evidence and assurances that actors’ concerns were

175

being considered. In the absence of an already established resource around which to organize, sometimes this meant incorporating actors’ preferences into the final design choices. By implication, incorporating these preferences also meant relaxing the project constraints. We, therefore, propose that having negotiable resource boundaries is crucial during collective development processes. The ultimate objective in this regard is not to have all actors agree on design choices with the same level of enthusiasm, but rather that they are ‘satisfied’ (Simon, 1972) with the design choices and do not oppose them in the end.

Our final proposition for fostering cooperation is the creation of a shared perception of the proportionality between costs and benefits. The perception of proportionality legitimizes the collective effort and promotes a sense of fairness among core actors and third parties (Ostrom, 1990). Such a sense of fairness is especially important when actors have to make trade-offs and compromises, as is the case when design choices are indivisible and the solution space is constrained. Without it, however, we can expect individual actors to be less willing to contribute toward the collective effort (Ostrom,

1990).

Our findings suggest that coordination and cooperation mechanisms contribute to consensus building during the collective development process. Coordination mechanisms allow for information processing (Galbraith, 1973; March and Simon, 1958; Tushman &

Nadler, 1978) and for the reciprocal predictability of action (Gulati, Lawrence & Puranam,

2005; Puranam et al., 2012; Camerer, 2003; March & Simon, 1958). To achieve coordination, we propose the use of inter-organizational technical working groups and boundary objects. In turn, cooperation mechanisms, in the form of rules and governance structures, are important for ensuring that actors’ private interests are not advanced at the expense of the collective effort, and that opportunistic behaviour is kept at a minimum

176

(Ostrom, 1990; Gulati et al., 2012). To achieve cooperation, we propose the use of polycentric structures, negotiable resource boundaries, and appeals to the proportionality of costs and benefits.

To that end, our findings also suggest that failures in collective development can partly be traced back to failures in governance. In other words, if governance structures do not allow for the sufficient exchange and assimilation of information among actors, or if actors feel that their preferences are not represented in the emerging artefact, they will tend to advance their private interests at the expense of collective effort.

Overall, we appreciate that organizational forms that facilitate coordination and cooperation typically evolve at a slow rate, and that the process of designing these is not a trivial one (Ostrom, 1990, 2005; Gulati And Puranam, 2009). Considerable friction can arise when attempting to create organizational forms if they do not already exist and to readapt existing organizational forms to technological and economic change (Ostrom,

2005). In addition, we understand that “no humanly designed, complex system can be fully planned to achieve optimal performance” (Simon, 1990 paraphrased by Ostrom, 2005;

2007) and that “there is no guarantee that we can take the steps - including exercises in political design and management - needed to alter the operation of prevailing arrangements in a well-planned fashion” (Ostrom & Dietz, 2001: 286). Nevertheless, we argue that efforts be made towards robust organizing with a view to improving the quality of the collective development process. Otherwise, if core actors remain “unorganized”, it is unlikely that they will achieve a joint return as high as they can achieve if they organize in some way to undertake collective action.

177

Contributions to Theory and Practice

Organization Design Our first major theoretical contribution, we argue is to organization design literature.

There is a trend in new forms of organization to encourage cooperation between multiple, formally independent firms and non-contractually linked individuals (Puranam et al., 2014).

These collaborative pursuits require inter-organizational structures that facilitate coordination and encourage cooperation among an array of actors with heterogeneous interests, abilities and knowledge. Existing theories of organizing are, however, lagging behind the radically changing world of practice and continue to have a heavy intra-firm focus (Puranam, Gulati and Singh, 1998; Gulati, Lawrence, and Puranam, 2005; Reuer,

2004). Said differently, existing theories are “inextricably rooted in bygone features of economic production” (Walsh, Meyer, & Schoonhoven, 2006: 661). To that end, recent calls from organization design scholars have explicitly encouraged more research into new forms of organizing. These calls suggest a need to look beyond known mechanisms, such as employer-employee relationships, hierarchy, formal roles, and incentives such as salaries, bonuses, benefits, and promotion opportunities, for integrating the effort of multiple interdependent actors (Lakhani, Lifshitz-Assaf, and Tushman, 2013). Because in inter-organizational contexts that unify legally independent actors, such mechanisms are

“inconspicuous, if not entirely absent” (Gulati et al.,2012: 6). This study, therefore, responds to the call by identifying mechanisms for integrating effort in pluralistic settings with a particular focus on polycentric forms of organizing.

We also extend ideas from collective action to meta-organizations formed in pluralistic settings. Recent studies have had a strong bias toward meta-organizational forms enabled by technological progress such as new business ecosystems (Baldwin, 2012), and collaborative communities such as Linux or Wikipedia (Garud, Jain & Teurtscher,

2009; Zammuto, Griffith, Majchrzak, Dougherty, & Faraj, 2007; Miller, Greenwood, and 178

Prakashm 2009). However, we ground our study in the planning of a large megaproject, a form of meta-organization underexplored in organization design (Lundrigan et al., 2015).

By adding to the pool of examined inter-organizational contexts, we illuminate mechanisms to integrate effort that have remained elusive in organization design literature and contribute to advancing our empirical and theoretical knowledge of organizing in inter-organizational contexts.

More specifically, our contribution extends our understanding of polycentric forms of organizing in pluralistic settings. We draw on the conceptualization of polycentric structures as ones with many interdependent centres of decision-making created through the addition of local structures (Ostrom and Ostrom, 1977). In so doing, we contribute to the relatively lacking empirical complement of existing general theorizing on polycentric meta-organizations. Illuminating how a polycentric meta-organization emerges and works is important because, as our findings suggest, it can be used as a strategic instrument to manipulate organizational boundaries and encourage collaboration.

In this research, we, therefore, contribute to advance both our empirical and conceptual understanding of polycentric meta-organizations in pluralistic settings. We do not argue here that polycentric forms of organizing are superior or inferior to monocentric forms of organizing. Instead, we show that polycentricity is a valid alternative for organizing collective development in pluralistic settings. Furthermore, we reveal mechanisms for achieving collective action through the integration of effort. These include technical working groups, boundary objects, polycentric structures, negotiable resource boundaries, and appeals to proportionality in costs and benefits.

The extent to which polycentric forms of organizing will be effective will ultimately be a function of a number of factors. These include, but are not restricted to, the degree of decomposability of the designed artefact that is being produced, the ability of the

179

organizational actors to create and appropriate individual and collective value, the extent to which actors contribute towards collective value creation, and the level of homogeneity among the actors involved in the collective development process. Still, while we can assume that coordination and cooperation problems will always exist, we can also expect that encouraging face-to-face interaction to develop common ground, and implementing rules and structures to discourage opportunistic behaviour, is necessary to make polycentric forms of organizing sustainable and effective.

Finally, we contribute to organization design literature by presenting an adaptive process of organization design in a pluralistic setting. Adaptability is particularly important in the context of infrastructure development given that both the organization and the technical artefact are inexorably linked and emergent over time. Collective structures and collective strategy can, therefore, be said to be mutually reinforcing in the context of collective development. From extant studies on organizational adaptation (Gulati and

Puranam, 2009; Levinthal and Posen, 2007; Meyer, Gaba, and Colwell, 2005; Puranam and Swamy, 2016; Rivkin and Siggelkow, 2003; Siggelkow and Rivkin, 2005), we know that an important source of heterogeneity in organizational performance lies in the differential ability of unitary organizations to adapt to changes in their environment

(Agarwal and Helfat, 2009; Teece et al., 1997). However, less is known about the nature of adaption for polycentric forms of organizing. In this research, we, therefore, contribute to our understanding of adaptation in the context of polycentric meta-organizations and in the context of organizations more generally. We do this by empirically showing how organizing in pluralistic settings is driven by an adaptive search process, in line with the seminal works Simon (1955) and March and Simon (1958).

180

Projects and Project Management

Despite their prevalence in practice, there is much ambiguity in the projects related literature about megaproject organizing and performance (Pinto and Winch, 2016). This is captured in conflicting views about the degree to which projects are temporary (Bakker et al., 2011; Cattani et al., 2011), the bounds between the project organization and the environment (Engwall, 2003), and how to characterize project capabilities and learning

(Davies and Brady, 2000; Nightingale and Baden-Fuller, 2011). Differences also exist with regards to how megaproject performance should be defined and measured (Miller &

Lessard, 2001; Samset & Volden, 2015; Shenhar et al., 2001, Flyvbjerg et al., 2003 Morris,

1994, 2013). Complicating matters further is that megaprojects are studied in intellectual silos using multiple disparate theoretical and empirical lenses. Thus, at best the relevant literature is described as broad with a varied terminology-in-use (Winch, 2015) and at worst the project management discipline is described as being in crisis (Ansar, 2017).

Despite this state of affairs, we hope that this research can contribute to the field by providing empirical insight into the organization design perspective of megaprojects. This perspective captures the challenges associated with the temporariness of projects and the blurred organizational boundaries between the project organization and its surrounding environment. It also contributes to a broader understanding of project performance and the capabilities and learning that contribute to it.

Generally, project management research that looks beyond the organizational boundaries of project management organizations is underdeveloped. In the existing body of literature, when a broader view of projects is adopted, research tends to be concerned with understanding how projects engage with the environment given that projects are seen to be ‘‘inextricably interwoven with a...social context, which provides key resources of expertise, reputation, and legitimization” (Grabher 2004: 1492). This research, therefore,

181

builds on this conceptualization by contributing to a deeper understanding of the inter- organizational context within which planning decisions occur, and illuminating the interdependence between the project organization and the project environment (Maoret et al., 2011; Grabher, 2002). Furthermore, taking this approach, we support Engwall’s (2003) assertion that “No project is an island” and that no project operates in a social vacuum.

Crucially, the project stakeholder management literature, and stakeholder management literature more generally, fail to account for polycentricity. Instead, this literature has tended to focus on projects from the perspective of the focal firm or project organization (Winch, 2017). Thus, particularly on construction projects, the process of defining the project mission has been framed as one undertaken by one decision maker in isolation (Winch, 2007). Stakeholders are, therefore, seen as a part of the organizational

“environment,” and as groups or individuals that affect, or are affected by, the organizations pursuit of its objectives (Freeman, 1984; Ansoff, 1968; Cleland and King

1968; Laplume, Sonpar and Litz, 2008; South, Levitt, Dewulf, 2015). To that end, stakeholders are categorized as either cooperative (opportunities) or threats and are subsequently dealt with accordingly (Freeman, 1984). Conceding to stakeholder demands is, however, considered the last resort (Laplume et al., 2008; Walsh, 2004).

Through this monocentric lens of project planning, delays, bias or communication breakdowns are often attributed to the challenges associated with communicating up and down the hierarchy. These include difficulties with accessing knowledge or a disconnect between local problems and the global optimal solutions developed at the top of the hierarchy. In contrast, by framing front-end planning as a process undertaken by a polycentric structure, and foregrounding stakeholders’ actions and interests, we offer an alternative perspective. Perhaps poor project performance is not solely a problem of information, but also one of divergent objectives, goals and interests, among project

182

participants with a direct influence on project outcomes. This suggests that there is an opportunity to extend project stakeholder management theory for megaprojects both empirically and theoretically (Winch, 2017).

We, therefore, contribute to this research stream by extending our understanding of polycentric forms of organizing in pluralistic settings. We draw on the conceptualization of polycentric structures as ones with many interdependent centres of decision-making created through the addition of local structures (Ostrom and Ostrom, 1977); and show how the HS2 planning process involved multiple decision centres at different levels of governance. We also contribute to the relatively lacking empirical complement of existing general theorizing on polycentric meta-organizations. Illuminating how a polycentric meta- organization emerges and works is important because, as our findings suggest, it can be used as a strategic instrument to manipulate organizational boundaries and encourage collaboration.

We not only show that projects are embedded within a complex societal web of structures, resources, values, and players (Granovetter 1985); but also, how the promoter can deliberately include nominally independent stakeholders in the project organization irrespective of wider interpersonal networks and epistemic communities within which core actors are embedded (Bakker, 2010; Bechky, 2006). Thus, we agree with the idea of depicting projects as context-dependent, organizationally and environmentally embedded, history-dependent units of analysis (Engwall, 2003). We subscribe to this, as opposed to the view of projects as “lonely phenomenon” independent of history, contemporary context and future (Engwall, 2003; Kreiner, 1995, Eisenhardt and Tabrizi, 1995; Katz and Allen,

1982).

Extraordinarily, in the project's literature, the planning process, sometimes called the project front-end (Morris, 1994) is relatively understudied. Developing a multi-level

183

conceptual framework, therefore, contributes to filling this gap. It also serves to advance debates about (mega) project performance. Both the academic literature on project management and project management in practice stress efficiency as a salient measure of performance (Samset, 2010). In so doing, these bodies of literature “promote the value of formal calculations, risk modelling, clean optimization and rigorous, formal method”

(Nightingale & Brady, 2011: 85). However, as our research shows, carrying out the necessary estimations can be challenging during the planning stage given that uncertainty is particularly high and information is also most limited during this stage (Morris, 2013;

Morgan 1987). High uncertainty appears to have increasingly significant implications as the magnitude of capital investments increases, the number of stakeholders directly impacted by the project grows, and regulatory frameworks coupled with reduced communication costs put pressure on promoters to attend to the interests of stakeholders in the environment (Davies et al., 2011; Lessard & Miller, 2000; Flyvbjerg et al., 2003).

To that end, ambiguity in how we define, measure and evaluate performance makes it difficult to monitor and anticipate project outcomes accurately (Pinto and Slevin, 1988;

Lundrigan et al., 2015). Thus, by providing an insight into the processes through which megaproject organizational networks emerge over time and the structures that govern these networks, our research demonstrates that ambiguity in performance is rooted in collective efforts to resolve coordination and cooperation problems. Building on Lundrigan et al.’s

(2015) conceptualization, we show the interdependence between the meta-organization membership, the constraints on the solution space and thus the project scope and corresponding cost and schedule targets, in the language of practitioners. By implication, this calls into question whether project megaproject failures are a result of strategic misrepresentation and optimism bias as conjectured by Flyvbjerg et al. (2003), or whether such failures are rooted in limited professional and academic norms around how

184

performance is defined and measured, and the expectations they create. Is ‘failure’ sometimes beyond the powers of the promoter?

In other words, we ask if iterations in the planning process that are perceived to be

‘inefficient’ are actually inevitable for agreeing on a consensual plan that the promoter cannot produce ex-ante. If this is the case, as this research suggests, then there is scope within the project management discipline to emphasize the promoter’s ability to design adaptable meta-organizations and to delineate boundaries between the project meta- organization and the environment. Such an approach would augment the age-old mantra that revolves around “on time, on budget”, value-for-money and benefits realization.

As such, our work complements extant research on planning that has been biased towards legal contracts, a governance mechanism that is inadequate for managing planning challenges since the promoter and the stakeholders are legally independent at this stage.

Instead, our study reveals that the promoter has to adopt mechanisms that garner the support of key actors. Furthermore, this has to be done without challenging existing administrative structures more than necessary lest the promoter risk losing legitimacy. Our study also confirms that solving the coordination problem is crucial during project planning, and thus so is the problem of providing knowledge and information across organizational boundaries (Winch, 2015). However, we argue that this is only half of the picture. The other half entails advancing the collective interest and attenuating conflicting private interests.

Specifically, our findings show that governing megaprojects during the planning stage involves guiding an array of legally independent yet interdependent (and often politically inclined) actors towards a common goal. Achieving this goal requires accumulating and sharing information while taking a sequence of decisions (Winch, 2015;

Samset & Volden, 2015) and navigating complex power dynamics often underrepresented

185

in the existing literature. Complicating matters further, the environment surrounding the project meta-organization itself can be turbulent. Thus, the project organization needs to be capable of coping with this turbulence (Williams & Samset, 2010). What this suggests, in turn, is that that trying to control the system completely is futile (Lampel, 2011). A better strategy, as presented in this research, may be to adopt both ‘restraining’ and ‘enabling’ rules and structures that support a recursive process in which projects interact and coalesce into an emergent strategy (Lampel, 2011; Lindkvist, 2004, Sydow et al., 2004).

In practice, this suggests that the technical tools currently promoted by project management professional institutions are inadequate for dealing with the uncertainty and turbulence endemic to megaprojects. Project promoters require an array of skills during the project planning process. Apart from being trained in the technical aspects of large-scale projects, they also require a broad understanding of organization design and redesign in politically charged, dynamic inter-organizational contexts; and the capabilities to mobilize resources and collective action across a range of local conditions. In other words, and as in the context of common pool resources (Ostrom, 1990), to manage projects sustainably, project managers require a firm understanding of both the social system and the technical system, coupled with an understanding of the system-level conditions more generally.

Project promoters, therefore, need to be equipped to achieve both efficient and effective measures of project performance (Miller & Lessard, 2001) where effectiveness is measured in terms of economic performance, technical functionality, social acceptability, political legitimacy and economic development (Miller & Lessard, 2000; Williams &

Samset; Shenhar et al. (2001)). As argued by Pinto and Winch (2016) the project manager must become “more fully qualified, not only in the technical details of the project but also as an individual who can interface with top management and critical stakeholders early in

186

the project’s development; in fact, far earlier and with greater responsibilities than in the outdated execution model” (Pinto & Winch, 2016: 240).

Furthermore, as Morris et al. (2006) suggest, in certain instances, no matter how vigorously technical project management tools are enforced, if the project manager cannot navigate the institutional landscape, there is an increased likelihood of project failure (at least as perceived in light of deep-seated norms). In the absence of such know-how, opportunities to create the social, political and intellectual capital necessary to advance project goals may be undermined (Samset & Volden, 2015; Merrow, 2011; Morris, 2013).

A better understanding of the institutional context within which megaproject organizations exist is therefore also needed from a research perspective. Such an understanding would contribute to the creation of project organizations that are more capable of defining negotiated outputs between the promoter and other stakeholders. It would also contribute to the fulfilment of agreed objectives (Samset & Volden, 2016). In sum, our view is that it is worthwhile to explore various ways that the project planning stage can be organized, so that project stakeholders move in a coherent, self-reinforcing direction. Improvements in this regard will help stakeholders to better monitor how evolving technical design choices impact the evolution of other critical variables, notably cost and schedule.

Limitations and Future Research

The generalizability of these research findings is limited in part because the research is based on a single “case” of infrastructure project planning embedded in a specific social setting. In the UK, individuals and multiple organizations are legally permitted to partake in the process of designing new infrastructure in various ways. Promoters are legally obliged to consult and show that they have considered alternatives before proposing a plan, and stakeholders materially impacted by the project can petition against particular aspects

187

of proposed plans even after being formally consulted. The Government itself has no legal rights to acquire land compulsorily, but rather the Government needs to gain this right from Parliament. Violating the laws in the environment can make the development process vulnerable to judicial challenge by aggrieved stakeholders. Hence, the UK’s institutional framework influenced the way in which the collective development process unfolded and arguably contributed to the promoter’s decision to include local authorities in the meta- organization.

It remains indeterminate whether robust governance structures for collective development can be created in an institutional environment where the legal frameworks and means of enforcement are less robust. While project organizations may be able to influence their surrounding environment, we can expect that the institutions in the environment also constrain the range of organizational choices available or applicable to project participants. By the same token, since HS2 gained traction in 2009, the project has been supported by all three of England’s major political parties - Conservatives, Labour and the Liberal Democrats - in government and in opposition. Thus, in our study of the polycentric megaproject meta-organization, we were able to restrict conflict to the operational level of the scheme. We conjecture that introducing conflict or efforts to subvert a project at the collective choice or constitutional level (on the grounds of its environmental or social impact, for example) will introduce another dimension of complexity to the governance problem. Opportunities, therefore, exist to study how project-based meta-organizing differs across institutional and political contexts.

The generalizability of these research findings is also limited given the type of project under consideration. While HS2 is a single, very large programme, each station is in a separate city. Thus, the basic problem is amenable to decomposition (Gil and Baldwin,

2014; Levitt, 1984). As is the case with other “longitudinal projects”, such as a pipeline or

188

a highway, a railway line can be converted, for management purposes, into what can be viewed as more manageable ‘subprojects’ (Levitt, 1984). During the HS2 planning process, this conversion was carried out geographically, with a local design decision-making process carried in each city along the route. The promoter was also able to convert HS2 into manageable subprojects by dividing the Y-shaped route into two phases one in the

‘south’ and another in the ‘north’. Carrying out the same exercise would be more challenging for a more complex non-decomposable megaproject such as a power plant or industrial plant, for example. In such cases, we can expect that a high level of (reciprocal and sequential) interdependence will persist at the project level (Thompson, 1967). Thus, we conjecture that the project promoter will have to contend with a more complex division of labour. By implication, the promoter will also have to resolve more complicated coordination problems including synchronizing and scheduling tasks in the absence of hierarchical authority. Opportunities, therefore, also exist to study how project-based meta-organizing differs under a variety of technical conditions.

189

REFERENCES

Abdallah, C. & Langley, A. (2014). The Double Edge of Ambiguity in Strategic Planning. Journal of Management Studies, 51(2), 235-264. Adner, R. & Kapoor, R. (2010). Value Creation in Innovation Ecosystems: How the Structure of Technological Interdependence Affects Firm Performance in New Technology Generations. Strategic Management Journal, 31(3), 306-333. Agarwal, R. & Helfat, C. E. (2009). Strategic Renewal of Organizations. Organization Science, 20(2), 281-293. Aggarwal, V. K. (1998a). Institutional Designs for a Complex World: Bargaining, Linkages, and Nesting. Ithaca, N.Y.: Cornell University Press. Aggarwal, V. K. (1998b). Reconciling Multiple Institutions: Bargaining, Linkages, and Nesting. In: Aggarwal, V. K. (ed.) Institutional Designs for a Complex World: Bargaining, Linkages, and Nesting. Ithaca: Cornell University Press. Alchian, A. A. & Demsetz, H. (1972). Production, Information Costs, and Economic Organization. American Economic Review, 62(5), 777-795. Allen, T. J. & Cohen, S. I. (1969). Information Flow in Research and Development Laboratories. Administrative Science Quarterly, 14(1), 12-19. Altshuler, A. A., Luberoff, D. & Lincoln Institute of Land Policy. (2003). Mega-projects: The Changing Politics of Urban Public Investment. Washington, D.C.: Brookings Institution Press. Anderies, J. M., Janssen, M. A. & Ostrom, E. (2004). A Framework to Analyze the Robustness of Social-ecological Systems from an Institutional Perspective. Ecology and Society, 9(1). Ansar, A. (2018). The Fate of Ideals in the Real World: A Long View on Philip Selznick's Classic on the Tennessee Valley Authority (TVA).International Journal of Project Management, 36(2), 385-395. Arrow, K. J. (1974). The Limits of Organization. New York: John Brockman Associates. Astley, W. G. (1984). Toward an Appreciation of Collective Strategy. Academy of Management Review, 9(3), 526-535. Astley, W. G. & Fombrun, C. J. (1983). Collective Strategy: Social Ecology of Organizational Environments. Academy of Management Review, 8(4),576-587. Baer, M., Dirks, K. T. & Nickerson, J. A. (2013). Microfoundations of Strategic Problem Formulation. Strategic Management Journal, 34(2), 197-214. Bakker, R. M., Cambre, B., Korlaar, L. & Raab, J. (2011). Managing the Project Learning Paradox: A Set-theoretic Approach Toward Project Knowledge Transfer. International Journal of Project Management, 29(5), 494-503. Baland, J.-M. & Platteau, J.-P. (1999). The Ambiguous Impact of Inequality on Local Resource Management. World Development, 27(5), 773-788. Baldwin, C. Y. (2012). Organization Design for Business Ecosystems. Journal of Organization Design, 1(1), 20-23. Baldwin, C. Y. (2015). Bottlenecks, Modules and Dynamic Architectural Capabilities. Harvard Business School Finance Working Paper, No. 15-028. Baldwin, C. Y. & Clark, K. B. (2000). Design Rules. Cambridge, Mass. ; London: MIT Press. Balliet, D. (2010). Communication and Cooperation in Social Dilemmas: A Meta- Analytic Review. Journal of Conflict Resolution, 54(1), 39-57. Bechky, B. A. (2003). Sharing Meaning Across Occupational Communities: The Transformation of Understanding on a Production Floor. Organization Science, 14(3), 312-330.

190

Bechky, B. A. (2006). Gaffers, Gofers, and Grips: Role-based Coordination in Temporary Organizations. Organization Science, 17(1), 3-21. Benner, M. J. (2002). Process Management and Technological Innovation: A Longitudinal Study of the Photography and Paint Industries. Administrative Science Quarterly, 47(4), 676-706. Biernacki, P. & Waldorf, D. (1981). Snowball Sampling: Problems and Techniques of Chain Referral Sampling. Sociological Methods & Research, 10(2), 141-163. Birmingham City Council. (2015). Birmingham City Plan: Birmingham Curzon HS2 Masterplan for Growth. Birmingham: Birmingham City Council. Blau, P. M. (1964). Exchange and Power in Social Life. New York: John Wiley & Sons, Inc. Boland, R. J. & Tenkasi, R. V. (1995). Perspective Making and Perspective Taking in Communities of Knowing. Organization Science, 6(4), 350-372. Breene, R. T. S., Nunes, P. F. & Shill, W. E. (2007). The Chief Strategy Officer. Harvard Business Review, 85(10), 84. Bresnahan, T. F. & Trajtenberg, M. (1995). General-Purpose Technologies: Engines of Growth. Journal of Econometrics, 65(1), 83-108. Brix, J. & Peters, L. S. (2015). Exploring an Innovation Project as a Source of Change in Organization Design. Journal of Organization Design, 4(1),29-43. Brunsson, N. (1982). The Irrationality of Action and Action Rationality: Decisions, Ideologies and Organizational Actions. Journal of Management Studies, 19(1), 29- 44. Brunsson, N. & Jacobsson, B. (2000). A World of Standards. Oxford, England: Oxford University Press. Brusoni, S., Prencipe, A. & Pavitt, K. (2001). Knowledge specialization, organizational coupling, and the boundaries of the firm: Why do firms know more than they make? Administrative Science Quarterly, 46(4), 597-621. Burgelman, R. A. (1983). A Process Model of Internal Corporate Venturing in the Diversified Major Firm. Administrative Science Quarterly, 28(2),223-244. Burns, T. & Stalker, G. M. (1961). Management of Innovation. London: Tavistock. Burton, R. M. & Obel, B. (1984). Designing Efficient Organizations: Modelling and Experimentation. Amsterdam: Elsevier Science Publishers B.V. Camden Council. (2014a). HS2 Phase 2 Consultation, Comments of London Borough of Camden. London: Camden Council. Camerer, C. & Knez, M. (1996). Coordination, Organizational Boundaries and Fads in Business Practices. Industrial and Corporate Change, 5(1), 89-112. Camerer, C. & Knez, M. (1997). Coordination in Organizations: A Game-theoretic Perspective. In: Shapira, Z. (ed.) Organizational Decision-making. Cambridge, UK: Cambridge University Press. Camerer, C. & Russell Sage, F. (2003). Behavioural Game Theory: Experiments in Strategic Interaction. Princeton, NJ: Princeton University Press. Carlile, P. R. (2002). A pragmatic view of knowledge and boundaries: Boundary objects in new product development. Organization Science, 13(4),442-455. Carlile, P. R. (2004). Transferring, translating, and transforming: An integrative framework for managing knowledge across boundaries. Organization Science, 15(5), 555-568. Carlson, J. M. & Doyle, J. (2002). Complexity and robustness. Proceedings of the National Academy of Sciences of the United States of America, 99,2538-2545. Cattani, G., Ferriani, S., Frederiksen, L. & Taube, F. (2011). Project-Based Organizing and Strategic Management: A Long-Term Research Agenda on Temporary Organizational Forms. Advances in Strategic Management, 28, 3-26. 191

Chandler, A. D. & Massachusetts Institute of, T. (1962). Strategy and Structure: Chapters in the History of the Industrial Enterprise. Garden City, New York: Doubleday. Clark, K. B. (1985). The Interaction of Design Hierarchies and Market Concepts in Technological Evolution. Research Policy, 14(5), 235-251. Clark, K. B. (1996). Competing through Manufacturing and the New Manufacturing Paradigm: Is Manufacturing Strategy Passé? Production and Operations Management, 5(1), 42-58. Clark, K. B. & Fujimoto, T. (1991). Product Development Performance: Strategy, Organization, and Management in the World Auto Industry. Boston, MA: Harvard Business School Press. Clark, K. B. & Wheelwright, S. C. (1993). Managing New Product and Process Development. New York: Free Press. Clement, J. & Puranam, P. (2017). Searching for Structure: Formal Organization Design as a Guide to Network Evolution. Management Science,(Articles in Advance), 1- 17. Coase, R. H. (1937). The Nature of the Firm. Economica-New Series, 4(16), 386-405. Colfer, L. J. & Baldwin, C. Y. (2016). The Mirroring Hypothesis: Theory, Evidence, and Exceptions. Industrial and Corporate Change, 25(5), 709-738. Cramton, C. D. (2001). The Mutual Knowledge Problem and its Consequences for Dispersed Collaboration. Organization Science, 12(3), 346-371. Crawford, S. E. S. & Ostrom, E. (1995). A Grammar of Institutions. American Political Science Review, 89(3), 582-600. Crilly, N. (2010). The Roles that Artefacts Play: Technical, Social and Aesthetic Functions. Design Studies, 31(4), 311-344. Cyert, R. M. & March, J. G. (1963). A Behavioural Theory of the Firm (2 ed.). New York: Blackwell. Davies, A. & Brady, T. (2000). Organisational Capabilities and Learning in Complex Product Systems: Towards Repeatable Solutions. Research Policy, 29(7-8), 931- 953. Davies, A., Brady, T., Prencipe, A. & Hobday, M. (2011). Innovation in Complex Products and Systems: Implications for Project-Based Organizing. Project-Based Organizing and Strategic Management, 28, 3-26. Dawes, R. M., Van De Kragt, A. J. C. & Orbell, J. M. (1988). Not Me or Thee but We: The Importance of Group Identity in Eliciting Cooperation in Dilemma Situations: Experimental Manipulations. Acta Psychologica, 68(1-3), 83-97. Dearborn, D. C. & Simon, H. A. (1958). Selective Perception: A Note on the Departmental Identifications of Executives. Sociometry, 21(2), 140-144. Denis, J.-L., Langley, A. & Rouleau, L. (2006). The Power of Numbers in Strategizing. Strategic Organization, 4(4), 349-377. Denis, J. L., Dompierre, G., Langley, A. & Rouleau, L. (2011). Escalating Indecision: Between Reification and Strategic Ambiguity. Organization Science, 22(1), 225- 244. Denis, J. L., Langley, A. & Rouleau, L. (2007). Strategizing in Pluralistic Contexts: Rethinking Theoretical Frames. Human Relations, 60(1), 179-215. Denis, J. L., Langley, A. & Rouleau, L. (2010). The Practice of Leadership in the Messy World of Organizations. Leadership, 6(1), 67-88. Dietz, T., Ostrom, E. & Stern, P. C. (2003). The Struggle to Govern the Commons. Science, 302(5652), 1907-12. Dougherty, D. (1992). Interpretive Barriers to Successful Product Innovation in Large Firms. Organization Science, 3(2), 179-202.

192

Doz, Y. L. (1996). The Evolution of Cooperation in Strategic Alliances: Initial Conditions or Learning Processes? Strategic Management Journal, 17,55-83. Doz, Y. L. & Kosonen, M. (2008). Fast Strategy: How Strategic Agility Will Help You Stay Ahead of the Game. Harlow: Wharton School Publishing. Dunbar, K. (1995). How Scientists Really Reason: Scientific Reasoning in Real-world Laboratories. The nature of insight, 18, 365-395. Eisenhardt, K. M. (1989). Building Theories from Case-Study Research. Academy of Management Review, 14(4), 532-550. Eisenhardt, K. M. & Brown, S. L. (1999). Patching: Restitching Business Portfolios in Dynamic Markets. Harvard Business Review, 77(3), 72-82. Eisenhardt, K. M. & Graebner, M. E. (2007). Theory Building from Cases: Opportunities and Challenges. Academy of Management Journal, 50(1),25-32. Eisenhardt, K. M. & Tabrizi, B. N. (1995). Accelerating Adaptive Processes: Product Innovation in the Global Computer Industry. Administrative Science Quarterly, 40(1), 84-110. Engwall, M. (2003). No Project is an Island: Linking Projects to History and Context. Research Policy, 32(5), 789-808. Eppinger, S. & Ulrich, K. (2016). Product Design and Development (6 ed.). New York: McGraw-Hill Higher Education. Eriksson, K., Wikstrom, K., Hellstrom, M. & Levitt, R. (2017). Governance of Business Ecosystem Projects. In: Mahalingam, A., Shealy, T. & Gil, N., eds. 15th Engineering Project Organization Conference with 5th International Megaprojects Workshop, 2017 Stanford Sierra Camp, California. Ethiraj, S. K. & Levinthal, D. (2004). Modularity and Innovation in Complex Systems. Management Science, 50(2), 159-173. Feldman, M. S. & March, J. G. (1981). Information in Organizations as Signal and Symbol. Administrative Science Quarterly, 26(2), 171-186. Fischer, A., Greiff, S. & Funke, J. (2012). The Process of Solving Complex Problems. Journal of Problem Solving, 4, 19-42. Fjeldstad, O. D., Snow, C. C., Miles, R. E. & Lettl, C. (2012). The Architecture of Collaboration. Strategic Management Journal, 33(6), 734-750. Flyvbjerg, B., Bruzelius, N., Rothengatter, W., Bruzelius, N. & Rothengatter, W. (2003). Megaprojects and Risk: An Anatomy of Ambition. Cambridge UK: Cambridge University Press. Frischmann, B. M. (2005). Infrastructure Commons. Michigan State Law Review, (89), 121-136. Frischmann, B. M. (2012). Infrastructure: The Social Value of Shared Resources. New York: Oxford University Press. Galbraith, J. R. (1973). Designing Complex Organizations. Reading, Massachusetts: Addison-Wesley Publishing Co. Galbraith, J. R. (1974). Organization Design: An Information Processing View. Interfaces, 4, 28-36. Galbraith, J. R. (1977). Organization Design. Reading, Massachusetts: Addison-Wesley. Gann, D. M. & Salter, A. J. (2000). Innovation in Project-based, Service-enhanced Firms: The Construction of Complex Products and Systems. Research Policy, 29(7- 8), 955-972. Garud, R., Jain, S. & Tuertscher, P. (2009). Incomplete by Design and Designing for Incompleteness. Design Requirements Engineering: A Ten-Year Perspective, 14, 137-156.

193

Garud, R. & Karnoe, P. (2003). Bricolage Versus Breakthrough: Distributed and Embedded Agency in Technology Entrepreneurship. Research Policy, 32(2), 277- 300. Gavetti, G. (2005). Cognition and Hierarchy: Rethinking the Microfoundations of Capabilities' Development. Organization Science, 16(6), 599-617. Gavetti, G. (2012). Toward a Behavioural Theory of Strategy. Organization Science, 23(1), 267-285. Gavetti, G., Levinthal, D. & Ocasio, W. (2007). Neo-Carnegie: The Carnegie School's Past, Present, and Reconstructing for the Future. Organization Science, 18(3), 523- 536. Gehman, J., Glaser, V. L., Eisenhardt, K. M., Gioia, D., Langley, A. & Corley, K. G. (2017). Finding Theory–method Fit: A Comparison of Three Qualitative Approaches to Theory Building. Journal of Management Inquiry, Published online ahead of print. Ghemawat, P. & Khanna, T. (1998). The Nature of Diversified Business Groups: A Research Design and Two Case Studies. Journal of Industrial Economics, 46(1), 35-61. Gibbons, R. (1998). Incentives in Organizations. Journal of Economic Perspectives, 12(4), 115-132. Gibson, C. C. (2001). Forest Resources: Institutions for Local Governance in Guatemala. Protecting the Commons, 71-89. Gil, N. & Baldwin, C. Y. (2013). Creating a Design Commons: Lessons from Teachers' Participation in the Design of New Schools: Harvard Business School. Gil, N. & Beckman, S. (2009). Introduction: Infrastructure Meets Business: Building New Bridges, Mending Old Ones. California Management Review, 51(2), 6-29. Gil, N., Lundrigan, C., Pinto, J. & Puranam, P. (2017). Megaproject Organization and Performance: The Myth and Political Reality. Pennsylvania: Project Management Institute, Inc. Gil, N. & Tether, B. S. (2011). Project Risk Management and Design Flexibility: Analysing a Case and Conditions of Complementarity. Research Policy, 40(3), 415-428. Gil, N. A. & Pinto, J. (2016). Collective Action at the Complex Systems Project Front-end: Governance and Performance Implications. In: Academy of Management Proceedings, 2016. Academy of Management. Grabher, G. (2004). Temporary Architectures of Learning: Knowledge Governance in Project Ecologies. Organization Studies, 25(9), 1491-1514. Grandori, A. (1997). An Organizational Assessment of Interfirm Coordination Modes. Organization Studies, 18(6), 897-925. Granovetter, M. (1985). Economic-Action and Social-Structure: The Problem of Embeddedness. American Journal of Sociology, 91(3), 481-510. Gray, B. (1989). Collaborating: Finding Common Ground for Multiparty Problems. San Francisco: Jossey-Bass. Gray, B. (2011). The Complexity of Multiparty Negotiations: Wading into the Muck. Negotiation and Conflict Management Research, 4(3), 169-177. Gray, B. & Clyman, D. R. (2003). Difficulties Fostering Cooperative Agreements in Multiparty Negotiations: Cognitive, Procedural, Structural, and Social. In: West, M., Tjosvold, D. & Smith, K. (eds.) International Handbook of Organizational Teamwork and Cooperative Working. Chichester: Wiley. Gray, B. & Clyman, D. R. (2008). Difficulties Fostering Cooperative Agreements in Multiparty Negotiations. International Handbook of Organizational Teamwork and Cooperative Working, 401. 194

Greve, H. R. (1998). Performance, Aspirations, and Risky Organizational Change. Administrative Science Quarterly, 58-86. Greve, H. R. (2008). A Behavioural Theory of Firm Growth: Sequential Attention to Size and Performance Goals. Academy of Management Journal, 51(3), 476-494. Gulati, R. (1995). Does Familiarity Breed Trust: The Implications of Repeated Ties for Contractual Choice in Alliances. Academy of Management Journal, 38(1), 85-112. Gulati, R. (2007). Silo busting: How to Execute on the Promise of Customer Focus. Harvard Business Review, 85(5), 98-108. Gulati, R., Lawrence, P. R. & Puranam, P. (2005). Adaptation in Vertical Relationships: Beyond Incentive Conflict. Strategic Management Journal, 26(5), 415-440. Gulati, R. & Puranam, P. (2009). Renewal Through Reorganization: The Value of Inconsistencies Between Formal and Informal Organization. Organization Science, 20(2), 422-440. Gulati, R., Puranam, P. & Tushman, M. (2012a). Meta-organization Design: Rethinking Design in Interorganizational and Community Contexts. Strategic Management Journal, 33(6), 571-586. Gulati, R. & Singh, H. (1998). The Architecture of Cooperation: Managing Coordination Costs and Appropriation Concerns in Strategic Alliances. Administrative Science Quarterly, 43(4), 781-814. Gulati, R. & Sytch, M. (2007). Dependence Asymmetry and Joint Dependence in Interorganizational Relationships: Effects of Embeddedness on a Manufacturer's Performance in Procurement Relationships. Administrative Science Quarterly, 52(1), 32-69. Gulati, R., Wohlgezogen, F. & Zhelyazkov, P. (2012b). The Two Facets of Collaboration: Cooperation and Coordination in Strategic Alliances. Academy of Management Annals, 6, 531-583. Gutwin, C., Schneider, K., Paquette, D. & Penner, R. (2005). Supporting Group Awareness in Distributed Software Development. Engineering Human Computer Interaction and Interactive Systems, 3425, 383-397. Hardin, G. (1968). Tragedy of Commons. Science, 162(3859), 1243-1248. Hardy, C., Langley, A., Mintzberg, H. & Rose, J. (1983). Strategy Formation in the University Setting. The Review of Higher Education, 6(4), 407-433. Heath, C. & Staudenmayer, N. (2000). Coordination Neglect: How Lay Theories of Organizing Complicate Coordination in Organizations. Research in Organizational Behavior, Vol 22, 2000, 22, 153-191. Henderson, R. & Newell, R. G. (2010). Accelerating Innovation in Energy: Insights from Multiple Sectors. Cambridge MA: Harvard Business School Press. Henderson, R. M. & Clark, K. B. (1990). Architectural Innovation: The Reconfiguration of Existing Product Technologies and the Failure of Established Firms. Administrative Science Quarterly, 35(1), 9-30. Henisz, W. J., Levitt, R. E. & Scott, W. R. (2012). Toward a Unified Theory of Project Governance: Economic, Sociological and Psychological Supports for Relational Contracting. Engineering Project Organization Journal, 2(1-2), 37-55. Henkel, J. & Baldwin, C. Y. (2009). Modularity for Value Appropriation: Drawing the Boundaries of Intellectual Property: Harvard Business School Boston, MA. Hess, C. & Ostrom, E. (2006). A Framework for Analysing the Microbiological Commons. International Social Science Journal, 58(188), 335-349. High Speed Two Limited. (2012a). Corporate Plan 2012-2015 and Business Plan 2012- 2013. London: High Speed Two Limited. High Speed Two Limited. (2012b). Options for Phase Two of the High-Speed Rail Network: A Report to Government by HS2 Ltd. London: High Speed Two Limited. 195

High Speed Two Limited. (2013a). High Speed Rail: Investing in Britain’s Future: Consultation on the Route from the West Midlands to Manchester, Leeds and Beyond. London: High Speed Two Limited. High Speed Two Limited. (2013b). Route Engineering Report: West Midlands to Manchester. London: High Speed Two Limited. High Speed Two Limited. (2014a). Rebalancing Britain: From HS2 Towards a National Transport Strategy. London: High Speed Two Limited. High Speed Two Limited. (2014b). HS2 Phase 2b: Manchester Piccadilly Factsheet 2013. London: High Speed Two Limited. High Speed Two Limited. (2014c). HS2 Phase 2b: Sheffield Meadowhall Factsheet 2013. London: High Speed Two Limited. High Speed Two Limited. (2015). The Yorkshire Hub: An interim report on the redevelopment of Leeds station. London: High Speed Two (HS2) Limited. High Speed Two Limited. (2017a). From Crewe to Manchester, West Midlands to Leeds and Beyond. Phase 2b Route Decision. London: High Speed Two Limited. High Speed Two Limited. (2017b). High Speed Two: Phase 2b West Midlands to Leeds Route Engineering Report. London: High Speed Two Limited. Hobday, M., Davies, A. & Prencipe, A. (2005). Systems Integration: A Core Capability of the Modern Corporation. Industrial and Corporate Change,14(6), 1109-1143. Hyden, G. (1990). Reciprocity and Governance in Africa. In: Wunsch, J. & Olowu, D. (eds.) The Failure of the Centralized State: Institutions and Self-governance in Africa. Boulder: Westview Press. Iansiti, M. & Levien, R. (2004). Strategy as Ecology. Harvard Business Review, 82(3), 68-81. Innes, J., Gruber, J., Neuman, M. & Thompson, R. (1994). Coordinating Growth and Environmental Management through Consensus Building. In: California Policy Seminar, 1994 Berkeley, California. Jarzabkowski, P. & Fenton, E. (2006). Strategizing and Organizing in Pluralistic Contexts. Long Range Planning, 39(6), 631-648. Jarzabkowski, P., Sillince, J. A. A. & Shaw, D. (2010). Strategic Ambiguity as a Rhetorical Resource for Enabling Multiple Interests. Human Relations, 63(2), 219- 248. Kahneman, D. (1992). Reference Points, Anchors, Norms, and Mixed Feelings. Organizational Behavior and Human Decision Processes, 51(2), 296- 312. Kahneman, D. & Tversky, A. (1979). Prospect Theory: Analysis of Decision under Risk. Econometrica, 47(2), 263-291. Kaplan, R. S. & Norton, D. P. (2005). The Balanced Scorecard: Measures that Drive Performance. Harvard Business Review, 83(7-8), 71-79. Katz, R. & Allen, T. J. (1982). Investigating the Not Invented Here (NIH) Syndrome - A Look at the Performance, Tenure, and Communication Patterns of 50 R&D Project Groups. R & D Management, 12(1), 7-19. Kivleniece, I. & Quelin, B. V. (2012). Creating and Capturing Value in Public-Private Ties: A Private Actor's Perspective. Academy of Management Review, 37(2), 272- 299. Klag, M. & Langley, A. (2013). Approaching the Conceptual Leap in Qualitative Research. International Journal of Management Reviews, 15(2), 149-166. Klein, B., Crawford, R. G. & Alchian, A. A. (1978). Vertical Integration, Appropriable Rents, and the Competitive Contracting Process. Journal of Law & Economics, 21(2), 297-326.

196

Knorr-Cetina, K. (1999). Epistemic Cultures: How the Sciences Make Knowledge. Cambridge, Ma: Harvard University Press. Knudsen, T. & Srikanth, K. (2014). Coordinated Exploration: Organizing Joint Search by Multiple Specialists to Overcome Mutual Confusion and Joint Myopia. Administrative Science Quarterly, 59, 409-441. Kogut, B. & Zander, U. (2003). Knowledge of the Firm and the Evolutionary Theory of the Multinational Corporation. Journal of International Business Studies, 34(6), 516-529. Kreiner, K. (1995). In Search of Relevance: Project Management in Drifting Environments. Scandinavian Journal of Management, 11(4), 335-346. Lakhani, K. R., Lifshitz-Assaf, H. & Tushman, M. (2013). Open Innovation and Organizational Boundaries: Task Decomposition, Knowledge Distribution and the Locus of Innovation. In: Grandori, A. (ed.) Handbook of economic organization: Integrating Economic and Organizational Theory. Northampton, Ma: Elgar. Lampel, J. (2011). Institutional Dynamics of Project-Based Creative Organizations: Irving Thalberg and the Hollywood Studio System. Project-Based Organizing and Strategic Management, 28, 445-466. Langley, A. (1989). In Search of Rationality: The Purposes Behind the Use of Formal Analysis in Organizations. Administrative Science Quarterly, 34(4), 598-631. Langley, A. (1991). Formal Analysis and Strategic Decision-Making. Omega- International Journal of Management Science, 19(2-3), 79-99. Langley, A. (1995). Between Paralysis by Analysis and Extinction by Instinct. Sloan Management Review, 36(3), 63-76. Langley, A. (1999). Strategies for Theorizing from Process Data. Academy of Management Review, 24(4), 691-710. Langley, A., Smallman, C., Tsoukas, H. & Van de Ven, A. H. (2013). Process Studies of Change in Organization and Management: Unveiling Temporality, Activity, and Flow. Academy of Management Journal, 56(1), 1-13. Lawrence, P. R. & Lorsch, J. W. (1967). Differentiation and Integration in Complex Organizations. Administrative Science Quarterly, 12(1), 1-47. Lawrence, T. B., Hardy, C. & Phillips, N. (2002). Institutional Effects of Interorganizational Collaboration: The Emergence of Proto-institutions. Academy of Management Journal, 45(1), 281-290. Lazzarini, S. G. (2015). Strategizing by the Government: Can Industrial Policy Create Firm-Level Competitive Advantage? Strategic Management Journal, 36(1), 97- 112. Lee, E. & Puranam, P. (2015). The Nature of Expertise in Organization Design: Evidence from an Expert-Novice Comparison. Cognition and Strategy, 32, 181-209. Lee, E. & Puranam, P. (2016). The Implementation Imperative: Why One Should Implement Even Imperfect Strategies Perfectly. Strategic Management Journal, 37(8), 1529-1546. Lee, S. & Puranam, P. (2017). Incentive Redesign and Collaboration in Organizations: Evidence from a Natural Experiment. Strategic Management Journal, 38(12), 2333-2352. Leeds City Council. (2016). Council Leader Welcomes HS2 Preferred Route and Revamped Leeds Station Confirmation. Leeds: Leeds City Council. Lessard, D. & Miller, R. (2000). Mapping and Facing the Landscape of Risks. The Strategic Management of Large Engineering Projects: Shaping Institutions, Risks, and Governance. Cambridge, Massachusetts: The MIT Press.

197

Lessard, D., Sakhrani, V. & Miller, R. (2014). House of Project Complexity: Understanding Complexity in Large Infrastructure Projects. Engineering Project Organization Journal, 4(4), 170-192. Levinthal, D. & Posen, H. E. (2007). Myopia of Selection: Does Organizational Adaptation Limit the Efficacy of Population Selection? Administrative Science Quarterly, 52(4), 586-620. Levinthal, D. A. & March, J. G. (1993). The Myopia of Learning. Strategic Management Journal, 14, 95-112. Levitt, R. (2015). An Extended Coordination Theory for Governance of Tasks, Projects, Firms and Business Ecosystems. GPC Working Paper. Stanford, CA: Global Projects Center; Stanford University. Levitt, R. E. (1984). Superprojects and Superheadaches: Balancing Technical Economies of Scale Against Management Diseconomies of Size and Complexity. Project Management Institute. Levitt, R. E. (2011). Towards Project Management 2.0. Engineering Project Organization Journal, 1(3), 197-210. Levitt, R. E. & Scott, W. R. (2017). Institutional Challenges and Solutions for Global Megaprojects. In: Flyvbjerg, B. (ed.) The Oxford Handbook of Megaproject Management. Oxford: Oxford University Press. Levitt, R. E., Thomsen, J., Christiansen, T. R., Kunz, J. C., Jin, Y. & Nass, C. (1999). Simulating Project Work Processes and Organizations: Toward a Micro- Contingency Theory of Organizational Design. Management Science, 45(11), 1479-1495. Levy, B. (2014). Working with the Grain: Integrating Governance and Growth in Development Strategies. New York: Oxford University Press. Lindkvist, L. (2004). Governing Project-based Firms: Promoting Market-like Processes within Hierarchies. Journal of Management and Governance, 8(1), 3-25. Locke, K. (2007). Rational Control and Irrational Free-play: Dual-thinking Modes as Necessary Tension in Grounded Theorizing. The SAGE handbook of grounded theory, 565-579. Lundrigan, C. P., Gil, N. A. & Puranam, P. (2015). The (Under) Performance of Mega- projects: A Meta-Organizational Perspective. In: 75th Academy of Management Conference, 2015 Vancouver. Academy of Management Proceedings. Mahalingam, A. & Levitt, R. E. (2007). Institutional Theory as a Framework for Analyzing Conflicts on Global Projects. Journal of Construction Engineering and Management-ASCE, 133(7), 517-528. Manchester City Council. (2014a). Central Manchester Strategic Regeneration Framework. Manchester: Manchester City Council. Manchester City Council. (2014b). A City Council Response. Manchester: Manchester City Council. Manchester City Council Executive. (2016). Manchester City Council Report for Resolution. Manchester: Manchester City Council. Maoret, M., Massa, F. G. & Jones, C. (2011b). Toward a Projects as Events Perspective. In: Cattani, G., Ferriani, S. & Frederiksen, L., Täube, F (eds.) Project-Based Organizing and Strategic Management. Emerald Group Publishing Limited, 28, 427-444. March, J. G. (1991). Exploration and Exploitation in Organizational Learning. Organization Science, 2(1), 71-87. March, J. G. & Simon, H. A. (1967). Organizations. New York: John Wiley & Sons Inc.

198

Martinsuo, M., Hensman, N., Artto, K., Kujala, J. & Jaafari, A. (2006). Project-based Management as an Organizational Innovation: Drivers, Changes, and Benefits of Adopting Project-based Management. Project Management Journal, 37(3), 87. Maxwell, J. A. (1996). Qualitative Research Design: An Interactive Approach. Thousand Oaks, CA: Sage. McAdam, D., Boudet, H. S., Davis, J., Orr, R. J., Richard Scott, W. & Levitt, R. E. (Year). “Site fights”: Explaining Opposition to Pipeline Projects in the Developing World. In: Sociological Forum, 2010. Wiley Online Library, 401-427. McKean, M. A. (1992). Success on the Commons: A Comparative Examination of Institutions for Common Property Resource Management. Journal of theoretical politics, 4(3), 247-281. Merrow, E. W. (2011). Industrial Megaprojects: Concepts, Strategies, and Practices for Success. Hoboken, NJ: Wiley Online Library. Meyer, A. D., Gaba, V. & Colwell, K. A. (2005). Organizing Far from Equilibrium: Nonlinear Change in Organizational Fields. Organization Science, 16(5), 456-473. Miles, M. B. & Huberman, A. M. (1984). Qualitative Data Analysis. Beverly Hills, CA: Sage. Miller, R. & Hobbs, B. (2009). The Complexity of Decision-making in Large Projects with Multiple Partners: Be Prepared to Change. In: Williams, T., Samset, K. & Sunnevag, K. (eds.) Making essential choices with scant information. Basingstoke, UK: Palgrave MacMillan. Miller, R., Lessard, D. R., Michaud, P. & Floricel, S. (2001). The Strategic Management of Large Engineering Projects: Shaping Institutions, Risks, and Governance. Cambridge, MA: The MIT press. Mintzberg, H. (1979). The Structuring of Organizations: A Synthesis of the Research. New Jersey: Englewood Cliffs, NJ: Prentice-Hall. Mintzberg, H. (1989). The Structuring of Organizations. Readings in Strategic Management. Springer. Mintzberg, H., Ahlstrand, B. W. & Lampel, J. (1998). Strategy Safari : A Guided Tour through the Wilds of Strategic Management. New York: Free Press. Mintzberg, H., Lampel, J., Quinn, J. B. & Ghoshal, S. (2003). The Strategy Process: Concepts, Contexts, Cases. Essex, UK: Pearson Education Limited. Mintzberg, H. & Waters, J. A. (1985). Of Strategies, Deliberate and Emergent. Strategic Management Journal, 6(3), 257-272. Morgan, B. V. (1987). Benefits of Project Management at the Front-end. International Journal of Project Management, 5(2), 102-119. Morris, P. W. (1997). The Management of Projects. London: Thomas Telford. Morris, P. W. (2013). Reconstructing Project Management. Chichester: Wiley-Blackwell. Morris, P. W., Pinto, J. K. & Söderlund, J. (2012). The Oxford Handbook of Project Management. Oxford: Oxford University Press. Mounce, H. O. (1997). The Two Pragmatisms: From Peirce to Rorty: Routeledge. Msulwa, R. S. (2013). Using a Design Commons to Develop New Infrastructure Projects: The Case of the UK's High Speed Two Project. MSc Dissertation, University of Manchester. Nadler, D. A. & Tushman, M. L. (1997). A Congruence Model for Organization Problem Solving. Managing Strategic Innovation and Change: A Collection of Readings, 159-171. National Audit Office. (2013). High Speed 2: A Review of Early Programme Preparation. London: NAO. National Audit Office. (2016). Progress with Preparations for High Speed 2. London: NAO. 199

Nelson, R. R. & Winter, S. G. (1982). An Evolutionary Theory of Economic Change. Cambridge, MA: Belkap Press. Nightingale, P., Baden-Fuller, C. & Hopkins, M. M. (2011). Projects, Project Capabilities and Project Organizations. In: G. Cattani, S. F., L. Frederiksen, F. Täube (ed.) Project-Based Organizing and Strategic Management. Advances in Strategic Management. Emerald Group Publishing Limited. Nordqvist, S., Hovmark, S. & Zika-Viktorsson, A. (2004). Perceived Time Pressure and Social Processes in Project Teams. International Journal of Project Management, 22(6), 463-468. O'Mahony, S. & Ferraro, F. (2007). The Emergence of Governance in an Open Source Community. Academy of Management Journal, 50(5), 1079-1106. Oakerson, R. J. (1993). Reciprocity: A Bottom-up View of Political Development. Rethinking Institutional Analysis and Development: Issues, alternatives, and choices, 141-58. Okhuysen, G. A. & Bechky, B. A. (2009). Coordination in Organizations: An Integrative Perspective. Academy of Management Annals, 3(1), 463-502. Olson, M. (1965). The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge, MA: Harvard University Press. Ostrom, E. (1986). An Agenda for the Study of Institutions. Public choice, 48(1), 3-25. Ostrom, E. (1990). Governing the Commons: The Evolution of Institutions for Collective Action. New York: Cambridge University Press. Ostrom, E. (1998). A Behavioural Approach to the Rational Choice Theory of Collective Action: Presidential address, American Political Science Association, 1997. American Political Science Review, 92(1), 1-22. Ostrom, E. (2005). Understanding Institutional Diversity. Princeton, NJ: Princeton University Press. Ostrom, E. (2007). Institutional Rational Choice: An Assessment of the IAD Framework. In: Sabatier, P. (ed.) Theories of the Policy Process. 2nd ed. Cambridge, MA: Westview Press. Ostrom, E. (2008). Institutions and the Environment. Economic affairs, 28(3), 24-31. Ostrom, E. (2009). Design Principles of Robust Property-rights Institutions: What Have We Learned? In: Ingram, G. K. & Hong, Y. (eds.) Property Rights and Land Policies. Cambridge, MA: Lincoln Institute of Land Policy. Ostrom, E. (2011). Background on the Institutional Analysis and Development Framework. Policy Studies Journal, 39(1), 7-27. Ostrom, E. (2012). Nested Externalities and Polycentric Institutions: Must We Wait for Global Solutions to Climate Change Before Taking Actions at Other Scales? Economic Theory, 49(2), 353-369. Ostrom, E., Dietz, T., Dolšak, N. E., Stern, P. C., Stonich, S. & Weber, E. U. (2002). The Drama of the Commons. Washington, DC: National Academy Press. Ostrom, E., Gardner, R. & Walker, J. (1994). Rules, Games, and Common-pool Resources. Ann Abor: University of Michigan Press. Ostrom, E. & Janssen, M. A. (2004). Multi-level Governance and Resilience of Social- ecological Systems. In: Spoor, M. (ed.) Globalisation, poverty and conflict. Dordrecht, The Netherlands: Kluwer Academic. Ostrom, E. & Walker, J. (2003). Trust and Reciprocity: Interdisciplinary Lessons for Experimental Research: Russell Sage Foundation. Ostrom, V. (1999). Polycentricity (Part 1). In: Polycentricity and Local Public Economies: Readings from the Workshop in Political Theory and Policy Analysis, 1999. Ann Arbor: University of Michigan Press, 52-74.

200

Ostrom, V. & Ostrom, E. (1999). Public goods and public choices. Polycentricity and Local Public Economies. Readings from the Workshop in Political Theory and Policy Analysis. [originally published in 1977] ed. Ann Arbor: University of Michigan Press. Ostrom, V., Tiebout, C. M. & Warren, R. (1961). The Organization of Government in Metropolitan Areas: A Theoretical Inquiry. American Political Science Review, 55(4), 831-842. Pettigrew, A. M. (1973). The Politics of Organizational Decision-making. London: Tavistock. Pettigrew, A. M. (1990). Longitudinal Field Research on Change: Theory and Practice. Organization Science, 1(3), 267-292. Pfeffer, J. & Nowak, P. (1976). Joint Ventures and Interorganizational Interdependence. Administrative Science Quarterly, 398-418. Pfeffer, J. & Salancik, G. R. (2003). The External Control of Organizations: A Resource Dependence Perspective. Stanford, CA: Stanford University Press. Pinto, J. K. & Slevin, D. P. (1987). Critical Factors in Successful Project Implementation. IEEE Transactions on Engineering Management, (1), 22-27. Pinto, J. K. & Winch, G. (2016). The Unsettling of “Settled Science:” The Past and Future of the Management of Projects. International Journal of Project Management, 34(2), 237-245. Pitsis, T. S., Clegg, S. R., Marosszeky, M. & Rura-Polley, T. (2003). Constructing the Olympic Dream: A Future Perfect Strategy of Project Management. Organization Science, 14(5), 574-590. Porter, L. W., Lawler, E. E. & Hackman, J. R. (1975). Behavior in Organizations. New York: McGraw-Hill. Posen, H. E. & Levinthal, D. A. (2012). Chasing a Moving Target: Exploitation and Exploration in Dynamic Environments. Management Science, 58(3), 587-601. Prencipe, A., Davies, A. & Hobday, M. (2003). The Business of Systems Integration. Oxford: Oxford University Press. Prendergast, C. (1999). The Provision of Incentives in Firms. Journal of economic literature, 37(1), 7-63. Public Accounts Committee. (2013). High Speed 2: A Review of Early Programme Preparation. London: House of Commons. Public Accounts Committee. (2016). Progress with Preparations for High Speed 2. London: House of Commons. Puranam, P., Alexy, O. & Reitzig, M. (2014). What's “New” About New Forms of Organizing? Academy of Management Review, 39(2), 162-180. Puranam, P., Goetting, M. & Knudsen, T. (2010). Interdependence, Information Processing and Organization Design: An Epistemic Perspective. In: Academy of Management Proceedings, 2010. Academy of Management, 1-6. Puranam, P., Raveendran, M. & Knudsen, T. (2012). Organization Design: The Epistemic Interdependence Perspective. Academy of Management Review, 37(3), 419-440. Puranam, P., Stieglitz, N., Osman, M. & Pillutla, M. M. (2015). Modelling Bounded Rationality in Organizations: Progress and Prospects. The Academy of Management Annals, 9(1), 337-392. Puranam, P. & Swamy, M. (2016). How Initial Representations Shape Coupled Learning Processes. Organization Science, 27(2), 323-335. Puranam, P. & Vanneste, B. S. (2009). Trust and Governance: Untangling a Tangled Web. Academy of Management Review, 34(1), 11-31. Rail Technology Magazine. (2015). Higgins to Look at Other Sites for Leeds HS2 Station. Rail Technology Magazine. London: Cognitive Publishing, Ltd. 201

Raymond, E. (1999). The Cathedral and the Bazaar: Musings on Linux and Open Source by an Accidental Revolutionary (Vol. 12). Sepastopol, CA: O'Reilly Media Inc. Reuer, J. J. (2004). Strategic Alliances. New York: Oxford University Press. Rivkin, J. W. & Siggelkow, N. (2003). Balancing Search and Stability: Interdependencies Among Elements of Organizational Design. Management Science, 49(3), 290-311. Rogers, E. M. & Shoemaker, F. F. (1971). Communication of Innovations: A Cross- Cultural Approach. New York: Free Press. Ross, J. & Staw, B. M. (1993). Organizational Escalation and Exit: Lessons from the Shoreham Nuclear Power Plant. Academy of Management Journal, 36(4), 701-732. Salancik, G. R. & Pfeffer, J. (1978). A Social Information Processing Approach to Job Attitudes and Task Design. Administrative Science Quarterly, 224-253. Sally, D. (1995). Conversation and Cooperation in Social Dilemmas: A Meta-analysis of Experiments from 1958 to 1992. Rationality and Society,7(1), 58-92. Samset, K. & Volden, G. H. (2016). Front-end Definition of Projects: Ten Paradoxes and Some Reflections Regarding Project Management and Project Governance. International Journal of Project Management, 34(2), 297-313. Sanchez, R. & Mahoney, J. T. (1996). Modularity, Flexibility, and Knowledge Management in Product and Organization Design. Strategic Management Journal, 17(2), 63-76. Schelling, T. C. (1960). The Strategy of Conflict. Cambridge, Massachusetts: Harvard University Press. Scott, W. R. (2003). Institutional Carriers: Reviewing Modes of Transporting Ideas Over Time and Space and Considering Their Consequences. Industrial and Corporate Change, 12(4), 879-894. Shapiro, A. & Lorenz, C. (2000). Large-Scale Projects as Complex Systems: Managing Scope Creep. The Systems Thinker, 11(1), 1-5. Sheffield City Council and South Yorkshire Passenger Executive. (2012). Maximizing the Economic Impact of the HS2 Investment in Sheffield. Sheffield: Sheffield City Council. Sheffield City Council. (2014b). Sheffield Victoria Station Development Prospectus. Sheffield: Sheffield City Council. Shenhar, A. J., Dvir, D., Levy, O. & Maltz, A. C. (2001). Project Success: A Multidimensional Strategic Concept. Long Range Planning, 34(6), 699-725. Siggelkow, N. & Rivkin, J. W. (2005). Speed and Search: Designing Organizations for Turbulence and Complexity. Organization Science, 16(2), 101-122. Simon, H. A. (1947). Administrative Behavior: A Study of Decision-making Processes in Administrative Organisations. New York: Free Press. Simon, H. A. (1951). A Formal Theory of the Employment Relationship. Econometrica: Journal of the Econometric Society, 293-305. Simon, H. A. (1955). A Behavioural Model of Rational Choice. The quarterly journal of economics, 69(1), 99-118. Simon, H. A. (1957). Models of Man: Social and Rational. New York: Wiley. Simon, H. A. (1962a). New Developments in the Theory of the Firm. The American economic review, 52(2), 1-15. Simon, H. A. (1962b). The Architecture of Complexity. Proceedings of the American Philosophical Society, 106(6), 467-482. Simon, H. A. (1972). Theories of Bounded Rationality. Decision and organization, 1(1), 161-176. Simon, H. A. (1996). The Sciences of the Artificial. Cambridge, MA: The MIT Press.

202

Sosa, M., Gargiulo, M. & Rowles, C. (2007). Component Connectivity, Team Network Structure and the Attention to Technical Interfaces in Complex Product Development. Journal of Mechanical Design, 129(11), 1118-1129. South, A. J., Levitt, R. E. & Dewulf, G. (2015). Dynamic Stakeholder Networks and the Governance of PPPs. In: Proceedings of the 2nd International Conference on Public-Private Partnerships, 2015. 26-29. Srikanth, K. & Puranam, P. (2011). Integrating Distributed Work: Comparing Task Design, Communication, and Tacit Coordination Mechanisms. Strategic Management Journal, 32(8), 849-875. Srikanth, K. & Puranam, P. (2014). The Firm as a Coordination System: Evidence from Software Services Offshoring. Organization Science, 25(4),1253-1271. Star, S. L. (1989). The Structure of Ill-Structured Solutions: Heterogeneous Problem- solving, Boundary Objects and Distributed Artificial Intelligence. Distributed Artificial Intelligence, 2, 37-54. Star, S. L. & Griesemer, J. R. (1989). Institutional Ecology, Translations' and Boundary Objects: Amateurs and Professionals in Berkeley's Museum of Vertebrate Zoology, 1907-39. Social Studies of Science, 19(3), 387-420. Stinchcombe, A. L. (1990). Information and Organizations. Berkeley, CA: University of California Press. Sydow, J., Lindkvist, L. & DeFillippi, R. (2004). Project-based Organizations, Embeddedness and Repositories of Knowledge. Organization Studies, 25, 1475- 1488. Teece, D. J., Pisano, G. & Shuen, A. (1997). Dynamic Capabilities and Strategic Management. Strategic Management Journal, 18, 509-533. Thomas, J. B. & Trevino, L. K. (1993). Information Processing in Strategic Alliance Building: A Multiple‐case Approach. Journal of Management Studies, 30(5), 779- 814. Thompson, J. D. (1967). Organizations in Action: Social Science Bases of Administrative Theory. New York: McGraw-Hill. Tuertscher, P., Garud, R. & Kumaraswamy, A. (2014). Justification and Interlaced Knowledge at ATLAS, CERN. Organization Science, 25(6), 1579-1608. Tushman, M. & Murmann, J. P. (2002). Dominant Designs, Technology Cycles, and Organizational Outcomes. Managing in the Modular Age: Architectures, Networks, and Organizations, 316. Tushman, M. & Nadler, D. (1986). Organizing for Innovation. California Management Review, 28(3), 74-92. Tushman, M. L. (1978). Technical Communication in R&D Laboratories: The Impact of Project Work Characteristics. Academy of Management Journal, 21(4), 624-645. Tushman, M. L. & Nadler, D. A. (1978). Information Processing as an Integrating Concept in Organizational Design. Academy of Management Review, 3(3), 613- 624. Tushman, M. L. & Rosenkopf, L. (1992). Organizational Determinants of Technological- change: Towards a Sociology of Technological Evolution. Research in Organizational Behavior, 14, 311-347. Ulrich, K. T. (2011). Design is Everything? Journal of Product Innovation Management, 28(3), 394-398. Van de Ven, A. H. (2007). Engaged Scholarship: A Guide for Organizational and Social Research. Oxford, UK: Oxford University Press. Van de Ven, A. H. & Huber, G. P. (1990). Longitudinal Field Research Methods for Studying Processes of Organizational Change. Organization Science, 1(3), 213- 219. 203

van der Westhuizen, J. (2017). Megaprojects as Political Symbols. In: Flyvbjerg, B. (ed.) The Oxford Handbook of Megaproject Management. Oxford: Oxford University Press. Von Hippel, E. (1988). The Sources of Innovation. New York: Oxford University Press. Von Hippel, E. (1990). Task Partitioning: An Innovation Process Variable. Research Policy, 19(5), 407-418. Wachs, M. (1989). When Planners Lie with Numbers. Journal of the American Planning Association, 55(4), 476-479. Wade, R. (1987). The Management of Common Property Resources: Collective Action as an Alternative to Privatisation or State Regulation. Cambridge journal of economics, 11(2), 95-106. Walker, J. & Ostrom, E. (2009). Trust and Reciprocity as Foundations for Cooperation. In: Cook, K., Levi, M. & Hardin, R. (eds.) Whom Can We Trust? How Groups, Networks, and Institutions Make Trust Possible. New York: Russell Sage Foundation. Walsh, J. P., Meyer, A. D. & Schoonhoven, C. B. (2006). A Future for Organization Theory: Living in and Living with Changing Organizations. Organization Science, 17(5), 657-671. Weick, K. E. (1969). The Social Psychology of Organizing. Reading, MA: Addison- Wesley. Weick, K. E. (1974). Amendments to Organizational Theorizing. Academy of Management Journal, 17(3), 487-502. Welde, M. & Odeck, J. (2017). Cost Escalations in the Front-end of Projects: Empirical Evidence from Norwegian Road Projects. Transport Reviews, 37(5), 612-630. Williams, T. (2008). How do Organizations Learn Lessons from Projects—And Do They? IEEE transactions on engineering management, 55(2), 248-266. Williams, T. & Samset, K. (2010). Issues in Front‐end Decision-making on Projects. Project Management Journal, 41(2), 38-49. Williamson, O. (1985). The Economic Institutions of Capitalism: Firms, Markets, Relational contracting. Free Press. Williamson, O. E. (1975). Markets and Hierarchies: Analysis and Anti-Trust Implications. New York: The Free Press. Williamson, P. J. & De Meyer, A. (2012). Ecosystem Advantage: How to Successfully Harness the Power of Partners. California Management Review, 55(1), 24-46. Winch, G. & Bonke, S. (2002). Project Stakeholder Mapping: Analyzing the Interests of Project Stakeholders. In: Slevin, D. P., Cleland, D. I. & Pinto, J. K. (eds.) Frontiers of Project Management Research. Project Management Institute. Winch, G. M. (2010). Managing Construction Projects (2nd ed.). West Sussex, UK: John Wiley & Sons. Winch, G. M. (2015). Project Organizing as a Problem in Information. Construction management and economics, 33(2), 106-116. Winch, G. (2017). Megaproject Stakeholder Management. The Oxford Handbook of Mega-project Management. Oxford University Press, Oxford, UK, 339-361. Winch, G. M. & Courtney, R. (2007). The Organization of Innovation Brokers: An International Review. Technology Analysis & Strategic Management, 19(6), 747- 763. Yin, R. K. (1994). Case Study Research: Applied Social Research Methods Series. California: Newbury Park, 5. Youker, R. (1999). The Difference Between Different Types of Projects. In: Newton Square, P., ed. Proceedings of the Project Management Institute Annual Seminars & Symposium, 1999. Project Management Institute. 204

Young, O. R. (2002). Institutional Interplay: The Environmental Consequences of Cross- scale Interactions. In: Ostrom, E., Dietz, T., Dolsak, N., Stern, P., Stonich, S. & Weber, E. (eds.) The Drama of the Commons. Washington, DC: National Academy Press. Zammuto, R. F., Griffith, T. L., Majchrzak, A., Dougherty, D. J. & Faraj, S. (2007). Information Technology and the Changing Fabric of Organization. Organization Science, 18(5), 749-762. Zander, U. & Kogut, B. (1995). Knowledge and the Speed of the Transfer and Imitation of Organizational Capabilities: An Empirical Test. Organization Science, 6(1), 76-92. Zeng, M. & Chen, X.-P. (2003). Achieving Cooperation in Multiparty Alliances: A Social Dilemma Approach to Partnership Management. Academy of Management Review, 28(4), 587-605.

205

APPENDICES

Appendix A Interview Details

Appendix A 1 Interviewees, job roles and employers

Organizational Description of Number of interviewees and job roles actors Organizational Actor #17 interviewees

Job roles : CEO, Chairman, Senior Route Engineer (North West), Director Program Cost and commercial arrangements, Technical Director, Head of Route Engineering, Area Manager West High Speed 2 The public agency Midlands/Manager of Design and development Limited (HS2 responsible for the HS2 Birmingham,Leeds, Manchester and Heathrow Project Director, Ltd) planning and delivery Head of Bill Preparation, Country North Community and Stakeholder Manager, Commercial Director, Area Manager (Euston), Country North Community and Stakeholder Manager, Director Program Cost and Commercial Arrangements, London West Midlands Phase 1 Programme Director, Area Manager (Euston), Head of Route Engineering #2 interviewees Engineering Delivery CH2MHill Partner of HS2 Ltd Job Roles : Cost and Efficiency Manager ; HS2 Phase One Project Director Owner of the UK’s railway #2 Interviewees network (priva te company Network Rail limited by a public Job Roles : Head of High Speed Rail Development, Euston guarantee) Development Director, Top-tier administrative #1 Interviewees Greater London body for Greater London, Authority England Job Roles : Senior Manager Planning Decisions, Local government body #1 Interviewees Transport for responsible for the London transport system in Greater Job Roles : Lead Sponsor on HS2 London, England Public body responsible for #1 Interviewees Transport for co-ordinating transport Greater services throughout Manchester Job Roles : Rail Programme Director Greater Manchester #5 Interviewees Responsible for local Local strategic planning and the Job Roles : London Borough of Camden : Director for Authorities provision of a range of Regeneration and Planning Leeds City Council : Transport public services Strategy Manager ; Manchester City Council : Head of Regeneration, (; Leader of the Council Sheffield City Council : Executive Director Places,; Liverpool : Chair Merseytravel #12 Interviewees

Other Stakeholders with an Job Roles : Strategic Urban Design, Regeneration & Planning organizational interest in HS2 Principal, AECOM ; Land and Property Director, Crossrail ; CEO, actors London First ; Former Chief Planner, Planning Director, Manchester Airport ; Project Director, Infrastructure UK (HM Treasury)

206

Appendix A 2 Number of Interviews and Job Roles by Unit of Analysis

Number of Unit 1 Job Roles interviews

Chairman HS2 Ltd., CEO HS2 Ltd., Head of Route Engineering HS2 Ltd., Techncial Director HS2 Ltd., London West Midlands Phase 1 Director HS2 Ltd., Cost and Effieciency Manager CH2MHill (Seconded to HS2 Ltd), Senior Route #14 Engineer (North West) HS2 Ltd., Head of Bill Preparation HS2 Ltd., Area Manager West Midlands/Manager of Design and development of Birmingham,

Birmingham Project Director Infrastructure UK (HM Treasury) Chairman HS2 Ltd., CEO HS2 Ltd., Head of Route Engineering HS2 Ltd., Director Program Cost and Commercial Arrangements, Techncial Director HS2 Ltd., London West Midlands Phase 1 Programme Director HS2 Ltd., Commercial Director HS2 Ltd., Cost and Effieciency Manager CH2MHill (Seconded to HS2 Ltd., Senior Route Engineer (North West) HS2 Ltd., Country North Community #22 and Stakeholder Manager HS2 Ltd., Area Manager (Euston) HS2 Ltd., Project

London Director Infrastructure UK (HM Treasury), Senior Manager (Planning Decisions) Greater London Authority, TfL/HS2 Lead Sponsor, Assistant Director for Regeneration and Planning, London Borough of Camden, CEO London First, Head of High Speed Rail Development Network Rail Chairman HS2 Ltd., CEO HS2 Ltd., Head of Route Engineering HS2 Ltd.,

Director Program Cost and Commercial Arrangements, Techncial Director HS2

Ltd., London West Midlands Phase 1 Director HS2 Ltd., Commercial Director HS2 Ltd., Senior Route Engineer (North West) HS2 Ltd., Rail Programme #22 Director Transport for Greater Manchester (TfGM), Head of Rail (Interim) TfGM, Head of Manchester City Council Regeneration, Deputy Mayor Manchester City

Manchester Council, Planning Director Manchester Airports Group, Former Senior Transport Official North West Development Agency, Executive Director (Places) Sheffield City Council

Chairman HS2 Ltd., CEO HS2 Ltd., Head of Route Engineering HS2 Ltd., Director Program Cost and Commercial Arrangements, Senior Route Engineer #18 (North West) HS2 Ltd., , Former Senior Transport Official North West Leeds Development Agency, Transport Strategy Manager Leeds

Chairman HS2 Ltd., CEO HS2 Ltd., Head of Route Engineering HS2 Ltd., Techncial Director HS2 Ltd., , Former Senior Transport Official North West #11 Development Agency, Executive Director (Places) Sheffield City Council

Sheffield In 13 interviews with senior officials we discussed the same issues across the five units of analysis as opposed to focusing on a single unit of analysis, and some officials were interviewed more than once.

207

Appendix B Exploratory Questions for HS2 Ltd. Informant Interviews

1. What is the HS2 story? 2. Does this story apply to both phases of HS2? 3. At what stage of development is HS2 in? How likely is it to succeed? 4. HS2 Ltd. is described as an arm’s length delivery body- what does this mean? 5. What is the company’s ultimate goal? 6. How can you set up a company such as HS2 to win whilst playing a fair game? 7. In your role, how are you helping the company achieve its goal? 8. Who are your allies in the project ecosystem? Who are your opponents in the ecosystem? What capabilities does HS2 Ltd. outsource? Who are your complementors? (other unconventional sources of value?) 9. Who makes up the core of HS2? Which players are at the periphery? 10. To what extent can HS2 influence the interactions with allies, opponents? 11. To what extent are you engaged in processes such as public consultations? 12. Who is the ultimate client of HS2? 13. What is the role of politicians? What is the role of Government (Cabinet)? What is the role of Parliament? What is the role of local authorities? 14. Does having cross party support make achieving HS2’s goals easier? 15. What criteria were used to assess the different route options (design)? 16. Why was the final route design chosen over other options? 17. How is the design going to evolve over time? What parts can you freeze? 18. What is your timescale? Will that change over time? 19. What is the budget? How likely it is to evolve over time? 20. There has recently been a judicial review into whether Governments consultation of the first phase was flawed. If the High Court rules against government, how will this affect the timescale of the project? 21. There have been reports that HS2 could provide faster broadband to rural areas. Has this feature always been a part of the design? 22. How can you sell HS2 to the country in an age of austerity? 23. Where is the finance for HS2 coming from? Is private finance an option? 24. Who will operate HS2? Why does the UK need HS2, as opposed to upgrading existing rail lines or roads, which arguably have much higher returns?

208

Appendix C Teaching Case Studies

32 Appendix C 1 Britain’s High Speed 2: All Aboard? (A)

The biggest transport project Britain’s done in centuries... Complicated? Yes. But fun. [HS2 Chairman]

By early January 2013, Doug Oakervee, the shrewd chairman of HS2 Ltd. had reasons to be optimistic. HS2 Ltd. was the company wholly owned by the UK’s Department forDfT), and created in January 2009 to promote a new high-speed railway network. The goal was to connect London to Birmingham in a first phase by 2026, and then connect Birmingham to two major conurbations in the North of England, Manchester and Leeds, by 2032. The scheme’s P50 budget33 was £32.7bn (in 2011 prices) and included a robust £8bn contingency, but was yet to include a projected budget for acquiring over 170 train cars.

The goal was ambitious and perhaps justifiably so—the last major railway line North of London had been built over 120 years ago. Still, HS2 had enjoyed strong cross-party consensus since the idea gained traction in 2008. The UK was going through a severe economic downturn, and the political leaders shared the conviction the time was right to modernize the national railway network. And in May 2010, when a Tory-led coalition ousted the Labour government, the scheme continued to enjoy governmental support despite fierce lobbying against it from interest groups in the Tory heartland north of London, Doug said:

The people who want HS2 are Manchester, Leeds, Sheffield, Birmingham; Londoners are a bit indifferent ...but the people between London and Birmingham say ‘well, what’s in it for me?’— huge lobby against us, the Chilterns...we’ve got other pockets, Warrington, Cheshire

Through a process that Doug facetiously dubbed ‘too democratic’ (‘nothing is too democratic, but it feels like that sometimes’, he quipped), the HS2 route was specified. By January 2009, consultation for phase one was completed, but the outcome of legal challenges taken against HS2 Ltd. was yet to be announced. Still, Doug and Alison, the chief executive of HS2 Ltd., were optimistic that the outcome would be favourable to them. Furthermore, details of the route for phase two were due to be published later in January, and plans to publish a draft Environmental Statement for consultation were also on track.

The political pressure to keep the scheme on track was enormous since the government had pledged to get the first phase (approximately 225km long) approved by the UK Parliamentby May 2015, ahead of the next national elections. To meet this target, HS2 Ltd. had been instructed to lodge the HS2 bill for phase one with the Parliament by autumn 2013.

HS2 Ltd. officials were dancing to the government’s tune, but many recognised that the 2015 deadline was not realistic. Only the UK Parliament could give legal powers for the government to compulsory purchase the land required for the scheme. And history suggested things would take time. First, the Parliament needed to listen to petitions from everyone who felt materially disadvantaged by the bill. A previous bill for the £5.8bn 108km HS1 connecting London to the

32 Rehema Msulwa, PhD student and Professor Nuno Gil, The University of Manchester, prepared this case as the basis for class discussion. The case does not intend to serve as endorsement, source of primary data, or illustration of effective or ineffective handling of an administrative situation. The authors are solely responsible for any factual inaccuracies. Copyright © 2014 (May), Rehema Msulwa and Nuno Gil All rights reserved 33 A P50 budget meant that the probabilities of the final costs exceeding or not £32.7bn was 50:50 based on a risk assessment that quantified the likelihood of occurrence of all the foreseeable risks and their cost impacts in the event of their occurrence.

209

Channel Tunnel received 1,000 petitions and took 25 months of discussions (and the addition of 647 undertakings and assurances) to become law.34

Doug, Alison and their highest ranked officials spent countless days in meetings all over the country. Doug said, ‘I was engaged as a two day a week job, but usually find it more like a 4-day a week job.’ For the first phase, the most difficult issue to resolve was how to integrate HS2 into London’s transport network. The plan was to develop two stations. One would be located on a derelict industrial site (Old Oak Common) for which the London leaders had great ambitions. And the other would involve overhauling the dilapidated and congested Euston station in central London that was operated by Network Rail, the state monopoly that owned the UK railway infrastructure. The plan was ambitious, explained one official: a bit like a fairy-godmother scheme where we came along with our magic wand and we sort of obliterate everything and start again with everything new and shiny

As discussions continued, demands from the London stakeholders kept coming in. At Old Oak Common, key local stakeholders—four local councils, Greater London Authority35, and Transport for London36—wanted to seize the opportunity that a new high-speed line would bring. According to them, the scheme could catalyse the development of over 19,000 new homes and thus address the shortage of affordable housing in London. In their view, the scope of the HS2 station had to augment commensurately. This would put however massive pressure on the budget that HS2 officials had earmarked for the station.

Things were even more complicated at Euston. First, the political leaders of the local council opposed HS2 because of the land take that was required and the impacts in the borough. Even so, the Council wanted to redevelop the Euston station, and thus joined forces with the GLA and TfL to develop a so-called Euston Area Plan. As discussions progressed, the cost estimate to rebuild Euston doubled from £1.1bn to over £2bn, and the time required to complete all works spiralled to 10 years away from the initially planned 7 years.

What should Doug and Alison recommend to the government? ‘If we keep on with that, one HS2 official said, ‘we’re going to so severely breach our targets that the whole reputation of HS2 will be in doubt’. Given that parts of Euston were still fit for purpose, they could backtrack from the fairy- godmother’s scheme, and propose a modest intervention. This would surely infuriate the local stakeholders who had already bought into the idea of lowering all the platforms and tracks at Euston, the plan that was circulated for consultation.

The HS2 executives were also troubled that following the consultation the secretary of state agreed to mitigation measures to resolve a conflict over the design for the Chilterns, an area of outstanding beauty. The deal committed HS2 Ltd. to design in extra tunnelling through the Chilterns and other sensitive locations at an extra cost of £700m, thus increasing the price of the whole scheme. The decision could potentially set a tricky precedent:

34 And 365 petitions were lodged against another bill, the £15bn, 118km Crossrail, Whilst the the issues were settled outside Parliament for 252, it still took 40 months of discussions to agree around 700 undertakings and assurances before the bill became law.

35 Greater London Authority was the top-tier administrative body for Greater London, mostly funded by government grant. It consisted of a directly elected executive Mayor of London, Boris Johnson at the time, and an elected 25-member Assembly with scrutiny powers.

36 Transport for London was a local government body responsible for most aspects of the transport system in Greater London, and was part of the GLA.

210

It’s a bit of a poker game here... because if you give it away in the beginning the next thing is they want something else ... and I’m not too certain what real difference it’s made... you’re never going to satisfy the Chilterns, they’re always an opposition group [Doug]

Why couldn’t things always be as simple as developing the Manchester stations, remarked Doug? In Manchester, although there were lots of local players at the negotiating table, so far settling the issues regarding critical decisions had been relatively straightforward.

Doug and Alison’s calendar was jam-packed with meetings with their sponsor, the Department for Transport, the London stakeholders, and many others affected by the project. Under pressure to submit the bill for phase 1 by the end of 2013—the whole thing was expected to exceed 50,000 pages!—the HS2 leaders needed to sketch a strategy to keep the development on track. One thing they were sure about. The London stakeholders and many others would petition against the bill to try to extract more concessions in Parliament.

HIGH-SPEED RAILWAY TECHNOLOGY

The history of High-speed rail (HSR) technology dates as far back as 1940 when the Tokaido Shinkasen, the world's first high-speed railway line, was conceptualised in Japan. Once completed in 1964, the Shinkasen, operating between Tokyo and Osaka, was designed to run trains at a top speed of 210 km/h (130 mph), but since then the trains on the line were reaching top speeds of 270 km/h. The Tokaido Shinkasen was the most heavily travelled high-speed rail route in the world with a cumulative ridership of over 5 billion passengers.

Following the success of HSR in Japan, countries around the world began to invest in research and development of similar technologies. In 1967, France launched the world’s first regular service by classic train at 200km/h between Paris and Toulouse. And in 1969, America inaugurated the Metroliner rail service between New York and Washington D.C.

Over time, HSR trains became faster. In 1981, the French inaugurated the first section of the Paris- Lyon High-Speed line with a 260km/h top speed, raised to 270km/h soon after. Following France, Germany launched a Hannover-Würzburg high-speed railway, operating at a top speed of 280 km/h. And in 2000 Express was launched in America and could reach a maximum speed of 241 km/h on a small section of a line along the East coast.

For four decades the Japanese was the only high-speed rail service in Asia. But in the 2000s things started to change. In 2004, South Korea launched services on the Seoul-Busan corridor, Korea’s busiest traffic corridor, which could reach top speeds of 305 km/h in regular service though the infrastructure was designed for 350 km/h. In 2007, Taiwan launched its first and only HSR line. And between 2003 and 2012 China opened four HSR lines. The Wuhan– Guangzhou high-speed line was launched at 350 km/h, an unprecedented speed. In December 2012 China set yet another record by opening the world's longest high-speed rail line, the 2,208 km (1,372 mi) Beijing–Guangzhou–Shenzhen–Hong Kong Railway service. By 2011, China already had the world's longest HSR network with 8,358 km of tracks, and was on target to fully complete a National High Speed Rail Grid by 2015.

HIGH-SPEED RAIL IN THE UK Having launched the world’s first locomotive-hauled public railway in 1825, the UK boasted the world’s oldest railway system. In the 1970s introduced high-speed when it pursued two development projects. These projects included the development of a technology, the '', and the development of a conventional high-speed diesel train, the 'High Speed Train'. The former was eventually abandoned while the latter broke a world speed record for diesel trains of 230km/h. 211

High Speed rail in Britain was augmented when the was electrified. British Rail Class 91 was then introduced, branded as InterCity 225 and intended for passenger service. The line was designed for a maximum service speed of 225 km/h. The first high-speed rail line actually delivered in the UK was the Rail Link (HS1). HS1’s first phase opened in 2003 and the line hosted an international passenger service between London’s St Pancras and Continental Europe, reaching speeds of up to 300km/h. The motivating factors for high-speed rail were clear. It offered fast and efficient transport, increased capacity, and its advocates argued, it encouraged economic growth by joining up various cities. Still, the UK, an Anglo-Saxon democracy with a strong regime of property rights and a stringent planning process, was lagging behind in the global race to build high-speed railway infrastructure. To catch up, the UK Government would have to secure the powers to construct a new line through Parliament by the passage of a “hybrid bill”, which would need to consider all the public and private interests affected by the scheme.

THE EARLY YEARS

As the first phase of the HS1 scheme approached completion in 2003, the idea of developing a new line linking London with northern England and Scotland started to gain traction. To kick-off a discussion, the Strategic Rail Authority (SRA) launched a consultation exercise based on existing studies commissioned to transport consultant WS Atkins. To prepare for the consultation the Commission for Integrated Transport (CfIT), a statutory body, commissioned a study from transport consultants Steer Davies Gleave. The study was to evaluate the adequacy of the UK’s approach to economically appraise large investments. One of the main arguments from the study was that a lack of strategic investment in transport infrastructure could lead to severe transport bottlenecks in the medium term and lasting negative effects on the national economy. Thus the case for building a new rail line was a strong one. The study did however highlight that HS1 had been the world’s most expensive railway scheme per kilometre even when ignoring financial costs [Exhibit 1], and warned that the expansion would be ‘difficult, disruptive, and very expensive’ (pp. 62). It also noted that the UK was the only country in the world that increased expected capital costs by up to 66% and timescales by up to 25% to account for risk and so-called ‘optimism bias’ factors; contingencies as low as 5 to 10% were the norm elsewhere. These reports did not stop Network Rail from continuing to work on the so-called HS2. But in July 2007, a white paper published by the Department for Transport poured cold water on the scheme. The report, which received backing from Ruth Kelly, the transport secretary at the time, suggested that a decision on the new line could wait for another 15 years. By September 2007, the world had changed beyond recognition. The markets were hit by a financial tsunami that brought memories of the 1930s Great Depression. The UK economy had been hit particularly hard, and the British pound lost substantial value. The UK saw the first bank run in 150 years, and by February 2008 the government was taking banks into state ownership. Investment in the construction industry collapsed, and lobby groups asked the government to accept Keynesian arguments that investment in new infrastructure schemes was vital to shore up the British economy and weather the storm. Lobbying was effective, and by October 2008, the transport secretary established a National Networks Strategy Group to look into strategic investment in transport. Geoff Hoon said: The Minister of State, Lord Adonis, will chair a National Networks Strategy Group, with senior partners from the Highways Agency, Network Rail, HM Treasury and other Government Departments as required. ...the study will focus on longer-term solutions for the strategic corridors. This will include consideration of wholly new rail lines, including high-speed rail. We are committed to developing a modern sustainable rail system that supports economic growth, including housing development, and the climate change agenda.... However, it is crucial that the case for such investment is underpinned by robust evidence on long-term demand projections and a 212

clear understanding of the capacity of the existing networks, and takes full account of relevant geographical, technical and environmental considerations. By 2009, the National Networks Strategy Group’s work revealed a strong economic case for promoting a new high-speed rail line from London to the West Midlands.

HS2 LTD Lessons from previous mega infrastructure projects in the UK suggested that establishing a company wholly independent from governmental departments was essential to accelerate development. Thus in January 2009 the government created a new company, High Speed Two Ltd. (HS2 Ltd). In the company’s remit, detailed in a letter from the Secretary of State to the Chair [Exhibit 2], HS2 Ltd. was tasked with establishing the case for new high-speed lines in the UK, and assessing the environmental impact and business case of different routes in order to narrow down the options for the Government to assess. HS2 Ltd. was owned by the Department for Transport (DfT), but spending approvals required direct approval of the Cabinet office and the HM Treasury. A HS2 Programme Board was also created and made accountable for the programme meeting its objectives and realizing the expected benefits. This board included representatives from the DfT, HS2 Ltd., HM Treasury and Infrastructure UK37, and received authority to carry on development work up to £100m [Exhibit 3]. Internally, HS2 Ltd. had its own executive board comprising a chief executive and six directors appointed by the DfT [Exhibit 4]. The Board’s responsibilities included:  Effectively governing the company  Making company decisions at the right time  Properly managing risks  Shaping, challenging and directing the agenda for the company delivering stated priorities  Monitoring performance and risk  Making choices (or recommendations to ministers) on priorities / risk appetite  Overseeing the health of relations with stakeholders and commercial partners.

HS2 Ltd. subsequently set up a Professional Services Framework (PSF) to gain access to services that would support development work. The framework comprised a shortlist of organizations eligible to tender for work packages that would be awarded in secondary competitions. HS2 Ltd. awarded framework agreements in four categories or “lots” of work [Exhibit 5a]. The framework was set to last for six years, and companies on the framework received the option to extend the framework duration if the scheme went ahead. In August 2011, after issuing the framework agreements, HS2 launched the procurement for a Development Partner. The successful partner would scoop a lucrative contract worth between £50m and £70m over six years, and would be responsible for project managing the firms carrying out the design, environmental and land referencing work for Phase 1. By November 2011, HS2 revealed that the shortlist of bidders in the running included Turner & Townsend, Mace, CH2MHill and Parsons Brinckerhoff. In January 2012, it announced that CH2MHill38 - a global leader in programme and project management – had won the contract.

Between October 2010 and January 2013, HS2 let around £110m of contracts [Exhibit 5b]. The contracting sphere had been dominated by a handful of top consultants, e.g., £34.5m for Arup,

37 Infrastructure UK was a unit within the Treasury that worked on the UK's long-term infrastructure priorities and on securing private sector investment. 38 CH2M Hill had just acquired the UK consultant Halcrow, which had experience of working on HS1 and Crossrail; CH2M Hill had also been involved in the delivery of the 2012 London Olympics.

213

£17.5m for Mott MacDonald, and £19.1m for Atkins. By January 2013 more than 1,500 people were working on HS2, whilst HS2 Ltd. itself had recruited over 200 people DESIGNING AND DEVELOPING HIGH-SPEED 2

In 2009, after a 12-month study, Network Rail made a strong case for investing in a new high- speed route with city centre terminal stations at London, Birmingham, Manchester, Glasgow and Edinburgh, and connecting to Heathrow by a spur [Exhibit 6]:

We’re running one million more trains per year than just five years ago... By 2020 the main rail line to Birmingham and the North West will be full...high-speed rail is now widespread throughout the rest of Europe and across the globe...a new high-speed line to Scotland more than pays for itself [NR 2009 Meeting the Capacity Challenge: The Case for new Lines]

As it turned out, the design layout changed rapidly after HS2 Ltd. took the helm of the development, even though several NR employees were seconded to the company. The new plan was to provide even more links to the UK’s major economic centres and communities, releasing capacity on regional and local services. And through a permanent link to High Speed 1 at London St Pancras, HS2 would also provide links into the European Network.

In March 2010, the government published the first command paper detailing the proposal produced by HS2 Ltd. A new line would be built between London and the West Midlands, with connections to HS1 at London St Pancras, and to Heathrow airport via a spur. The service would host 400m long trains, and would be future-proofed 39 to run trains at up to 400km/h and carry 1,100 passengers. A Y-shaped route would link London to Birmingham, and from there it would go on to Manchester and Leeds [Exhibit 7a]. Further connections onto existing tracks would allow the services to travel to Newcastle, Liverpool, and Scotland. The vision fell short of the expectations of Greengauge 21, a lobby group [Exhibit 7b] but still the new line would stretch along 539km in total. To reduce noise and visual impact, the line would hug the contours of the land. Overall journey times would improve as follows: Before HS2 After HS2 London to Birmingham: 1 hour 24 minutes 49 minutes London to Sheffield: 2 hours 1 minute 1 hour 15 minutes London to Manchester: 2 hours 8 minutes 1 hour 20 minutes London to Leeds: 2 hours 15 minutes 1 hour 20 minutes

The total route was estimated to cost £30bn (in 2009 prices). The first 192km stretch from London to the West Midlands would cost between £15.8bn and £17.4bn and would treble rail capacity. Tunnels made up only 10% of the route, but they contributed to some 25% of the base construction cost, and were most important in two sections: first, in and around London; and second in the Chilterns where they would mitigate the environmental impact on the Chiltern’s Area of Outstanding Natural Beauty (AONB). In the AONBs, an estimated 32% of the route would be tunneled, while 37% would follow existing route corridors as closely as possible. For the second phase, the cost of the route per km was expected to halve, primarily because of the reduced amount of tunneling likely to be needed.

Another idea embedded in the first command paper was that city centre super-stations could catalyze urban regeneration and economic growth opportunities. Sir Terry Farrell, a leading architect and a contributor to the government’s command paper, argued:

39 Guarding against a technology becoming obsolete by protecting it from consequences in the future

214

The potential of stations to transform cities is critically important…Stations have become pre- eminent in their role as place-makers. This is all a far cry from the origins of rail in the 19th century, where it was initially associated with goods transportation (NCE 22 April 2010)

Whilst conservatives and Labour shared the goal, with general elections looming, the Tories opted out of cross-party discussions over HS2. Accused by the government of playing ‘cheap politics’ and jeopardizing the HS2 timescale, the Tories hit back. They slammed the government for not including a hub at Heathrow airport, and suggested an alternative S-shaped route [Exhibit 8]. The shadow transport secretary Theresa Villiers said:

Labour has not focused strongly enough on the need for a top class rail hub for Heathrow to connect it to Europe via the HS1 to provide an alternative to thousands of short haul flights.

Concerned that the scheme was becoming a political football, Sir David Rowlands, the chairman of HS2 Ltd. was critical of the political leaders and insisted that the scheme be a key part of a comprehensive, integrated transport policy. He also warned that an effective planning process was crucial to efforts to get the high-speed rail project built:

No high-speed railway will be built unless government finds a way of balancing the national need with the local interest [Sir David Rowlands, HS2 chairman]

In the midst of the political fighting, the government instructed HS2 Ltd. to address the criticisms surfacing against the command paper. HS2 Ltd. was tasked to look at both a line that went through Heathrow, as well as assessing the options for linking HS1 with HS2.

By May 2010, power had changed hands, and a Tory-led coalition led the country. A report released in July 2010 commissioned by the previous government, High Speed Rail Access to Heathrow, so-called the Mawhinney’s report, insisted that proposals for a Heathrow hub station had not a compelling case. Unexpectedly, the Mawhinney report also recommended that “serious consideration” be given to making Old Oak Common in west London the primary London terminus instead of the HS2 Ltd-favored London station at Euston:

Connecting at Old Oak Common on to Crossrail would give a quicker and more convenient overall journey to many destinations in and around London rather than would travelling via Euston

As the scheme continued to move forward, opposition started to mount from campaigners such as “Stop HS2” a campaign group whose convener told the Parliament:

There is no business case, no environmental case and no money for HS2. The more we find out (about) the project, the worse it seems to be (Joe Rukin, NCE, 30 November 2010)

By December 2010, the Coventry City Council voted unanimously to oppose HS2, becoming the first council not directly affected by the location of the line to take a position against the plans. Network Rail in turn came publicly in defense of the scheme after it released its own Route Utilization Strategy. In it, NR argued that HS2 was essential notably because passenger demand was set to grow by as much as 61% between London and Manchester, and thus the route would run out of capacity by 2024.

Amid an increasingly hotter debate about the business case for HS2, the new government made a U-turn on ideas that the Tories had raised during the electoral campaign. Hence, the government dropped plans for a Heathrow Hub endorsing instead the link to Heathrow via a spur and an interchange station between HS2 and Crossrail at Old Oak Common. The new government also insisted on including a direct link to the HS1 line at an estimated cost between £2.5bn and £3.9bn.

215

In February 2011 HS2 Ltd. launched a 6-month public consultation on the proposed £16.8bn (2011)40 first stage due to be completed by 2026. The plan was to organize over 30 public road shows. The budget for the overall network was now at £32.2bn (in 2011 prices), a cost increase attributed to the extra engineering done to improve the route.

Opposition to the project by some groups continued to mount. In July 2011, for example, an economic think tank called the Institute of Economic Affairs (IEA) argued:

The wasteful allocation of resources is demonstrated by the ‘gold-plating’ of the HS2 route...The first 8km from Euston to Old Oak Common, for example, will add almost 25% (about £4 billion) to the cost of the first phase but deliver negligible time savings....

In response, a pro-HS2 group ‘Campaign for High Speed Rail’ e-mailed subscribers urging supporters to come forward since the number of negative responses [to the public consultation] was far greater than the number of positive responses. And 70 top bosses also gave their public backing for HS2. Aware of the controversy, the Parliament ’s Commons Transport Committee commissioned an independent review, and the chairman said: The review is a guide for us. We felt we needed it.

Despite the controversy, the government forged ahead. In October 2011, HS2 Ltd. published a contract notice to recruit up to 19 firms to do preparatory work. And in January 2012, after the end of the consultation for phase one, Justine Greening, the new transport secretary, announced the government was proceeding with phase one. In its new form, the scheme would include extra tunnelling through the Chilterns and other sensitive locations at an extra cost of £700M; HS2 would thus be a £32.7bn scheme. The line was being designed to run up to 18 trains an hour in each direction, travelling at up to 362km/h.

The cost of the first phase was estimated based on a high-level process to be £16.8bn (2011 prices, undiscounted) after a review by Infrastructure UK concluded it was possible to reduce costs by £1.5bn given that UK construction costs on high-speed rail were 20% higher than in Europe41. (Other concomitant DfT documents still showed however a range between £15.4bn and £17.3bn). In addition, the estimated cost of the trains for phase one was £2.6bn (2011 prices). Whilst the government admitted it had revised down the economic benefits of the scheme, it confirmed its commitment to go ahead:

I have decided Britain should embark upon the most significant transport infrastructure project since the building of the motorways ...By following in the footsteps of the 19th century railway pioneers [Justine Greening 2012]

The government also pledged to release preliminary plans for Phase two of the scheme that included stations at Manchester, East Midlands, Sheffield, and Leeds. Having announced the intention to proceed with phase one and further develop phase two, the government and HS2 Ltd. had to refine the plans with local authorities, Network Rail, Highways Agency, and other actors, e.g., environment and heritage organisations. Agreeing on a compensation package for homeowners along the route was another important task, but one that fell under the remit of the Department for Transport 100-people HS2 team.

In January 2012, HS2 chose CH2MHill for development partner, and selected 13 consultants for its professional services six-year framework. Pleased with her performance, in March 2012, the government reappointed Alison Munro as HS2 chief executive, and appointed Doug Oakervee for

40 Economic Case for HS2 The Y Network and London – West Midlands, Department for Transport, UK February 2011, table 7; all costs and benefits are shown in 2011 prices 41 HM Treasury and Infrastructure UK, Infrastructure Cost Review, December 2010

216

the role of HS2 chairman. Doug was seasoned in these battles, having been executive chairman for the Crossrail project during the hybrid bill phase. In June 2012, HS2 Ltd. awarded the final professional service contracts (Land Referencing) under Lot 4 of its Professional Services Framework covering land access and land referencing services.

Amidst controversy over value for money, in June 2012 the HS2 technical director publicly insisted that his team preferred to think of HS2 as a ‘£25bn project’, with none of the £8bn earmarked as optimism bias added to it—“We’ve no intention of going near [the £8bn]” (NCE, June 2012). Doug and Alison seemed to take a more hedged position. Doug said:

The bill gives you the limits of deviation where you can build within and some outline design... provides the power to build it and outline planning permission.... The real design has got to be done after you’ve got powers, and then the detailed design is approved by the local authorities... So really you’re talking about route and opportunity and problems.

There was a lot of work ahead before the bill for first phase was ready to submit. Things at London were complex. At times, it seemed that key players such as Camden Council, Transport for London and the Mayor were collectively holding the government to ransom. And even at Birmingham, development for two stations was not clear-cut.

To complicate matters further, four groups of campaigners launched a legal challenge against the government’s consultation. They hoped to overturn the decision to forge ahead with phase one. The groups had overlapping claims and between them agreed to focus on specific areas to challenge the transport secretary’s January 2012 decision.

 One group, Heathrow Hub, insisted that the Heathrow Hub should be built;  51M, a collection of fifteen councils along the route, argued the DfT failed to: 1. provide an adequate environmental assessment; 2. properly consider alternatives to increasing capacity; and 3. consider the impact of looming capacity crunches at Euston Underground station and on the North London Line which would be used to connect HS1 and HS2.  HS2 Action Alliance (HS2AA), a not for profit organisation working with over 70 local community groups, alleged HS2 failed to comply with Strategic Environmental Assessment (SEA) regulations; and that HS2 was in breach of the Habitats Directive  And Aylesbury Park Golf Club alleged that the compensation payments to be made to affected landowners along the route were insufficient.

Against all of this, Alison and Doug breathed a sigh of relief when it came to Manchester; things seemed to be going smoothly there…at least for the time being.

CHILTERNS

HS2’s analysis confirmed that the most direct route from London to Birmingham would pass through the Chiltern Hills, a protected Area of Outstanding Natural Beauty (AONB). In fact, any viable line of route between London and Birmingham would necessarily traverse some part of the Chilterns. When routes avoiding the Chilterns were considered it was found that their alignments were less direct resulting in longer journey times. Furthermore, routes avoiding the Chilterns would involve significantly more demolition.

Substantial tunnelling would enable to mitigate the negative impacts of HS2 along the route. But balancing the extra costs of tunnelling with the noise and visual impacts that a surface route would have on communities was difficult. Tunnelling definitely had some appealing benefits. Long tunnelled sections would be less obtrusive for the communities. In contrast, surface routes could have visual and noise impacts, required compulsory purchases and demolitions of property, and could cause severance if it created a new transport corridor.

217

But tunnelling also had negative impacts. It would involve higher and more uncertain costs than surface alignments, and had environmental costs too. For one, tunnelling included higher ‘embedded’ carbon in construction, and created a need to dispose of a substantial volume of excavated material. It also required developing surface works for ventilation purposes; and (relatively small) risks could arise of surface properties being affected by subsidence during construction and by ground-borne noise and vibration afterwards.

In addition to devising recommendations for the government regarding tunnelling, HS2 leaders had to consider how close to existing motorway alignments the line should follow. Following existing motorways closely could reduce the potential impacts of a new rail line. But because high-speed rail required shallower curves than either conventional rail or motorways, it would not be possible to follow many existing routes without requiring frequent speed restrictions. This in turn would undermine the benefit of high-speed rail. Taking these issues into consideration, HS2 Ltd. identified six main corridors through the Chilterns before short-listing three potential routes:

 A route leaving London via the existing Chiltern Line corridor, and then using a combination of tunnelling and the existing A413 corridor to reduce impacts (‘Route 3’);  A route following the same corridor to Ruislip, but then passing in tunnel beneath Gerrards Cross before crossing the Chiltern AONB through a combination of tunnelling and surface routes (‘Route 2.5’);  A route leaving London via a 28km tunnel towards Kings Langley, before passing through the Chilterns and close to the town of Berkhamsted (‘Route 4’).

HS2 Ltd. eventually recommended ‘Route 3’ because it was shorter, provided a faster journey time, and was more economical [Exhibit 9]. Its estimated cost of £3.7bn (without provision for risk) was lower than Route 2.5 (~£4.3bn) and Route 4 (~£5.1bn). Further, route 3 offered sustainability and environmental advantages over the other options given its potential for isolation from existing settlements and ground-borne noise. Its visual intrusion was reduced through the use of cuttings and screening with vegetation and embankments.

In line with HS2 Ltd., the Government agreed that route 3 was the best option going forward. The route would, however, be subject to consultation. The aim would be to mitigate its impact as judged against a full range of its objectives. The new line would still despoil the Chilterns, damaging four Wildlife Trust reserves, 10 Sites of Special Scientific Interest and over 50 ancient woodlands; furthermore it would run close to a National Trust property. All this would spawn no small amount of local objection in Tory heartland.

Campaigners demanded the abandonment of the entire scheme so as to leave the Chilterns AONB untouched. Unashamedly NIMBYs, the campaigners in the Chilterns argued that the case for HS2 was weak and transitory; and that every effort should be made to provide the needed rail capacity without causing long-term damage to the environment.

In reaction to the influx of campaigns the Transport Secretary delayed any announcement over the future of HS1 until January 2012 so as to gain time to consider options for upping environmental protection in the Chilterns. This reflected the political reality of the project, and some argued it was akin to political pressure giving way to “parochial pressure groups”. When the government announced the intention to proceed, the amended HS2 proposal included a medley of mitigation measures proffered to the naysayers:

1. A longer, continuous tunnel from Little Missenden to the M25 through the Chilterns 2. A new 4.4 km bored tunnel along the Northolt Corridor to entirely avoid major works to the Chilterns Line and impacts on local communities in the Ruislip area 3. A longer green tunnel past Chipping Warden and Aston Le Walls, and to curve the route to avoid a cluster of important heritage sites around Edgcote

218

4. A longer green tunnel to significantly reduce impacts around Wendover, and an extension to the green tunnel at South Heath

Doug believed that the tunnels had been a premature concession. The Chilterns’ dwellers would not necessarily back down because they had received tunnels. In fact, they could well request more concessions in the future. Building HS2 through the Chilterns was proving to be a contentious issue. Equally contentious, however, were the discussions at London.

LONDON

As early as 2009 Sir Terry Ferrell, a leading UK architect, argued that though expensive, Euston was an obvious choice for locating the London terminus [Exhibit 10] together with another station at Heathrow. But in the summer 2010, an independent report commissioned by the government “High Speed Rail Access to Heathrow” argued that running HS2 via Heathrow at an additional cost of £2bn to £4bn was not likely to represent value for money. The report also recommended serious consideration to making Old Oak Common the initial London terminus, allowing an interpretation that plans for a Euston station could be ditched. HS2 Ltd. had however undertaken an extensive ‘Options Sifting Process’ [Exhibit 11] before arriving to Euston as its station of choice. Doug and Alison were therefore confident in the decision to stick with the station. The sifting process started with an extensive long list of 27 sites in London, which included inner and outer London locations, both surface and underground options. Various terminal structures were considered including a single London terminus, two smaller termini or two configurations of a central London through-station.

During this process, all the locations were assessed against various criteria including: overall fit with the remit, operational/engineering feasibility and demand. A shorter list of potential stations was then developed. And a single-level station at Euston was then identified as the most promising option based on sustainability, cost and a superior economic case.

The new Euston station would comprise 10 high-speed platforms and 14 classic platforms [Exhibit 12a, b]. It would be able to accommodate the proposed new high-speed line’s long-term maximum service pattern of 18 trains per hour. To avoid blighting lots of sites during the sifting process for selecting the London sites, HS2 Ltd. consulted the Transport for London, Network Rail, and the Greater London Authority, but did not involve the local authorities in the discussions. Still, HS2 Ltd. was aware of Camden Council’s plans to improve the connection between the communities on either side of the Euston station. Thus, in its preferred design for Euston, all the platforms would be built 2m below the current track level trains to almost the full length of the trains. This would create an opportunity to introduce new streets and public squares above connecting the two communities. Beyond the station itself, there would be significant redevelopment potential at the Euston area.

Alongside Euston, an interchange between HS2 and Crossrail would be another big-ticket item in the proposed London infrastructure [Exhibit 13]. The interchange would allow for primary connections to Crossrail, Heathrow airport and the Great Western Mainline. It would create a connection to the airport with an 11-minute journey time and would include a direct link to the European high-speed rail network High Speed 1. These direct connections would ease crowding at Euston and improve links to important London destinations.

If the Interchange was introduced, a third of all HS2 passengers were expected to use it instead of Euston, relieving pressure on the Tube lines around Euston, an issue raised by Transport for London. From a construction perspective, Old Oak Common was the only site in West London suitable for launching the tunnel boring machines needed to create the tunnels to reach Euston. Old Oak was also suitable as it was in an area of London identified as a priority for regeneration. A new interchange station could contribute to this goal.

219

The Government accepted HS2 Ltd.’s recommendations for an HS2-Crossrail Interchange, at Old Oak Common. The financing of the station was however unclear as the Government expected equitable significant contribution from local beneficiaries. Still, the Old Oak Common station would form part of the public consultation on the route.

In December 2010 when the Transport Secretary first unveiled plans for Phase one, the plans were met with significant opposition. The more significant tension points were:

Ditching plans for Euston. Some respondents raised concerns as to whether there was sufficient evidence supporting Euston Station as the central London station. It was suggested that instead of the two separate stations, there would be only one station in London, and that it be St Pancras near the HS1 terminus. Others proposed Old Oak Common to be the only London station. Stopping HS2 short of the city center was however an unacceptable option for Boris Johnson, the capital’s mayor, and other influential parties.

Crossrail 2. For those who insisted HS2 needed to go to Euston, concerns were mounting about the large volume of passengers who would descend on Euston Station, and how to disperse the pressure that the new line would bring on the local Tube stations. In August 2011, the Greater London Authority (GLA) proposed that Crossrail 2, the new underground line, be completed before the second phase for HS2 became operational. GLA was not alone in this appeal. TfL, the London Mayor, the business lobby group London First42 , and others all agreed that the Crossrail-HS2 Interchange at Old Oak Common was essential to minimize the pressure that HS2 would bring to Euston. But they also insisted that Crossrail 2 (which had been on the drawing board for several decades) needed to be built. For the London stakeholders, Crossrail 2 was indispensable to relieve congestion on the Tube even if HS2 was not built, thus the lobbying for the line gained momentum. Financing was, however, an issue. Potential options included a mix of government grants, fares and contributions from developers and local businesses. By June 2012, TfL demanded Crossrail 2 be safeguarded in the HS2 bill for phase one. This would avoid having Euston “dug up twice” and ensure that Euston station was not forced to close under the pressure of overcrowding. To put pressure on the Government, TfL and London First joined forces to consult Londoners on two alternatives for Crossrail 2:  a shorter, Tube-style metro from Clapham Junction to Seven Sisters via Tottenham Court Road, Euston and King’s Cross St Pancras, estimated to cost £6bn [Exhibit 14a]  a regional metro like Crossrail 1 with a route extending connecting the South West with the North East. This longer route would come in at around £10bn [Exhibit 14b]

For some at HS2 Ltd., this was about logrolling43. In some sense the London stakeholders were politically blackmailing the government to secure financing for Crossrail 2, and perhaps even dump part of the Crossrail 2 costs onto the HS2 budget. Others argued however that the London stakeholders had a good point, and were simply doing their jobs by protecting the interests of their constituencies. Either way, it was clear that the London Mayor was flexing his political muscles and that his important support for HS2 was subject to various conditions. Boris Johnson was certainly proving to be a thorn in the government’s plans.

Demolition and disruption at Euston. The headaches at London did not stop with London’s insistence that the government should finance Crossrail 2 and safeguard the Euston plans for that connection. Camden council was also aggrieved with the HS2 plans for Euston and felt excluded from the decision–making process. A HS2 official recalled:

42 London First had been created by business to “agitate for London’s infrastructure”. 43 The exchange of political favours, especially the trading of influence or votes among legislators to achieve passage of projects that are of interest to one another.

220

things weren’t shared as openly in the early days with the thinking about the London terminus.. Camden would argue that the decision-making about where the station was in London wasn’t something that they were particularly involved in...they were sort of told. Clearly, the Camden elected officials were not happy to see all their plans for Euston squandered by HS2. The proposed demolition around Euston and in surrounding ‘urban green’ areas such as The Regents Park and Primrose Hill was a major concern for them. The proposal would require moving over 100 families and demolishing at least five blocks of flats occupied by Council tenants. The Council elected officials insisted that appropriate housing should be created in advance of demolitions, and further expressed concerns about the impact that constructing the line would have on local business and trade. To address the housing issue, the government committed to prioritizing the development of proposals for re-accommodating people, and to consider rehousing options on-site as part of the redevelopment. But for the Council politicians, the government proposals were inadequate, and thus they legally challenged the Government’s decision. Sarah Hayward, the Council’s elected leader, wrote in her blog in February 2012: The impact in Camden is colossal, devastating and ill considered by Government. ... we really were left with nowhere else to go... In Camden nearly 500 families will be forced from their homes. Currently those people have no idea where or what they will be offered as an alternative... this is one of the poorest areas of the country... We’re not nimbys... we oppose HS2 because it will devastate our borough. If at a political level the fight was gory, things were better at an official level. And the HS2 Ltd. officials were working with the officials of the Council as well as of Transport for London and Greater London Authority on the best possible plan for Euston. From the onset, HS2 Ltd. ruled out the idea of a double-decker station. Although this could reduce the footprint of the revamped Euston station, HS2 Ltd. insisted it was not affordable and its implementation would create horrendous disruption for the train services running into London Euston. Thus, the discussions centred on rebuilding the existing Euston station. The Council officials liked the idea of fully modernizing Euston, thus creating conditions for exciting over ground developments. But as Transport for London, the Council, NR and others added things to the scope, by January 2013, the budget had just about doubled from the initial £1.1bn estimate, and the works would last at least three extra years. This created a problem for HS2 Ltd.

If HS2 Ltd. recommended the government to ditch the plans for a complete rebuild of Euston station and opt instead for a more cost effective and less disruptive solution, it would infuriate the local stakeholders. The new scheme [Exhibit 15] would retain rather than rebuild 13 of the existing platforms, and only the 11 new platforms for the high-speed trains would be sunk into the ground. But if this scheme went ahead, the costs would be around £1.6bn, and it could be ready by 2026 as initially planned. The revised station would include:  A sub-surface pedestrian link between Euston and Euston Square Tube stations  Better connections with the Underground, including a new Underground ticket hall  New facilities for all passengers in a redeveloped, integrated station with a new, combined concourse and façade  East-west pedestrian routes across the station, helping to link communities on either side of the station

A more modest £1.6bn scheme still offered potential for unlocking development opportunities for the area. But it remained more expensive than the initially estimated cost of £1.1bn and therefore HS2 Ltd. argued that they should not finance all the costs. The question, as one HS2 official said, thus remained ‘how to carve it [budget] up?’

221

MANCHESTER

Greater Manchester was one of the UK’s largest conurbations and acted as the principal economic hub for the wider Northern region. Manchester was also one of the largest centers of inter-city rail demand in the country. The core of the Greater Manchester conurbation included two cities, Manchester and Salford, which operated as two separate authorities. Any station developed in the centre would therefore have to serve both authorities.

In March 2012, whilst Phase 1 of HS2 was being consulted, HS2 Ltd. received a remit from the Secretary of State to develop route proposals and options for the Phase 2. A relevant aspect of the remit was to develop station options that would serve the Manchester city centre, plus consider an interchange station in Greater Manchester and provide access to major airports. With regards to the latter, Manchester Airport warranted proper analysis as the largest airport outside London in terms of passengers served.

To decide where to locate the stations, HS2 Ltd. again adopted its ‘Sifting Process’ going from the initial generation of options to long listing then short listing before selecting options for refinement. Given the blight risks, HS2 Ltd. opted to consult but not to invite any local authorities to actively participate in the sifting process for determining the line of route for the two legs of the phase two. Instead, internal technical working groups would delineate the line after sifting through the various options for the routes.

However, to nail down the location of the stations, HS2 Ltd. decided to directly involve the local players with local planning and transport responsibilities. Meeting attendants would be asked to sign confidentiality agreements, and could not take any paperwork away from the meeting rooms. The HS2 Ltd. leaders anticipated that this would reduce the risks of leaks of information and thus unnecessary blight on property. HS2 Ltd. asked the Manchester City Council to identify local stakeholders who should attend the options sifting meetings, and proposed to create two working groups, one to explore a city centre station, and another to explore an interchange station for Greater Manchester [Exhibit 16].

Following much deliberation with local stakeholders, in March 2012 the Government announced its preferred route and choices for the four remitted stations that would form part of phase two of HS2. To arrive at its final proposal, the Government reviewed the recommendation prepared by HS2 Ltd., and also considered the reports prepared by local players. In addition, ministers and officials visited the proposed route and station options and met with the local political and economic leaders to hear their views first hand. The western leg of the route presented by Government would serve proposed stations at Manchester Airport and Manchester Piccadilly [Exhibit 17].

In the heart of the city, a new station would be built alongside the existing Piccadilly station, and the HS2 platforms would be parallel to and immediately alongside the existing platforms [Exhibit 18]. The redevelopment of Piccadilly would enhance connectivity with the local transport network and add new car parking. A bigger station with more facilities and better links with public transport would help the city. It would allow easy connections with regional rail services, as well as easy access to the Manchester public transport network.

The Manchester Airport station, on the other hand, would serve the airport and the local market for rail demand. With improved transport links, the airport would be capable of increasing its passenger volumes. Additionally, an HS2 station alongside Manchester Airport would create opportunities for passengers to interchange directly between high-speed rail services and the airport.

222

Resolving the location for the city centre station had been straightforward. The process started with 42 potential sites after refinements44, but the choices quickly boiled down to two options, one near the existing Piccadilly station and another in Salford, not far from the BBC headquarters in the North West. Local leaders then quickly converged on Piccadilly as the right location, which had one local official muse whether ‘they had it too easy’.

Still, discussions continued between the government, HS2 Ltd., and local stakeholders. For one, the Manchester stakeholders were unhappy with the HS2 Ltd. proposal for developing a simple station adjacent to the existing station in the city centre. This was a once-in-a-life-time opportunity to regenerate the area around Manchester Piccadilly, and the Council planned to commission independent studies to develop plans for a grandiose scheme that would integrate the existing station with the new facility for receiving the high-speed two trains. Clearly, such a vision would cost more than the £350m that HS2 Ltd. had earmarked for the station, but Manchester was digging its heels in. To bolster its case, the council announced it would establish a working group to further explore how to seize the opportunity to catalyse a multi-billion pound regeneration plan around a new-look Piccadilly station. The council also had a ‘strategic regeneration framework’ in the works, which drew inspiration from the likes of Amsterdam and New York. A Council official explained: This is a once in a century opportunity to have a major transport interchange … We want ...one fantastic station that is iconic in design.... they [HS2 Ltd]’ve got an idea of what they want to see and we’ve got an idea about what we want to see, and I think we’re both respecting of each other’s position.... broadly we want to get to the same end result, a station that is successful and that is generating growth and development

Issues were also emerging around the plans for the Manchester airport station. HS2 Ltd. and the local stakeholders disagreed on the economic case for the station. For HS2 Ltd., it was a weak one especially since the route connecting the airport station to the city centre station would require significant tunnelling as the line approached the city centre. If such tunnelling were undertaken, it would further push up the estimated costs for the second stage.

The local stakeholders disagreed, and argued that there was a strong business case for the airport station. This was argued in view of the local vision championed by the Manchester Airport Group (MAG), a holding company that was owned by the ten metropolitan borough councils of Greater Manchester and an Australian investment fund. The ambitious airport company, which was in the process of purchasing the Stansted airport in London for £1.5bn, had confirmed plans to develop a £800m Manchester Airport City in January 2013. The Manchester site had been announced as an Enterprise Zone45, and MAG was initiating discussions with international investors to fund the development of the Enterprise Zone. The HS2 leaders were not convinced, however, and would likely recommend the government to ask local stakeholders to finance significant elements of the airport station.

Still, compared to the tumultuous developments in the Chilterns and London, for the HS2 leaders, the systematic, respectful way in which development had been unfolding in Manchester was refreshing. The region was living up to its reputation of taking the ‘bull by the horns,’ and being very well prepared for opportunities that arose. Doug said:

You’ve got first class people at the top of it, and you’ve got the structure to make decisions quite quickly; they work collaboratively within the northwest much more so. Certainly when I go up there, you’re in a very organised lobby. They just grasp the opportunities

44 (Temple-ERM, 2013) High Speed Rail: Consultation on the route from the West Midlands to Manchester, Leeds and beyond Sustainability Statement (A Report by Temple-ERM for HS2 Ltd)

45 Urban Enterprise Zones are areas where companies can locate free of certain local, state, and federal taxes and restrictions to encourage development in blighted neighbourhoods

223

Indeed, in support of its proposition for an affordable airport station, the Manchester stakeholders committed to leading the work required to draw up a package of innovative funding measures. Coupled with this, the Government committed to working with the Manchester Airport group and the wider region to develop clear funding principles and outline a package, which would be sufficient for the Airport station to go ahead.

********* Doug and Alison had less than 10 months to submit the HS2 phase one bill to the UK Parliament. The good news was that apart from a handful of Ministers of Parliament(MP) who represented the interests of NIMBYs North of London, cross-party political support remained very strong. Furthermore, there was plenty of time to resolve the issues around phase two since that bill would not be submitted before 2015. This interval was important because pressure was mounting to add another station on the Western leg at Crewe without which lobbyists argued Liverpool and Warrington would be disadvantaged. And conversations on the Eastern leg, with three remitted stations, appeared to be way more complicated than those on the western leg up to Manchester. Less good news was that significant concessions had already been made for the Chilterns. Even more unsettling was that on some days it seemed that the HS2 plans for London were about to implode given the ongoing political fights—“HS2 isn’t an organisation that puts the world right”, said Doug, “we’re the infantry...[we] get all the fights... we’re the bad boys, but we’re really only doing what we’re instructed to do”. Be that as it may, it seemed that everyone expected a lot more from HS2 Ltd.

Ultimately, a few significant questions remained: should HS2 Ltd. continue to plough ahead with a scheme that fit within their budget, with the best information they had at the time, and draft a bill that would not preclude the opportunities of others that were emerging? Things could always be amended through additional provisions, and people could also deal with emergent pieces of work integral to HS2 through Transport Works Acts.

Or instead, should the controversy give them pause, and thus should they recommend the government to delay the whole thing by another year or so? More time would be helpful to resolve the differences in London and other locations, and reconcile whatever decisions would come with a revised and more robust budget and timescale.

And how detailed did the bill actually need to be? It seemed the key piece of the law was to define the scheme’s limits of deviation46. In the end, it was within those limits that the bill would grant legal powers.

HS2 Ltd. also needed to focus attention on the scheme’s business case. The scheme needed to get to get a majority in Parliamentto get approved, and thus HS2 Ltd. needed supporting evidence to persuade as many MPs as possible of its merits. But concerns were mounting about the reliability of the benefit-cost ratio, first announced in March 2010 at 2.4 to 1, but restated in August 2012 at 1.9 to 1 for the whole network and 1.4 to 1 for phase one.

For one, the National Audit Office, the government’s watchdog, was preparing its first review of the HS2 business case. The House of Commons Committee of Public Accounts was also planning to look at the HS2 business case. And another watchdog, the Major Projects Authority (MPA), had raised concerns about risks regarding the timetable, governance arrangements and resources in late 2011, and noted the £32.7bn estimate was yet to include VAT and inflation. Clearly, it was not enough to appeal to intuition in arguing that HS2 was a springboard for economic growth, job creation, and securing Britain’s prosperity

46 The maximum extent within which the railway and ancillary works would be built

224

EXHIBIT 1 – High Speed Line Construction Costs per KM (Steer Davies Gleave 2004, High- speed rail: international Comparisons. February)

225

EXHIBIT 2: HS2 Remit (Letter from the Secretary of State to the HS2 Chairman, 2010)

226

227

228

229

EXHIBIT 3- HS2 Programme Governance Structure (NAO 2013, High Speed 2: A review of early programme preparation)

230

EXHIBIT 4: The HS2 Ltd. Governance Structure (HS2 Ltd., 2012)

231

EXHIBIT 5a: Overview of ‘Lots’ Professionals Services Framework (NCE, 2012)

232

EXHIBIT 5b: HS2 Professional Service Framework Agreements (NCE, 2012; HS2, 2013)

233

EXHIBIT 6: The Network Rail proposal (NR 2009, Meeting the Capacity Challenge)

234

EXHIBITS 7a, b HS2 Proposed Y Route and Greengauge 21 proposal (NCE, 2010)

235

Exhibit 8: Plan floated by Tories in 2010 for discussion (NCE, 2010)

236

Exhibit 9: Chilterns Route 3 (Chilterns Conservation Board, 2012)

Exhibit 10: Sir Terry Farrell’s Euston Station Design (DfT Command Paper, 2010)

237

Exhibit 11: Options Sifting Process (HS2 Ltd., 2013. HS2 Phase 2 Initial Preferred Scheme Sustainability Summary)

Sifting Phase Phase Overview

Initial generation of The initial options for approaches were formed by looking at the possible options and long listing ways of linking the main route options and the city centre station options that had already been generated Initially approach routes were proposed

where there was an existing transport, utilities or natural corridor.

Long listing As with the main lines of route all options were progressed to the next stage as long listed options.

The long listed options were developed in more detail from an engineering, Short listing sustainability, journey time and cost perspective. The short listing of approach options was driven both by the relative performance of approaches and on the fact that some connecting route and station options had not been progressed.

Selecting options for Some options did not progress further because of high demolition numbers, refinement high cost and poor sustainability performance.

A more detailed assessment was undertaken of the remaining options, including design optimisation and mitigation. Although it was clear that Finalising our options there are differences between the costs, journey time, complexity and sustainability performance, no options were parked at this stage. There

were no clear all round better performers between the two approach options to Salford and the two approach options at Piccadilly.

238

EXHIBITS 12a: Proposed Design Euston Station, 2010 (DfT Command Paper, 2010)

239

Exhibit 12b Proposed Footprint for Euston Station, 2010 (DfT Command Paper, 2010)

240

Exhibit 13: Euston and Crossrail interchange (DfT Command Paper, 2010)

241

EXHIBIT 14a: Crossrail 2 Route (urban option) (TfL, 2012)

Exhibit 14b: Crossrail 2 Route (regional option) (TfL, 2012)

242

Exhibit 15: A Scaled down Plan for the Euston Station (HS2, 2013)

243

Exhibit 16: Consultation Stakeholder Groups for the Four Remitted Stations in the second phase (HS2 Ltd., 2012. Record of stakeholder engagement for phase two of the high-speed rail network: A report to Government by HS2 Ltd)

244

EXHIBIT 17: Manchester Stations (HS2 Ltd., 2013. High Speed Rail: Investing in Britain’s Future: Consultations on the route from the West Midlands to Manchester, Leeds and beyond)

245

EXHIBIT 18: Manchester Piccadilly Visualization (HS2 Ltd., 2013. High Speed Rail: Investing in Britain’s Future: Consultations on the route from the West Midlands to Manchester, Leeds and beyond)

246

Appendix C 2 Britain’s High Speed 2: All Aboard? (B)47

By the end of 2013 the hybrid bill for Phase 1 of the high-speed 2 (HS2) scheme was deposited in Parliament. This was a major achievement for the management of HS2 Ltd. considering how much things had changed in the last ten months. The most proverbial change, and one which dangerously rocked the boat, was a sharp increase in the budget to £50.1bn (2011 prices) 48.

In the face of uncertainty around how the scheme’s design would develop in the upcoming years (it was expected that thousands of objections against particular aspects would be lodged in Parliament) the Department for Transport, the scheme’s main sponsor, held private negotiations with the UK Treasury and Cabinet. From the negotiations a substantially more robust budget envelope emerged.

The estimated costs cropped up slightly to £34bn (2011 prices) in February 2013. But by June 2013 the scheme was budgeted at £50.1bn made up of £42.6bn for fixed infrastructure and £7.5bn for rolling stock. Included in the new budget was a contingency of £14.4bn for the infrastructure and a further £1.7bn in contingency for rolling stock.

The leadership of HS2 Ltd. was also about to change. In September 2013, the government announced that Doug Oakervee, the chairman at HS2 Ltd., would leave at the end of the year to be replaced by Sir David Higgins the former chief executive of the Olympic Delivery Authority, the agency that led the design and delivery of the London Olympic park. To this purpose, Sir David Higgins would step down from his job as chief executive of Network Rail.

The design of the HS2 scheme had also evolved. To attend to the concerns of many communities, e.g., Cheshire, Warrington, and Liverpool, who were frustrated that no station had been planned between Birmingham and Manchester, a station at Crewe was added to the second phase. In addition, to prevent slippages in the estimated costs and schedule that would arise from a complete overhaul of the Euston station, HS2 Ltd. opted for a more modest plan. As expected, the proposed scaled down development at Euston station infuriated many influential London stakeholders.

Crucially, fragilities in cross-party political support started to emerge after the budget for HS2 shot up. In response, some influential Labour politicians out rightly removed their support. In a letter to the Financial Times entitled ‘Why I no longer support a high-sped railway line for Britain’, Lord Mandelson, a former business secretary, urged all political parties to reconsider ‘binding themselves irrevocably to HS2’: In 2010, when the then Labour government decided to back HS2, we did so based on the best estimates of what it would involve. … Probably the most glaring gap in our analysis were the alternative ways of spending the £30bn cost...[HS2] is not all it seems and has the potential to end up a mistake, damaging in particular to those people that it was intended to help.

The stir lost momentum over the summer with political leaders in the Midlands and the north of England still backing the scheme. The Labour party was, however, left divided. Approval of the bill was therefore no longer clear-cut as was evident in the final voting of the so-called HS2 paving

47 Rehema Msulwa, PhD student and Professor Nuno Gil, The University of Manchester, prepared this case as the basis for class discussion. The case does not intend to serve as endorsement, source of primary data, or illustration of effective or ineffective handling of an administrative situation. The authors are solely responsible for any factual inaccuracies. Copyright © 2014 (May), Rehema Msulwa and Nuno Gil All rights reserved

48 How £50bn translated into final prices (cash costs) remained unclear but the press floated rumours that the Treasury was pencilling down a figure somewhere around £70bn excluding VAT

247

bill in October 2013. This bill had been approved by the House of Commons, which allowed the government to continue spending to develop the scheme and buy property. But in a House of 650 elected MPs, the bill was approved with only 350 votes and received 34 votes against it. Only a quarter of the Labour MPs voted, and 17 Conservative MPs voted against the bill.

Aware that HS2 could not go ahead without strong cross-party political support, the Government tasked the incoming chairman Sir David Higgins to fully review the scheme by March 2014. The aim was to get the Bill passed the second reading scheduled for 28th April 2014. At least after this date, HS2 could no longer be opposed in principle when debating the petitions against scheme.

David and his team had 3 months to develop a politically adept strategy to help rebuild cross-party consensus. In doing so a number of issues had to be managed. Critically, pressure was mounting for David to give away some of the contingency built into the budget. The Labour-controlled cities that would receive HS2 stations in the second phase were lobbying fiercely to accelerate the delivery, and for other local changes. Furthermore, influential stakeholders in London were threatening to remove political support unless they could extract major concessions from the government. *********** DEVELOPMENTS IN 2013

Since Labour’s transport secretary Lord Adonis released the first design for the high speed 2 scheme in March 2010, the scheme had seen some major modifications and survived the tenure of four different transport secretaries. Changes notwithstanding, the key principles had largely remained the same. The plan was for the government to build a new high-speed railway serving London and Birmingham in Phase 1, and extending it further north to Manchester and Leeds in Phase 2.

Since its inception, the HS2 scheme had been met with a range of responses. On the one hand there was a positive and buoyant response from the railway and civil engineering world, as well as business and political leaders in the Midlands and the North of England. Sir Richard Leese, Labour leader of Manchester City Council, for example, said:

[without HS2 the northwest] will be left stranded, unable to compete with the likes of Munich, Milan and Copenhagen who are already well ahead in the high-speed stakes49

On the other hand opponents included outraged communities and the elected politicians representing those constituencies. Whilst some opposed to the scheme in principle, others appeared to be politicking. Given the right incentives perhaps it would be possible to get them to fall in line with the view that the HS2 scheme was one of national importance.

An important change in the design early on in 2013 was the addition of a major hub north of Birmingham at Crewe, a historic railway town that connected northwest Britain, the Midlands regions and Wales. The hub at Crewe had been lobbied by the local business community and politicians and would put 6m people within two hours of London. The leader of the Cheshire East council said: “We can connect UK Plc through Crewe. It will transform the whole North West. We have more businesses in Cheshire East than there are in Manchester.”50

Controversy remained in other parts of the design. Things were still very much up in the air in London (discussed later) whilst discussions were ongoing as to the exact location of the two stations in Birmingham, and how to best connect them. With regards to the second phase, Manchester was digging its heels in by insisting on a grandiose station for the city centre and by

49Pickard, J., Bounds, A. (2013). HS2 route opponents smell blood as costs head north, Financial Times, 3 July 50 Bounds, A. (2013). Change at Crewe as HS2 hub plan gathers pace. Financial Times, 4 January

248

arguing that the business case for the airport station was stronger than HS2 Ltd. was stating.

Things were also complicated at Sheffield. HS2 Ltd. argued against locating a station right in the middle of Sheffield in favour of a stop at Meadowhall, a shopping area three miles outside the city centre that some locals said had eviscerated Sheffield. Some local politicians supported the idea but others were opposed to this plan, and insisted on a city centre-to-centre links. They argued that a HS2 link outside of Sheffield’s city centre could accelerate the decline of the centre unless a link was provided to the city centre. HS2 officials insisted, however, that linking Meadowhall to the centre of Sheffield would cost at least £600m and was thus unaffordable largely due to the geography of the city with its hills and rivers: “We had to apply various criteria: cost, engineering complexity, sustainability, journey times and scope for following existing travel corridors...[Meadowhall] offered a better balance”, said the HS2 project director for the northern station.

LEGAL CHALLENGES THROWN OUT

In March 2013 Judge Justice Ouseley who was presiding over the legal challenges launched against HS2 by a group of four campaigners threw out all but one of the claims. The Judge ruled that the consultation process that the Government followed to arrive at the proposed route was ‘lawful’. Judge Ouseley found that due consideration had been given to the impact of the scheme on habitats and protected species. The Judge also ruled that constructing a new high-speed rail line instead of upgrading the existing network was well justified as the latter would fail to meet the government’s objectives of providing a long-term boost to capacity and economic growth.

Much to the relief of the HS2 Ltd. leadership, the one area where a challenge was upheld would not affect the HS2 construction timetable. It concerned the way in which the property compensation consultation had been carried out, rather than the merits of the policies. In accordance with the Judge’s findings, the Secretary of State for Transport decided to re-run the property compensation consultation rather than launch an appeal on the judgment. Notwithstanding the comprehensive defeat in the High Court, campaigners were adamant on overturning the Transport Secretary’s decision to proceed with HS2, and thus lodged an appeal to the Court of Appeal. The Court again ruled in the government’s favour on all issues raised. In response, campaigners decided to fire more ammunition: It sets a massively worrying precedent for other projects, like HS2, that people can ride roughshod over regulations by using a Hybrid Bill… voted on by MPs who are whipped… we will be making a complaint to the Aarhus Convention Compliance Committee … we will make a complaint to the European Commission [Emma Crane, campaigns director for HS2 Action Alliance (HS2AA)]

A £10BN LEAP IN COSTS The first indication that the costs of the scheme were increasing was at the end of January 2013 when the government announced the preferred route and stations for the second phase. At the time, the government discretely disclosed that the total cost had risen to £34.5bn (from £32bn a month before). The estimated costs of the second phase, due to open in 2033, had jumped from £16.4bn to £18.2bn. The first phase, due to open in 2026, was expected to cost £16.3bn.

As the costs kept rising, the business case for HS2 came under attack. In March 2013, the National Audit Office, the government-spending watchdog, published findings suggesting that there was ‘little’ evidence to support the claim that building HS2 would bridge the north south divide and rebalance the UK economy. The NAO argued that the business case lacked ‘clarity’ - a comment that was picked up by the chair of the Public Accounts Committee: “There is virtually no evidence in this business case to support claims that HS2 will deliver regional economic growth.”51 The

51 Odell, M. (2013). Watchdog rejects strategic case for building HS2 link. Financial Times. March 16.

249

supporters of HS2 hit back, with the transport secretary declaring “I do not accept the NAO’s core conclusion...it depends too much on out of date analysis and does not give due weight to the good progress that has been made since last year.”

Prevailing controversy aside, the real bombshell dropped in June 2013 when the government announced that the HS2 budget had shot up to £42.6bn. On top of the £42.6bn a further £7.5bn (of which £1.7bn was contingency) would be set aside for rolling stock. A breakdown of the costs revealed a Phase 1 budget of £21.4bn including £5.7bn contingency, and a Phase 2 budget of £21.2bn including £8.7bn contingency. The budget was quoted at a P95 level of certainty, which meant that there was 95 % certainty that HS2 would be delivered to the budget assigned to it.

Announcing costs at a P95 level of certainty was unprecedented at this early stage of design. Typically, budgets would be announced at a P50 level, as was the case with the previous budget. The revelation that the government had ratcheted up its level of certainty undermined public confidence in the scheme, and forced the government and senior HS2 officials onto the defensive.

The Chair of the Public Accounts Committee slammed the £14.4bn contingency, and warned that the government had failed to prove that HS2 was worth its price tag:

HS1 will continue to cost the taxpayer money … so before going ahead with HS2 we need a robust cost benefit analysis…That is a heck of a lot of public money set aside as a contingency…Why so much?…You will not get the hybrid Bill through in the time period that you have set yourself… there will be cost implications which will take us beyond the £15.7bn and the £5.7bn contingency just for phase 1.

In the DfT’s defence, the DfT permanent secretary Philip Rutnam insisted that the previous figure was only an “early estimate” and that the cost of the scheme had increased “for very good reasons”. Regarding the contingency, Rutman argued that a risk assessment calculation was used, with an optimism bias added for Phase 2. But it remained unclear what proportion of the cost increase was due to specific design changes versus increases in contingency funds.

The change between the amount for phases 1 and 2 together – the figure of £32.7bn, which was used in 2012 – and the figure released last week reflects two things…much more detailed design work and an increased amount of contingency [DfT Permanent Secretary] The new budget still assumed that the hybrid bill would gain Royal Assent by March 2015. The budget made no allowance for delays in getting the enabling legislation through Parliament, a goal that most senior leaders (privately) believed was virtually impossible to meet.

CRACKS IN CROSS-PARTY POLITICAL SUPPORT

Since its inception, HS2 had been under political attack. Lord King, a one-time Tory transport secretary, for example, remarked that “[HS2] could be a wonderful technological achievement that actually in the end doesn’t have a commercial outcome”. In a similar tone, former chancellor Lord Lawson, a long-time critic of the project warned that it would be “madness” to go ahead with a “prestige project which makes no economic sense at all”. After the announcement that the budget had shot up to £50bn and the revelation in the press that the economic benefits would be revised downwards, attacks on HS2 intensified. The attacks were coming from Labour grandees such as Alistair Darling, the former chancellor, as well as Lord Mandelson, the former business secretary. And for the first time there was even an attack from a member of the shadow cabinet namely, Mr Ed Balls.

Lord Mandleson’s forceful intervention crystallised growing doubts about the scheme. He said he feared that HS2 could be an expensive mistake. He had changed his mind and urged other politicians to rethink their support for HS2. He admitted that in 2010 the Labour government’s decision to go ahead with HS2 was “partly politically driven” and estimates at the time were 250

“almost entirely speculative”. He also argued that alternative ways of spending the £30bn cost had not been actively considered. Lord Mandleson also revealed that the business case failed to account for the full extent to which building HS2 would threaten existing regional rail services, and risked giving way to “railway deserts”. Also sceptical, Mr Balls, the shadow chancellor, questioned HS2’s value:

The question is not just whether the HS2 line is a good idea or a bad idea but whether it is the best way to spend £50bn from the future [Ed Balls, Shadow Secretary]

As a retort to the opposition, key actors reaffirmed the government’s commitment to HS2. Doug Oakervee asked for the government’s authorization to respond to Lord Mandelson, and two days later published an article titled ”Britain needs a new high-speed line”:

HS2 will strengthen our country and support our economy – but like all big infrastructure projects it is also controversial…. This is not to dismiss questions being put by people such as Lord Mandelson...But we’re ready with good answers. Do we need HS2 to serve the transport needs of the future? Is the economic case really there? To each of these my answer is yes.

In further attempts to neutralise opponents, the Transport Secretary launched a review of the cost benefit analysis. The Prime Minister David Cameron also insisted that his own enthusiasm for the project ‘remained undented’, and argued that if Labour failed to back HS2 it would be a betrayal of the North. Still, when the consultation for the second phase started in July, the Labour shadow chancellor said that he was “concerned about rising costs... as a Leeds MP I can see the benefits for the region and the north of England, but it is not a blank cheque”.

To further convince sceptics that the HS2 economic benefits would far outweigh the costs, the government created an “HS2 growth task force”. The task force, led by the Infrastructure Minister, was instructed to investigate regeneration opportunities around the stations, the creation of new towns, potential areas for skills development, and job opportunities. Convening for the first time on July 31, its membership included local council chiefs along the route of the line, industry leaders, union representatives, and academics.

Over the summer, HS2 opponents continued their fight against the scheme: the Institute of Economic Affairs claimed the overall cash costs could hit £80bn; reports were floated in the press that Treasury officials were estimating the cash cost of HS2 would hit £73bn; and the Institute of Directors called the scheme “a grand folly”. The London issues

The integration of HS2 in London was very complicated on a number of fronts [Exhibit B.1a-c]. Throughout 2013, HS2 Ltd. faced slippages in the estimated costs and schedule on the baseline plan for Euston, but was reluctant to use the planning contingency or move the completion targets. Thus HS2 Ltd. recommended that the government drop initial plans to overhaul Euston station. Instead, the Bill was now proposing a less disruptive intervention that fit well within the initial targets. This decision did not surprise some London stakeholders—‘very early on, people’s eyes are much bigger than their stomachs’, said a TfL officer. But it infuriated others; a Camden council officer said:

They [HS2 Ltd] said ‘look, this baseline scheme’s not deliverable... so that really annoyed us... retaining the classic platforms and adding the high-speed platforms is the worst-of-all-worlds for us; we get the demolition, but retain the bad bit of the current Euston... deliver a high-speed rail station into central London by 2026, that is their overriding ambition – anything that gets in the way in terms of alternatives is not considered if it challenges that.

Another controversy was the proposed £2bn link between the HS2 interchange at Old Oak Common and HS1. This link was important for the political leaders in the North particular Birmingham who insisted HS2 should connect directly to Continental Europe. It would require 251

running a single track in a 6.1km tunnel before connecting to an existing 2.4km over ground section that needed to be enlarged. Such enlargement required the demolition of eight 150-year old bridges in Camden Town, a landmark of commerce and youth culture. The Camden Council wasn’t happy; one official explained it: we feel hard done...the HS1-HS2 link through Camden town could have a massive impact on the economic wellbeing of Camden town ...it’s incredibly ill-thought out... we would argue ‘Don’t spend any money on a link for which there’s no business case; spend that money in Euston on developing a proper solution

A level up, the issues in London involved Boris Johnson, the powerful Mayor of London, and the agencies working with the Mayor’s office, Transport for London (TfL) and the Greater London Authority (GLA). Unlike Camden Council, these entities were not opposed to the high level principles of HS2, including the geographic line of the route, the intermediate station at Old Oak Common, the terminus station at Euston, and the link to HS1. But from early discussions with HS2 Ltd. officials and government, it was obvious that the Mayor of London and other London stakeholders would flex their political muscles to get the best deal for the capital. The Mayor insisted that HS2 should go into Euston. TfL led the discussions with HS2 Ltd., and TfL senior officials complained that HS2 Ltd. was not always good at listening to their claims; one TfL senior official said: at times [HS2 Ltd] treated us like a stakeholder rather than a partner ... what it really should be is that the two of us are in the same room discussing the design... it has been a bit like tennis rather than a process of actually working on the same team.

TfL was however happy that HS2 Ltd. had adapted the design for Euston to meet their concerns: the [Euston] design has changed...because of input from our team. So that’s a good thing. Still, the Mayor sided with the Council in opposing the HS2-HS1 link in its proposed form, arguing that the construction of the over ground section was too disruptive, and insisting on a tunnel solution.

The Mayor also joined a chorus of voices, including the business lobby group London First, which insisted that it was vital to build Crossrail 2 before completing the second phase of HS2 to relieve pressure on Euston [Exhibits B2a, b]. Finance for Crossrail 2 was a major issue, which the Mayor and others acknowledged, and suggested a mix of government grants, bonds, and direct or indirect contributions from private developers and local businesses. But the Government was reluctant to make a concession because it did not want to further stretch its capacity to borrow money, and seen to excessively favour London in detriment of other parts of the country. Some officials at TfL understood that and would be happy if HS2 Ltd. simply agreed to design in some passive provisions into their scheme. Although this would cost a couple extra million pounds it would serve as a safeguard so that in the future TfL could easily plug a Crossrail 2 station in at Euston:

We’re quite pragmatic...it’s always difficult to get any government... to commit to a project of this scale, let alone to add ‘oh, by the way, let’s get another major project’… after the next general election in 2015, we will then be presenting our case to government... We will then say we think you should...help us fund Crossrail 2... But that will be on its own merit.

But other London stakeholders were more forceful, and were pressing the government for an immediate commitment. This insistence was a concern for the HS2 leaders. Doug Oakervee said:

[Crossrail 2] needn’t be part of this bill – there’s no reason why they can’t have their own act. There’s no reason why they can’t do that at all. But it’s really a question of trying to push the cost of it into HS2. … this is saying ‘I want something far bigger than I’ve got’

Political leaders in London were unarguably putting pressure on the government to build Crossrail 2 before opening the second phase. They were adamant given their concerns that London’s population was projected to grow by 1.5m over the next 2 decades spelling further rail capacity 252

issues. Thus, in February 2013, TfL announced that it was starting to study two options for Crossrail 2: a new and cheaper line under central London costing about £9.5bn; and a longer option crossing central London but linking into commuter lines outside London, projected to cost at least £12bn.

Another controversial idea that had been kicking around since 2010 was adding even more noise to the discussions. Billions of savings could be achieved if HS2 terminated at Old Oak Common— this idea could be enacted by submitting an amendment to the bill. The idea was opposed by TfL and the Mayor who dubbed it the “Ryanair solution”. But terminating HS2 at Old Oak Common was an attractive option for resolving the problem of the cost and schedule slippages with remodelling Euston—the principle that suggests that all the works at Euston station are on the HS2 credit card has been accepted, said a TfL officer.

Furthermore, terminating the line at Old Oak Common would release monies to fund the entire redevelopment of this site, another controversial issue. HS2 Ltd. was proposing to simply develop a new HS2 station. But the Mayor, GLA, and the local councils wanted £0.5bn of extra investment to fully redevelop the site and had already commissioned a master plan. A GLA official said:

[HS2 Ltd] have been asked to design a station which will provide an interchange with Crossrail and keep the cost down ... [our masterplan documents] are means by which a process of negotiation is enabled ... they might not agree with them... but we are looking for a contribution from the government and we want [the government] to include it in the Bill.

In summary, the Mayor and other London stakeholders threatened to withdraw support to HS2 unless: 1. The line terminated in a fully modernised Euston station; 2. The HS2-HS1 link was entirely tunnelled; and 3. Both Old Oak Common and Euston stations were properly integrated with London Underground and other Transport for London services.

CHANGING THE LEADERSHIP

As Doug Oakervee passed the baton to Sir David Higgins, ‘who is twenty years younger than me’, quipped Doug, he expressed confidence that David was the right person for the job—Sir David is the right man to continue... I’m proud of all that I have achieved, particularly that we will be bringing the hybrid bill to Parliament by the end of the year. David, who would join HS2 Ltd. on 1 January 2014 in a part-time role before becoming full-time chairman in March, was passionate about HS2. Leveraging his NR chief executive credentials, he said:

Everything is being done 20 years late. Like the water industry, the railways have been dining out on the legacy of Victorian investment… I know the challenges we have ahead with capacity on the railways … HS2 presents a strategic opportunity for this country

Ahead of David and his team lay the challenge of maintaining cross-party support for the scheme. HS2 Ltd. needed to help the government convince MPs and the public in general that the scheme could be done responsibly and cost effectively. This required sharpening the story around HS2. To this purpose, the government tasked David with producing a full review of the scheme and report in March. The prime minister asked David to use his “penetrating eye” to cut costs and give consideration to bringing the construction of phase two forward. David expressed sympathy with this last point—phase 1 will open in 2026 with phase 2 opening six years later…if you are in the North you want to see the benefits earlier.

The government and others were therefore expecting the Sir David Higgins’ review to come up with ideas to deliver HS2 faster and more economically. Although David’s remit was popular some treated it with caution. The HS2 Ltd. technical director, for example, agreed that if the second hybrid bill could be approved quicker than planned, then “you could see a number of years being dropped off the scheme”. But he noted that there were constraints associated with this notably the 253

Treasury interest in smoothing out the annual spend on the project to minimize pressure on the public purse. The government was also quick to play down talk of building HS2 from its northern end stating:

We are starting with the London to Birmingham section because it is the southern end of the West Coast Mainline where the capacity crunch needs tackling first. However, we will work hard to complete the northern sections more quickly than 2032/33 if possible [DfT spokesperson]

***** David was bringing a wealth of experience to HS2 Ltd., but had never led a hybrid bill process. The strategy needed to ensure that a majority of MPs continued to support the bill. Strong cross- party support was imperative not only for the HS2 bill to survive in April 2014, but for the scheme to survive an eventual change of the party in power after the 2015 national elections.

David and others would have to cultivate confidence in businesses, communities and the MPs who represented them in Parliament. How could they best sell HS2? Should the narrative be crafted around capacity and connectivity, or evolve to include urban regeneration and growth?

The new HS2 leadership team also needed to focus on the costs and schedule. Should they keep their cards close to their chests until the bill had been approved in principle in April, or continue the discussions so as to try to dissuade the Councils, the London Mayor and others to petition? HS2 Ltd. could try to craft legal agreements that would assure the affected parties that the scheme would do what it was promising to do. HS2 Ltd. could even submit amendments to the bill before the vote in April. But if it did this, which concessions should be made? The political environment was such that another major slip in cost or schedule could kill the whole scheme.

HS2 Ltd. was also sitting on top of a huge and controversial contingency. Should they devolve some contingency to the government to neutralise the argument that they were sitting on a fat budget, or keep the contigency intact to accommodate all the incoming design deals? And how far should they push back on demands for concessions to ring fence the contingency for late construction risks?

At this point, not only were the design plans for the first phase up in the air, but discussions with the Northern cities were still on-going for the second phase. Liverpool, for example, was lobbying for a spur between Crewe and Liverpool. Pressure from the Northern cities to speed up the construction was also mounting. And although it wasn’t part of the current discussions, stakeholders in the capital still wanted a spur to Heathrow. This could eventually be up for reconsideration if the government approved the airport’s request for a third runway (a decision not due before 2015).

The stakes at play were high. Both political parties were urging their MPs to support the bill, which reassured to the HS2 officials that the bill would get a favourable vote in April. But with the upcoming national elections in 2015, things could always take an unexpected turn. Politics aside, the risk of further slippages as design discussions progressed and petitions were debated remained high.

254

EXHIBIT B.1a - High Speed 2 Proposed Scheme for London (Overground) (TfL Presentation, 2012)

255

EXHIBIT B.1b, c - HS2 Ltd. and TfL Proposed Scheme’s for London (Overground) (TfL Presentation, 2012)

256

EXHIBIT B2a, b- CROSSRAIL 2 SCHEME ALTERNATIVES (London First, 2012)

257

Appendix D HS2 Ltd’s wider route network options

Source: DfT, 2010

258

Appendix E HS2 Route

Appendix E 1 Proposed High Speed 2 Route Map, 2013

Source: HS2 Ltd., 2013

259

Appendix E 2 Route Map, 2017 (includes changes in Sheffield)

Source: HS2 Ltd., 2017

260

Appendix F Longitudinal Chronology of the HS2 Planning Process

Major events and related Performance targets (cost, Time Scope description documents time)

Major events Estimated Cost A High-Speed Rail Line HS2 Ltd. established by Labour running north from London to Government to examine the case £34 billion (in 2009 prices, Birmingham, Manchester, for a new high speed rail line including optimism bias of 66% or and Glasgow and Edinburgh Idea generation for routes £13.5 billion) (Network Rail, 2009) between London and Birmingham £15 billion (in 2011 prices) for Phase 1 2009 Related Documents DfT, Britain’s Transport Infrastructure High Speed Two DfT, The role and funding of High Speed Two Ltd Network Rail, Meeting the Capacity Challenge : The Case for New Lines Major events Estimated Cost Scope : Unchanged Preferred route and other In the region of £30bn (in 2009 Y-shaped network of around possible routes published prices, including risk, spread out 335 miles linking London to Formation of Coalition over 20 years or more) Birmingham, Manchester, Government. the East Midlands, Sheffield Slightly revised line resembling Phase 1: Between £15.8 billion and Leeds capable of

original Y-shaped route and £17.4 billion (in 2009 prices) carrying trains of up to 250

published for consultation Phase 2: Uknown given miles per hour. (DfT, 2010)

Confirmation that HS2 would be significantly less detailed 2010 2010 built in two phases planning therefore significantly higher uncertainty Related Documents DfT (2010) High Speed Rail Timescale Command Paper Phase 1 (London to West HS2 Ltd. for DfT, High Speed Midlands) : 2026 Rail London to the West Phase 2 (Manchester and Midlands and Beyond: A Report Leeds) : 2032-33 to Government by High Speed Two Limited, March 2010 Major event Estimated Costs Scope :Unchanged DfT publishes route configuration Estimated cost of £32.2 billion to ‘The broad plan for the for Phase 1 for National construct (2009 prices, including network to Manchester and Consultation 64 % in allowances for risk and Leeds including the section Strong responses from those in optimism bias) between London and the areas of outstanding national West Midlands is roughly the beauty and on the area around Additional estimations of capital shape of a letter Y.’ (HS2

Euston station, particularly costs for rolling stock Ltd., 2011) Camden procurement and replacement of

2011 £5.3 billion (in 2009 prices Related Documents inclusive of optimism bias) HS2 (2011) Consultation document DfT, High Speed Rail : Investing Timescale : Unchanged in Britain’s Future – Consultation, February 2011, p16 HS2 Ltd., Economic case for HS2: The Y network and London-West Midlands

261

Major events Estimated Costs Scope : Government Announces Plans to £32.7 billion (2011 prices), capital ‘The HS2 Y network will Proceed with Y-shaped route cost including significant provide direct, high capacity, HS2 Judicial Review (High Court allowances for risk and optimism high speed links between Hearing) bias. In addition there will be London, Birmingham, Leeds

Public consultation on blight rolling stock costs of around £8.2 and Manchester, with HS2 Growth Taskforce created billion (2011 prices) intermediate stations in the

2012 Phase 1: £16.3 billion (in 2011 East Midlands and South Related Documents prices) Yorkshire. There will also be DfT (2012) High Speed Rail : Phase 2: £16.4 billion (in 2011 direct links to Heathrow Investing in Britain’s Future – prices) Airport and to the continent Decisions and Next Steps via the HS1 line.’ (DfT, 2012) DfT (2012) High Speed Rail : Timescale : Unchanged Investing in Britain’s Future – Decisions and Next Steps Major events Estimated Cost and Budget Scope Changes: Initial Preferred Route 14 significant changes configuration for Phase 2 Costs : £ 38.4 billion (in 2011 proposed to the outline Announced ( prices) project design, including to Consultation on environmental redevelop rather than knock Statement for Phase 2 Phase 1 : £19.4billion (in 2011 down and rebuild Euston commences prices, including contingency of station : ‘Work has led HS2 High Court Rules in favour of £3.7 billion) Ltd. to refine the design of govt on 9 of 10 areas of the railway and in a small challenge, and upon appeal court Phase 2 £19.0bn (in 2011 prices, number of locations propose rules in favour of Government on including contingency of £6.5 refinements to the design all areas of challenge billion) that go beyond simply Introduction of Hybrid Bill to developing the January 2012 Parliamnent (Dec) An additional £6.9bn (in 2011 design’ [HS2 Ltd., 2013) Contingency funds increase prices) for HS2 rolling stock. 1 of the 14 major proposed NAO and PAC publish reports amendments was to replace raising concerns over reliability of (All at ‘P50’ confidence level, the January 2012 proposal HS2 value forecasts excluding VAT) to rebuild Euston Station Increase in the amount of with a revised proposal for a tunnelling required Budget : £50.1 billion total new station providing new budget envelope (in 2011 prices) shared passenger facilities Related Documents to the frontof the station HS2 (2013) Design Refinements £42.6 billion budget envelope for while retaining 13 existing Consultation Leaflet fixed capital costs including £14.4 platforms, with only minor 2013 Mcloughlin, P (2013) We can’t billion of contingency across the modi cation, at their current afford not to build HS2. scheme (All in 2011 prices) level on the eastern side of www.conservativehome.com the site and 11 new DfT, The Government Response Phase 1 : £21.4 billion budget platforms for high speed to the HS2 Design Refinement envelope (in 2011 prices, trains on the western side, Consultation, 2013 including contingency of £5.75 Scopeconstructed at a lower DfT, High speed rail: investing in billion) level Britain’s future – Phase Two: the route to Leeds, Manchester and Phase 2 : £21.2 billion (in 2011 Beyond, 2013 prices, highest estimated) DfT, High Speed Rail: investing in Britain’s future - Consultation An additional £7.5 billion on the route from the West (including £1.7 billion Midlands to Manchester, Leeds contingency) for HS2 rolling and beyond, 2013 stock. Environment Statement, November 2013; Timescale : Unchanged PAC, High Speed 2: A review of early programme preparation (twenty-second report of session 2013-14), 2013, NAO, High Speed 2: A review of early programme preparation, 2013

262

Major Event Estimated Cost and Budget Scope Change : Sir David Higgins, Chairman of Change to Phase One from HS2 LTd., publishes a report Costs and budget unchanged London-Birmingham to

proposing, among other things, a since 2013 London-Birmingham-Crewe more ambitious plan for Euston

2014 Related Documents Timescale HS2 Ltd., HS2 Plus: A report by Delay full opening of London David Higgins, 2014 station to 2032 HS2 Ltd., Rebalancing Britain: from HS2 towards a national transport strategy, 2014 Major events Estimated Cost and Budget Scope Changes : Upwards of 120 amendments to Euston redevelopment: 11 Hybrid Bill tabled by Government Cost : Phase 1 : £24.3 billion (in new platforms for HS2, to be in June 2015 2015 prices, and including built in two stages, with ‘at Revised plans for the contingency) least’ 11 platforms remaining development of Euston. in the current station to serve Government announces intention Phase 2 : Under refinement the existing network, and to bring forward the route to new public spaces for shops, Crewe. restaurants and cafes Government updates the HS2 Budget envelope : £55.7bn (in Plans for an HS2 spur to cost figures to 2015 prices 2015 prices) or £50.1 billion( in Heathrow Airport dropped stating that changes do not 2011 prices) including rolling relfect a budget increase but stock adjustments for inflation National Audit Office and Public Phase 1 : £27.18 billion budget accounts committee publish envelope (in 2015 prices) reports flagging concerns about the possibility that contingency Phase 2 a: £3.72 billion budget funds being used to hide cost envelope (in 2015 prices, overruns (PAC) including 40 % optimism bias)

Related Documents Phase 2b : £27.18 billion budget

PAC, Lessons from major rail envelope (in 2015 prices, 2015 infrastructure programmes including 40 % optimism bias) (twenty-eighth report of session 2014–15), HC 709, 12 January Flexible budget for Phase 2 2015 (based on NAO, Lessons agreed between DfT, HS2 Ltd from major rail infrastructure and HM Treasury. Scope can programmes 2014 and related budget can be DfT, High Speed Rail transferred between Phase 2a (Preparation) Act 2013 and Phase 2b Expenditure Report 21 November 2013 - 31 March Timescale 2015, e2015 Phase 1 to open in 2026 The Economic Case for HS2, (London-Birmingham) p85; and DfT, High Speed Two: East and West: The next steps to Phase 2a to open in 2027 Crewe and beyond, 2015 (Birmingham-Crewe) Lords EAC, The Economics of High Speed 2 (first report of Full Phase 2b to open in session 2014-15), 2015 2032/2033 (Crewe-Birmingham House of Lords Economic Affairs and Birmingham-Leeds) Committee: The Economics of HS2 Government Response, 2015

263

Major Event Cost Estimates and Budget Scope Changes: National Audit Office (NAO) Conflicting Cost estimates for Main HS2 route to run east publishes report on HS2’s Phase 1 of Sheffield, with a separate progress and preparations £24.3 billion (with contingency) spur to take passengers to stating that the £55.7bn funding (DfT, 2016) Sheffield city centre (instead envelope is not sufficient to of a new station at Sheffield deliver all the growth and £27.4 billion (in 2015 prices, Meadowhall). regeneration promised including contingency and rolling ’ Route for Phase 2 confirmed stock) (NAO, 2016)

Related Documents Phase 2 : under refinement 2016 DfT, High Speed Two Phase 2b: Phase 1: Route-wide budgte Strategic Outline Business Case envelope unchanged - Economic Case, 2016 NAO, Progress with preparations for High Speed 2, 2016 Budget envelope Phase 2a is £3.48bn and Phase 2b is £25.07bn (in 2015 prices and Inclusive of 40 % optimism bias)

Timescale : Unchanged Major Event Cost Estimates and Budget Scope :Unchanged Hybrid Bill authorising construction of Phase 1 was Costs : Unchanged since 2016 granted Royal Assent Hybrid Bill authorising works for Phase 2a published Budget : £55.7 billion (in 2015 Begin construction for Phase 1 of prices, including £14.9 billion the scheme contingency

Related Document Phase 1 : £24.3 billion (in 2015 DfT, High Speed Two: Phase prices, including £6.3 billion

Two Economic Case, 17 July contingency) 2017

2017 DfT, High Speed Two Phase 2 Phase 2a : £3.48 billion (in 2015 Financial Case, 17 July 2017 prices)

Phase 2b : £25.07 billion (in 2015 prices, including £6 billion contingency)

Additional £7 bilion (in 2015 prices, including £1.6 billion contingency)

Timescale : Unchanged since

264

Appendix G Birmingham Appendix G 1 West Midlands Sifting Process

Source: DfT, 2010

265

Appendix G2 Possible Stations Locations in Birmingham – Stage Two

Source: Dft, 2010

266

Appendix G3 Birmingham Curzon Street Connections from High Speed Two

Source: DfT, 2010

267

Appendix G4 Virtual Image Showing of Birmingham Curzon Street HS2 Station

Source: Birmingham City Council, 2015

268

Appendix H Leeds

Appendix H 1 Leeds Sifting Process

Source: HS2 Ltd., 2012

269

Appendix H 2 Proposed Leeds City Centre Station Final Options

Source: HS2 Ltd., 2012

270

Appendix H 3 Proposed New Lane Station Layout

Source: HS2 Ltd., 2012

271

Appendix H 4 New Lane Indicative Station Illustration

Source: HS2 Ltd., 2012

272

Appendix H 5 Leeds Station North Layout

Source: HS2 Ltd., 2012

273

Appendix H 6 Leeds Station North Indicative Illustration

Source: HS2, 2012

274

Appendix H 7 Options for Leeds Station

Source DfT, 2015

275

Appendix H 8 Integrated Leeds Station

Source: HS2 Ltd., 2017

276

Appendix I London Appendix I 1 London Sifting Process

Source: DfT, 2010

277

Appendix I 2 Proposed Footprint for a Completely Rebuilt Euston Station Footprint

Source: DfT, 2010

278

Appendix I 3 London Rail and Underground Connections for High Speed 2

Source: DfT, 2010

279

Appendix I 4 Proposed Plan for a Scaled Down (Redeveloped) Euston Station

Source: HS2 Ltd., 2013

280

Appendix J Sheffield

Appendix J 1 South Yorkshire Sifting Process

Source: HS2 Ltd., 2012

281

Appendix J 2 Sheffield Victoria Station Connectivity

Source: Sheffield City Council, 2014

282

Appendix J 3 Proposed Sheffield Victoria Station Layout

Source: HS2 Ltd., 2012

283

Appendix J 4 Sheffield Victoria Station Indicative Illustration

Source: HS2, 2012

284

Appendix J 5 Sheffield Meadowhall Connectivity Map

Source: HS2 Ltd., 2014

285

Appendix J 6 Proposed Meadowhall Station Layout

Source: HS2 Ltd., 2012

286

Appendix J 7 Meadowhall Station Indicative Illustration

Source: HS2 Ltd., 2012

287

Appendix K Manchester

Appendix K 1 Manchester City Centre Station Long List to Final Options

Source: HS2 2012

288

Appendix K 2 Manchester Connectivity Map

Source: HS2 Ltd., 2014

289

Appendix K 3 Manchester Piccadilly Station - Platform Level Plan

Source: HS2, 2013

290

Appendix K 4 Manchester Piccadilly Indicative Station Illustration

Source: HS2 Ltd., 2012

291

Appendix K 5 Architects Rendition of HS2 Piccadilly Station

Source: Manchester City Council, 2014

292

Appendix K 6 Proposed Salford Central Middlewood Station Layout

Source: HS2 Ltd., 2012

293

Appendix K 7 Salford Central Middlewood Indicative Station Illustration

Source: HS2 Ltd., 2012

294

Appendix K 8 Proposed Salford Central combined station layout

Source: HS2 Ltd., 2012

295

Appendix K 9 Salford Central combined indicative station illustration

Source: HS2 Ltd., 2012

296