The strategic front-end of large infrastructure projects
a process of nesting governance
The paper is organized into seven sections: 1) the characterization of front-end processes in terms of sources of complexity and potential explanations of governance failures, 2) the contrasting of the mental models of “project planning” and “project shaping,” 3) the roles and requirements of successful governance structures, 4) contingency factors in the design of governance structures for the front end, 5) creating a nested project governance structure, 6) modeling the front-end processes, and 7) conclusion and further research.
The Characterization of Front-End Processes in Terms of Sources of Complexity and Potential Explanations of Governance Failures
Large engineering projects are complex systems that are usually led by a sponsor but include other players such as regulators, bankers and lenders, affected parties, government officials, and a range of contractors. The complexity induced by the number of parties is only a starting point. Each party has objectives, responds to particular incentives, and can contribute to risk management as it controls an area of activities where it has a comparative advantage. The multiple components interact over time to produce planned but also emergent processes. On the average the front-end phase of the projects in the sample took seven years. As parties interact, they give birth to a series of processes that can lead to system adaptation, but also to its demise. The governance structure must integrate the many parties involved and control processes. The authors identified the following processes and characteristics of the front end that give rise to complexity and provide explanations for the high failure rates observed.
1. Processes of over optimism: Many managers and groups tend to be unrealistically optimistic, exhibit illusions of control, and display blind spots in their analyses. By contrast, however, some groups relish pessimism. As a consequence of over optimism, cost, delays, and risks tend to be underestimated. Decision biases make it very difficult for parties to assess risks, particularly downside risks, adequately. There is even a tendency to escalate commitments with respect to bad choices in the hope of correcting past errors. (Dosi and Lovallo 1997) To bring reason and healthy pessimism to projects analysis, one needs to counter such overconfidence. With the tendencies toward over optimism, attempts to analyze worst-case scenarios often become an analysis of minor variations on the most likely future. A thorough analysis of possible catastrophes is not likely to take place.
2. Processes of reductionist strategic focusing: Over optimism is associated with a narrow focus in terms of the possible solutions to problems, but more importantly, leading to the late identification of stakeholders, issues, and risks. We have observed that experienced sponsors adopt a systemic perspective and attempt to build a prototype of the project which focuses on a wide range of strategic issues: 1) financial justification, 2) agreement with the state, 3) social legitimacy and acceptability, 4) technical feasibility, 5) survivability of future streams of income, 6) abilities of sponsor to meet emerging crises, 7) potential for savings and innovation through generative contracting forms, and so on (Miller and Lessard 2000).
3. Processes of increasing complexity of issues: In the early phases of the front end of projects, only a few parties are involved and the problem to be solved involves generally high-level strategic decisions. Concerns for the inevitable details of practical reality are left for later action. Decisions taken in the early phases entail logical links to other issues and external effects on parties. Quite often, these are not dealt with immediately. Eventually, the development of the project triggers the entry of additional parties into the decision processes. The new arrivals redefine the nature of the problem to be solved. As parties participate in decisions, issues get more complex. This implies extensive negotiations and requires agreements that are difficult to reach (Miller and Lessard 2000).
4. Processes of shaping to face uncertainty: Projects cannot be specified in advance. It is impossible to bring all parties to the table, gather all pertinent risk estimates or information, and conduct a negotiation to arrive at a final agreement. Not only do new parties enter decision processes, but also the nature of the context changes, over time. Some risks remain constant, but new opportunities and risks emerge, and unexpected turbulence is brought by unforeseen events. In the sample studies, an average of four major unforeseen events were encountered. In such situations of high emergent complexity and uncertainty, the project development process is not linear and continuous, but is characterized by a series of episodes or rounds of negotiation. We have observed that on average five to seven distinct shaping episodes are necessary before arriving at a commitment to finance and build.
5. Processes of influence, negotiation, and closure: During each episode, the sponsor will provide data and analysis in an attempt to convince parties. Methods of influence such as offering incentives, involving public officials, and informing parties about the consequences of actions will also be used. Eventually closure will be achieved by an agreement on a formulae or a memorandum of understanding. Unless they are framed by legitimate institutional frameworks, agreements can unravel. The process of negotiating an agreement, therefore, often involves shaping the institutional framework within which the project will be carried forward (North 1990).
6. Processes of creating and sharing value: The value that an engineering project can create is the result of lengthy negotiations to arrive at agreements with the relevant parties that have a legitimate or powerful right to participate in the process. The smart sponsor knows that transaction costs will be high and that sometimes extra transactions will allow new value creation. To gain additional value, the sponsor(s) may have to share some part of the returns with affected parties or participants with power (Cabral 2000).
7. Process of search for projects characterized by uncertainty: Projects that create significant value are almost by necessity risky endeavors. Less complex projects in more well established domains invite many players and expected profits are low. There are few barriers to entry and thus quickly diminishing returns. On the other hand, seeking out high-risk projects with little hope of high returns in foolhardy. The best returns seem to be with moderately complex projects in incomplete but transformable institutional frameworks. The art is to identify these potential projects and to have the skill to bring a significant portion of them to fruition.
8. Processes of project approval through episodes: Few large projects have a single approval process with simple criteria. Most projects must pass multiple tests on a variety of criteria before they acquire the legitimacy necessary to ensure their survival. In addition, the tests of the projects’ legitimacy will take place in multiple forums. In a fashion similar to investments by venture capitalists, decisions to invest and proceed are made through rounds (Trigeorgis 1996). We have observed that the shaping of large engineering projects is very similar but more complex than venture capital investments in high-technology firms. Project shaping proceeds through episodes that depend heavily on the specific context of the project.
9. Processes of interactions between parties can become highly dysfunctional and even destructive. Parties do not always behave rationally. Barriers, meetings, or arenas where parties could reconsider their actions may not have been designed. Moves and countermoves can lead to a vortex that can cause project demise. Without a set of institutional and governance devices, otherwise viable projects entire into deadlock or demise (Stinchcombe 1990).
We employ the expression “project shaping” to describe the iterative and evolutionary process of front-end development. Project shaping is not a passive process, but requires active guidance by a strong project sponsor or sponsors. The perspective in project shaping is, however, quite different from the traditional project management viewpoint.
Contrasting Mental Models of Project Planning and Project Shaping
The thrust of the project management profession to date has been the design and delivery of a solution to a predetermined problem. It is assumed that good project management practice requires advanced specification and programming. The project management profession tends to assume the existence of a pre-established business plan, in which the objectives and constraints are clearly identified. An illustration of this is the very summary treatment of the project initiation process that is offered in A Guide to the Project Management Body of Knowledge (PMBOK® Guide) – 2000 Edition.
In this perspective, the project life cycle is a set of foreseeable decisions that will progressively refine the solution and deliver it. The project trajectory is seen as a program, the details of which are yet to be determined but the overall program is defined in the project plan. The level of uncertainly is high at the beginning of projects, but efforts are spent to successively reduce it. Major changes are seen as signs of failure in project management. The implicit or explicit assumption is that the path of well-managed projects is there to be discovered. If the project is well managed, it will unfold in an orderly fashion. There is one best path, and if we do things correctly, we will discover it. The traditional project management perspective is in many ways a planning perspective.
Our observations have shown that this image does not fit the reality of the front ends of large and complex projects. Shaping projects is a journey with a starting point, a vision of the end point but with an uncharted path. The trajectories of the front-ends of large complex projects are, to a large extent, indeterminate and ill structured. It is not possible to foresee the journey ahead with any degree of accuracy. There are multiple paths that the project might take and as some are explored others emerge. Along the path, new players and issues are encountered. As parties with legitimate claims enter into the process, the definition of the problem to be solved evolves.
There is a fundamental difference in the perceptions of control between the project planning and shaping perspectives. In project planning, the proportion of things under control in high, and efforts are focused on confirming the project concept and further reducing levels of uncertainty in those elements that are not yet under control. In the project-shaping mode, it is recognized that a great deal is not yet under control. Efforts are invested to discover possible scenarios that might lead to a feasible project. This is done through probes, many of which involve interaction with stakeholder to explore possible common ground. This involves redefining the project concept and the incentive system that accompanies it. Within a shaping perspective, the project sponsors accept that they are putting into place a semi-autonomous system to explore possibilities. Project sponsors must accept the fact that they will have to live with the unforeseeable consequences of their own actions. They accept the paradox that they must accept losing some control early on, in order to reach a destination that is acceptable to them and that is sufficiently under control to make the project feasible.
From a planning perspective, the shaping of projects is a process that looks very messy and unorderly. This is not to say that it is entirely random. Some sponsors do better than others in developing projects. But in order to succeed, they must adopt a mental model of the project development process that is radically different from the traditional project management planning perspective. In order to come to grips with the characteristics of the front-end development processes, a major paradigm shift is required. This paper attempts to identify the basis of such a model.
The Roles and Requirements of Good Governance Structures
Many of the projects analyzed were successful. In these projects, a governance structure was in place that focused decisions to be made and forced the involvement of many players. There is considerable variation from one context to another and from one project to another as to the specific governance arrangements or structures that were put in place. However, when investigating the role of these successful governance structures and what they accomplished, many of the same themes were common throughout the sample. These are presented in the present section.
1. Controlling the opportunism of parties: Projects are submitted to risks, emergent turbulence, and opportunistic behavior that can trigger degeneration processes. The institutional and regulatory arrangements created within in country can be a bulwark to control opportunistic actions. In fact, the presence of well-thought institutional arrangements is the factor that has the highest impact on the probabilities of success (Miller and Lessard 2000). Over the years, it has become clear that institutional frameworks and practices form the basic determinants of competitiveness.
2. Creativity and scrutiny: Effective governance structures provide scrutiny, by providing a process that identifies the major issues and makes certain that they are addressed. The key is to identify all the major issues, to put in place mechanisms to address them and not to allow commitment to the project to get out of step with the resolution of key issues. To do this, the governance structure must identify all the relevant perspectives from which the project may be viewed, and all the tests it will be subjected to. This involves setting up a decision structure that will make sure all the right questions are being asked, initiate processes to develop answers to them, and set hurtles the project must negotiate. This in turn, requires the identification of the parties that are well suited to the task and structuring a decision process to integrate them.
3. Systemic perspective: Identifying all the key issues, the key players, and putting in place appropriate processes requires a systemic perspective. The focus must be on identifying the major systems and subsystems that will affect or be affected by the proposed project, and identifying the interaction effects among them. A systemic perspective will allow the project sponsors to identify the key issues and players, and to foresee possible reactions to proposed solutions. It must be recognized that not all positions are “rational,” that some reactions to proposed project concepts are emotional, symbolic, and/or political. A systemic perspective will also help set the appropriate time frame for the project and support the timing of initiatives and decisions.
4. Multiple scenarios and paths: Effective governance processes are almost always based on the simultaneous pursuit of multiple scenarios. This happens for several reasons. First, early in the project it is recognized that a feasible concept has not yet been produced and that it is better to have a small set of competing concepts. Second, the governance structure initiates the examination of improbable but possible scenarios, optimistic scenarios, and particularly scenarios that represent significant downside risk. The serious examination of downside risk is one of the discriminating characteristics of good governance. It is closely related to the question of project scrutiny. Third, stakeholders in the project context develop and promote their own concepts or variations of concepts. Four, effective front-end structures tend to avoid premature commitment to a concept. The governance process must not only tolerate multiple proposals but must initiate them. Obviously, it must also decide to kill all but one concept. The question is the timing of decisions.
5. Collective risk and reward sharing: The process seeks to align the interests of the players and to put in place an accountability framework. The accountability framework is based on the identification and allocation of the important risks inherent in the project. The functioning of the venture capital market is a good analogy for the process of progressive commitments observed in front-end governance. As scenarios are developed and tested in successive rounds, the players develop strategies for anchoring commitments, while at the same time, developing exit strategies, in case the project becomes unfeasible or no longer represents a strategic interest for the stakeholder.
Contingency Factors in the Design of Governance Structures for the Front End
A great variety of specific governance structures and governance mechanisms were observed among the projects analyzed. With the benefits of hindsight, it is possible to observe how effective governance structures have been. However, designing a governance structure for a specific project is a very complex matter. During our investigation we observe that both the options that were available and the mechanisms that proved to be effective tend to vary according to the context within which the project is developed. Because of contextual variations, a contingency approach will need to be applied to the design of front-end governance structures. Five dimensions of the project context have very significant impacts on both the challenges that the governance structure must be designed to meet and the design options from which project sponsors will be able to choose.
1. Nature of the need the project is intent to fill: The nature of the demand behind a project will largely determine the appropriate governance framework. For example, a project where market demand dominates will likely be operated as an economic unit, while an initiative undertaken to fill a social need will have a governance structure skewed toward public ownership.
2. Coalitional nature of project sponsorship: There are two archetypes of project sponsorship. On the one hand, one large firm often dominates projects sponsorship. On the other hand, many projects are sponsored by coalitions of organizations. Very different governance structures will necessarily be put in place in each of these contexts.
3. Stability, completeness, and appropriateness of the institutional framework: In contexts where the institutional framework is unstable, incomplete, and/or inappropriate, the governance structure will need to be designed to effectively influence the development and/or modification of the institutional framework.
4. Role of the state: Both in terms of the nature of the specific role played and its weight in decision processes, the role of the state will be an important aspect of the governance structure.
5. Significant impacts on third parties: Projects can have both negative and positive impacts. The analysis of this contingency variable needs to go beyond the simple identification of effective parties to include an analysis of their ability to influence the project.
Several of these contingency factors interact with and impact upon the project’s legitimacy.
When one thinks of contingency variables “uncertainty and turbulence” obviously come to mind. Each of the sets of contingency variables identified here can be seen as a source of uncertainty that may result in varying levels of turbulence.
Creating a Nested Project Governance Structure
The process of developing a viable project concept can be seen as a process of shaping the project concept in interaction with the social, political, and technical context, through a series of episodes or rounds. The process can also be seen as one to build legitimacy. In order to acquire sufficient legitimacy to ensure its viability, a project will be tested in several different forums, which we have called “arenas of legitimacy.” These arenas are inhabited by communities of stakeholders who have particular issues that they feel are important and that must be dealt with before they will lend their support to the project. These same communities of stakeholders have the potential to contribute to the project and may even initiate actions to mold the project to suit their interests. To build project legitimacy a number of issues have to be addressed in each arena. The major issues in the different arenas are nested into one another.
Projects cannot be anchored solely by one or two governance devices such as a concession or harmonization of rules. Should one of these fail, the project will enter spirals of degradation and demise. Our argument is that increasing the probability of success in projects requires a range of governance devices that must be nested. In this way, should problems arise at one level, the project is protected by the presence of other layers of governance. For example, in an inter-regional or international transportation concession project, the business model of the project, i.e., flows of revenues and repayment of expenditures, depends not only on effective demand but also on its legitimacy within a local context. Consent and mobilization are the two most important sources of legitimacy. A legitimate and viable concession project needs in turn to be supported by legal and policy structures in areas such as transportation and finance. Alignment with national policies is managed by different administrative agencies that exist within a country. Furthermore, in the case of multinational projects, agreements between states to coordinate efforts are necessary. Finally, transport systems, the effects of which are felt over many countries, need to be coordinated by a supra-national policy framework.
Exhibit 1 illustrates a nested governance structures. Each issue is coped with using a governance device which is thought to be appropriate: 1) insufficiency of demand is met either by a public-private partnership, subsidies, or public ownership, 2) pressures group challenges are countered by mobilization efforts, 3) the difficulty of establishing tolls is controlled through a regulatory process with probity and transparency, 4) nonconcordance of standards across nations is dealt with by harmonization of rules, 5) coordination between agencies is achieved by means of inter-governmental commissions, 6) systemic needs across a continent are coped with by a supra-national policy framework.
Exhibit 1. Nested Governance Structures
Modeling the Front-End Processes
The process of shaping the project and building legitimacy will involve many decisions-making episodes and will take time. Someone has to take the lead. This job may be done by 1) a sponsor who can take into consideration all perspectives: project, government, and regional integration, 2) who will design a framework that will consciously aim at producing systemic benefits, and 3) will not be based on ideologies but on practical rules to increase the probability of success. As Exhibit 1 indicates, there are many levels of governance that are nested one into the other. The authors have developed a model to represent the actions that have been observed on successful projects. The model is necessarily a simplification and a stylization of reality. It is presented as an iterative stepwise process as follows:
1. Identify the relevant “arenas of legitimacy” or “stakeholder communities” that will have to come on board for the project to be feasible.
2. Analyze each arena to identify the important players.
3. Analyze each arena to identify the issues that will need to be addressed and identify issues that are recurring and/or interrelated. This process will identify the major issues, tests, or legitimacy criteria.
4. Analyze the arenas to identify the potential stakeholder contributions to resolution of the major issues.
5. Develop a project concept or concepts that meets a reasonable number of criteria in a reasonable number of arenas.
6. Do a round of project development by initiating studies around important issues and by testing the concept with stakeholders.
7. Go back to step 1 with the results of each round.
Conclusion and Further Research
The front-end processes are fundamentally different from those of project planning and execution. They are necessarily messy, political, open to social interaction, and at least partially indeterminate. A paradigm shift is necessary to understand the mental models that are effective in font-end governance processes.
A clear and systematic approach to a priori design of front-end governance structures does not presently exist. However, many elements of the puzzle have been identified. The approach presented in this paper starts to come to grips with systematically analyzing front-end processes and governance structures. However, much remains to be done.
The authors are presently initiating a major research project to investigate front-end governance structures and processes. The project will add twenty-five case studies to the sixty+ already in the IMEC database with the objective of providing a priori guidance in the design and assessment of front-end governance structures and processes.
Cabral, Luis. (2000). Industrial Organization. Cambridge, MA: MIT Press.
Dosi, Giovanni, and Franco Malerba. (1995). Organization and Strategy in the Evolution of the Enterprise. London: Macmillan.
Miller, Roger and Donald R. Lessard. (2000). The Strategic Management of Large Engineering Projects: Shaping Institutions, Risks, and Governance. Boston, MA: MIT Press.
North, Douglas C. (1990). Institutions, Institutional Change and Economic Performance. Cambridge, UK: Cambridge University Press.
Project Management Institute. (2000). A Guide to the Project Management Body of Knowledge (PMBOK® Guide) – 2000 Edition. Newtown Square, PA: Project Management Institute.
Stinchcombe, Arthur L. (1990). Information and Organizations. Berkeley, CA: University of California Press.
Trigeorgis, Lenos. (1996). Real Options. Cambridge, MA: MIT Press.