Exercising Agency

Decision Making and Project Initiation

Mark Mullaly

• Publisher: Routledge
• Year: 2016
• eBook ISBN: 9781317138099
• Topic: Business
• Subtopic: Business General
• Index: Business

Exercising Agency is a book about decision making. In particular, it looks in detail at how a very important type of organizational decision gets made: whether or not to initiate a project. Making strategic decisions of this kind can never be a wholly rational and scientific process. And Exercising Agency lifts the lid on many of the important behavioural factors that inform project decisions: power and politics, personality, the 'rules' of an organization. Mark Mullaly draws on his research to provide practical guidance for decision makers; project shapers, approving executives and those responsible for how initiation decisions are made. By explaining the influence, value and risks associated with the elements that inform the way we make strategic decisions he will help you identify how individuals and organizations can best support the process to ensure project initiation decisions are effective and most closely underpin the priorities of the organization. If you are involved in framing or making decisions about the future of your organization; the projects that you do or don't decide to initiate, then read this book. It won't make the decisions any easier but it will help you improve the quality of the decisions you make and over time, the effectiveness of your organizational decision making.

---

Chapter 1
Introduction

---

Understanding Initiation

This is a book about decision making. In particular, it looks in detail at how a very important type of organizational decision gets made: whether or not to initiate a project. Projects are about how strategy is implemented. Projects are the means by which change is initiated. Projects are about how new capabilities get created. Understanding how projects are initiated is therefore extremely useful. Interestingly, it is not something that has been comprehensively explored. Project initiation decisions live in a middle space between strategy development on one hand and project management on the other. They represent the critical transition point of when strategy begins to gain traction, and projects that respond to strategy begin to move forward.

This is also a book about how politics, process and personal influence combine to influence the conduct of project initiation. These influences are not always obvious, and they do not necessarily work in the way that many of us might predict. While many organizations profess to employ robust processes and rational analysis, the reality is that a significant number do not. Politics can play a strong role in influencing project initiation outcomes, and at the same time politics can also derail effective decision making. Individual actors in organizations also play a role. Depending upon the actor, that role might augment other influences, it might confound them, and in some instances it is the only way in which things actually get done.

The up-front process of initiation has been identified as having a dominant influence in determining the success or failure of individual project efforts. This book focuses on the process of project initiation: it seeks to explore the rule systems that influence how projects are initiated, the roles that are involved in project initiation, and how individual actors perceive and approach their roles. The book offers a substantive theory of how agency and rule emphasis influence the effectiveness of project initiation decisions. Based upon these insights, it offers practical guidance for those who are involved in shaping projects, the executives who approve them and the organizations that endeavour to improve how initiation decisions are made.

The Need to Make Initiation Decisions

An important theme that has emerged in connecting the initiation of specific projects with the larger purpose of the organization is the assertion that projects are a vehicle for delivering organizational strategy. This is often phrased as the need to ensure that the right projects are being done in the right way (Artto & Wikström, 2005; Aubry et al., 2012; Cooper et al., 2000; Crawford et al., 2006a). In part, linking projects with strategy responds to a recognition that for projects to proceed, they should in some way be responsive to the objectives and goals of the organization. If we accept that projects are being undertaken for a purpose, then the linkage between the project and the organization’s strategy needs to be more broadly understood (Morris et al., 2006). This notion is also supported by the fact that the study of project failure shows that the causes of failure are more often strategic rather than technical, and are therefore likely the product of political processes within the organization (Cicmil & Hodgson, 2006). Linking projects with strategy not only connects projects with a sense of organizational purpose, but also firmly grounds project initiation within the political environment of the organization.

There is a fundamental question, however, of how this is done. There are many arguments about how the strategic alignment of projects should work. There are suggestions that project and organizational strategy need and should have two-way alignment processes that integrate views of policy, strategy and capability development (Maylor, 2001; Milosevic & Srivannaboon, 2006). Other studies go so far as to explicitly claim that projects are a means of implementing organizational strategy, even while recognizing that project management itself is not viewed from a strategic perspective (Aubry et al., 2012); this reinforces observations by Thomas et al. (2002) on the challenges of selling project management as a strategic capability to senior executives. When projects and project management do attain organizational focus, it tends to be more in response to crises than naturally perceived alignments; senior executives often fail to see the connection between project management and the goals of the organization (Thomas et al., 2002). The risk is that while projects are philosophically presented as proactive means of delivering strategy, they are more often practically seen as means of reacting to tactical and operational crises.

One means of attempting to reconcile these perspectives has been through the exploration of project governance. The concept of project governance largely addresses the oversight of projects, rather than being involved with the actual delivery. The role of governance in a project context includes choosing the right projects and establishing the correct objectives in response to organizational priorities and strategies; ensuring the appropriate allocation of resources; establishing appropriate strategies for reporting, and ensuring the projects and their results are sustainable (Morris et al., 2006; Williams et al., 2010). While these processes are again often expressed as being rational and normative in nature, there is also a strong political dimension that needs to be understood and investigated further (Flyvbjerg et al., 2003). As an example, one study found within one of its case studies that even where rational decision making was applied, the final decision was always a political one (Williams et al., 2010). This again reinforces the need to understand the political forces by which projects are initiated.

Exploration of the political and power dynamics that underlie purported rational approaches to governance include in particular their use as a source for legitimization of the project, or as a tool for reassurance of project owners (Williams et al., 2010). This notion is reinforced by observations that legitimization can be seen as a key focus for project management as a whole (Cicmil et al., 2009; Thomas, 1998), providing a façade for rationalism, power, efficiency and control. One of the rationales for project governance is that it serves to ensure that projects do not fail; it must ‘prevent their birth, weeding out those projects that do not adequately address strategic aims, and destroying the seeds of failure before they can germinate’ (Smith & Winter, 2010, p. 48). This is achieved through the introduction of stage-gating or gatekeeping mentalities. In many instances, however, these frameworks are seen more as boundary systems between executives and staff that enable executives to stay distant from the actual work of implementation (Artto et al., 2011). The danger, then, is that rational approaches are used to legitimize or justify project decisions without addressing the underlying complexities and political influences that are actually present, and idealized views of what should happen supplant an understanding of what actually occurs.

What We Already Know about Decision Making

While many in the field of project management have identified the need to better understand decision making within projects, and particularly how initiation decisions are made, there is a lot that we already know and understand about decision making. Decision making and project management actually share a common history, with both practices evolving out of the field of operations research in the years following the Second World War (Simon, 1965; Simon, 1987). Despite this common history, these fields have remained largely separate and distinct, with little overlap of research interest or subject matter content. Decision making has none the less developed into a rich field with much to offer in terms of insight and perspective on project initiation decisions.

Decision making in an economic and strategic context goes back more than 250 years, to a paper by Bernoulli to the 1738 proceedings of the Royal Academy of Science in St Petersburg. In a corporate and managerial context, the initial principles of decision making can be found in the work of Fayol (1949), who defined managerial activities as including planning, organization, command, co-ordination and control. The first direct definition of decision making, however, is found in the works of Barnard, who identified the role of decision making as one of the chief functions of the executive (Barnard, 1938), and made explicit the idea of decision as the delineation of ends – the objective to be realized – and means – the methods to be employed – that is the essence of much subsequent exploration of decision making. In particular, Barnard distinguished between the principles of decision making by the individual and those made on behalf of, or in the interests of, the organization. In a discussion of the decision making environment, Barnard states: ‘within organizations, especially of complex types, there is a technique of decision, an organizational process of thinking, which may not be analogous to that of the individual’ (139).

The principles of rational decision making in an economic context have their foundation in the work of von Neumann and Morgenstern (1944), who first advanced the notion of economic utility as a means of objectively measuring and quantifying the value of personal preferences. Fundamental to the models of rational decision making and choice is the idea of ‘economic man’ who, in being economic, is also ‘rational’. Faced with an array of different, specified options, each option of which has different consequences attached to it, ‘economic man’ has a system of preferences against which the consequences of each option are evaluated, from which the option with the highest expected value is selected (Cyert et al., 1956; Simon, 1955). Rational decision making has faced extensive criticism (admittedly, largely from advocates for other models of decision making, and particularly behaviouralist models). Chief among these criticisms is the contention that rational models of decision making do not reflect how decisions are actually made (Simon, 1955; Simon, 1959). In particular, advocates of behavioural decision making argue that rational approaches ignore the fact that decision makers possess modest calculation powers, and that a normative theory – if it is to be useful – should only call for information that can reasonably be obtained and calculations that can actually be performed (Simon, 1965). In fact, research showed attempts at rationality lead to less effective decisions; empirical studies found that the comprehensiveness of analysis called for in rational models had a consistently negative relationship with performance (Fredrickson, 1984; Fredrickson & Mitchell, 1984). The implication is that decision making is complex, difficult, subjective and inconsistent, and that accurate understanding of decision making requires explicitly embracing the psychological complexity and strategies for simplification that underlie how decisions are actually made by individual actors in real world situations.

Behavioural decision making models emerged as a reaction to the rational models that had previously dominated perspectives of decision making. A fundamental consideration in the development of behavioural models was that they were based upon the capacity and limitations of human perception, and that perceived reality was vastly different from the ‘real’ world (Simon, 1959; Simon, 1965). Unlike the idealistic presumptions of perfect data and comprehensive analysis associated with rational techniques, behavioural decision making approaches are rooted in the limitations and constraints faced by actors constrained by limited cognitive capacity. One of the early explorations of behavioural decision making principles was the landmark book Administrative Behavior by Herbert A. Simon (Simon, 1947/1997), who argued that the limits of knowledge regarding means and consequences meant that rationality was at best approximate. This laid the foundation for the later work of March & Simon (1993), who fully developed the concept of ‘bounded rationality’. Key concepts that are identified in bounded rationality are ‘satisficing’ – where decision makers accept a ‘good enough’ decision rather than seeking the best decision – and ‘sequential search’ – which recognizes that decision makers view available options one at a time, rather than comprehensively reviewing all options. Bounded rationality acknowledges that decision making is based upon incomplete information about alternatives and their consequences, and that information is not innocent: it is the product of different coalitions in organizations pursuing differing objectives (March, 1987). Unlike the idealistic presumptions of perfect data and comprehensive analysis associated with rational techniques, behavioural decision making approaches are rooted in the limitations faced by actors constrained by limited cognitive capacity.

While bounded rationality is designed to address the physical and practical limitations associated with decision making, the development of understanding of cognitive biases, heuristics and frames has been another significant area of development within behavioural theories of decision making. Historically, most of the literature dealing with risky choice assumed a decision maker who was risk-averse, an assumption that once again did not align with observed behaviour. This led to the development of cognitive decision models, which were particularly developed through the work of Kahneman & Tversky (1979) and their ground-breaking framing of ‘prospect theory’. Prospect theory endeavours to provide a cognitively realistic view of how individual actors approach decision making when faced with possible gains and losses, and the fact that each of these appear to result in preferences for different strategies. Prospect theory consciously breaks the decision making process into two distinct stages: editing and evaluation. Editing is the process of choosing what inputs into the decision making process will be used, while evaluation reflects the actual selection based upon the edited prospects. Prospect theory also involves processes of simplification, where preferences and outcomes are rounded rather than retaining their initial precision, and where extremely unlikely prospects tend to be eliminated. It explicitly allows for the theory of bounded rationality, recognizing the inherent challenges of making risk-based or ambiguous judgements (Kahneman, 2003) while embracing many of the principles of cognitive bias that have been observed, but otherwise not explained. The implication is that cognitive decision models in general, and prospect theory in particular, provide a complementary perspective to other views of behavioural decision making.

Many of the behavioural decision making theories that have been developed thus far reflect a decision maker trying to make the best decisions possible (in other words, to maximize the results of the decision making process) in the face of limitations of information, knowledge, cognition and calculative capacity. A number of alternative decision making models have also emerged which consciously rejected the underlying assumptions of both rational and behavioural decision making, and which can perhaps best, or at least generously, be described as ‘anarchic’. The best-known of these is the ‘garbage can model’ (Cohen et al., 1972), which was influenced in part by the experiences of March following his assumption of the position of dean of a university business school. The major feature of the garbage can model is the uncoupling of problems and choices, and throwing whatever else happens to be around at the time into a can to see what sticks to what. The garbage can model radically expanded on the assertion of Cyert and March (1992) that organizations do not have fully consistent goals; it developed the notion of ‘loose coupling’ among problems, participants, solutions and decisions (Gavetti et al., 2007). A temporal theory of decision making, the garbage can model deliberately rejected the ends–means model that had guided much of decision making.

The mix of garbage in a single can depends on the mix of cans available, on the labels attached to the alternative cans, on what garbage is currently being produced, and on the speed with which garbage is collected and removed from the scene. (Cohen et al., 1972, p. 2)

Problems were ‘resolved’ when any particular combination of problem, solution and decision maker interacted with each other in a decision making environment where there was a sufficient level of effort to get something done.

The garbage can model was viewed by some as going too far in its rejection of the essential features of decision making behaviour. While critics recognized that organizations do create problems, successes, threats and opportunities as a justification for their actions, they felt that there was a need to retreat from the full anarchy proposed by the garbage can model: ‘This backtracking occurs because the garbage can model understates cause-effect attributions, de-emphasizes the activities preceding decisions, and ignores the activities following decisions’ (Starbuck, 1983, p. 91). The anarchic models feel unfamiliar to some, in that they deliberately break the construct of means–ends that has been the hallmark of traditional perspectives since the earliest rational decision making models; what they do provide, however, is other insights into the dynamics of decision making encountered within the structural realities and limitations of organizations. Most importantly, they provide alternative perspectives for how decision making processes may be perceived, and how actors may view the dynamics underlying the making of decisions.

Considering the decision process associated with project initiation, none of the models discussed above fully offers a relevant framework. The process of initiation for any complex project clearly cannot be considered to be rational; too much is unknown and uncertain about both options and consequences. While the behavioural models consciously reflect the cognitive and capacity limitations inherent in project initiation choices, they do not provide contextual guidance as to how a decision maker would prefer one project over another. The anarchic model, while perhaps appealing in its description of decision making as a random intersection of problems, choices, decision makers and opportunities, removes the means–end focus that is still in part a consideration of evaluating projects. A sensible compromise within the literature as articulated would appear to be in part offered through an understanding of the principles of rule following originally articulated by Cyert & March (1992); rule following would appear to offer a middle ground between a purely means–end-based presumption of how individual decisions are made and a broader contextual understanding of the forces that influence decision making in organizational contexts.

The discussion of organizational routines, or decision making as rule following, presents a modification to behavioural decision making models. This concept first emerged in A Behavioral Theory of the Firm by Cyert & March (1992). The central principle of rule-following behaviour is that, in addition to the universally bounded nature of rationality, ‘behaviours get programmed through spontaneous habits, professional norms, education, training, precedents, traditions, and rituals as well as through formalized procedures’ (Starbuck, 1983, p. 93). Rule following in decision making emerged from the introduction of the principles of evolutionary theory to sociology. The essential premise was that a firm operates according to a set of decision rules that link a range of environmental stimuli to a range of responses on the part of firms (Nelson & Winter, 1974). The assumption that firms have decision rules, and that these are in turn retained or replaced through satisficing, provides a basis for both stability and ongoing evolution (Winter, 1971). Rule following therefore respects and reinforces the traditions of behavioural models, while providing a larger contextual appreciation of the influences of the organization on how decisions are ultimately made.

The development and use of rules in decision making draws from principles of bounded rationality. Decision making can be costly, and reliance upon simple rules to guide decision making is a form of cost minimization; it results in economies in terms of information collection, computation and communication, and provides frameworks in which actors throughout the organization are able to perform their roles with greater confidence and certainty (Winter, 1971). Burns & Dietz (1992) defined sets of rules as representing ‘institutions’: these entities collectively defined the settings or context of interaction, the actors who might take part and the rules for behaviour of roles within that context or setting. An important consideration in the understanding of rules and rule following is their application to the concept of ‘agency’. Agency tends to assume that social actors have limited room for decision making, autonomy or creativity; actors are ‘programmed’ by the culture, and their ability to operate is limited by these constraints (Burns & D...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Contents
• List of Figures
• List of Tables
• Acknowledgements
• Reviews of Exercising Agency
• 1. Introduction
• 2. Exploring How Projects Do (and Don’t) Get Initiated
• 3. The Influences on Project Initiation
• 4. When Process Drives Choices
• 5. When Politics Drives Choices
• 6. When Individuals Drive Choices
• 7. When Choices Go Off the Rails
• 8. Shaping Better Project Results
• 9. Making Better Project Decisions
• 10. Improving the Project Initiation Process
• Bibliography
• Index