Forth International Seville Conference on Future-Oriented Technology Analysis (FTA)
FTA and Grand Societal Challenges – Shaping and Driving Structural and Systemic Transformations
Seville, 12-13May 2011
Uncertainty, foresight, and strategic decision making: evidence from leading companies
Riccardo Vecchiato Claudio Roveda
Politecnico di Milano Politecnico di Milano
P.zza Leonardo da Vinci, 32 P.zza Leonardo da Vinci, 32
20133 Milano, Italy 20133 Milano, Italy
Ph.: +39 02 23993994 Ph.: +39 02 23993994
Keywords:Environmental uncertainty; decision making; foresight; planning; learning.
ABSTRACT
The external environment is a major source of uncertainty for strategic decision makers in charge of sustaining the advantage of the firm over time. Specific practices and tools (foresight techniques) have been developed for coping with uncertainty and enabling strategic planning in the face of major changes. However, they had uneven success. Scholars failed to clearly define their value added and to provide empirical evidence of their contribution to sustain the advantage of the firm. In this context, criticisms have addressed the theoretical foundations themselves of strategic planning and the tools and practices that are meant to cope with uncertainty, by pointing to the impossibility of making reliable predictions. Foresight scholars and practitioners generally respond to concerns about the reliability of future-oriented techniques by arguing that their role is not so much to predict the future but to prepare for the future, through a learning process that enhances the capability to handle new opportunities and threats. Still, the following questions remain largely unexplored: what different kinds of uncertainty (and drivers of change) may be faced in the business environment? What foresight processes and techniques might be used to address them? These are precisely the main questions we deal with in this paper. Their relevance is extremely high because deciding what firms should do in uncertain situations is a key issue in the literature on strategic management. In particular, our research questions are at the heart of the research streams that recently attempted to bridge the gap between planning and adaptive approaches to strategic decision making, by encouraging firms to carefully plan in order to quickly adapt.
In order to explore the relationships between environmental uncertainty, strategic planning, and the competitive advantage of the firm we performed multiple case studies of corporate organizations that recently faced major changes in their external environment and increasing turbulence. First, we expand our understanding of environmental uncertainty by defining the concept of ‘boundary uncertainty,’ which regards the core identity of the components of the business (micro) environment. We thus distinguish between ‘continuous’ and ‘discontinuous’ drivers of change. The former bring about ‘state’ uncertainty about their evolution, ‘effect’ uncertainty about their consequences on the organization, and ‘response’ uncertainty about the viable responses. The latter bring about boundary uncertainty as well regarding the identity of the main players of the industry and the activities of the value chain. We thus advance a conceptual framework of foresight efforts and practices as a contingent methodological and organizational approach.
1. Introduction
The strategic management literature (Hofer and Schendel, 1978; Miles and Snow, 1978; Teece, 2007) and the organization theory literature (Dill, 1958; Thompson, 1967) have long emphasized the role of the environment as a major source of uncertainty for strategic decision makers in charge of coping with emerging opportunities and threats.
A broad range of heuristic approaches for carrying out future studies have been developed in corporate organizations with the aim of promptly selecting drivers of change in the company’s outside environment (environmental scanning: see Hambrick, 1982; Day and Schoemaker, 2006) and of investigating their likely evolution and impact on the organization (future-oriented techniques: see Porter et al., 2004). Today the term ‘foresight’ is widely used to designate the activities and processes that assist decision makers in the task of charting the company’s future course of action by encompassing either scanning or future-oriented techniques (Bradley Mackey and Costanzo, 2009; Grupp and Linstone, 1999;Tsoukas and Shepherd, 2004; see Figure 1).
Figure 1: Foresight and strategic decision making
So far, foresight has had uneven success and popularity. On one hand, scholars have documented that in the last two decades many large firms in such diverse sectors as energy, automotive, telecommunications, and information technology were regularly applying future-oriented techniques (Becker, 2002; Coates, 2001). The wide interest in foresight seems to be confirmed by the growing number of consulting companies and networks in the field.[1] On the other hand, scholars failed to clearly define the added value of foresight and to provide empirical evidence of its contribution to sustain the advantage of the firm. Little research has been conducted regarding the performance of the different organizational and methodological frameworks that were adopted in corporate organizations by omitting, in particular, the ex post evaluation of the reliability and accuracy of the output of foresight. The most relevant example concerning the impact of foresight on the success of the organization still remains the case of Shell Scenarios and its anticipation of the forthcoming 1973 oil crisis (Wack, 1985; van der Heijden, 1996). In this context, some skepticism arose in the academic community regarding the soundness and appropriateness of foresight for making predictions about the future and, thus, for supporting predictive strategies (Bradley Mackey and Costanzo, 2009; Grant, 2003;Wiltbank et al., 2006). The major evidence of this skepticism may be the fact that today it is not specifically addressed by most MBA curricula[2] and so far only a limited number of validated analyses of future studies have been hosted by leading academic journals.
Supporting scholars and practitioners generally respond to concern about the reliability of foresight by arguing that its role in corporate organizations is not so much to predict the future, but to prepare the firm for the future through a learning process (Tsoukas and Shepherd, 2004; van der Heyden et al., 2002). Still, the following questions remain largely unexplored: what different kinds of uncertainty (drivers of change) may be faced in the business environment? how should a firm design the right foresight approach, i.e. what type of techniques and analysis should be conducted depending on the kind of environmental uncertainty?
These are precisely the main questions we deal with in this paper.Their relevance is extremely high because deciding what firms should do next in uncertain situations is a key issue in the literature on strategic management (Wiltbank et al., 2006), and at the center of the debate between the planning and the learning schools (Ansoff, 1979; 1991; Mintzberg, 1990; 1994).
In order to explore the relationships between environmental uncertainty, foresight and strategic decision making we performed multiple case studies of corporate organizations (Vecchiato and Roveda, 2010a; Vecchiato and Roveda, 2010b). In this paper we focus on BASF in the chemical industry, Daimler in the automotive industry, Philips in the consumer electronics industry, and Siemens in the information and communication industry.
First, we expand our understanding of environmental uncertainty by defining the concept of ‘0-Level uncertainty’, which regards the core identity of the components of the business (micro) environment. We then distinguish between ‘continuous’ and ‘discontinuous’ drivers of change, and shed light on their implications for foresight and strategic decision making.
2.Literature Review
2.1 Conceptualization of environmental uncertainty
One feature of uncertainty is the inability of managers to make accurate predictions (Milliken, 1987). Early conceptualizations of uncertainty go back to pioneering management scholars such as Knight (1921) and March and Simon (1958), who argued that the business environment of the firm is inherently unstable and this instability creates uncertainty for rationally bounded managers who are not able to fully collect, process, and comprehend information about changes and new events. More specifically, ‘environmental’ uncertainty arises when managers lack accurate information about organizations, activities, and events in their external environment; namely, when they are not confident that they can predict what the major changes are or will be (Duncan, 1972; Lawrence and Lorsch, 1967). Milliken (1987) distinguished between three types of uncertainty that act together to determine the overall uncertainty faced by strategic decision makers. The first is ‘state’ uncertainty and refers to the inability to understand how components of the environment might change (e.g., in the case of the automotive industry, the driver of change of ecological concern by public policy makers in Europe: will new laws be enacted in the next 10 years? If so, how strict these will be?). The second is ‘effect’ uncertainty and refers to managers’ inability to predict what the consequences of drivers of change will be on their organizations (e.g., will customers switch from a traditional product—fuel-based car—to an innovative one—hybrid car; and if so, how should car-makers redefine their product portfolio?). Finally, ‘response’ uncertainty is the inability to formulate response options and/or to predict the consequences of a response choice (e.g., how should we acquire the resources we need to develop environmentally friendly products?). Courtney (2001) further expands the concept of ‘state’ uncertainty by identifying four levels of uncertainty.
Concerning the classification of the different components of the external environment that bring about uncertainty, Dill (1958) made the distinction between ‘general’ and ‘task’ environment. The latter one is made up of elements and sectors with which the firms has direct contact and that directly affect business strategy, day-to-day operations, and goal attainment. According to the organization theory, the task environment was initially defined to include the sectors of competitors, suppliers, customers, and regulatory bodies. The strategic management theory expanded the concept of task environment by defining the broader concept of business microenvironment, which identifies the key forces (sectors) that govern competition in an industry. These forces are competitors, customers, suppliers, potential incomers and substitute products (Porter, 1980), and providers of complementary products (Porter, 1985).
The general environment refers, instead, to sectors that affect firms indirectly; these are the political, economic, ecological, societal, and technological landscapes that surround the business microenvironment and are today commonly referred to as the business macroenvironment.
2.2Foresight practices and techniques
Foresight encompasses two main tasks: the first regards environmental scanning’ and the detection of new events and drivers of change.[3] The second one regards the future-oriented techniques and practices for investigating drivers of change in relation to their likely evolution (state uncertainty), their consequences on the organization (effect uncertainty), and the most suitable responses (response uncertainty).Roadmaps[4] and scenarios[5] are by far the most popular future-oriented techniques (Cuhls and Johnston, 2008);but there are many others.For example, Delphi, relevance trees, trend-impact analysis, cross-impact analysis, simulation modeling and systems dynamics, and game theory (Glenn1999; Porter et al., 2004).
Foresight practitioners and scholars generally agree that the role of foresight and its value added in handling uncertainty is not so much to predict the future, but to prepare the organization for dealing with it by activating a learning process about the future (Tsoukas and Shepherd, 2004; Bradley Mackey and Costanzo, 2009). In any case, literature on strategy and foresight fails to provide a sound explanation of the ways this learning process takes place or even empirical evidence about it. So far, the most relevant description of the learning process entailed by foresight concerns scenarios and is based on the concept of ‘memory of the future’ (Ingvar, 1985). According to Ingvar (1985), human brains are constantly occupied with making up plans and programs for the future, which the author defines as alternative time paths. The set of these time paths makes up the memory of the future. Human brains then store these various time paths: the more time paths are stored, the more individuals are able to recognize and make sense of changes in their external environment (van der Hejden et al., 2002). As Arie de Geus[6] illustrates:
You personally and your company are being bombarded by an overload of signals from the outside world. The hypothesis of Ingvar is that the function of the memory of the future is to allow the brain to select those signals that are relevant for you. The test of relevance is your memory of the future. If a signal comes in, it passes through this memory of the future. If it finds a store in an alternative time path, meaning that it is relevant for you, then the signal is translated into data: it becomes information and then information becomes understanding.
De Geus[7] thereby argues that the value of scenarios lies in adding to the memory of the future of the managers, by enabling them to visit and to experience different time paths ahead of time:
If you have only one possible alternative path into the future, you see—or hear—very little. This is the real importance of scenario planning. It stretched the time horizon from one or two years, to ten or twenty years. And paradoxically, while increasing the time horizon, at the same time, in the present, it increases the power of perception. You hear more signals that are relevant to you.
However, this explanation still entails that, in order to prepare for the futuresuccessfully, foresight has to bring new information and knowledge about the future, even if in the form of some alternative scenarios (time paths): as long as one of these scenarios turns out to be the right one, the foresight learning process allows managers to recognize the real future as it begins to emerge. But under what conditions can firms achieve useful information about the future? Putting id differently, what kind of foresight techniques and approaches should a firm adopt in different business environment and thus under different conditions of uncertainty?
The main goal of this paper is to link the research stream on environmental uncertainty with that on foresight by setting the groundwork for a more rigorous analysis of the role and added value of strategic foresight.
3. Methods and data
The research design is based on in-depth, inductive, and multiple case studies of selected firms that long ago started to be systematically engaged in foresight. These firms are BASF, Daimler, Philips, and Siemens[8]. Given the inadequate analysis in the literature and the open-ended nature of our questions, we felt that this methodological approach would be the most useful for theory building (Eisenhardt and Graebner, 2007; Yin, 2003). Table 1 provides an overview of our empirical setting.
Table 1: Overview of case studies
Firm / Business / Foresight activities startedPhilips / Consumer Electronics / Early 1990s
BASF / Chemicals / Mid 1990s
Daimler / Automotive / Late 1970s
Siemens / Consumer Products, ICT / Mid 1990s
The unit of analysis was twofold. On one hand, we examined the historical evolution of each industry where these firms compete since they started engaging in foresight, and in particular throughout the 2000s. On the other hand, we analyzed their foresight activities in relation to their impact on strategic decision making, the management of significant drivers of change, and the long-term results of foresight-based decisions.
Data were collected through the combination of various sources and throughout an iterative process. Firstly, we collected publicly available data on the industry and the selected firms, historical annual reports including, financial analysts’ reports, conference presentations by top managers,and articles and prior studies in the business press and scientific journals. Secondly, company archives such as internal memos and technical papers supplemented publicly available data. Thirdly, we interviewed a sample of senior and mid-level managers, in particular the heads of the organizational unit in charge of foresight. We also interviewed external consultants who were involved in the foresight process. Finally, we interviewed a sample of leading experts from academia and industry who had extensive knowledge of each company and their industry. Overall, we conducted over 45 personal interviews that were semi-structured, and lasted from one hour to half a day (a number of these interviews are quoted in the following).
Data analysis was highly iterative and used traditional approaches for inductive research (Eisenhardt, 1989; Yin, 2003). Analysis began with detailed written accounts and schematic representations of the historical evolution of each industry where the selected firms operate foresight approachthey developed. After constructing the case histories, we conducted within-case analysis, which was the basis for developing early constructs and hypotheses. Cross-case analysis and theory triangulation with different bodies of literature on environmental uncertainty, foresight, and strategic management produced our conceptual framework.
4. Results
Our data suggest that companies facing different kinds of drivers of change and thus different conditions of uncertainty in their business environments adopt different strategic foresight approaches to identify such drivers and investigate their long term evolution (state uncertainty), consequences on the organization (effect uncertainty), appropriate responses (response uncertainty).
4.1 Foresight at Philips and Daimler
Environmental uncertainty and methodological approach. The chemical and automotive industries (i.e. BASF and Daimler business) throughout the last decade were typically mature and global industries where trajectories of technologies and customer needs are well-established and companies compete for market share at the international level: the situation of the boundaries between the micro and macro environments were blurred in these industries; the huge number of drivers of change in their PEEST landscapes, their strong mutual influences and the slow overall pace of evolution have contributed to high complexity.
The structure of the chemical industry resulted from an as yet uncompleted consolidation process, and also from the rise of new competitors in Asia, Middle East and Eastern Europe that entered the market in the last decade by producing reliable, good-quality commodities at low cost. Since the 1990s the demand of chemicals has been characterized by low growth and considerable cyclicality (which is likely to increase in the near future). Capacity cannot be adjusted easily, so there is a constant danger of over capacity. The industry has also been increasingly exposed to rising raw materialprices, steep rises in energy costs, growing ecological concerns and stricter environmental rules, while, at the same time, the rapid development of ICT tools made the market far more transparent, and increased the pressure to optimise commodity production.