Managerial perception and assessment of catastrophic supply chain risks

TABLE OF CONTENTS

Chapter 1 - Introduction
1.1. Research motivation
1.2. Research goals
1.3. Academic and managerial contribution
1.4. Thesis outline

Chapter 2 - Literature review
2.1. Understanding risk
2.2. Supply chain risk management
2.3. Classification of SCRM research
2.3.1. Categorization of supply chain risk sources
2.3.2. The elements of the SCRM process
2.4. Catastrophic supply chain risks
2.5. A conceptual framework for supply chain risk analysis
2.5.1. The risk assessment stage
2.5.2. Managerial risk perception

Chapter 3 - Theoretical model
3.1. Risky decision making model
3.1.1. Supply chain risk conditions
3.1.2. Probability and magnitude of potential catastrophic supply chain events
3.1.3. Overall perception of catastrophic supply chain risks
3.1.4. Assessment of catastrophic supply chain risks
3.2. Research propositions

Chapter 4 - Methodology
4.1. Multiple case study approach
4.2 Literature review as a starting point
4.3. Data collection
4.4. Data analysis
4.5. Shaping prepositions and enfolding literature
4.6. Achieving case study rigor

Chapter 5 - Findings
5.1. Within-case results
5.1.1. Case Company A
5.1.2. Case Company B
5.1.3. Case Company C
5.1.4. Case Company D
5.1.5. Case Company E
5.1.5. Case Company F
5.1.6. Case Company G
5.2. Synopsis of findings - a cross-case display
5.2.1. Supply chain risk characteristics
5.2.2. Frequency and magnitude of past catastrophic supply chain events
5.3.3. Managerial perception of catastrophic supply chain risks
5.3.4. Assessment of catastrophic supply chain risks: procedures & barriers

Chapter 6 - Discussion
6.1. Supply chain risk levels
6.2. Formative elements of managerial risk perception
6.3. Effects of managerial risk perception
6.4. Barriers to a formal assessment of catastrophic supply chain risks
6.5. Methods and tools for the assessment of catastrophic supply chain risks

Chapter 7 - Conclusion
7.1. Summary of findings
7.2. Theoretical and managerial implications
7.3. Limitations and suggestions for future research

References

Appendix

Appendix A: Case study research protocol

Appendix B: Interview protocol

LIST OF FIGURES AND TABLES

List of tables

Table 1: Overview literature review (1/5)

Table 2: Overview literature review (2/5)

Table 3: Overview literature review (3/5)

Table 4: Overview literature review (4/5)

Table 5: Overview literature review (5/5)

Table 6: Overview of findings

Table 7: Findings: market & technological turbulence

Table 8: Findings: supply chain complexity

Table 9: Findings: dependence and riskiness of sourcing strategy (1)

Table 10: Findings: dependence and riskiness of sourcing strategy (2)

Table 11: Findings: frequency and magnitude of catastrophic supply chain events

Table 12: Findings: Overall perception of catastrophic supply chain risks

Table 13: Findings: assessment and obstacles to the assessment of catastrophic supply chain risks

List of figures

Figure 1: Classification of supply chain risks. Source: adapted from Wu, et al. (2006)

Figure 2: Risk categorization scheme. Source: Adapted from Knemeyer, et al. (2009)

Figure 3: Conceptual framework for supply chain risk analysis. Source: adapted from Cohen and Kunreuther (2007)

Figure 4: Model of managerial perception and assessment of catastrophic supply chain risks

CHAPTER 1 - INTRODUCTION

The field of supply chain risk management (SCRM) has received a lot of attention in the recent years by academia as well as by practitioners. Modern supply chains are more and more vulner- able to catastrophic events, such as natural disasters or major accidents. Not only are supply chains more exposed to those events but they are also operated with less “slack” (Hendricks & Singhal, 2005; Knemeyer, Zinn, & Eroglu, 2009). This trend, to create leaner and more cost effi- cient supply chains, comes with the price of increased risk levels (Norrman & Jansson, 2004; Wagner & Bode, 2006).

Catastrophic events in supply chains are characterized as events that occur with low probability, but have severe consequences (Knemeyer, et al., 2009). Failing to plan for low probability - high consequence (LP-HC) risks can have disastrous consequences. In 2001 a fire at one of Ericsson’s key suppliers stopped the whole mobile phone production and caused lost revenues of about $400 million (Norrman & Jansson, 2004). September 11 attacks and the resulting delays in crossborder shipments forced Ford to close five plants for several days (Martha & Subbakrishna, 2002). Another, very recent example is the Volcano eruption in Iceland in 2010, which seriously affected supply chains of various industries.

1.1. Research motivation

Previous research primarily focused on operational supply chain risk. Vanany, Zailani, and Puja- wan (2008) for example conclude, after an extensive literature review of existing SCRM litera- ture, that the catastrophic risk type has found relatively less attention in comparison with other risk types, i.e. operational risks. The previous introduced examples show how severe conse- quences of an underestimation of this risk type can be. Many firms assess risks with low impact that frequently occur, but tend to neglect the LP-HC risks (Chopra & Sodhi, 2004). Therefore, this study wants to contribute to SCRM research by exploring the risk type catastrophic supply chain events.

Literature typically offers countermeasures that companies may adopt to prepare and respond to catastrophic events, but lacks an implementable process to assess the risk (Sheffi, 2001). Also Zsidisin, Ellram, Carter, and Cavinato (2004) notice that few studies focused on “the key con- structs necessary for assessing supply [chain] risk”. In order to contribute to the previously less emphasized assessment stage of the SCRM process, which was also suggested by Jüttner, Peck, and Christopher (2003) to be an area for future research, the assessment stage is the focus of this study.

Existing literature provides evidence that risk perceptions play an important role in the context of managing supply chain risks. Due to the special characteristics of LP-HC risks, this role might be of even higher importance in the assessment phase of catastrophic supply chain risks. Vana- ny, et al. (2009) and Zsidisin (2003b) explicitly mention the managerial perception of risk in a supply chain context as a field for future research. By addressing this issue and the before dis- cussed ones, the study attempts to make a useful contribution to the SCRM literature by follow- ing this call for future research.

1.2. Research goals

As previously motivated, supply chain managers and researchers have focused predominantly on operational risks in a supply chain context. While there have recently been efforts to explore the risk type catastrophic in a supply chain context, for example by Knemeyer, et al. (2009), more explorative research is necessary.

Due to the lack of previous quantitative research findings, this study takes an explorative case study approach to shed more light on the managerial perception and assessment techniques of catastrophic supply chain risks. The goal of this research is twofold. It shall explore the construct of managerial risk perception in the context of catastrophic supply chain risks and shall also identify tools and methods applicable for the assessment of those risks. To achieve this goal, a set of research questions has been formulated.

Research question 1:

Which elements play a formative role to the managerial perception of catastrophic supply chain risks? How are those risks perceived by practitioners?

Research question 2:

Which methods and tools are applicable for the assessment of catastrophic risks in a supply chain context?

Research question 3:

How are managers’ perceptions of catastrophic supply chain risks linked to the decision to assess those risks?

Research question 4:

What are barriers to the assessment of catastrophic supply chain risks?

1.3. Academic and managerial contribution

This thesis attempts to contribute to the SCRM field by further exploring the risk type catastrophic events in the supply chain. It shall identify useful assessment tools by reviewing existing literature of different fields, i.e. catastrophe research and SCRM research. Furthermore, through empirical research, the construct of risk perception was explored. The thesis gives further evidence that catastrophic supply chain risks cannot be treated analogously to operational supply chain risks, happening more frequently with lower impacts. The study will be part of an hopefully evolving research on non-operational supply chain risks.

Also from a practitioners’ view, this paper makes some important contributions. In general, the awareness among supply chain managers of the risk type catastrophic could be increased. It al- so may provide some evidence of the potential negative consequences of ever higher integrated and cost effective supply chains. For executives that attempt to include the assessment of catastrophic risks in their SCRM process, the identified tools and techniques could serve as a starting point for doing so.

1.4. Thesis outline

To answer the introduced research questions, a literature review was performed as a first step. It covers SCRM research as well as catastrophe research. To structure the empirical research in form of multiple case studies, a theoretical model based on Yates and Stone (1992) was devel- oped and research propositions were derived. Chapter 4 describes in a more detailed way the applied research methodology in terms of research construct and applied instruments. After this, the findings of the empirical research are presented in intra-case and cross-case format. The findings are discussed in the light of existing literature in Chapter 5. Finally in the last sec- tion, the thesis closes with a summary of results as well as some implications and limitations pointing out to future research.

CHAPTER 2 - LITERATURE REVIEW

2.1. Understanding risk

The concept of risk is dealt with in various disciplines, such as decision theory, insurance, finance, marketing, management or psychology. These disciplines all look at different aspects of risk (Wagner & Bode, 2008). The area of investments and portfolio management has a totally different angle than for example the medical field (Spekman & Davis, 2004). Consequently, the concept of risk is highly context specific and no single way of definition can be appropriate in all circumstances (Zsidisin, 2003a). Therefore, it is crucial for each study that deals with risk to define the understanding of risk as a first step.

In literature a tension about the nature of risk exists (Rao & Goldsby, 2009). The main question hereby is, whether risk is seen as a purely danger or as both: danger and opportunity (Wagner & Bode, 2008). Kahneman and Tverksy (1979) for example see risk as the process of choice between different outcomes (not necessarily negative outcomes).

Jüttner, Peck, and Christopher (2003) define supply chain risk as the “variation in the distribu- tion of possible supply chain outcomes, their likelihood, and their subjective value”. Whereas Harland, Brenchley, and Walker (2003) conclude that supply chain risk is the “chance of danger, damage, loss, injury or any other undesired consequences”. For the purpose of this study, by considering the assessment and managerial perception of catastrophic supply chain risks, the author finds that the second view of seeing risk as purely negative reflects best the reality of supply chain business. In adaption from Wagner and Bode (2008), supply chain risk is defined as: the chance of any undesired consequences (such as harm or loss) resulting from any kind of dis- turbance of the focal firm’s supply chain.

This definition can be formalized as risk being the product of probability (of a given event) and its severity (negative business impact): = ∗ ; whereas ‘P’ reflects the probability and ‘l’ the significance of the loss (Mitchell, 1995; Peck, 2005).

2.2. Supply chain risk management

Supply chain risk management (SCRM) is rapidly developing into a favorite research area for academicians. Also practitioners are increasingly concerned about risks in supply chains, as recent disasters have vividly demonstrated the fragility of global supply chains (Rao & Goldsby, 2009; Sodhi, Son, & Tang, 2010).

Companies have become aware of the critical importance of SCRM. A study by Hillman and Keltz (2007) reports that 46% of asked executives believe that better SCRM is urgently needed. Supply chain executives at IBM put SCRM as top priority on their agenda (Hillman & Keltz, 2007). Contrary to the perceived importance of the topic, a global survey by McKinsey identified that few companies have taken significant action (Muthukrishnan & Shulman, 2006).

There are several drivers that make SCRM a critical element of companies’ future supply chain strategies. One reason is the increasing complexity of supply chains. As noticed by Christopher and Lee (2004): “Organizations have often overlooked the critical exposures along their supply chains. Supply chains that run to hundreds if not thousands of companies over several tiers present significant risk”. Modern supply chains have to be seen as complex networks in which no company can be seen isolated (Christopher & Peck, 2004). Literature argues that also higher levels of globalization are a main driver for the increased supply chain complexity. The integra- tion with supply chain partners adds further to this. Reduction of inventory levels, shorter lead times, outsourcing, shorter product life cycles, and global operations are all trends that occur in supply chains since the 90s and are likely to continue. The consequences are increased levels of risks which call for integration of risk management into the theories of supply chain manage- ment (SCM) (Sorensen, 2005; Wagner & Bode, 2008).

For this study the understanding of SCRM shall be based on the definition of SCRM provided by Kajüter (2003): “Supply Chain Risk Management is a collaborative and structured approach to risk management, embedded in the planning and control processes of the supply chain, to handle risks that might adversely affect the achievement of supply chain goals”.

2.3. Classification of SCRM research

As mentioned before, research about SCRM has increased significantly over the last few years. This interest seems to be a logical consequence of the growing importance of the topic and the recognition that general risk management concepts of other disciplines can be applied or adopted to SCM (Rao & Goldsby, 2009; Seshadri & Subrahmanyam, 2005; Wagner & Bode, 2008). This section shall therefore give a short overview over the different topics that are of in- terest to academia, in order to make it easier to relate this study to overall SCRM research. By reviewing the general SCRM literature and also by reviewing works that attempted to provide an extensive categorization and literature review of existing SCRM studies (Pfohl, Köhler, & Thomas, 2010; Rao & Goldsby, 2009; Sodhi, et al., 2010; Vanany, et al., 2008), the author comes to the conclusion that a broad classification of the literature according to the source of supply chain risk and to the discussed stage in the SCRM stage, is most appropriate.

2.3.1. Categorization of supply chain risk sources

Early research in SCRM has followed the general trends in pioneering research to define and classify concepts (Oke & Gopalakrishnan, 2009). Therefore, various ways to categorize supply chain risks can be found in literature. Christopher and Peck (2004) offer the following three cat- egories: internal to the firm (with the sub-categories process and control), external to the firm but internal to the supply chain network (with the sub-categories demand and supply) and envi- ronmental. Different classifications are provided by various other articles (Jüttner, et al., 2003; Peck, 2005; Wagner & Bode, 2008). A lot of research captures only specific risks or tries to iden- tify risks that occur in certain situation.

Based on an extensive literature review and numerous interviews Wu, Blackhurst, and Chidambaram (2006) created a new hierarchical supply chain risk classification.

illustration not visible in this excerpt

Figure 1: Classification of supply chain risks. Source: adapted from Wu, et al. (2006).

They distinguish on the highest level between internal and external types of risk. Further they introduce the following sub-categories: internal controllable, internal partial controllable, internal uncontrollable, external controllable, external partial controllable and external uncontrollable. Figure 1 gives an overview of the classification scheme. In this study the risk type external uncontrollable is discussed and more precisely natural/manmade disasters and partly political instability, following the classification of Wu, et al. (2006).

The likelihood of occurrence and business impact are two dimensions that are frequently used in literature to categorize risks occurring in a supply chain context. Many authors see this categorization according to high-probability, low-consequence (LP-HC) and low-probability, highconsequence (HP-LC) risks as an important starting point in order to identify relevant mitigation strategies (Oke & Gopalakrishnan, 2009).

Risks that occur frequently and have only little impact are often classified as operational risks and are not in the scope of this research.

2.3.2. The elements of the SCRM process

Next to risk sources, the different steps of the SCRM process provide a useful categorization criterion for literature on the topic. Many authors focus on specific steps of the risk management process, like the development of mitigation strategies. Most of the literature describes four elements of the SCRM process, which might differ slightly from author to author: (1) risk identification; (2) risk assessment; (3) risk mitigation; and (4) responsiveness to operational and catastrophic risks (Vanany, et al., 2008).

According to Kouvelis, Chambers, and Wang (2006), the literature on managing risk is well developed, whereas research on risk assessment and quantification is still in its early stage. The authors argue that firms have to invest resources to build systematic risk assessment approaches. Similarly, Zsidisin, et al. (2004) mention that “few studies exist that explore the key constructs necessary for assessing supply [chain] risk”. This study will focus on the risk assessment stage in the context of catastrophic supply chain risks.

Table 1 to Table 5 provide an overview of the reviewed literature. This study draws from literature of different disciplines, such as SCRM, risk modeling and assessment, catastrophe research and behavioral research. For the most relevant studies, the main contributions/ findings, the reason to include it and the type of research method are indicated.

illustration not visible in this excerpt

illustration not visible in this excerpt

illustration not visible in this excerpt

illustration not visible in this excerpt

illustration not visible in this excerpt

2.4. Catastrophic supply chain risks

During the last decade numerous disasters around the world have occurred. Research has shown that in fact the risk of human and economic losses from natural disasters has grown tremendously within the last decade (Grace, Klein, Kleindorfer, & Murray, 2001). But supply chains are not only increasingly vulnerable because of the higher exposure to these events. The characteristics of modern supply chains add to this. Literature speaks of supply chains that are operated with less “slack”. Enormous efforts have been undertaken to reduce human and capital resources, in particular inventory. In addition, through global-spanning supply chains, locally occurring catastrophes have indirect effects worldwide (Knemeyer, et al., 2009; Pfohl, Köhler, & Thomas, 2010). Increased levels of risk are the consequence.

Supply chain disruptions through catastrophes have tremendous impact on companies’ perfor- mance. Ericsson lost over €400 million after a fire on one of their sub-supplier’s plant. Similarly, a fire at Toyota’s break supplier forced Toyota to stop its production and caused an estimated damage of $40 million per day. After an earthquake in Taiwan in 1999, Apple was not anymore able to fulfill some customer orders due to a supply shortage of microchips (Norrman & Jansson, 2004). In Europe, September 11 terrorist attacks lead to severe problems of the 3PL industry (Wagner & Bode, 2008). Hendricks and Singhal (2005) found that companies that experience supply chain disruptions have 33-40% lower stock returns in comparison to the industry bench- mark. Additionally these disruptions can also mean a loss of reputation for the companies (Sod- hi, et al., 2010).

The literature suggests several classifications for catastrophic events. A very broad classification is offered by natural versus man-made catastrophes. Mitroff and Alpaslan (2003) identified three types of threats: natural accidents , normal accidents, and abnormal accidents.

Regarding the first category, natural accidents, the literature discusses the following catastro- phes in a supply chain context: earthquakes, extreme weathers, diseases or epidemics, and vol- cano eruptions. Especially the Florida series of hurricanes in 2004, the tsunami of 2004, and the SARS disease are mentioned frequently (Finch, 2004; Jüttner, et al., 2003; Norrman & Jansson, 2004). Warren (2010) warns that climate change most probably will have a significant effect on the frequency with that extreme weather events and natural disasters occur. Therefore, man- agers need to prepare in order to mitigate these risks. Norrman and Jansson (2004) report that by the foot-and-mouth disease in the UK in 2001, due to more international supply networks, not only the agriculture industry was affected much more heavily than during its last outbreak 25 years ago, but also that other industries, like luxury car manufacturers, were strongly af- fected. Among normal accidents, the most common ones that are mentioned in literature are fires and chemical accidents (Cohen & Kunreuther, 2007; Norrman & Jansson, 2004). Abnormal accidents are defined as ill-will by insiders or outsiders (Mitroff & Alpaslan, 2003). Terrorism is the most salient topic, which has been discussed extensively in the supply chain context. Terror- ism can affect supply chains in a direct or indirect way (Sheffi, 2001; Sheffi & Rice, 2005).

For several reasons it is important to investigate the before discussed risks in a supply chain context. This kind of events can have dramatic effect on companies’ short term performance. Therefore, it is kind of surprising that SCRM literature has focused much more on the high- likelihood, low-impact type of risk and kind of neglected catastrophic risks (Chopra & Sodhi, 2004; Oke & Gopalakrishnan, 2009; Vanany, et al., 2008; Zsidisin, et al., 2004). Sodhi et al. (2010) suppose that researchers probably rather cover supply chain risks that are more easily quantifiable. Understanding the special characteristics and challenges of catastrophic events in a supply chain is crucial to be able to mitigate those risks appropriately. Therefore, it will be dis- cussed in more detail what makes catastrophic events special in a SCRM context and how this can help companies to deal more effectively with the process of assessing and managing them.

Catastrophic risks are environmental risks. This means they impact the overall business context across industry sectors. The magnitude may vary, but every company is therefore affected to some degree by this risk category. For each company a set of typical risks exists according to the nature of the supply chain and the special circumstances of the firm. The general characteristic of catastrophic risks is, that they have a severe impact once they occur (Rao & Goldsby, 2009). While the catastrophe may impact only a particular value stream or one node of the supply chain, through rippling effects many companies may be affected.

illustration not visible in this excerpt

Business impact

Figure 2: Risk categorization scheme. Source: Adapted from Knemeyer, et al. (2009).

What all the discussed catastrophic risks have in common is that managers have no control over these risks. As Peck (2005) puts it: “Disruptions emanating at this level are likely to be beyond the direct control of supply chain managers and business strategists”.

The risk class is further characterized by rare occurrence, but severe business impact once occurred (Vanany, et al., 2008; Wagner & Bode, 2008). As the risk potential is quite uncertain regarding frequency and timing of occurrence, special strategies to assess and monitor these risks have to be applied. Christopher and Peck (2004) add that although the timing of catastrophes might be not predictable, the impact may be assessed. This classification according to probability and impact can be shown in a risk matrix. Catastrophic events are located in the bottom right corner of the matrix as demonstrated in Figure 2 (Knemeyer, et al., 2009).

In addition to this, these extreme events are not only of LP-HC nature, but also share some cha- racteristics as high degree of dread and uncertainty, their involuntary nature and their occur- rence outside the normal range of experience of the respective system (Bier, Haimes, Lambert, Matalas, & Zimmerman, 1999; Slovic, Fischhoff, & Lichtenstein, 1979). To adopt this to a SCRM context, the supply chain or supply network can be interpreted as the system. Zsidisin et all.

(2005) add that firms often lack confidence in assigning probabilities to these events or have no awareness of those risks at all.

The discussed issues make it difficult to assess probabilities and impact of catastrophic events. For obvious reasons, little historic data exists for those events, which makes it much harder to quantify catastrophic risks than for example operational risks.

This does not mean that informed decision making is not possible in this area. The exposure of a supply chain to known phenomena can in many cases be assessed in advance. Geological or meteorological risks for example are identifiable, though the exact time and place of occurrence is very hard to predict. Other categories of catastrophic events such as the Foot and Mouth Disease or SARS are much more difficult or even impossible to predict (Peck, 2005).

This study draws from two streams of literature to help to identify suitable risk assessment techniques that can be applied in a supply chain context. On the one hand SCRM literature is used to account for the system elements of supply chains. On the other hand, a body of re- search exists that puts focus on the general assessment of LP-HC risks. The topic of business continuity management in general has a wider scope than SCRM and covers elements such as crisis management, disaster recovery, business recovery, or contingency planning (Vanany, et al., 2008). Those are normally not focus of SCRM and are also not in the scope of this research.

2.5. A conceptual framework for supply chain risk analysis

Literature provides several frameworks for analyzing and managing supply chain risks. Often a four step approach is proposed. The single steps may differ from study to study, but in most cases contain in some form risk identification, identification of risk drivers, risk assessment, mitigation strategies and ongoing monitoring, feedback or a learning loops.

Jüttner et al. (2003) suggest the four steps: (1) assessing the risk sources for the supply chain; (2) defining the supply chain adverse consequences; (3) identifying the risk drivers; and (4) mitigating risks for the supply chain.

illustration not visible in this excerpt

Figure 3: Conceptual framework for supply chain risk analysis. Source: adapted from Cohen and Kunreuther (2007).

For this research a framework based on the works of Kleindorfer shall be used. Kleindorfer has done widely recognized works on catastrophic supply chain risks from chemical emergencies. Moreover, Kleindorfer includes also a behavioral aspect, namely risk perception to his frame- work. This aspect is neglected by most of the other frameworks. The risk type of catastrophes is in the opinion of the author only covered sufficiently, when also discussing managerial risk per- ception.

Figure 3 shows the main components of the framework for risk analysis, based on the research of Paul Kleindorfer. Risk perception and risk assessment are linked to build risk management strategies. Finally, they are evaluated for their effectiveness. Knemeyer et al. (2009) define the risk assessment stage as “the evaluation of the likelihoods and consequences of prospective risks”. Risk perception focuses on psychological aspects of risks, which have an enormous influence on behavior. As a result from interaction of risk assessment and risk perception, strategies to manage the risk are developed. Finally the risk management strategies have to be assessed for effectiveness and should be evaluated (Cohen & Kunreuther, 2007).

[...]