HLSSWk8

Journal of Business Continuity & Emergency Planning Volume 12 Number 1

Understanding risk in an emergency management context

Bob Manson Received (in revised form): 26th January, 2018 Health Emergency Management British Columbia (HEMBC), Provincial Health Services Authority, 2nd Floor, 1770 West 7th Ave., Vancouver, British Columbia V6J 4Y6, Canada Tel: +1 604 829 2539; E-mail: [email protected]

Bob Manson earned his MA in disaster and emergency management from Royal Roads University and his Certificate in Emergency Management from the Justice Institute of British Columbia. He has been a volunteer firefighter, municipal emergency programme coordinator, and continues to be a search and rescue volunteer and search and rescue instructor. As an emergency man­ agement coordinator with Heath Emergency Management British Columbia, he works with a number of healthcare facilities in the Provincial Health Services Authority to build resilience to emergencies, and prepare for and respond to disasters.

A bstract

Understanding o f risk is important fo r emer­ gency managers. The variety o f definitions, expressions, assessment techniques, efforts to measure, perceptions, levels o f acceptability and the role o f probability consideration can create challenges when considering and communicating risk. This paper presents a brief overview o f some important factors with regard to risk, along with some case studies to inform emer­ gency managers. It also introduces an alternative approach to risk that includes and retains the three dimensions o f hazard, exposure and vulnerability.

Keywords: risk, hazard, exposure, vul­ nerability, emergency management

INTRODUCTION Risk is a concept that pervades all aspects of life. There is nothing we do that does not or cannot have associated with it, some form of risk. The word has different meanings to different people, and different disciplines address risk in different ways. In the discipline of emergency manage­ ment, there is a need to understand risk in a way that allows decisions to be made, information communicated and desired outcomes to be achieved for individuals, businesses and communities. Emergency managers may find themselves thinking about risk while standing in front of an immediate hazard to life and property or while in a corporate boardroom discussing the long-term potential for a develop­ ment. However, risk can be a nebulous concept that can lead to very different understandings within groups that make it difficult for those in the emergency management field to reach agreements and understand risk in meaningful ways. This paper will explore that four-letter word in an attempt to help those involved in emer­ gency management to learn how to think and communicate risk more effectively.

DEFINING RISK There are many different definitions of risk used in industry, finance, business,

Bob Manson

Journal of Business Continuity & Emergency Planning Vol. 12, No. 1, pp. 27-39 © Henry Stewart Publications, 1749-9216

Page 27

Understanding risk in an emergency management context

government and emergency management. Vlek1 identifies 11 different definitions of the term, including those with posi­ tive, neutral and negative connotations. Vlek goes on to group these into three categories:

• a single possibility of accident, loss or disease;

• a collection o f accident possibilities; and • an activity with accident (and other)

possibilities.

These categories serve to demonstrate the difficulty that can be had when engaged in discussions about risk as participants in the discussion may be referencing their under­ standing o f the meaning of risk to different categories. Although an awareness and understanding o f each o f these may be beneficial in an emergency management context, another useful definition of risk is offered by Kadvany:2 adverse consequences under uncertainty. This definition is useful in its brevity while demonstrating the lack of certainty that is associated with risk. If this definition is adopted, it may be useful to begin risk discussions by reviewing and accepting this definition within the group.

Risk must have something to which the term is applied; something must be at risk. This encompasses things that people care about.3 It may include tangibles, such as their health and safety and that o f loved ones, property or other assets, businesses, infrastructure, professional qualifications, etc. It may also include less tangible things such as relationships, sense o f commu­ nity, culture, history, belief in the future, way of life, etc. That people have dif­ ferent things that they care about and value demonstrates, in part, why they can determine very different levels of risk for the same hazard and why emergency managers, when communicating risk with the public, can face great challenges. It is valuable to understand the things valued

within an audience when engaging in risk discussions.

Risk must also be consciously separated from hazards. The media commonly use terms such as ‘risk of a thunderstorm" or ‘risk of an earthquake’. Both of these exam­ ples are hazards and, by themselves, do not constitute risk. It is not until they come in conflict with things that people care about and offer the potential for those things to be damaged or destroyed, or people to be hurt or killed as a result that there is risk. Risk addresses the adverse outcome. It should be remembered that a hazard can be seen or otherwise sensed. Risk cannot. The perception of risk is created by our own individual mental frameworks.4,5

As risk involves uncertainty, it must be accepted that there is no such thing as real or actual risk. Risk retains the uncer­ tain. Although some aspects of probability associated with a risk may be calculated in some circumstances with some degree o f accuracy, it should be noted that prob­ ability is only one dimension in some formulas for calculating risk, when it is considered at all. O ther variables in those calculations remain uncertain, and often highly subjective in their assessment.

EXPRESSION OF RISK How risk is expressed offers another chal­ lenge to emergency managers due to the variety and incompatibility of certain dif­ ferent means o f expression. One common expression to describe risk is as a ratio. ‘One in 10,000’ is a common form and value. However, this example is missing an important qualifier in order to put it into proper context, ie whether it is per year, per certain hours o f exposure, per event, etc. Such qualifiers must be included in any ratio expression o f risk. W ithout it, the expression loses meaning.

Risk is also commonly expressed as a single, unidimensional value. Such values

Page 28

Manson

are often determined through calculations in an effort to produce values that can be compared in order to make ‘informed’ decisions. However, as will be shown, there may be little meaning in a single value if it has been produced from sub­ jective variables. That lack o f meaning is especially acute when someone assigned to manage risk is asked to do something about an especially high and unacceptable risk value.

Other expressions of risk exist, such as loss of life expectancy, deaths per unit measure, or financial costs per event, dura­ tion, etc. Should any of these expressions of risk be mixed in any conversation about risk they will serve to confuse the discus­ sion. Care must be used to ensure that meaningful expressions of risk are used consistently. Again, prefacing risk discus­ sions by offering definitions of the terms that will be used and how they will be used can be highly beneficial.

ASSESSING RISK Before a risk can be expressed, it must be assessed in some way. Assessing risk involves predicting possible future out­ comes if everything remains unchanged from its current state, or things follow an expected path to a future state. This process involves gathering relevant data about relevant factors. In the financial sector, for example, relevance comes from economic indicators, certain values and trends, labour disputes, etc. For emer­ gency managers, there can be many factors that bear influence, such as a hazard’s type and potential magnitude, the people and the things they care about that may be in the path of a hazard, and the degree to which their interaction will cause injury. Each of these may require additional classification and analysis in order to iden­ tify the elements that are relevant and the ones that may be changed. Figure 1

Hazard Probability

Consequences Impact

Vulnerability Human Factors

Safety Mitigation

Costs Complexity

Linearity Exposure Capacity

Acceptability Liability

Likelihood? Severity?

Figure 1 Examples of the many things considered in risk assessments — Unpacking a traditional risk assessment effectively may be difficult where one is constrained to believe that risk is ultimately composed of only two dimensions, namely likelihood and severity

demonstrates one challenge that may arise when many different factors are considered in determination of risk but the resulting unidimensional value that is passed on for risk managers to address lacks meaning and cannot facilitate action.

When performing a risk assessment, consideration should always be given to what will be done with the answer so that it may be framed in a meaningful way. Providing meaning involves answering questions before they are asked: what, where, why, who, how and when. If the risk assessment will only be used for com­ parison with the risk assessments of other options, then it should be framed as such. If it is to be communicated to the public, then appropriate factors must be assessed and the communication of risk carefully planned. If the risk assessment is to be used to plan and direct actions for mitigation and prevention, or to plan for response and recovery, then consideration must be given to ensure the assessment provides meaning in those contexts so that the answer will provide direction.

Some of the major challenges with assessing and communicating risk to the

Page 29

Understanding risk in an emergency management context

public for emergency managers include the abundance of technical language used and the overwhelming belief that lack of agreement from the public results because they ‘just don’t get it’. However, risk is not in of itself a highly technical subject. All members of the public have some under­ standing of the term. Challenges come from assessments that consider different things as being at risk than the public does. It is also possible to create problems by using language that is unnecessarily complex and implies that the things people care about are not important to the assessors. For example, description of a probability distribution has little meaning to a homeowner concerned if her prop­ erty value will change. The public needs to have risk issues communicated to them in ways that have meaning to them.

MEASURING RISK Risk does not have properties that can be measured. It has no mass, density, colour, shape, volume or any other property that can be detected and measured with test instruments. It is a social construct deter­ mined when people consider the loss of things that are important to them. However, in efforts to provide quanti­ fication of risks in various industries, disciplines and government, a number of techniques have been used to determine an assessed value.

The following briefly describes some techniques of assessing and quantifying risk:6

• Expert intuition: No formal or structured system, just reliance on the outside expert’s opinion.

• Expert audit: Applying some structure, possibly in the form of checklists, to facilitate the expert’s opinion.

• Simple stratification method: Using simple scales, such as high, medium, low, or

green, amber, red, often for two dimen­ sions as might be seen on risk matrices.

• Weighted scores: A lengthy checklist is used with an assessed value deter­ mined for each identified risk criterion with an overall score determined. A scale may be used for each criterion to provide weighting (eg 1-3) with some criteria having positive values, others having negative values and some used as multipliers. These can allow considera­ tion for such things as the influence of mitigation.

• Traditional financial analysis: The use of traditional financial analysis tools to determine a risk value.

• Calculus of preferences: More typically used in decision analysis according to their risks with experts using specialised tools.

• Probabilistic models: Consideration of the probabilities of certain outcomes deter­ mined mathematically.

As the simple stratification method and, to some degree, probabilistic modelling are the two most predominant in emergency management, only those will be discussed further here.

Efforts to measure risk are often done to quantify a risk value in an effort to make decisions more effective, such as choosing which of two options presents the least risk. There are challenges, however, with attempting to determine a unidimen­ sional representation of risk. For example, the calculation itself does not determine risk, but instead, produces an expecta­ tion value,7 the magnitude of which is often used to represent risk with the hope that it does so accurately. Unidimensional expectation values have, among their chal­ lenges, the problem of producing results that may not represent particularly con­ cerning issues that have been rolled into the final product. For example, if deter­ mining risk values from two dimensions,

Page 30

Manson

the calculations 2 x 4 and 1 x 8 both return the same risk value, yet do not reflect the prominence of one of the vari­ ables should it be present and indicating something in dire need of attention. Using this type of calculation, one risk in this example is determined to be no greater than the other.

The following formula has been iden­ tified as the most common in use in emergency management to calculate a risk value:8

risk = likelihood x consequences.

Although it may appear simple and attractive, the major challenge comes in quantifying the variables, and doing so consistently and with meaning. In some circumstances it may be possible to deter­ mine a likelihood, or probability, with a certain amount of accuracy when suf­ ficient data are available. At other times, likelihood may only be quantified through a rough, highly subjective and potentially misleading estimate. An example may be the effort to determine the characteristics of a 500-year flood event when consid­ ering only 50 years of river data and not adequately considering the changes made to the landscape over those 50 years that will have affected runoff and drainage characteristics. For the emergency manager making such a calculation with such delusive data, the values determined for probability, magnitude and ultimately impact may be wildly inaccurate. This paper will discuss probability assessments as a part of risk assessments in due course.

In emergency management, the con­ sequences of a hazard manifesting itself can also be very difficult to quantify, leaving users with few options other than monetary impacts, such as replacement costs, cost of disruption of businesses, cost of litigation, among others as the indicator of the magnitude of an impact.

As noted previously, these are not always what people value, and relying on such valuations can place emergency managers at odds with stakeholders when com­ municating risk. Such conflicts can be particularly acute after a disaster.

When one chooses to ‘measure’ risk in an effort to quantify it, one can look only at those factors that have been subjec­ tively chosen as partial indicators of risk. Table 1 provides a random selection of risk-calculating formulas from emergency management and transportation literature, showing examples of various formulas either in use or recommended. Note that many of them do not consider probability, although all include variables that at times must be determined subjectively. It should also be noted that the selection of a par­ ticular formula to use for determining a risk value is itself subjective.

Although such formulas can be used to determine risk values for comparison, their use comes with several cautions: users require calibration in order to con­ sistently produce the same value for the same observed condition. Training and standards are required to maintain such consistency. They can also only be used to compare decision options where risk has been calculated using the same process and considered the same conditions. Using dif­ ferent variables or inconsistent values will make the results incompatible and efforts to make comparisons problematic. The results are also highly subjective. Even with calibration they produce a value that may be different from that of another assessor, making effective comparison potentially unreliable. They are influenced by biases, heuristics and experience, the very things that can, at times, lead to irrational deci­ sions. They are also meaningless outside of their context, so they cannot be used to produce values that can be compared with risk values from other disciples or those cal­ culated with another formula. And finally,

Page 31

Understanding risk in an emergency management context

Table 1: Examples o f som e formulas used or recom m ended for calculating risk values

Formula to calculate risk Source

Vulnerability x Hazard Abbott, P. L. and Samson, C. (2009) ‘Natural Disasters, Canadian Edition’, McGraw- Hill Ryerson’, Toronto, O N

[Hazard x Vulnerability] - Capability Alexander, D. (2011), ‘Natural hazards’, in M. Sala (ed.) ‘Geography’, Vol. 1, Eolss Publishers, Oxford, pp. 364—387

Hazard x Vulnerability [ X Exposure] Alexander, D. E. (2008, Apr. 24) ‘Planning and management o f complex emergencies’, available at: http://emergency-planning.blogspot.ca/2008/04/planning-and- management-of-complex.html (accessed 29th October, 2013)

Severity x Probability x Exposure US Coast Guard (2011) ‘SPE risk assessment model work sheet’, available from h ttp :// www.dco.uscg.mi1/Portals/9/DCO%20Documents/National%20Strike%20Force/ foscr/ ASTFOSCRSeminar/Presentations/Safety/SPERISKASSESSMENTMODEL. pdf?ver=2017-09-14-144539-567 (accessed 6th December, 2017)

Likelihood x Consequences Coppola, D. P. (2011) ‘Introduction to International Disaster Management’ (2nd edn), Elsevier, Burlington, MA

Threat x Vulnerability x Consequence Cox Jr, L. A. (2008) ‘Some limitations o f “Risk = Threat X Vulnerability x Consequence” for risk analysis o f terrorist attacks’, Risk Analysis, Vol. 28, No. 6, pp. 1749-1761

Probability x Severity Drottz-Sjoberg, B. M. (1991) ‘Perception o f risk: Studies o f risk attitudes, perceptions and definitions’, Centre for Risk Research, Stockholm.

Probability x Vulnerability50" 1 Vllucs Ferrier, N. and Haque, C. E. (2003) ‘Hazards risk assessment methodology for emergency managers: a standardized framework for application’, Natural Hazards, Vol. 28, No. 2, pp. 271-290

Hazard x Exposure x Vulnerability Kron, W. (2002) ‘Keynote lecture: Flood risk = hazard X exposure x vulnerability’, in ‘Flood Defence 2 0 0 2 ’, Science Press, New York, NY, pp. 82-97.

Probability x Intensity x Exposure x Vulnerability Power, W. (2013) ‘Tsunami impacts’, in ‘Review o f Tsunami Hazard in New Zealand (update)’, GNS Science Consultancy report, Wellington, N Z

(Probability + Magnitude) / 2 X (Exposure + Fragility + Resilience) / 3

Rottach, P. (2010) ‘Background and components o f disaster risk reduction’, ACT Alliance (ACT), Diakonie Katastrophenhilfe. available at: www.preventionweb.net/files /24122_24122backgroundcomponentsofdrractdi.pdf (accessed 11th June, 2018)

Probability X Consequences x Exposure Transport Canada (2011) ‘Module 5 — risk management’, available at: https://w w w . tc.gc.ca/eng/civilaviation/publications/tp 13897-module5-1904.htm (accessed 20th November, 2016)

they can be misleading by failing to include important considerations that could shift the risk value in a dramatically different direction, perhaps leading users to believe that the risk is lower by hiding the magni­ tude of certain variables that should serve as strong indicators of potential problems.

RISK PERCEPTION Many things influence the perception of risk. These include cultural and organisational

influences, attitudes, training and educa­ tion, and motivation, among others. For example, Starr9 notes that the public will generally accept voluntary risks to a degree 1,000 times greater than they will involun­ tary risks, demonstrating the difficulty in assessing risk perception in others. Although the topic is a rich one, only a brief explo­ ration of the influences on individual risk perception of interest to emergency man­ agers is offered here as understanding of them can be highly beneficial.

Page 32

O f particular importance is the under­ standing that different people can have very different perceptions of the risks presented to them by the same hazard,10 and that they can each be entirely correct in that individual determination. The highly subjective nature of risk determi­ nations will always see broad differences in opinion. Some of the things that shape perception of risk are people’s own heuris­ tics and biases.

Heuristics are the mental models and strategies that aid in making judgments and decisions by allowing us to use pre­ determined mental cues to facilitate quick assessments in response to certain signals." Heuristics are learned and moderated through social and cultural learning, expe­ rience and beliefs.12,13 They can promote survival by helping us avoid danger with quick responses in the face of hazards without the necessity of time-consuming cognitive assessments. However, they do not guarantee an accurate response. Where heuristics provide cues for action, biases influence cognition, producing results that may be different from reality14 leading to unrealistic perceptions of risk, underesti­ mation of risk, unwarranted confidence in risk decisions or the misjudging of risks.1516 Although a complete exploration of heuristics and biases is not possible here, Table 2 offers brief descriptions of some key heuristics and biases.

ACCEPTABLE RISK Acceptable risk, also called risk appetite or risk tolerance, is the level of risk at which it has been decided that efforts at mitiga­ tion and prevention will cease. The belief is that any efforts to reduce risk further will have prohibitive costs, especially where the law of diminishing returns is applied, and the likelihood of an occurrence has been determined to be acceptably small. It should be noted that there is a significant

difference between acceptable risk and risk that has been accepted. In the former, one may define a clear threshold for where acceptability lies and in the latter it may be necessary to decide between two or more options, especially in an emergency situation where a decision must be made quickly, and accept the one with lower risk without consideration of where it lies in relation to the acceptable threshold. In such situations, consideration must be given to whether any of the available options are below the level of risk the organisation has defined as its acceptability threshold.

Making decisions in the face of risk can be one of the most difficult decisions an emergency manager can make. When determining acceptable risk the question should be asked, ‘what are we accepting?’. In her treatment of the loss of the space shuttle Challenger, Vaughan17 described the decisions made by various engineers who were aware of the problem with the shuttle’s solid rocket booster O-rings but determined it to be an acceptable risk. Those engineers expressed dismay after the shuttle was lost, as though it was something that could not or must not have happened. This demonstrates one of the challenges with acceptable risk decisions: it should not be numbers that are being accepted but an outcome. The engineers who made the launch decision, when asked, described the level of risk that they had accepted as one in 10,000. The senior management of NASA described the level of risk that they believed they were accepting as one in 100,000. The engineer in charge of the solid rocket motor testing programme had assumed an acceptable level of risk of one in 25.18 Although the dramatically different levels of acceptable risk within the same organi­ sation that different people thought they were operating at should not be ignored, it may be regarded that the differences

U n d e rs ta n d in g r is k in an e m e rg e n c y m a n a g e m e n t c o n te x t

Table 2: B rief descriptions o f som e heuristics and biases relevant in understanding risk perception

Heuristics and biases Description

Affect heuristic Perception o f risk is influenced by either a positive or negative emotional response to a stimulus. A positive emotional association tends to influence perception to see the benefits high and the risks low. A negative association tends to see the benefits as lower and the risks higher. See: Slovic, P. (2000) ‘Introduction and overview’, in Slovic, P. (ed.), ‘The Perception o f Risk’, Earthscan, London, pp. xxi-xxxvii.

Availability heuristic Influences from our experiences and how easily a recent similar situation, or the number o f similar situations come to mind when compared with a current situation. See: Coppola, D. P. (2011) ‘Introduction to international Disaster Management’ (2nd edn), Elsevier, Burlington, MA; Tversky, A. and Kahneman, D. (2000) ‘Judgment under uncertainty: heuristics and biases’, in Connolly, T., Arkes, H .R . and Hammond, K.R. (eds), ‘Judgment and Decision Making’ (2nd edn), Cambridge University Press, Cambridge, pp. 35-52. Experience includes those from training, operational events, or even from exposure to media coverage which can skew estimates o f probability. See: Tversky, A. and Kahneman, D. (2000) ‘Judgment under uncertainty: Heuristics and biases’, in T. Connolly, H .R . Arkes, and K.R. Hammond (eds.), ‘Judgment and Decision Making’ (2nd edn), Cambridge University Press, Cambridge, pp. 35-52.

Representative heuristic Seeking an exact match between the current situation and a situation found in memory to use as reference to assess risks for the current situation. See: Slovic, P., Fischhoff, B. and Lichtenstein, S. (1981) ‘Perceived risk: psychological factors and social implications’, Proceedings o f the Royal Society A , Mathematical and Physical Sciences, Vol. 376, No. 1764, pp. 17-34. Problematic if the match is not exact, possibly leading to an inappropriate risk assessment.

Optimism bias Perception that for one individual or group the level o f risk is lower than for others exposed to the same hazard. See: Stranks, J. W. (2007) ‘Human Factors and Behavioural Safety’, Butterworth-Heinemann, Elsevier, New York, NY. Strong in those who have had little or no direct exposure to a hazard. See: Powell, C. (2007) ‘The perception o f risk and risk taking behavior: implications for incident prevention strategies’, Wilderness and Environmental Medicine, Vol. 18, No. 1, pp. 10—15.

Confirmation bias Once information is gathered and a decision made, there is a tendency to ignore any new conflicting information while accepting only new information that supports the decision. See: Hunter, P. R . and Fewtrell, L. (2001) ‘Acceptable risk’, Water Quality: Guidelines, Standards and Health. Risk Assessment and Management for Water-related Infectious Disease’, IWA Publishing, London, pp. 207-227.

Risk homeostasis After efforts are made to reduce risk, such as through mitigation, new factors and behaviours are accepted that raise the risk back to the previous level. See: Stranks, J. W (2007) ‘Human Factors and Behavioural Safety’, Butterworth-Heinemann, Elsevier, New York, NY.

only reflect something of a numbers game and not an indication of what exactly was being accepted. The outcome was not a completely random event and its acceptance as a very real possible outcome needed to be considered as the thing that was being accepted.

Accepting a level of risk should be acknowledged as accepting the undesired outcome, as the accident or event unfolds. It may be painful, unwanted, expensive and/or emotionally devastating, but it should be regarded as acceptable only

when it can be accepted. The term ‘accept­ able risk’ should only indicate a price that is willing to be paid. Professional gamblers understand this concept as they will not place a bet on the table that they cannot afford to lose, otherwise they will not be professional gamblers for long.

Yet another challenge that emergency managers may face with regard to accept­ able risk is the belief that doing anything about a risk is doing enough about the risk. Efforts such as mitigation, protec­ tive measures, public education, rules and

P age 34

regulations, and many others are often regarded by some as sufficient to reduce risk to an acceptable level simply because action has been taken. It is important to resist the assumption that addressing risk with some action automatically reduces the risk to an acceptable level. Further re­ assessment must be conducted afterwards to analyse the effectiveness of the effort and ensure it has not created new risks in order to understand the residual risk, the risk that remains after action has been taken.

PROBABILITY AND RISK As noted previously, probability is featured in some risk calculations used in the dis­ cipline of emergency management. The probability of an event occurring may be determined in one of three ways:

• as one of a number of optional out­ comes, either weighted or unweighted (eg a game die has six sides and being unweighted, each has an equal oppor­ tunity for outcome);

• frequency where, given sufficient data from past observations, a performance of x outcomes for every n events can be expected to con tinue in the future if the parameters remain unchanged; or

• scientifically derived where there are too few or no observations from the past and interactions and outcomes can only be estimated, eg with new technology.

Each of these may have a place in emer­ gency management when and where determination of probability is desired. Each, however, has some challenges in returning reliable values in an emergency management context. In many cases, the number of influences on probability cannot be known, leaving the number of optional outcomes and potential weighting of variables unknown. From a frequency

perspective, the number of observed and recorded iterations may need to be in the hundreds to thousands in order to deter­ mine a frequency confidently, but only if it is known that parameters that influence the frequency rate remain unchanged. If a frequency rate is known and regarded as sufficiently low for the risk to be ignored solely for that reason, decision makers can be left unprepared when it does occur. In the case of scientifically-derived risk determinations, Perrow'9 notes that some technology can be so complex and tightly coupled that some accidents can defy prediction.

Some examples can illustrate that prob­ ability as a property of risk may not always be a useful way of determining and con­ sidering risk. On 30th June, 1987, Delta flight 810 departing LAX experienced a complete loss of engine thrust, due to the pilot inadvertently shutting off both engines in-flight. United Airlines flight 232 in July 1989 experienced an engine disintegration in-flight that disabled all three redundant hydraulic systems, leaving the aircraft with no aerodynamic controls. On 3rd July, 1988, the guided missile cruiser USS Vincennes mistakenly shot down an Iranian passenger jet with the loss of all 290 aboard. Barings Bank, one of the oldest and largest banks of its kind, collapsed through the actions of one individual. The Exxon Valdez grounded in Prince William Sound while sailing under ideal conditions. Mount St. Helens erupted in 1980 in a way that was unknown to volcanologists for that type of volcano and who had not recom­ mended a larger evacuation area as a result. On 21st January, 1968, a USAF B-52 bomber acting as a routine radio relay between the Thule US Air Force base in Greenland and the rest of the world in an effort to prevent accidental nuclear war, accidentally caught fire and, after being abandoned by its crew, flew directly over

Understanding risk in an emergency management context

the airbase and crashed outside the fence with four hydrogen bombs on board, fortunately not precipitating the very acci­ dental nuclear war it was intended to prevent. In each of these cases, if, before they occurred, experts had been consulted on their probability of occurrence, the answers would have almost certainly been that they were so unlikely or impossible that their consideration was not worth­ while or realistic; they could not happen. From an emergency management perspec­ tive, it is of little value to place too much emphasis on thinking about risk from a probabilistic perspective when one can acknowledge that things that have never happened before happen all the time.

A DIFFERENT WAY OF LOOKING AT RISK To address risk in meaningful ways that better allow communication of risk and preserve values from a risk assessment so that they can individually be acted upon in mitigation and prevention planning, it is proposed that risk not be assessed and expressed as a single unidimensional value but as three separate and distinct dimensions: hazard, exposure and vulner­ ability. Each can be assessed and quantified individually once a means o f doing so is agreed upon. There is value to emergency managers to think about and express risk in these three dimensions. Figure 2 illus­ trates the three dimensions.

If any one of the dimensions of hazard, exposure or vulnerability are absent, there can be no risk. Risk can be reduced when any or all of the three can be reduced. Efforts at managing risk can be better focused when the risk is expressed in these separate dimensions — ones that can be individually or collectively addressed. People can see hazards. They can see expo­ sure. They can see vulnerability. Expressing risk in these three dimensions can make

Hazard Figure 2 Looking at risk through its components of hazard, exposure and vulnerability may provide greater meaning Source: Reese, S. and Schmidt, J. (2008) Tsunami and flood hazard exposure of city council infrastructure in Christchurch City’, NWA Client Report WLG-2008-67, unpublished report; Yan, J (2010) ‘Disaster risk assessment understanding the concept of risk', in ‘Training Workshop on Drought Risk Assessment for the Agricultural Sector, Ljubljana, 20th-24th September’, Global Risk Identification Programme (GRIP).

it easier to assess and communicate risk factors to stakeholders. Doing so can avoid the problem associated with two identical risk values being determined from vastly different input variables, as happens when considering a unidimensional value for risk. Each of the three dimensions will be briefly expanded upon here.

Hazard Hazards should be regarded as presenting the potential to do harm. They may be wholly or partially controllable or, as is the case with many natural hazards, such as earthquakes, entirely uncontrollable. Understanding a hazard as a discrete entity allows mitigation and prevention efforts to be better focused and aspects of the other dimensions to also each be addressed more effectively. Various types of hazards have been defined, to include physical,

Page 36

structural, chemical, biological and psy­ chological. 20 To this list one can add executive failure, as defined by Perrow. 21 It should be noted that in some cases, a single threat can form more than one hazard. A flood, for example, presents erosion, accretion, biological and chemical contam­ ination hazards, among others. Emergency managers should already be familiar with the common means of addressing hazards; elimination, substitution, engineering controls, and administrative controls such as detection and warning systems. Each of these can be enhanced when exposure and vulnerability to the hazard are also known and considered separately.

Exposure Exposure offers an opportunity, occasion or opening for an unwanted event to have impact, typically the manifestation of the hazard. Whereas the hazard is active, exposure is passive, and can exist in time and space. Anything that people care about can be exposed, such as people, buildings, equipment, reputation, liability, etc. The exposure can be one of short-term, long­ term or cumulative. To address exposure, some controls are available: communica­ tions, barriers and boundaries, alternative courses of action, rules and regulations, and redundancies.

Vulnerability Vulnerability represents weakness, fragility, sensitivity or susceptibility. Like exposure, it too is passive. From vulnerability one can determine the magnitude of the impact, should it occur. It is regarded as the lack of capacity to anticipate a hazard, cope with it, resist it and recover in its aftermath. 22 As there are different kinds of exposure, there are also different kinds of vulnerability, such as physical, social, economic, technological, biological, community and environmental, among others. Planning becomes essential in order to have contingencies for each,

that provide for the ability to prevent, mitigate, adapt, withstand and rebound. To primarily address vulnerability it is impor­ tant to build resilience. Other controls include giving stakeholders ‘tools’ to help them help themselves so that the hazard may be less of a threat and/or they may face less exposure. From an emergency management perspective, Perrow23 offers the ‘all target’ approach, where vulnerable targets are separated from their threatening hazards, or they are minimised so as to minimise the magnitude of any potential impact.

UTILIZING THE HAZARD, EXPOSURE, VULNERABILITY APPROACH EFFECTIVELY For the proposed three-dimensional model of addressing risk to be useful, there must first be agreement on the definition of each dimension. This is necessary not only to provide clear meaning for each term, but to also clarify what can be addressed when problems are found so that they can be effectively alleviated. Initial communication with stakeholders must clarify the terms as they will be used in all risk discussions. Assessment and measurement criteria, if desired, must also be clearly defined and established for the organisation and efforts made to use them consistently. Even when communicating with the public, as intuitive as the terms hazard, exposure and vulnerability can be, the emergency manager will benefit greatly from ensuring the terms are clearly understood before applying them to any situation. By addressing each dimension separately, the concept of risk can be better understood, applied and communicated.

CONCLUSION For emergency managers, risk is an impor­ tant aspect of the discipline. It is important

Understanding risk in an emergency management context

to understand that the term has many definitions and many different interpreta­ tions, all of which can exist in the face of the same hazard and all of which can be correct to those involved. At the same time, risk cannot be real or actual as it inherently reflects the uncertain. Any efforts to calculate risk return expecta­ tion values that must be regarded as such and used according to their limitations. Although probability can be used in risk calculations, its consideration may create the belief that certain low-probability events are not worth considering, resulting in vulnerability, and there will always be unanticipated occurrences for which a probability could not be calculated and which are often left out of risk assessments as a result.

Addressing risk in the three dimensions of hazard, exposure and vulnerability will allow each to be considered separately, in meaningful ways, so that each discrete property can be retained and addressed when action is planned as part of mitiga­ tion, prevention and response. The three dimensions are easily communicated and can have meaning to the public during risk communications.

R eferences

(1) Vlek, C. (1995) ‘Risk assessment, risk acceptance and risk management: A psychological decision theorist’s view’, in Contaminated Soil ’95’. Springer, Dordrecht, pp. 565—579.

(2) Kadvany, J. (1997) ‘Varieties of risk representations’, Journal cf Social Philosophy. Vol. 28, No. 3, pp. 123-143.

(3) Renn, O. (1998) ‘Three decades of risk research: Accomplishments and new challenges’, Journal of Risk Research. Vol. 1, No. 1, pp. 49-71.

(4) Ash, J. S. and Smallman, C. (2008) ‘Rescue missions and risk management: Highly reliable or over committed?’ Journal of Contingencies and Crisis Management, Vol. 16, No. 1, pp. 37—52.

(5) Wachinger, G. and Renn, O. (2010) ‘Risk perception and natural hazards’, WP3-Report of the CapHaz-Net Projekt, Synergien zwischen Naturschutz und Klimaschutz—Wasser/ Gewasser (-Management), available at: www.researchgate.net/profile/ Tracey_Coates/publication/228827276_ Risk_perception_offnatural_hazards/ links/00b49519f2905da3f5000000/ Risk-perception-of-natural-hazards.pdf (accessed 26th January, 2014).

(6) Hubbard. D. W. (2009) The Failure of Risk Management: Why it’s Broken and How to Fix it, John Wiley and Sons, Hoboken, NJ.

(7) Drottz-Sjoberg, B. M. (1991) ‘Perception of risk: Studies of risk attitudes, perceptions and definitions’, Centre for Risk Research, Stockholm.

(8) Coppola, D. P. (2011) Introduction to International Disaster Management (2nd edn), Burlington, MA: Elsevier.

(9) Star, C. (1969) ‘Social benefit versus technological risk: what is our society willing to pay for safety’, Science, Vol. 165, No. 3899, pp. 1232-1238.

(10) Stranks, J. W. (2007) Human Factors and Behavioural Safety, Butterworth- Heinemann, Elsevier, New York, NY.

(11) Slovic, P., Finucane, M. L., Peters, E. and MacGregor, D. G. (2004) ‘Risk as analysis and risk as feelings: Some thoughts about affect, reason, risk, and rationality’, Risk Analysis, Vol. 24, No. 2, pp. 311-322.

(12) Stranks, ref. 10 above. (13) Wachinger and Renn, ref. 5 above. (14) Pessemier, W. (2008) ‘Improving

safety performance by understanding perceptions of risk and improving safety management systems’, paper presented at ‘Reducing Firefighter Deaths and Injuries: Changes in Concept, Policy, and Practice’, symposium conducted by the Public Entity Risk Institute, 22nd September to 3rd October, available at: http://citeseerx.ist.psu.edu/viewdoc/ download?doi= 10.1.1.511.3458&rep =repl&type=pdf (accessed 11th June, 2018).

Page 38

Manson

(15) Ibid., ref. 14. (16) Powell, C. (2007) ‘The perception

of risk and risk taking behavior: Implications for incident prevention strategies’, Wilderness and Environmental Medicine, Vol. 18, No. 1, pp. 10—15.

(17) Vaughan, D. (1997) The Challenger Launch Decision: Risky Technology, Culture, and Deviance at NASA, University of Chicago Press, Chicago, IL.

(18) Ibid., ref. 17. (19) Perrow, C. (1999) Normal Accidents:

Living with High-risk Technologies (2nd

edn), Princeton University Press, Princeton, NJ.

(20) Stranks, ref. 10 above. (21) Perrow, C. (2011) The Next Catastrophe,

Princeton University Press, Princeton, NJ.

(22) Rottach, P. (2010) ‘Background and components of disaster risk reduction’, ACT Alliance and Diakonie Katastrophenhilfe, available at: www. preventionweb.net/files/24122_24122 backgroundcomponentsofdrractdi.pdf (accessed 11th June, 2018).

(23) Perrow, ref. 21 above.

Page 39

Copyright of Journal of Business Continuity & Emergency Planning is the property of Henry Stewart Publications LLP and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.