Contingency planning for crisis management: Recipe for success or political fantasy?

Abstract

Contingency planning is widely considered to be an essential role of public authorities. Anticipation of what may happen, coupled with the prior allocation of resources, personnel, equipment, crisis control rooms, tasks, responsibilities and decision guidance/rules, is assumed to maximise the chances of a successful response in the event of a crisis. However, this paper proposes that the relationship between crisis planning and crisis management outcomes is more complex and nuanced relationship the often assumed. Contingency planning which is successful in the pre-crisis stage, does not guarantee a successful crisis response. Correspondingly, contingency planning failures in the pre-crisis stage, do not automatically lead to a flawed crisis response. The reasons rest primarily with the multiple influences on crisis responses – only some of which can be anticipated and planned for. The conclusion provides policy-oriented and analytical reflections which recognise the value of contingency planning, while suggesting that we should not inflate our expectation of contingency planners or rush too quickly to vilify them for a lack of adequate preparations.

1 Introduction

All societies deal on occasion with the rapid emergence of threats, conditions of high uncertainty and the need for swift decision making. A state of normality becomes shattered at both the operational and political-strategic levels (Boin, 2004). Operationally, the response may include saving lives, treating the injured, creating security zones and restoring critical infrastructures (McEntire, 2007). At the political-strategic level, tasks can include communicating the scale and significance of the episodes to a wider public, declaring a state of emergency and approving emergency funding (Boin, ‘t Hart, Stern, & Sundelius, 2005).

The terms ‘emergency’, ‘crisis’, ‘disaster’ and ‘catastrophe’ are the key words used to capture the scale and significance of such episodes. Over the last decade, 9/11, Hurricane Katrina, SARS, foot and mouth disease, London and Madrid bombings, volcanic ash clouds, BP oil spill and Hungarian toxic sludge, are powerful reminders of the vulnerabilities of modern cities and nation states, to the drama, volatility (and at times) tragedy of extraordinary events. The aforementioned terms are often used interchangeably – understandably so because of a general lack of agreement on definitions (Kapucu, 2008; Tarrant, 2010). In this article, we will use ‘crisis’ as an umbrella term to help capture the multitude of threats beyond the ‘normal’ mode of operational responders and political-strategic elites.

Amid varying scales of threat and disputes over the appropriate language to use, one factor commands near universal agreement: planning in advance for extraordinary scenarios allows organisational responders (at both the operational and political levels) to shift gear, applying the procedures and rules of ‘crisis management’ rather than those of ‘business as usual’. Broadly, such preparations are often referred to as contingency planning. The term ‘contingency’ is notoriously hard to define but as Schedler (2007, p. 56) suggests in his detailed examination of the concept, that contingency ‘carries the mark of chance, uncertainty, unpredictability’. The simple logic of contingency planning is that a process leading towards the prior allocation of resources, personnel, equipment, crisis control rooms, tasks, responsibilities and decision guidance/rules, coupled with training and scenario planning and an array or simulation exercises in a ‘safe’ environment, ensures that the operational and political-strategic layers within public authorities are best placed to manage any crisis that emerges. Therefore, planning is an ongoing process but contingency plan itself is an outcome, or a ‘…snapshot of that process at a specific point time’ (Perry & Lindell, 2003, p. 33).¹ A successful response is not guaranteed. However, the absence of a planning processes or contingency plan is generally considered to be a recipe for chaos, confusion and crisis mis-management.

It is little wonder, therefore, that contingency planning for crisis is encouraged by academics specialists, crisis management consultants, and by governments who spend millions and have developed intricate and detailed plans for multiple scenarios (Alexander, 2002; Lagadec, 1993; Perry & Lindell, 2007). For public authorities faced with managing events beyond the comfort zone of their ‘business as usual’ mode, a 24-h global news media is there to provide forensic and highly public scrutiny of crisis management failures considered to have stemmed from inadequate crisis planning and preparedness.

The assumption that successful contingency planning maximises the likelihood of successful crisis management is a widely held view among public authorities, political elites and many crisis management specialists. Yet the positive link been planning and response is rarely examined and questioned in detail. Some authors do seriously doubt the utility of many plans, describing them as ‘fantasy’ documents (notably Clarke, 1999), and there are others (Boin & Lagadec, 2000; Kapucu, 2008) who recognise that plans can have serious shortcomings in facilitating successful crisis responses. However, the crisis management literature lacks a succinct and nuanced overview of contingency planning and its varying capacities (or otherwise) to produce successful crisis responses.

A useful direction in which to proceed lies in the point made by Handmer and Dovers (2007, p. 29) in their book on disaster and emergency management, who argue that ‘If society wishes to better understand, avoid, prepare for or cope with emergencies and disasters, then over the long term, this can only be achieved through effective policy processes operating within suitable institutional settings’. The main contention of this paper, therefore, is that by applying ‘public policy’ thinking to the capacity of contingency planning to produce successful crisis management, we can bring a much broader and deeper understanding of crisis planning than hitherto exists in the literature. The focus of this paper is principally on ‘crises’ in developed countries where contingency planning is generally well developed. Many of the issues will also apply to planning in developing countries, but such are their socio-economic political complexity, that the role of contingency planning in such countries would need to be the subject of further research.

The article proceeds as follows. First it separates the nature of successful pre-crisis contingency planning from successful crisis management. Doing so allows us to map out a range of possible links between pre-crisis planning and crisis responses. It suggests that contingency planning may be ‘successful’ in the pre-crisis stage (in simple terms, planners have done as much as can be reasonably expected) but there is no guarantee of a successful crisis response. Indeed, it is possible to conceive of successful crisis management, despite weak or ‘unsuccessful’ pre-crisis planning. Second, it examines the reasons why the link between planning and management is not as stable as we might assume. It does so by identifying a range of possible influences on crisis management – contingency planning being only one. Finally, it examines the analytical and policy implications of an argument that conventional assumptions about contingency planning are more complex and nuanced that often assumed.

2 Separating successful pre-crisis contingency planning from successful crisis management

The world is subject continually to surprising contingencies (Bostrom & ĆirkoviĆ, 2008; Shapiro & Bedi, 2007) and prediction of the future is impossible. Of course, we can operate on the basis of hunches, scenario building, modelling and risk assessments, but there is a limit to what can be achieved. As Pawson (2006, p. 167) argues: ‘one must be modest in reflecting faithfully the limited authority of evidence itself. Good science is not futurology: we should always be humble about predicting the path ahead of what we know about the one already trodden’.

The limits of our knowledge of what the future holds may be instinctively obvious but are often give less credence among public authorities who need to plan on the assumption that certain types of events and circumstances may arise. This is a crucial point because it opens up a crack in the near linear-type assumption that successful contingency planning is likely to produce successful crisis management. Planning for crisis involves strong elements of foresight and certainty, while crisis – but its nature – is surprising and generates high levels of uncertainty (McConnell & Drennan, 2006). Indeed, as will be shown in this article, contingency planning may be successful in the pre-crisis stage but fail to produce successful crisis management outcomes (because inter alia it has limited knowledge of the future). The complex nature of ‘successful crisis management’ is the subject of the lead article in this special issue and there is no need to retread the issues here (McConnell, 2011). As long as we are aware the widely supported positive outcomes many emerge from a crisis response, we are in the position to concentrate on different linkages between crisis management outcomes and prior planning processes and plans.

2.1 Successful contingency planning in the pre-crisis stage

Despite the limitations of contingency planning, what it can realistically provide is ‘worst case’ thinking and the development of processes and plans which are sufficiently coherent to act as blueprints for crisis managers, while still offering flexibility depending on specific scenarios and contingencies (Webb & Chevreau, 2006). Contingency planning may be ‘successful’ insofar as it reasonably can be, given that planning involves synergy, coherence and predictability, while crisis involves uncertainty, information deficits and time pressures (McConnell & Drennan, 2006). Notwithstanding the reflections on crisis management success and failure in the first article in this special issue (McConnell, 2011), a public policy perspective with its emphasis on policy making, policy instruments and repercussions in terms of political powers, allows us to conceive of different types of success: processes, programmes and politics. Let us consider each in terms of pre-crisis contingency planning.

Policy process refers broadly to the ways in which problems are dealt with through an examination of options, discussion and consultation with regard to alternatives, and a filtering process that includes and excludes certain ideas and interests until a policy ‘solution’ is arrived at (Bardach, 2009; Howlett, Ramesh, & Perl, 2009). This process is manifest in the development of crisis plans. It involves an attempt to deal with the potentially destructive capacity of crises, by examining contingency planning options through scenario planning, consultation with first responders and relevant public officials, leading finally to the developed of an authoritative plan. Hence, we can conceive of contingency planning processes as being successful, insofar as public authorities achieve their goals of having a reflective, engaged and widely accepted legitimate process, aimed at developing plans which are as robust as they reasonably can be, given the surprising and devastating capacities of crisis. As ten Brinke, Kolen, Dollee, van Waveren, and Wouters (2010) argue with respect to contingency planning for flood scenarios in the Netherlands (a country where roughly 60% of the population live below sea level), successful planning and worse case thinking needs to be based on pragmatic ‘worst credible scenarios’ where there is a realistic hope that planning will be of some use.

Policy programmes refer broadly to authoritative laws, instructions, regulations or guidance, involving the use of differing mixes of policy instruments (non-intervention, persuasion, financial incentives/penalties, regulation, direct provision) (Hood & Margetts, 2007; Howlett, 2010). Contingency plans are ‘bundles’ of policy instruments, marshalled towards preparedness for crisis. Policy instruments in this instance may refer to plans which are a legal requirement on municipal councils, which leave some tasks to volunteers, involve protocols for evacuation of citizens, provide for emergency funding relief and so on. Hence, in the pre-crisis stage, it is possible to conceive of a successful contingency planning programme as the culmination of successful process. A successful programme involves allocation of resources/tasks and responsibilities which are as ‘rational’ a means of planning for foreseeable futures as they realistically can be, given the potential for crisis to behave in unpredictable and destructive ways.

Politics refers broadly to the political repercussions of processes and programmes, in terms of the impact on governmental support, its capacity to govern and its long-term agenda. In translating this phenomenon into contingency planning terms, it is useful to dissect politics into organisational/institutional (politics with a small ‘p’) and broader governmental politics (Politics with a large ‘P’). In the pre-crisis stage, success is conceivable and possible in both. Hence, at the organisational level (from hospitals and schools to prisons and space agencies) contingency planning processes and programmes may boost the internal and external ‘capital’ of the institution, convey an ‘in control’ message which helps marginalise or remove concerns about levels of preparedness, and allow the organisation to continue with its desired long-term trajectory (whether it be continuity or change). One NHS Hospital Trust in England, for example, was exceptionally well-prepared in its plans to deal with a lethal infection in a neo-natal intensive care unit. The outcome was not only successful in preventing any further deaths (sadly a baby died from the initial infection) but the response boosted the Trust's reputation and led to the award of two major prizes for crisis communication (Stronach, 2008).

On a broader governmental/governance level, contingency planning processes and programmes may help secure confidence in government, quell significant public fears and the need to deal with them, and pave the way for government to continue on its desired policy trajectory. These points follow-on from Clarke (1999, p. 13) and his argument that many plans are ‘fantasy documents…rhetorical instruments that have political utility in reducing uncertainty…’. As Clarke (1999) suggests (and as the subsequent BP oil spill confirms) there has never been a major oil spill success story, but this has not prevented oil companies and governments from accruing the political benefits of plans, prior to the ‘worst’ happening.

2.2 Mapping the relationship between pre-crisis contingency planning and crisis management outcomes

It is difficult to say where an acute crisis response ends and the recovery or aftermath begins. Many crisis management initiatives such as restoring critical services, vaccinating infected cattle and isolating harmful viruses, can justifiably be both. In this article when we write of the influence of contingency planning on crisis management, we are referring to the ways in which planning and plans may help and/or hinder the entire crisis management response – typically unfolding over an extended period of time, with response initiatives exhibiting varying degrees of success/failure.

The key point to take forward here, is that ‘successful’ contingency planning prior to a crisis, does not necessarily produce a successful crisis response. As Quarantelli (1996, p. 228) suggests in his paper on disaster trends: ‘The future is not the past repeated’. This crucial point can be unpacked by mapping a range of possible linkages between pre-crisis contingency planning and crisis management outcomes. The introductory article in this special edition deals in a fine-grained way with degrees of success and degrees of failure (McConnell, 2011). The analysis in the present article is much more parsimonious in order to illustrate a point. It based on the assumption that the link between pre-crisis contingency planning and crisis management outcomes is much looser than often conceived. An explanation can be provided by mapping out four broad types of linkages between pre-crisis planning and crisis management outcomes.

2.2.1 Successful pre-crisis contingency: planning successful crisis management outcomes

This scenario would be the classic justification for contingency planning (Alexander, 2002; Perry & Lindell, 2003). The crisis would be judged as well-handled, and the roots of this effectiveness, based on subsequent scrutiny and investigation, would seem to rest with prior planning and its anticipatory allocation of resources, roles and decision guidance/rules. For example, the response by first responders and decision makers to the 1996 bombing in Manchester, was widely considered to be effective in terms of establishing a crisis control centre within 30 min, creating synergies between multiple agencies and initiating swift evacuation. The key reason was considered the high level of planning preparedness (Williams, Bath, & Russell, 2000).

2.2.2 Successful pre-crisis contingency planning: unsuccessful crisis management outcomes

This scenario implies that despite a plan that was as robust as it could have been in the pre-crisis stage, the outcome nevertheless (to some degree) was a flawed crisis response. As Clarke (1999, p. 57) suggests:

Even the best laid plans can come to naught. Confusion can reign even when contingencies are thought through carefully. Organizations can fail to follow their mandates even when those mandates are clear. Organisational masters and experts can dissemble or even simply fail to understand that which confronts them.

In January 1982 a Boeing 737 crashed into ice-covered waters of the Potomac Rover in Washington, DC, in conditions so hostile than despite quite extensive and successful planning, responders struggled to cope in an efficient and timely manner. The crash was one of many severe problems facing the authorities that day: snow, ice, car accidents, trains packed to over capacity and road gridlocks (Clarke, 1999, pp. 57–61). The swift and destructive powers of some crises (especially on top of systems already stretched) can overwhelm plans in a matter of seconds.

2.2.3 Unsuccessful pre-crisis contingency planning: unsuccessful crisis management outcomes

In this scenario, weak planning or even the absence of a contingency plan, is a major causal factor in producing an ineffective crisis response. One example from New Zealand is the Auckland blackout in 1998. All four power cables supplying electricity to the city, failed and resulted in a power outage for over two months. Contingency planning was inadequate for several reasons (including a failure to plan for three or four cables failing) leaving a vacuum that was filled by buck-passing and confusion – leading not only to extended power outages, but also to reputational damage for the power provider Mercury Energy and the city itself (Newlove, Stern, & Svedin, 2003). A higher profile example is the 1989 Exxon Valdez oil disaster. The chair of the investigative commission, Professor Zygmunt Plater, described Exxon's contingency planning as ‘if it had been made on Mars. It was just complete fiction’ (www.necm.com 1 June 2010).

2.2.4 Unsuccessful pre-crisis contingency planning: successful crisis management outcomes

In this scenario, crisis management outcomes would be successful, despite contingency planning failures. Good fortune, creativity and improvisation can play a very strong role in shaping the crisis response. For example, in their study of the South-East Asian response to the 2004 tsunami where contingency planning and crisis preparedness infrastructures were weak or absent non-existent, Thévena and Resodihardjo (2010) show that despite many response flaws (especially at the initial stage) crisis outcomes were remarkably strong under the circumstances. The main reasons were an abundance of external aid and the availability of natural fruits. Even in episodes where no plans exist at all, response success is possible. For example, New York City had no plans for the waterborne evacuation of citizens from Lower Manhattan in the event disaster (the failure to plan being an extreme form of bad planning), yet a smooth evacuation is precisely what happened in the case of 9/11 – led by an ad hoc flotilla of vessels who rescued almost half a million individuals (Wachtendorf, 2004).

These four different scenarios illustrate the potential for a complex, non-linear and at times unpredictable relationship between contingency planning and crisis management outcomes. As Clarke (1999, p. 57) suggests: ‘planning and success do not coincide but are loosely connected or even decoupled entirely’. The simple but often forgotten reason amid the hopes and aspirations of many, is that contingency planning is only one factor among many, that impact on crisis management outcomes. A number of influences can be identified.

3 The role of non-contingency planning factors in shaping crisis management outcomes

The study of crisis management does reveal instances of crises being well-handled (the emergency response to the 2004 Madrid bombings; Australia's response to the 2002 Bali bombings), but also many which have attracted labels such as ‘mismanaged crises’ and ‘crisis management fiasco’ (Texan authorities handling of the Waco siege in 1993; Sydney's handling of the 1999 water contamination crisis; French authorities handling of the 2003 heatwave). Several key variables can be extracted from a wide a range of literature, allowing us to identify a multitude of factors which may influence crisis responses. They may be planned for to some degree and a common strand of ‘improvisation’ runs throughout. Improvisation may well be encouraged within the parameters of the plan (indeed it would be difficult to imagine a crisis response devoid of improvisation). However, factors influencing crisis responses are often beyond what was anticipated and planned for. Reponses can be shaped by ad hoc responses to specific circumstances, rather than falling into line with prior planning.

3.1 Nature of the crisis

Crises vary in terms of the speed of threat arrival, threat intensity, level of complexity, level of uncertainty and the timescale within which interventions need to be made. Cascading critical infrastructure breakdowns, hostage crises, rail crashes, oil spills and pandemic viruses, are but a few types of failures in a world where threats seem ever evolving. Of course, such contingencies may be discussed to some degree and examined in planning processes, but plans cannot cover every conceivable decision that needs to be taken for every conceivable crisis. Plans need an element of flexibility in order to cope with uncertainties (Webb & Chevreau, 2006). However, crises by their very nature, are chaotic and predictable. They do not always behave the way that planners might have expected in the relative comfort and safety of scenario planning (McConnell & Drennan, 2006). Hence, as Clarke (1999) suggests, the higher the level of uncertainty, the more likely it is that the contingency plan will not be relevant to the crisis resolution. All things being equal, plans are more suitable for crises where planning is based on relatively familiar and linear contingencies, allowing for experiential learning under conditions of fairly low uncertainty (such as airline disasters and rail crashes). By comparison, plans for unfamiliar and complex episodes are liable to have a high aspirational element, but be less useful because the nature of the crisis is likely to produce very high levels of confusion and improvisation (such as rapidly spreading pandemic viruses and terrorist biological attacks).

This variability in terms of crisis type, indicates that the nature of the crisis itself can be a significant influence on crisis management outcomes. Of course, it can be difficult to separate what should have been knowable and planned for, and what was not reasonably knowable and within prior capacities to plan for. Such distinctions are perhaps easier to make in publicised cases such as Hurricane Katrina where many of the recommendations from the Hurricane Pam exercise the previous year were never implemented, compared to the Kursk submarine tragedy where we do not really know what plans (if any) there were on the part of Russian authorities. More generally, the distinction between what should and should not reasonably have been planned for, is often blurred, fought over in the blame games that routinely follow crises (Boin, ‘t Hart, & McConnell, 2009; Boin, McConnell, & ‘t Hart, 2008; Brändström & Kuipers, 2003). Nevertheless, it is evident that crises come in many forms, helping produce substantial variation in crisis management outcomes.

3.2 Leadership, stress and decision making

The capacities and constraints on leaders to take tough crisis decisions (from political elites and chief executives to incident commanders and front line decision makers) are one of the most studied aspects of crisis management. Regardless of the level of planning and training, there are immense variations in personality types (Boin et al., 2005; Morris, 2002), comfort with crisis decision making under stress (Flin, 1996; Janis & Mann, 1977; Post, 2004), leadership styles (Hermann, Preston, Korany, & Shaw, 2001; Preston, 2001) reliance on advisors (Kowert, 2002), and types of interaction with others in group decision making situations (’t Hart, 1994; ‘t Hart, Stern, & Sundelius, 1997). All can play a role in shaping crisis outcomes. Cases ranging from the Cuban missile crisis and the 1980 Iranian hostage crisis to Hurricane Katrina and 9/11, have been shaped by complex mixes of personalities and leadership styles, interacting with stress-inducing threats. Crisis leadership is an extraordinary cognitive and emotional challenge. Sometimes the resulting interventions are praised widely as well-judged (Mayor Guiliani and the 9/11 attacks, German Chancellor Gerhard Schroeder and the 2002 floods, Chilean President Pinera and the rescue of 33 trapped miners). Other cases, involve accusations of absence at a critical time (Pakistani President Asif Ali Zardari and the 2010 floods), failures to accept outside assistance (George W. Bush and Hurricane Katrina), and ‘knee jerk’ reactions (UK Prime Minister Tony Blair and the 7/7 London bombings). More generally, for all the resources funnelled into contingency planning processes, the role of crisis leaders may be present in the programmatic ‘plan’ to some degree, but a large element of crisis leadership involves operating beyond the plan. As Hermann et al. (2001, p. 83) suggest: ‘Who leads matters’.

3.3 Institutional/organisational setting for crisis management

Contingency plans and planning processes neither exist nor are deployed in a vacuum. They operate within institutional settings (from organisational systems to political systems) exhibiting immense variations in goals, structures, procedures, power distribution, resources and culture (Pauchant & Mitroff, 1992; Roe & Schulman, 2008). Such variation applies just as much to ‘high reliability organisations’ such as air traffic control systems and nuclear power plants where safety and reliability are the indispensable goals which overrides issues of efficiency (Roe & Schulman, 2008), as it does to ‘regular’ public bodies such as those involved in education, road transport, social work and defence, where ‘safety’ has to compete and is often traded-off against the core imperatives of service provision. Of course, political systems could hardly by described as ‘high reliability’ because they have multiple and often conflicting goals, based on variations in constitutional configurations, power networks and so on. As a general rule, however, advanced economies are better prepared for crises and disasters than underdeveloped economies, by virtue of being able to allocate comparatively more resources to planning, training and crisis management infrastructures (Coppola, 2007). A more general point follows on. Regardless of the level of planning, variations across virtually every aspect of organisational and political life, can lead – all things being equal – to variations in crisis management outcomes.

3.4 Citizens, volunteers and extra-governmental organisations

In reality, the ‘human side of disaster’ (Drabek, 2010) is a major part of crisis/disaster responses, through a combination of volunteerism, ad-hoc involvement as a product of humanitarian acts or the sheer necessity of survival, leading to improvisation and assistance on a large scale. In addition to individual initiative there is the role of private companies with responsibility for crucial infrastructures in areas such as electricity, gas, transport, foodstuffs and telecommunications (Auerswald, Branscomb, La Porte, & Michel-Kerjan, 2006; Vineburgh, Benedek, Fullerton, Gifford, & Ursano, 2008). There is certainly a growing movement to pull such diverse interests into public sector contingency planning processes. But even the most thorough planning processes will find it impossible to capture the interest and harness the resources of every single citizen and private company. Crisis and disaster responses will always be shaped to some degree by non-governmental factors which were not anticipated. Sometimes the result is positive (such as Wal-Mart's rapid restoration of its supply chain and the distribution of emergency relief in response to the destruction of Hurricane Katrina) and other times disastrous (in the case of the 2004 Beslan school hostage crisis where traumatised parents breached security in an attempt to rescue their own children).

3.5 External power

For all that crisis decision makers may plan (or not), the arrival or seeming arrival of crisis may galvanise powerful interests who are threatened by the crisis, regardless of contingency planning (Boin et al., 2005; George, 1991). Crises destabilise and raise serious concerns over the legitimacy of policy sectors, sturdiness of public security measures, governing party credentials, leadership capabilities and more. At times, contingency planning rules and/or space for interpretation may be by-passed or interpreted in line with powerful interests. In the UK's 2001 foot and mouth crisis for example, a determined policy of cattle slaughter (and the refusal to vaccinate) were shaped predominantly by a long-standing coalition of food producer groups (the most notable being the National Farmers Union). This coalition marginalised other rural economy groups by successfully framing vaccination as detrimental to the farming industry and British exports (Wilkinson, Lowe, & Donaldson, 2010).

All of the factors mentioned above, indicate that crisis management outcomes are shaped my multiple factors – not simply contingency planning processes and plans. It is unsurprising, therefore, to conceive of ‘successful’ processes and plans, as leading to everything from successful to disastrous crisis management outcomes. Equally, it is unsurprising to conceive of ‘unsuccessful’ processes and plans (poor in conception or non-existent), leading not only to flawed crisis management outcomes, but also to positive outcomes by virtue of sheer good fortune. When further complexity is added by virtue of the different types of success across the process, programme and political aspects of crisis management, then it becomes apparent that the link between contingency planning processes and plans, and successful crisis response, is far more complex and nuanced than often assumed.

4 Analytical and policy implications

It is important, therefore that we now put forward some policy implications of our argument up to this point, as well as broader analytical reflections on contingency planning for crisis.

4.1 Contingency planning is still an essential role of public authorities

Under no circumstances are we advocating a policy of ‘no planning’ or treating it less seriously. For all the convoluted links between crisis process, plans and outcomes, it would be foolish, to coin a popular analogy, if we ‘threw the baby out with the bathwater’. If public authorities neglect contingency planning, the only barrier between responders and disastrous decisions, is sheer good fortune and the ad-hoc improvisational capabilities of responders. For example, no matter the level of criticism levelled at the city, state and federal response to Hurricane Katrina, the outcomes would in all likelihood have been much worse if New Orleans did not have an evacuation plan, or if the US Coast Guard had not conducted regular exercises for post-flood rescue (Derthick, 2007). Indeed, successful pre-crisis planning can produce successful crisis management outcomes. Australia's effective response to the 2002 Bali bombings, for example, was considered successful, precisely because all participants stuck rigidly to the plan in order to ensure a ‘joined up’ response (Paul, 2005).

4.2 ‘Sticking to the plan’ is not a guarantor of successful crisis management

The crisis management mantra of ‘sticking to the plan’ is likely to be a good instinct on balance. Even if the response has many shortcomings, managers can still say they ‘did what they had trained for’. However, sticking to the plan is not the pre-requisite for successful crisis management that is often assumed. Sticking to the plan may produce rigidity in the face of threats that are not playing out as hoped. The outcome is a crisis response unsuited to the crisis in hand.

4.3 Contingency planning processes are just as important as contingency plans/programmes

There are many instances of plans being ‘stuck on the shelf’ when a crisis hits, such as the 2005 storm in the South of Sweden (Eriksson, 2009). Indeed, at times one of the last resources in a crisis may be the ‘plan’ because the imperatives of quick decision making work against a time-consuming and line-by-line read of a plan that may not match the scenario that actually happened. Plans are ‘snapshots’ at particular points in time (Perry & Lindell, 2003, p. 338). However, even if the plan, the ‘snapshot’, sits on the shelf, the very process of planning is likely to been useful to some degree. It will, for example, have gathered together, individuals and organisations who would not normally work alongside each other, but makes them better placed to communicate in the event of a crisis. As Hillyard (2000) suggests in his work on public management preparedness for disasters, planning processes help develop common purpose and mutual learning among networks of professionals. The 9/11 Commission's examination of the response to the attacks on the Pentagon, praised the response for being largely successful – one reason being that the planning processes had helped build up trust and understanding between emergency responders (National Commission on Terrorist Attacks Upon the United States, 2004).

The process of contingency planning is also likely to have been useful, to some degree, in figuring out possible scenarios, the allocation of tasks, resources needed and so on. Processes also allow plans and to be formed and tested in a ‘safe’ environment.

4.4 Be cautious in viewing pre-crisis planning with post-crisis hindsight

We should be wary, with the benefit of hindsight, of over-harshness in identifying contingency planning failures. Yet it is easy to fall into such a ‘trap’. Contingency planning is routinely examined in a new light, once a crisis has hit. A process of sensemaking takes place, involving everyone from the media and first responders, to crisis executives and investigative committees. There is informal and formal backward mapping (through individual sensemaking and the sensemaking of an official inquiry) to assess the relevance of the plans and planning process to resolving the crisis in hand. Plans tend to be judged with hindsight in terms of relevance to the crisis that actually happened, rather than judged by what was fair and reasonable beforehand, given the capacity of crises to wreck even the most thorough plans. When backward mapping takes place to identify the causes of crisis and the effectiveness of the crisis response, the end point of the ‘story’ must be that certain events happened. No other conclusion can be reached. If a major city is struck by a serious water contamination episode where hundreds die and thousands fall ill, it is natural to seek an explanation in pre-crisis circumstances; assessing warning signs, regulations, level of planning and so on. In such an instance, and in others such as terrorist attacks, electricity grid failures and chemical explosions, post-crisis investigations must delve into past experiences, based on evidence, testimonies, expert witnesses and more, to construct a convincing storyline that leads from A (the ‘status quo ante’) to B (crisis). All other intervening variables, from plans to decisions, are scanned and assessed for their significance in generating the crisis outcome. Planning pathologies (such as failure to consider the ‘worst case’, failure to consult all relevant stakeholders, or neglecting to run a simulation) are routinely specified as causal factors in crisis outcomes (Hurricane Katrina being a highly public example).

Importantly, those involved in contingency planning should not be absolved of responsibility, simply because the plan planning process did not predict and prepare for the crisis that actually happened. But neither should they be damned for its weaknesses. The key issue raised here, refers to one of how much we can reasonably expect planners to predict the future – at least to a degree where they can provide an appropriate balance between pre-planned direction and pre-planned improvisation. This is a delicate balancing act which crisis planners have to contend with, amid the tension between the planning desire for order, stability and certainty and the capacity of crisis to generate disorder, chaos and uncertainty

4.5 Contingency planning has political rationality

It is, we would argue, a misjudgement of contingency planning to think that anything other than rational processes and plans to cope with future threats, is somehow illegitimate. Planning for threats is only one role of contingency planning. As the ‘politics’ of contingency planning illustrates, contingency planning processes and plans by organizations and public institutions, helps keep reputations intact, conveys impressions of being ‘in control’ in the event of the unforeseeable, and ensures that longer-term strategic goals proceed unhindered. If plans are to varying degrees ‘fantasy documents’ (Clarke, 1999), then the ‘fantasy’ performs a vital political role. Indeed, if higher degrees of uncertainty cultivate higher needs for fantasy documents (Clarke, 1999), then high levels of symbolic preparedness constitute ‘political success’ – a necessary goal for public authorities faced with the task of preparing for an unknowable enemy. Crisis management is concerned not just with managing objective threats, but also with managing public fears (Slovic, 2000). One broad qualifier is that we should not assume the political benefits of planning to be problem free. There is a paradox in highlighting risks and plans in order to dampen fears. Some such as Furedi (2005) argue that fear is politically paralysing rather than politically liberating.

4.6 Contingency planning can mask internal contradictions

If contingency planning has the capacity to produce three forms of success (process, programme, politics) we should not see them necessarily as compatible or mutually reinforcing. They may in fact conflict. A useful point here can be found with Edelman (1977) and Stone (2002) who suggest that ambiguity in the language of policy making is necessary in order for disparate interests with different values and aims, to coalesce. One could argue that ambiguity is a necessary condition of contingency planning. Planning processes and programmes provide certainty for those who seek prior direction and guidance, while also providing reassurance (through scope for improvisation in particular) for those who are aware that crises do not respect contingency planning. These two views may even be held at the same time by an individual. Yet ambiguity often masks conflict and it is not difficult to see the capacity of different ‘successes’ to conflict. Success with regard to contingency processes and plans in the sense that ‘political’ reassurance of vulnerable citizens and stakeholders, may actually involve plans which are operationally unrealistic, such as those of the US Postal Service to deliver mail immediately after a nuclear war (Clarke, 1999). Correspondingly, processes and plans which are so thorough and comprehensive, may provide to be politically unsuccessful because they highlight more fears than are necessary, continually rebound on government's agenda and derail government from a pathway it had hoped to pursue (Leiss & Powell, 2004).

5 Conclusion

Contingency planning is not a recipe for guaranteeing successful crisis management, but neither is it dispensable. In a world where the unpredictable and destructive tendencies of crisis never fail to surprise us, public authorities may need robust plans and planning more than ever before – not simply for operational reasons, but also for reasons that tend to be dismissed as dysfunctional. Governments need contingency planning for purposes of societal reassurance and political stability. Organisations also need them for similar purposes. The difficulty is that such ‘political’ success comes at a price. Contingency planning processes and plans, will often struggle to live up to the more politicised expectations. The result is that when a crisis hits, planning pathologies are often ‘discovered’ and held in full public view. Yet there are limits to what pre-crisis contingency planning processes and plans can achieve. It is a pity that those who struggle with these monumental tasks, are often considered scapegoats because they did not plan successfully for a phenomenon – crisis – which has no respect for contingency planning. Those officials and political actors involved in contingency planning should not be exempt from criticism and scrutiny, but neither should they be vilified for not living up the political expectations of contingency planning. Contingency planning for crisis is neither a simple recipe for success nor a crude political fantasy. It has elements of both. A more balanced understanding is needed if we are to have fair and realistic expectations of what public authorities can do to prepare for crises, disasters and catastrophes.

Acknowledgements

We would like to thank two anonymous reviewers, Arjen Boin, Alastair Stark, Mike Tarrant and Kurt Petersen.

Notes

1 Incident Commanders will also have individual incident-specific action plans, intended to complement broader contingency planning processes and plans.