Crisis-induced learning in public sector organizations

Abstract: How do public organizations manage crises? How do public organizations learnfrom crises? These seemingly basic questions still pose virtual puzzles for crisismanagement researchers. Yet, the interest of the academic and practitionerrealms in crisis management has grown in recent years. In this doctoral dissertationEdward Deverell sheds light on the problems regarding the lack ofknowledge on how public organizations manage and learn from crises, with anumber of critical knowledge gaps in contemporary crisis management as thestarting point. In the last few decades the interest in crisis management as a scholarly fieldhas grown. This developing field is composed of an increasing number of looselyconnected social science scholars concerned with issues of extraordinary events,their repercussions and the way in which they are managed by authorities,organizations, policy makers and other key actors. However, there are severallacunae to be dealt with in the emerging field of crisis management research.This dissertation sets the spotlight on four of these limitations of the crisis managementliterature to date. First, influential scholars within the field call for increased structuration andfeasible models to help us understand and explain various important factorsinfluencing the crisis management process. In this dissertation I try to bridgethis gap by developing theory on crisis response and learning. Crisis responsesignifies organized activities undertaken by a stakeholder when a community ofpeople – an organization, a town, or a nation – perceives an urgent threat to corevalues which must be dealt with under conditions of uncertainty. Crisis-inducedlearning refers to purposeful efforts, triggered by a crisis event and carried out bymembers of an organization working within a community of inquiry, that leadto new understanding and behavior on the basis of that understanding. Second, organizations play a key role in crisis management. Surprisinglyenough, however, crisis management research have only occasionally built theoryon how organizations respond to crisis. So far, the literature tells us moreabout crises as events than on how these events are actually managed. One reasonis the focus within crisis management research on highly unusual, big catastrophicevents and industrial accidents. Therefore, this dissertation explorescrisis episodes that affect specific organizations rather than entire communitiesor national governments. In addition, the dissertation brings together debateson crisis management and crisis-induced learning from a public managementand organizational perspective. Third, crisis management researchers have to date dealt mostly with acutecrisis response and issues of preparedness, while the issues of crisis aftermathsand crisis-induced learning are still relatively unknown. However, althoughthis study recognizes the importance of crisis planning and sense-making, thisshould not lead to a relative neglect of the issue of learning from crisis. Crisisinducedlearning is important as crises are rare events with huge repercussions.Thus crises are opportunities to draw lessons in order to improve future managementand crisis response, and to mitigate the risk of future crises. Fourth, the relatively few studies that have dealt with crisis-induced learninghave focused on learning after the crisis (intercrisis learning), while theoryon learning during crisis (intracrisis learning) is not as developed. My interestin both inter- and intracrisis learning obligates me to study crisis response andcrisis learning in conjunction. This means studying how organizations respondto crises and how they learn during and from these episodes. By focusing onprocesses of crisis response and learning under pressure – rather than pre-crisisplanning, threat perception, risk management and preparedness – the dissertationlooks into how organizations and their members manage the challenge ofcrises and how they take on, make use of and implement lessons learned fromone crisis to the next. The lacunae outlined above are theoretical points of departure for this dissertation’sinterest in the extent to which public organizations learn from crises.Accordingly, the overall objective of the dissertation is to increase understandingof crisis response and crisis learning in public organizations. In doing so, Iconduct an abductive study of how public organizations respond to crises andhow they learn during and after these events. The term ‘abductive’ refers toa research strategy which is characterized by continuous movement back andforth between theory and empirical data. The first step of the research process was grounded in the empirical world.The empirical contribution is a careful process tracing and case reconstructionof six cases involving Swedish public sector organizations. In the methodologychapter (Chapter 3) I describe the basis of the empirically bounded case study approach and case reconstruction and process tracing method. Six case studiesof organizational crisis management and learning were selected for furtheranalysis. The case studies were based on a variety of sources including posthoc accident investigations, articles, organizational documents and 129 extensivesemi-structured interviews with key crisis managers. The process tracingand reconstruction efforts led to case narratives, which were then dissected byidentifying dilemmas and critical decision-making occasions that were studiedin more detail. The following cases are explored in the dissertation: TheSwedish energy utility Birka Energi’s management of two cable fires that causedlarge-scale blackouts in Stockholm in March 2001 and May 2002; The cityof Stockholm’s management of the 2001 blackout and the repeated incidentin 2002; The Swedish Defence Research Agency’s (FOI) management of hoaxanthrax letters in 2001; and three Swedish media organizations’ (the Swedishpublic service radio Sveriges Radio, the Swedish private TV station with publicservice tasks TV4, and the Swedish public service TV station Sveriges Television)management of news work and broadcasting challenges on 11 September 2001(and to some extent following the murder of the Swedish Foreign MinisterAnna Lindh in September 2003). As the case selection reveals, all organizations under study are not puregovernment organizations. Rather three organizations (Birka Energi, SverigesRadio and Sveriges Television) are publically owned corporations, while one(TV4) is a privately owned media organization. Accordingly, this dissertationclaims that ownership is not the only measure of ‘publicness’. Media organizations,for instance, are of great importance for democratic societies. The term‘public organization’ is thus in this dissertation not used in the sense of equatingto government, but rather in reference to the degree of which political authorityand influence impacts on the organization. The theory generating approach that this dissertation takes on impliesthat the case studies are ‘heuristic’ case studies. The dissertation aims to promotenew hypotheses for further research rather than to produce generalizedknowledge. To this end the case studies are further analyzed by specific theoreticalapproaches suggested by prior research. This second step of the researchprocess is dealt with in some detail in the literature review. The literature reviewin Chapter 2 aims to bring an injection of organizational studies into the fieldof crisis management research. The review presents relevant studies from thefields of crisis management studies, organization studies (with special attentiongiven to organizational learning theory) and public administration and management.The review puts forth a twofold argument: There is a need of increasedknowledge not only about crises and how they develop, but also about how theyare actually managed by public organizations. However, prior crisis managementresearch with bearing on public management organizations are mostly based on either political executive foreign policy decision making or on veryspecific high reliability organizations operating in the pre-crisis phase. Hence,organization studies and public management studies should play a greater partin crisis management research. The review also provides an overview frame for the study by highlightingrelevant research. The chapter discusses the problems of defining, categorizingand operationalizing key concepts such as crisis, crisis management and organizationallearning. In the third step of the research process, the case studies are further analyzedusing theoretical approaches aimed at proposing propositions on how publicsector organizations may respond to crises, and how they may learn from theircrisis experiences. These analyses have been carried out with an aim to producestand-alone articles aimed for publication in international scholarly journals.Thus this dissertation differs somewhat from the typical public administrationdissertation as it is comprised of an analysis of several articles, as opposed to amonograph. The journal articles are published or accepted for publication inthe Journal of Contingencies and Crisis Management, the Journal of HomelandSecurity and Emergency Management, Public Management Review, and RiskManagement. The articles are reprinted in four empirical chapters (Chapters4-7), which make up the core of the dissertation. Introductory and concludingchapters aimed at bringing the discussion together have then been added.I present the first empirical analysis in Chapter 4. It looks into how organizationalculture affects strategy and adaptability in crisis management. The keyresearch question is: What mechanisms affect organizations’ ability to restructurein order to cope with acute crisis management challenges? In the study I propose atypology of temporal organizational responses to crises in public perception. Thetypology is based on organizations’ abilities to change strategy and adapt theirmanagerial and operational levels to deal with crises. The empirical data used toconstruct the typology covers three organizational crisis responses: 1) The utilityBirka Energi’s response to a cable fire that caused a thirty-seven hour blackoutin Stockholm in 2001; 2) The TV station TV4’s response in terms of how toreorganize and broadcast during the 11 September 2001 terrorist attacks; 3)FOI, the Swedish National Defence Research Agency’s response to the anthraxletter scare of 2001 and 2002. The different organizational outcomes featuredby the typology reveal distinct aspects of organizational crisis management.According to the typology, the Fully Adapting Organization (TV4) managesto adapt both its strategy and its managerial and operational levels to deal withthe crisis. The Semi-Adapting Organization (FOI) changes its strategy but lacksthe capacity to change managerial and operational levels according to the newstrategy. The Non-Adapting Organization (Birka Energi) does not grasp theimportance of strategy change in the first place. Based on three inductive case studies, the study concludes that organizational culture plays an important rolein this process where the Semi-Adapting Organization and the Non-AdaptingOrganization were dominated by strong expert cultures which proved to be lessinclined to change. In contrast, the Fully Adapting organization had deliberatelyfostered an organizational culture in which flexibility – understood as thecapacity to readily adapt to changing demands – was a cornerstone. The second empirical analysis is presented in Chapter 5. It deals with theissue of flexibility and rigidity in crisis response and crisis learning at two Swedishpublic organizations. The point of departure for the study is that the relationshipbetween crises, organizational crisis management response and learning hasto date been understudied. In an effort to broaden theoretical knowledge on therelation between crisis and learning, the study analyzes the crisis responses oftwo public organizations during a sequence of two failures. The empirical datais grounded in thorough process tracing and case reconstruction analyses ofhow the utility Birka Energi and the city of Stockholm managed two comprehensiveblackouts in March 2001 and in May 2002. The key research questionis: How does organizational rigidity and flexibility affect public organizations’ crisisresponse and crisis learning? A framework of rigidity versus flexibility in responseis utilized in the analysis. The findings are then discussed in relation to theirimplications for the nexus between crisis and learning. The study concludes byraising four propositions for further research. The third empirical analysis is presented in Chapter 6. This study aims tocontribute to the debate on organizational learning from crisis by sheddinglight on the phenomenon of crises as learning triggers. In the study I pose thefollowing key research question: How can we analyze organizational learningduring and after crisis and what criteria should be part of the analysis? In an effortto unveil patterns of how organizational crisis-induced learning may appearand develop, I suggest a conceptual framework based on conceptual categoriesand answers to four fundamental questions: what lessons are learned (single- ordouble-loop)?; what is the focus of the lessons (prevention or response)?; whenare lessons learned (intra- or intercrisis)?; is learning carried out or blocked fromimplementation (distilled or implemented)? In the analysis section I explorethe practical applicability of the framework by using the same empirical casestudies as in Chapter 5. The final section suggests four propositions for furtherresearch. The last empirical study is presented in Chapter 7. There I construct aframework of management, learning and implementation in response to crisis.My point of departure is a proposition from previous crisis managementresearch which posits that previous experience can shape crisis response as away of repeating former routines or as a precondition for improvisation. Thekey research question is: How do organizational management structures affect crisis response, learning and implementation? In the study I argue that flexibilityis closely connected to the way organizations learn – in behavioral or cognitivemodes. Moreover, these learning modes are connected to the role of managerialgroups, where I differentiate between centralized and decentralized top managerialgroups. In addition, two case studies of how two bureaucratic media organizations(Sveriges Radio and SVT) managed and learned from extraordinarynews events – most notably 9/11 and the assassination of the Swedish ForeignMinister Anna Lindh – are conducted. The findings show how the decentralizedmanagerial group learned in a behavioral fashion, by creating new formalpolicies and structures, while organizational members in the centralized managerialgroup relied on individual cognitive structures as a way of ‘storing’ lessonslearned. The study ends by discussing the findings from a crisis managementperspective, where I propose that the two modes of learning profoundly affectthe crucial issue of flexibility in organizational crisis response.The concluding Chapter 8 discusses and contrasts the findings and propositionsgenerated from the four separate empirical analyses. Here the role oforganizational structure and culture are highlighted by revisiting specific organizationalfactors that seem to impact on organizational crisis management andlearning processes, such as previous experience, flexibility and rigidity in crisisresponse and learning, and centralization and decentralization. These factorswere also outlined in the literature review. Further empirical evidence of howthe factors affect crisis response and crisis learning in organizations was foundin the four empirical analyses. In addition, findings from the empirical studies also related to different types of learning processes such as intra- and intercrisis learning and singleand double-loop learning. Consequently these concepts are also deliberated upon in the concluding sections of the dissertation. As a final attempt to bring the propositions and arguments together, a framework of the crisis management and learning process is proposed. In regard to this venture, it is important to acknowledge the limitations of the framework, and of the dissertation as a whole. As it is only based on data from six cases of Swedish public organizational responses to crisis, the framework is merely a visual schematic of a number of propositions to be further tested and validated by further research. However, the framework also has a few virtues. It is an attempt to approach the ambiguous nature of crises and crisis management processes. The framework may also assist in providing more sensible and practical conceptualizations, and thus bring us closer to definitions that remain close to everyday operations of practitioners involved in crisis management. This dissertation thus makes an effort to bridge the gap between crisis management scholars and practitioners. This is also an overall goal guiding research activities at the National Center for Crisis Management Studies (CRISMART) at the Swedish National Defence College, where the research behind this dissertation has been conducted.