Public administration and politics meet turbulence: The search for robust governance responses

Abstract

Public governance has always been challenged by turbulence, defined as “situations where events, demands, and support interact and change in highly variable, inconsistent, unexpected or unpredictable ways” (Ansell & Trondal, 2017, p. 1). Social conflicts, economic recession, war, and political leadership contests create turmoil and chaos to which governments aim to respond with a mixture of repression, concession, reform, and regime change to maintain or restore some form of social, economic, and political order. Political philosophers such as Machiavelli and Hobbes depicted history as a circular movement where the rise of relatively stable regimes predicated on the excise of hard power ultimately gives way to decadence, corruption, and ruin that erode the social and political order. In contrast, Hegel and Marx portrayed history as a linear trajectory governed by the rational unfolding of a dialectics whereby new, contradicting developments challenge the old, stable structures, thus leading to turbulent transitions that bring society to higher and higher stages. Despite their different views on history, they all agreed that order is temporary and invariably disturbed by short spells of crisis and heightened turbulence occurring at regular intervals but tending to foster a new period of relatively stable rule. Today, there seems to be a new sense that turbulence has become a chronic and endemic condition for modern governance. This new condition cannot be traced to any single factor, arising instead from multiple interacting developments. A first factor is that acute or creeping crises seem to be arising more frequently, affecting a wider range of sectors, spilling across political boundaries, and possibly producing multiple interacting crises (dubbed “poly-crisis” by Zeitlin et al., 2019). A second factor is how social, economic, and political interactions among widely distributed, multi-level parties are accelerating—producing interactions of surprising speed, scale, and scope (Hong & Lee, 2018). Communication and information technologies create lightning-fast information exchanges demanding a rapid and timely response to far-flung citizens, suppliers, stakeholders, and decision-makers who may not have even been part of the administrative picture until recently. Keeping up with potentially destabilizing, mediatized events can easily devolve into a constant stream of fire drills (Cottle, 2006). A third source of turbulence arises from an intensification of political conflict that challenges existing norms and mechanisms of conflict mediation. While public organizations are familiar with politics and conflict, they must now adapt to shifting political issues, polarized populations, rapid leadership turnover, clashing reform agendas, and uncertain planning horizons—sometimes all at once (Kriesi et al., 2012). Although the concept of turbulence implies the need to respond and adapt to change, it refers to a certain kind of adaptation and certain kind of change. When change is slow and steady, when shifts and trends can be clearly anticipated, and when important parameters change in observable, understandable, and relatively discrete ways, we are not in a turbulent world. When we have the leisure to respond to change through planned adaptation, when we seek to optimize structures or processes through comprehensive system reforms, or when we tweak operations to adapt to expected variance in resources, supplies, or personnel, we are generally not in a world of turbulence; instead, turbulence describes a state in which change is sudden, surprising and difficult to understand or track. It describes a world where we must deal with multiple, simultaneous changes, each demanding our immediate attention, often creating contradictions and dilemmas. Adaptation to turbulence can be like a group of strangers rapidly assembling a puzzle, where the picture is blurry and the pieces fit together poorly. While crises, speed, and political conflict are by no means novel challenges for public administration, our existing theories and analytical perspectives treat them as peripheral and exceptional rather than central and commonplace. Both political systems theory (Easton, 1965) and neoclassical economics (McKenzie, 1959) have focused on the occasional disturbance of a systemic equilibrium and the subsequent efforts to restore the equilibrium. Similarly, historical institutionalism talks about periodic crises that disrupt stable policy paths until new self-reinforcing paths are restored (Streeck & Thelen, 2005; Torfing, 2009). Even poststructuralist discourse theorists describe how hegemonic discourses are dislocated by events that they cannot domesticate, thus leading to turbulent political power struggles aimed at creating a new stable political and moral-intellectual leadership (Laclau & Mouffe, 1985). The common story here is that societal order and stable political rule prevail despite brief interregnums of crisis and heightened turbulence. The intellectual legacy of this sharp order–disorder dichotomy is reflected in the difficulty we have in understanding how governance structures and processes can be designed, led, and managed where complex interactive change is normal rather than exceptional. Our theories of public institutions tend to associate stability with the absence of change, while change is understood as a disruption of stability. As intuitive as this is, it often prevents us from understanding what makes governance robust in the face of complex and dynamic challenges. The key to addressing these challenges is to re-conceptualize how public administration theory views the stability–change relationship; we must focus more on how change enables stability and, reciprocally, how stability enables change. To be robust (as opposed to resilient, as explained below) essentially means to be able to continue providing public value in the face of variable, inconsistent, unexpected, or unpredictable events and demands. We argue that by paying attention to the interdependence—as opposed to opposition—of stability and change, we can illuminate some of the mechanisms contributing to robustness. This special issue suggests that we can address the challenges of turbulence by thinking about the robustness of our governing institutions and processes. To situate and frame the argument about the need for robust governance in turbulent times, we examine how different governance paradigms have conceptualized the order–disorder and stability–change relationships. We selectively focus on two prominent and well-established paradigms—public bureaucracy and network governance—and then contrast them with a third emerging paradigm: robust governance (we leave new public management out of the analysis because robust governance does not build as directly upon it). We account for the three governance paradigms one-by-one with a view to how they deal with turbulence. We then define robust governance, discuss its distinctiveness compared to crisis management, resilience, and agile management, and investigate the conditions and strategies for pursuing it. The following section reviews some prior theoretical claims about robust governance and relates them to the findings of the articles in this special issue. We conclude by returning to the notion that stability and change must be seen as interdependent, and we use it as a springboard for suggesting some additional strategies for achieving robust governance. Although the ideas about impersonal rule by civil servants have a long history (Kiser & Schneider, 1994; Nickerson, 1996), the growth of Weberian and Wilsonian bureaucracy was particularly pronounced in the interwar and postwar periods. The core ambition of modern bureaucracy was to create and maintain stable rules and societal order based on predictable administrative decisions (Du Gay, 2000). Separating private interests and personal whims from the public authority exercised by professionally trained and rule-bound civil servants provides a potent tool for securing a steady, controlled rule with few surprises. The bureaucratic quest for stability and predictability is motivated by an interest in maintaining sovereign political leadership while simultaneously protecting citizens against arbitrary administrative decisions. Hence, stability is seen as inherently good, thus making instability and disorder a problem. Since turbulence is regarded as exceptional, however, it poses no significant threat; indeed, bureaucracy was invented to tame and eliminate turbulence, which shows its ugly face if bureaucracy is incomplete or breaks down in extraordinary, unforeseen crisis situations. Modern bureaucracy was a child of the industrial revolution. Public bureaucracies administer numerous large-scale programs and systematically coordinate the actions of large numbers of people working at different levels to achieve common goals. They focus on economies of scale and delivering standardized service solutions to the masses. Built on centralized hierarchical structures, orders flow from top to bottom, and compliance is a key success criteria. Hierarchy along the vertical axis of public bureaucracy is complemented by a functional division of labor along the horizontal axis. The compartmentalization of public organizations into separate departments, agencies, and bureaus creates a specialization advantage, allowing a group of professionally trained public employees to become experts in undertaking a particular task while greatly enhancing the need for coordination provided by higher-level agencies. Together, the vertical hierarchy and horizontal division of labor help to keep everyone in check. In theory, employees should know exactly what people in their position should do, how they should do it, and what resources they have at their disposal. The formulation of written rules is the ultimate tool for securing compliance, predictability, and transparency, especially when combined with formal accountability systems based on monitoring, auditing, and legal sanctions (West, 2005). The bureaucratic governance paradigm relies on organizational structures and written rules to ensure that public employees perform as expected. Deviation from the organizational script is limited by professional norms and altruistic concerns for the public interest. Hence, there is little use for transactional and transformational leadership aiming to motivate public employees. Elected politicians lead by defining the overall policy goals and occasionally reshaping the public sector, whereas public managers lead the administration by overseeing the implementation of policies and dealing with occasional rule breaches. Being an integral part of liberal democracy, the bureaucratic governance paradigm is deeply concerned with securing free and fair elections based on universal suffrage, maintaining democratic control of the government, and preventing societal pressure groups from gaining excessive influence. Both democratic elections and the administrative implementation of legislation are regulated by rules, norms, and procedures aimed at guaranteeing fairness and preventing the abuse of power. Hence, in the bureaucratic governance model, input legitimacy and throughput legitimacy prevail over output legitimacy (Schmidt, 2013). It has long been recognized that the bureaucratic governance paradigm—with its rule-based command-and-control and formal legal accountability—may create a rigid, siloed organizational system that makes adaptation to changing events sluggish (Downs, 1967). This begs the question of how a large bureaucratic system can adapt to changing circumstances. The initial answer was “planning”; the planning of budgets and activities in the present and coming years is central to bureaucracy and relies on the forecasting of demographic developments, changing needs, and economic activities that largely determine tax revenues. Planning helps public bureaucracy to rationally predict and provide services, and it is undertaken by central departments responsible for producing comprehensive planning documents based on information from statistics bureaus requesting all parts of the public sector to submit relevant information (Friedmann, 1971). Comprehensive planning is supplemented by sector planning, and coordination emerges as a key instrument to align sector plans that may diverge and prompt a need to reduce conflicts and create synergies. Coordination is also needed to prevent overlaps and gaps in services and government activities sponsored by public agencies at different levels and in different sectors (Bouckaert et al., 2016). Bureaucratic coordination is top-down in the sense that higher-level bodies make authoritative decisions about who does what and how many resources they have at their disposal. Optimism for top-down comprehensive and sectoral planning peaked in the early 1960s, after which the assumptions underpinning rational and comprehensive planning were severely criticized (Lindblom, 1959; Wildavsky, 1973). Despite its fall from grace, planning did not disappear; instead, it assumed more modest and targeted forms that sought to facilitate adaptation to specific challenges. Strategic planning sought to anticipate challenges over the mid- to long-term and to provide context-dependent responses shaped by multiple stakeholders, thus differentiating itself from rational-comprehensive planning (Bryson, 2018). Contingency planning became fashionable in the late 1960s in the face of several environmental crises (Lentzos & Rose, 2009). Contingency plans are essentially decision-making tools determining who does what and takes which decision based on the anticipated circumstances possibly arising in a crisis. They are often mandatory, although the challenge is that planning assumptions are often foiled by actual events (Clarke, 1999). Organizational scholars also increasingly recognized that large bureaucratic organizations were open rather than closed systems and began questioning how they could be resilient in the face of external shocks. Thompson (1967) described how open-system organizations could buffer their “technical core,” coining the term “boundary spanner” to describe the specialized functions of managing external fluctuations that might disturb this core. Cyert and March (1963) described the importance of organizational “slack” as a resource that could be called upon when organizations undergo stress. Landau (1969) introduced the concept of redundancy as an important back-up system for ensuring organizational reliability. Finally, organizational scholars pointed to the need for special structural adaptations to respond to unique problems or challenges. For example, the term “adhocracy” was coined to describe a temporary organization that draws its membership from different parts of the bureaucratic organization and is sufficiently flexible and adaptable to cope with the heightened turbulence following in the wake of crisis events (Mintzberg & McHugh, 1985; Toffler, 1970). Large-scale organizations were urged to become “ambidextrous” by effectively exploiting well-known solutions while exploring the need to develop new solutions to cope with future demands (March, 1991). In sum, while Weberian and Wilsonian perspectives still dominated public administration thinking until the late 1980s, an increasing number of theoretical developments hinted that adaptation to change was becoming an increasingly important concern. However, change remained a monumental task for large public bureaucracies, and one that occurred only periodically when demands for change became urgent (Fernandez & Rainey, 2006). Often associated with the New Public Governance paradigm (Osborne, 2006, 2010; Torfing & Triantafillou, 2013), network governance emerged as a pluricentric alternative to the unicentric bureaucratic governance model (Kersbergen & Waarden, 2004), which is seen as limiting the mobilization of resources and exploitation of advantages associated with collaborative governance (Huxham & Vangen, 2013). The network mode of governing fits well with the postindustrial and post-Fordist logic of flexible specialization, since small groups of public and private stakeholders are brought together to design tailor-made solutions for particular target groups and subsequently amend them when circumstances and preferences change (Jessop, 2002). Governance networks thus focus more on economies of scope than economies of scale. Digital tools may support the sustained interaction between manifold network actors and facilitate knowledge-sharing in distributed settings. Collaborative governance in networks, partnerships, and so forth was initially perceived as a lender of last resort in the sense that it was only introduced when other bureaucratic or market-based forms of governance had failed (Ansell & Gash, 2008). This eventually became a standard tool in the public governance toolbox and was frequently used when public and private actors confronted complex and wicked problems (Rittel & Webber, 1973) and where there is a need to mediate conflicts between interdependent stakeholders (Rhodes, 1997). Wicked problems have both cognitive and political dimensions in the sense that the problems in question tend to be ill-defined and poorly understood due to the presence of tangled causalities, and the existence of tradeoffs between competing goals tends to hamper the formulation of solutions (Head & Alford, 2015). Complex problems vary in terms of their cognitive and political wickedness, thus calling for customized collaborative arrangements (Head & Alford, 2017). Much like its corporatism and policy network prequels, the overall ambition of network governance is to create islands of provisional stability in cognitively and politically complex—and therefore relatively unstable and challenging—policy contexts (Mayntz, 1993; Provan & Kenis, 2008; Sørensen & Torfing, 2007). Provisional stability is fostered by involving key stakeholders in networked governance through which distributed actors find ways of constructively managing their differences (Gray, 1989) and agreeing on particular understandings of the problems and challenges at hand and on a set of satisfactory governance solutions. Hence, network governance recognizes turbulence as a key challenge for governance but perceives it largely to be a matter of situational complexity. So how do governance networks manage to deal with cognitively and politically complex problems that easily spin out of control and cause heightened turbulence? First, they connect public and private actors across levels and sectors around specific problems that they all have an interest in solving. Second, the actors participate in trust-based collaboration in relatively self-regulated institutional settings. Collaborative interaction sometimes takes the form of hard-nosed bargaining, while at other times it involves open-ended deliberations through which the actors search for novel solutions and foster provisional agreement. Third, the actors contribute to the implementation of joint solutions over which they establish common ownership (Ansell & Gash, 2008; Emerson & Nabatchi, 2015). Compared to the modus operandi of bureaucracy, network governance replaces hierarchy, organizational insulation, and sovereign power with horizontal inter-organizational collaboration based on resource interdependency. Network governance tends to grant priority to output legitimacy, which prevails over input and throughput legitimacy. Pragmatic concerns for problem-solving and getting things done override concerns for impartiality and observing a fixed set of norms and procedures in the decision-making process. Nearly by definition, networks are understood to be “flexible”; a property that is clearly advantageous when responding to changing circumstances. In the absence of a centralized, hierarchical authority, planning becomes emergent and based on alignment as it aims to combine the plans of a diverse set of actors with jointly negotiated ideas about what should happen, when, and how (Innes & Booher, 2010). In network governance, the actors are interdependent but retain their operational autonomy. Hence, they bring their respective prefabricated ideas and plans to the negotiation table to explore congruencies and possible synergies, make mutual adjustments, and develop alternative joint planning scenarios contingent upon different priorities and future conditions. Thus, as Provan and Kenis (2008) observe, networks often face a “flexibility–stability” tension. This tension is well represented in the crisis management literature. Networks are now typically understood as critical for crisis response (e.g., Comfort & Zhang, 2020; Kapucu, 2006; Moynihan, 2008; Nohrstedt, 2018). They provide flexible and adaptive frameworks for responding to unique situational demands. However, crisis situations often demand rapid authoritative decision-making and often appeal, by necessity or fiat, to the importance of hierarchy, both as “command center” and as an orchestrator of networks in hybrid forms of governance (Christensen et al., 2016). Network coordination is often described as a “self-organizing” process (Innes et al., 2010), meaning that organizational actors come together in an emergent (rather than mandated) fashion and tend to be self-regulating. Research in the tradition of complexity theory particularly stresses this self-organizing nature of networks, underscoring how self-organization makes them adaptive to changing circumstances (Duit & Galaz, 2008; Koliba, 2013). In contrast to the bureaucratic governance paradigm, the network governance paradigm also tends to stress the informal nature of social networks (Hawkins et al., 2016). Still, leadership is important for convening actors, supporting their collaboration, and mediating conflict. This leadership tends to be facilitative rather than directional (Kickert et al., 1997) and may be organized in more or less distributed ways (Provan & Kenis, 2008). By stressing the values of flexibility, self-organization, and informality, the network governance paradigm is in many respects the mirror image of the bureaucratic governance paradigm. Indeed, networks have been seen as responding to the failures of public bureaucracies—particularly the significant limits placed on coordination by bureaucratic formality and hierarchical authority or by the fragmentation produced by the horizontal division of labor and contracting out public services. Although less studied than bureaucratic failures, networks also encounter failures (Koolma, 2013; Schrank & Whitford, 2011; Sørensen & Torfing, 2007; Teubner, 2009), which can occur through failures of collaboration; or, as Schrank and Whitford (2011, p. 170) vividly put it, when “exchange partners either screw each other or screw up.” Transaction costs may be high in governance networks (Lubell et al., 2017), and the interdependence associated with governance networks can increase the risk of failure. Governance networks often face paradoxes of how to manage conflicting demands, such as unity and diversity (Ospina & Saz-Carranza, 2010). Network failures can also arise due to the challenge of managing accountability (Koliba et al., 2011) and securing democratic anchorage (Sørensen & Torfing, 2005). While networks are considered flexible and are often called upon in crises, the temporal dimensions of network governance have rarely been explicitly considered. Yet effective negotiation and trust-building takes time, particularly as the number of stakeholders and the complexity of the issues grow (Ansell & Gash, 2008; Johnston et al., 2011; Klijn et al., 2010). Networks often imply shared leadership, which itself takes time to develop (Ulibarri et al., 2020). Failure to negotiate, build trust, and develop shared leadership can produce “collaborative inertia” (Huxham, 2003). If the traditional bureaucratic governance paradigm confronts a number of challenges related to its emphasis on order, control, and stability, the network governance paradigm equally confronts challenges related to its very flexibility, self-organization, and informality. Moreover, neither public bureaucracies nor governance networks are likely to simply replace one another. A more likely scenario is that they will simultaneously compete and co-exist (Christensen et al., 2016). As Agranoff (2014) rightly observes, bureaucracy tends to be reconstructed and transformed by growing societal turbulence and the rise of network governance. Moreover, bureaucratic actors may increasingly play the role of metagovernors aiming to initiate, support, and influence network governance processes without reverting unduly to command and control that is likely to constrain the built-in flexibility and scare off the network actors (Sørensen & Torfing, 2009; Torfing et al., 2012). While pursuing a flawless governance approach is quixotic, we suggest that considering the drawbacks of both bureaucratic and networked governance in light of the challenges of turbulence prompts us to search for a viable alternative. The new, embryonic public governance paradigm—what we call “robust governance”—aims to face up to the heightened turbulence of present societies and enhance the capacity of public organizations to permanently engage in the production of robust solutions through the creation of agile and developmental organizations capable of fostering improvisation, experimentation, and rapid learning and systematic involvement of relevant and affected actors beyond the narrow group of organized stakeholders involved in existing policy and governance networks. Envisioning a new approach to governance in the face of turbulence requires rethinking the stability–change relationship and the institutional and political modalities through which they interact. While public bureaucracies often exhibit an impressive stability, they may occasionally be characterized as “galloping elephants” (Rainey & Steinbauer, 1999) and are not generally known for their adaptability. Governance networks are known for their flexibility and self-organizing qualities, but while they may undergo processes of institutionalization (Ulibarri et al., 2020), they are not generally touted for their stability. As these two paradigms suggest, stability and change are generally viewed as opposing attributes: stability is about preventing change, change is about disrupting stability. On an abstract level, building governance systems that can meet the challenges of turbulence requires examining the stability–change relationship in a new light: rather than posing them as mutually opposing conditions, we must explore how stability requires change and change requires stability. We argue that the key property for relating stability to change in a new way is robustness. The notion of robustness is used in a wide range of scientific disciplines—from biology, engineering, and statistics to economics and sociology—to signify an ability to carry on or hold up across a wide range of demanding conditions (Anderies & Janssen, 2013; Carlson & Doyle, 2002; Holling, 1973; Huber, 1981; Kitano, 2004; Leeson & Subrick, 2006; Lempert et al., 2010; Schupbach, 2018). Robustness may simply refer to the ability of key features of a system to persist across a range of different circumstances and across temporal flux, implying some basic stability in the face of change. However, robustness is different from its cousin, resilience, in the sense that it does not refer, as does the latter, to the ability of a system to bounce back to a stable state after a shock. Although not necessarily part of its definition, robustness tends to point to the ability to uphold basic systemic functions (stability) through continuous transformations (change) that are supported by particular institutional infrastructure (stability). Hence, robustness is an instance of dynamic conservatism through which a system bounces forward to maintain some of its key functions in new and perhaps more attractive ways (Ansell et al., 2015). The robustness concept that stresses the need for agile adaptation in the face of turbulence has first recently found its way into the study of public governance, policy, and administration (Ansell et al., 2021; Capano & Woo, 2018; Ferraro et al., 2015; Howlett, 2019; Trondal et al., 2021). It all began with a growing interest in building societal resilience, which refers to the ability of a system to bounce back to its previous state of equilibrium in the wake of disruption triggered by external events or internal malfunctions (Duit, 2016; Juncos, 2017; Lindbom & Rothstein, 2006; Walker et al., 2004). Recent strands of governance research distinguish between static and dynamic resilience; whereas static resilience is the ability to restore the systemic equilibrium that was disturbed by external or internal turbulence, dynamic resilience is the ability to adapt the system of governance continuously so that it can function well under changing circumstances and perhaps even improve its performance in relation to a particular set of key goals, values, and functions (Ansell & Trondal, 2018; Howlett & Ramesh, 2022). There seems to be a clear resemblance between the notion of dynamic resilience and the new concept of robust governance advanced by Ferraro et al. (2015), Capano and Woo (2017), and Howlett (2019). Explaining the difference between resilience and robustness, Ansell and Trondal (2018) capture how robust governance aims to build flexibility into organizational and institutional arrangements and aims to absorb complexity, not only by improving the fitness of an organization to its new environment but also by incorporating requisite variety. Hence, robustness emphasizes the importance of maintaining multiple repertoires that can be flexibly redeployed to meet