Debate: What is complex government and what can we do about it?

‘Complex government’ relates to many factors: the size and multi-level nature of government; the proliferation of rules, regulations and public bodies; a crowded arena with blurry boundaries between policy-makers and the actors who influence them; and general uncertainty when people interact in unpredictable ways within a changeable policy environment. Complex government is difficult to understand, control, influence and hold to account. This article considers it from various perspectives: scholars trying to conceptualize it; policy-makers trying to control or adapt to it; and scientists, interest groups and individuals trying to influence it.


Paul Cairney
'Complex government' relates to many factors: the size and multi-level nature of government; the proliferation of rules, regulations and public bodies; a crowded arena with blurry boundaries between policy-makers and the actors who influence them; and general uncertainty when people interact in unpredictable ways within a changeable policy environment. Complex government is difficult to understand, control, influence and hold to account. This article considers it from various perspectives: scholars trying to conceptualize it; policy-makers trying to control or adapt to it; and scientists, interest groups and individuals trying to influence it.

Complex government as a concept
For scholars, a key aim is to distinguish between the intuitive meaning of complex government, as big, complicated and difficult to understand, and the specific meaning of complex systems. Policy theory breaks down the intuitive idea into five key elements: actors, institutions, networks, ideas, and context. The task is to make a complex process simple enough to understand, by focusing on one or more elements.
When we focus on actors, we examine who they are and how they act. Actors can be individuals or collectives, including private companies, interest groups and governments bodies (Weible, 2014). The academic literature explores a shift from an early post-war period characterized by centralized and exclusive policy-making towards a fragmented multilevel system with a much larger number of actors. This development could change the meaning of 'policy-making', from an association with central government action towards a wider policy-making system containing more key players.
Things get complicated further when we compare 'rational' action with other explanations for behaviour. Most theories identify 'bounded rationality': people do not have the time, resources and cognitive ability to consider all information, possibilities, solutions, or consequences of their actions. They use informational shortcuts or heuristics to produce good-enough decisions (Simon, 1976). Actors may be 'goal-oriented', but also use emotional, intuitive and often unreliable heuristics associated with 'fast' thinking (Kahneman, 2012). For example, policy-maker attention may lurch dramatically from one issue to another, 'advocacy coalitions' may 'demonize' their opponents, and policy-makers may draw on quick, emotional judgements to treat different social groups as deserving of government benefits or sanctions (Kingdon, 1995;Baumgartner and Jones, 2009;Jenkins-Smith et al., 2014;Schneider et al., 2014).
When we examine 'institutions', we want to know the rules, norms, and practices that influence behaviour. Some are visible and widely understood-such as constitutions which shape other institutional activity, by establishing the venues where decisions are made, and the rules that allow actors to enter the policy process (Ostrom et al., 2014). Or, institutions are informal rules, often only understood in particular organizations. This wide definition allows us to compare formal understandings of how people should act, with informal 'rules of the game'. Crucially, different rules develop in many parts of government, often with little reference to each other. This can produce unpredictable outcomes when people follow different (often contradictory) rules when they interact; a multiplicity of accountability and performance management processes which do not join up; 'international regime complexity' when agreements, obligations and bilateral deals overlap (Alter and Meunier, 2009); or, a convoluted statute book, made more complex by the interaction between laws and regulations designed for devolved, UK and EU matters (Cabinet Office and Office of the Parliamentary Counsel, 2013).
When we identify policy networks ('subsystems'), we begin with the huge reach and responsibilities of governments, producing the potential for ministerial 'overload'. Governments divide responsibilities into broad sectors and specialist subsectors, and senior policy-makers delegate responsibility to civil servants. 'Policy community' describes the relationships that develop between the actors responsible for policy decisions and the 'pressure participants', such as interest groups, with which they engage (Jordan and Cairney, 2013). For example, civil servants seek information from groups. Or, they seek legitimacy for their policies through group 'ownership'. Groups use their resources-based on what they provide (expertise, advice, research) and/or who they represent (a large membership; an important profession; a highstatus donor or corporation)-to secure regular access to government.
In some cases, the relationships between policy-makers and pressure participants endure, and policy becomes the 'joint product of their interaction' (Rose, 1987, pp. 267-268). Consequently, we use the term 'governance' to describe a messy world in which it is difficult to attribute outcomes simply to the decisions of governments (Rhodes, 1997). Jordan et al. (2004) also use the term 'pressure participant' to remind us that 'lobbying' to government is not done simply by interest groups; the most frequent lobbyists are businesses, public sector organizations, and other types of government body at various levels of government. 'Multilevel governance' captures this messy process involving the blurry boundaries between policy produced by elected policy-makers and civil servants, and the influence of a wide range of governmental, non-governmental and quasinon-governmental bodies (Bache and Flinders, 2004).
When we focus on 'ideas'-a broad term to describe ways of thinking, and the extent to which they are shared within groups, organizations, and networks-we identify two main types. The first describes the ways of thinking that people accept to such an extent that they are taken for granted or rarely challenged (Cairney and Heikkila, 2014). The second is the more intuitive, 'I have an idea', meaning, which refers to the production of new ways of thinking, combined with the persuasion necessary to prompt other actors to rethink their beliefs. The policy process involves actors competing to raise attention to problems and propose their favoured solutions. Not everyone has the same opportunity. Some can exploit a dominant understanding of the policy problem, while others have to work harder to challenge existing beliefs. A focus on ideas is a focus on power: to persuade the public, media and/or government that there is a reason to make policy, and to keep some issues on the agenda at the expense of others (Bachrach and Baratz, 1970;Cairney, 2012a, p. 62).
'Context' describes a policy-maker's environment. It includes the policy conditions that policy-makers take into account when identifying problems, such as a political system's geography, demographic profile, economy, and mass behaviour (Cairney and Heikkila, 2014). It can refer to a sense of policy-maker 'inheritance'-of laws, rules, and programmes-when they enter office (Rose, 1990). Or we may identify events, either routine, such as elections, or unanticipated, including social or natural crises or major scientific breakthroughs and technological change (Weible, 2014). In each case, we consider if a policy-maker's environment is in his/her control and how it influences his/her decisions. In some cases, the role of context seems irresistibleexamples include major demographic change, the role of technology in driving healthcare demand, climate change, extreme events, and 'globalization' (Cairney, 2012a, pp. 113-114). Yet, governments have shown that they can ignore such issues for long periods of time.

Complex policy-making systems
Each of these five elements could contribute to a sense of complexity. When combined, they suggest that the world of policy-making is too complex to predict or fully understand. They expose slogans such as 'joined-up' or 'holistic' government as attempts to give a sense of order to policy-making, in the face of cross-cutting or 'wicked' issues, when we know that policymakers can only pay attention to a small portion of the issues for which they are responsible.
We can go one step further to describe government as a complex system. Complexity theory explains outcomes in terms of the 'whole' policy-making system: 'greater than the sum of its parts'. It identifies, in policy-making systems, the same properties found in complex systems in the natural and social world, including: 'non-linear dynamics' when some forms of action are amplified and others dampened, by positive and negative feedback; 'sensitivity to initial conditions', or the cumulative effect of early decisions and events; 'strange attractors' or regularities of behaviour despite the unpredictability of complex systems; and 'emergence' (Geyer and Rihani, 2010;Cairney, 2012b, pp. 124-125).
Many of these concepts can be linked to established policy concepts. For example, nonlinear dynamics are caused partly by bounded rationality and the tendency of policy-makers to ignore most issues and promote a few to the top of their agenda (Baumgartner and Jones, 2009). Sensitivity to initial conditions is the focus of historical institutionalism, which traces current institutions to the cumulative effects of decisions made in the past (Pierson, 2000). 'Emergence' refers to the systemic outcomes of interactions between people based on local rules, in the absence of central control (Cairney and Geyer, 2015). This may require some translation when we consider political systems. Although there is a well-established literature on 'bottom-up' implementation (from Lipsky, 1980to Nilsen et al., 2013, and central government does not control local policy delivery in an absolute sense, few of us would reject its role and influence on local outcomes entirely. Complex government as a challenge for policy-makers 'Complex government' can be used to reject the 'Westminster model'-which describes the concentration of power in the hands of a small number of people in central government-or a 'British political tradition' based on a top-down, 'government knows best' approach (Blunkett and Richards, 2011). Complexity theory invites us to consider a more realistic policy-making philosophy, and strategies including: relying less on centrally-driven targets and punitive performance management in favour of giving local bodies more freedom to adapt to their environment; trial-and-error projects, that can provide lessons and be adopted or rejected quickly; and to teach policy-makers about complexity so that they are less surprised when things go wrong (Room, 2011;Cairney, 2012b).
Yet, there is a profoundly important tension between the reality of complex government and the assertion of government control and accountability. For example, UK policy-makers have to justify their activities with regard to the Westminster model's narrative of accountability to the public via ministers and parliament (Rhodes, 2013, p. 486). We expect ministers to deliver on their promises, and few are brave enough to admit their limitations (until they leave government). Civil servants also receive training to encourage them to use management techniques to exert control over their policymaking tasks (Cairney, 2014a). Squaring this circle is not easy. Sanderson (2009) suggests that important strides have been made by the Scottish Government, which sets a broad national strategy, invites local bodies to produce policies consistent with it, and measures performance using broad, long-term outcomes. This is consistent with a Scottish system designed to contrast with Westminster culture, but important tensions still remain about the government's dual aim to encourage discretion and produce nationwide aims. We can also identify tensions (in case studies) in countries such as the US, where policy-makers present an image of strong performance management, partly to mask their frustrations with key organizations and a lack of implementation success (Radin, 2006).

Complex government as a challenge for participants and reformers
Most pressure participants have the same choice when seeking to engage with complex government: to bemoan and seek to reform, or to be pragmatic and adapt. This is a feature of the interest group world, in which we identify a tendency for groups to follow the action (Mazey and Richardson, 2006), often maintaining multi-level lobbying strategies, either directly or as part of networks (although the willingness and ability of groups to do so varies markedly-see Cairney, 2009;Keating et al., 2009;Keating and Wilson, 2014).
In contrast, we can identify in some scientific circles a naïve attachment to the ideal of 'evidence-based policy-making' in which we should seek to minimize the gap between the evidence-based identification of a problem and a proportionate government response (Cairney and Studlar, 2014). This idea relies on a concentration of power at the centre, and a direct link between scientists and elected policymakers. 'Complex government' prompts scientists to be pragmatic. First, they might adapt their strategy to help produce the dissemination of evidence throughout a messy policy process (such as by working with local governments, public bodies and stakeholders to 'co-produce' meaningful measures of effective interventions in particular areas). Second, they may recognize that policy-relevant knowledge is not just about the evidence of a problem; it also requires knowledge of how the policy process works and how any solution will fare (Cairney, 2014b).
Complex government also prompts us to consider how we can hold policy-makers to account if the vast majority of the population does not understand how the policy process works; if policy outcomes seem to emerge in unpredictable or uncontrollable ways; or the allegation of complexity is used to undermine popular participation or obscure accountability. The aim of political reformers, to go beyond representative government and produce more participatory forms of democracy, may solve a general sense of detachment by the political class, but it will not necessarily increase the transparency, and a popular understanding, of government.