469C Bukit Timah Road Oei Tiong Ham Building Singapore 259772 Tel: (65) 6516 6134 Fax: (65) 6778 1020 Website: www.lkyspp.nus.edu.sg Lee Kuan Yew School of Public Policy Working Paper Series The Achilles Heel of Governance: Critical Capacity Deficits and Their Role in Governance Failures Michael Howlett Department of Political Science Simon Fraser University Canada and Lee Kuan Yew School of Public Policy National University of Singapore Singapore Email: [email protected] And M Ramesh Lee Kuan Yew School of Public Policy National University of Singapore Singapore Email: [email protected] April 05, 2014 Working Paper No.: LKYSPP14-03 Electronic copy available at: http://ssrn.com/abstract=2423166 Abstract This study assesses the usefulness of conceptions of policy capacity for understanding policy and governance outcomes. In order to shed light on this issue, we revisit the concept of governance, derive a model of governance types and discuss their capacity pre-requisites. A model of capacity is developed combining competences or skills over three levels of activities with analysis of resource capabilities at each level. This analysis is then applied to common modes of governance. While each mode requires all types of capacity if it is to be high functioning and match its theoretical optimal potential, most on-the-ground modes do not always attain their highest potential. Each mode has a critical type of capacity which serves as its “achilles heel‟. That is, without high levels of this specific capacity it is unlikely to perform as expected. While some hybrid modes can serve to supplement or reinforce each other and cover off gaps in critical capacity sectors other mixed forms may exacerbate single mode issues. Switching between modes or adopting hybrid modes is therefore a non-trivial issue in which considerations of capacity issues in general and Achilles heel capacities in particular should be a central concern. Introduction: Policy Capacity as the Achilles Heel of Governance Efforts at policy reform have been omnipresent in many developed and developing countries over the past several decades. Many of these efforts have featured waves of management reforms and administrative re-structuring, privatizations, de-regulation and re-regulation and the like (Ramesh and Howlett, 2006). ―Anything but the government‖, for example, has been a popular sentiment in public policy reform for at least two decades (Christensen and Laegreid 2008). These reforms can often be characterized as efforts to shift governance styles between different modes of governing (Treib et al 2007). Initially, for example, the sentiment behind many reform efforts and coalitions in the 1980s and 1990s favoured transitions from government service delivery and regulation to more market-based types of governance regimes. Similarly in more recent years the tilt has shifted towards transitions from hierarchical and market forms of governance to more network-oriented governance relationships (Lowndes and Skelcher 1998; Lange et al 2013; Weber, Driessen and Runharr 2011). 2 Electronic copy available at: http://ssrn.com/abstract=2423166 Even more recent efforts at reform in many countries and sectors have sought to correct for or reverse excesses in ‗de-governmentalization‘ from this era, often introducing hybrid elements into existing governance modes (Ramesh and Howlett 2006; Ramesh and Fritzen 2009). Many proponents, for example, claim ‗collaborative governance‘ combines the best of both government- and market-based arrangements by bringing together key public and private actors in a policy sector in a constructive and inexpensive way (Rhodes 1997).1 Many key sectors from health to education and elsewhere now feature elements of either or both hierarchical approaches – regulation, bureaucratic oversight and service delivery – as well as both market and network-based hierarchical and non-hierarchical approaches such as markets, voluntary organizations, and increasingly co-production (Brandsen and Pestoff 2006, Pestoff 2006 and 2012; Pestoff et al 2006). Not all of these reforms have been successful (Ling 2002) and whether and how well such different modes of governance perform, we argue, is based in large part on their capacity requirements or pre-conditions (Howlett 2009). That is, each form of governance requires a high level of state and actor capacity in order to function effectively (Bullock et al 2001). Whether such capacity exists and how it is mobilized is a significant but little understood factor affecting the effectiveness and efficiency of any single governance mode (Canadian Government 1996). In order to shed light on this issue, we revisit the concept of governance, derive a model of governance types and discuss their capacity pre-requisites. A model of capacity is developed which combines competences or skills over three levels of activities with analysis of resource capabilities at each level. This analysis is then applied to each main mode of governance and hypotheses related to key capacity needs is set out. As the analysis will show, each mode requires all types of capacity if it is to be high functioning and match its theoretical optimal potential. However while most on-the-ground modes do not always attain 3 Electronic copy available at: http://ssrn.com/abstract=2423166 their highest potential, each mode has a critical type of capacity which serves as its ―achilles heel‖. That is, without high levels of this specific capacity it is unlikely to perform as expected. While some hybrid modes can serve to supplement or reinforce each other and cover off gaps in critical capacity sectors others mixed forms may exacerbate single mode issues. The consequences for policy making and governance reforms are then spelled out. Governance Modes in Theory and Practice: Ideal Types, Hybrid Forms and Their Performance Governing is what governments do: controlling the allocation of resources among social actors; providing a set of rules and operating a set of institutions setting out ‗who gets what, where, when, and how‘ in society; and managing the symbolic resources that are the basis of legitimacy (Lasswell 1958). In its broadest sense, ―governance‖ is a term used to describe the mode of government coordination exercised by state actors in their efforts to solve familiar problems of collective action inherent to government and governing (Rhodes, 1997; de Bruijn and ten Heuvelhof, 1995; Kooiman, 1993 and 2000; Majone 1997; Klijn and Koppenjan, 2000). That is, „governance‟ is about establishing, promoting and supporting a specific type of relationship between governmental and non-governmental actors in the governing process. In modern capitalist societies this means managing relationships with businesses and civil society organizations also involved in the creation of public value and the delivery of goods and services to citizens (Hall and Soskice 2001). Governance thus involves the establishment of a basic set of relationships between governments and their citizens. Although early models such as Pierre (2000), for example, distinguished between two only modes - state-centric ―old governance‖ and society-centric ―new governance‖ – and many economists similarly compared and contrasted only two types of ―market‖ and ‗hierarchical‖ relationships (Williamson 1975) - even with just these three 4 basic sets of actors and relationships governance arrangements can take many shapes (Treib et al 2007). Other significant modes have been proposed by others such as Peters (1996), Considine and Lewis (1999), Newmann (2001), Kooiman (2003) and Cashore (2002) including types such as community-based ‗network‘ governance and pure ‗private‘ governance with little if no state involvement and which operate on ‗network‘ as opposed to hierarchical or market-based relationships. As Steurer (2013) suggested, the three basic governance actors can be portrayed as a interacting within a set of inter-related spheres of activity (see Figure 2) generating at least four ideal governance types (see Figure 1). Figure 1 – Logic of Ideal Types of Governance after Steurer Government(cid:9) Legal(cid:9) Market(cid:9) Governance(cid:9) Governance(cid:9) Network(cid:9) Governance(cid:9) Private(cid:9) Governance(cid:9) Civil(cid:9)Society(cid:9) Business(cid:9) Source: Modified from Steurer 2013 5 Beyond these ‗ideal types‘, however, other studies also identified a range of intermediate or ‗hybrid‘ governance modes or styles existing between the two ends of the state-society or state-market spectrums put forward by Pierre and Williamson (Bevir and Rhodes 2003; Rhodes 2007). Considine (2001) for example, proposed four ideal types of ‗public‘ governance styles by distinguishing between ―legal‖ governance and ―corporatist‖ governance while including market governance and network governance ideal types in his model (see Figure 2). Figure 2 – Ideal Type Modes of Governance following Considine Form of State Control Prime Service Key Procedural Mode of Central Focus of Overall of Delivery Tool for Policy Governance Governance Activity Governance Aim Governance Mechanism Implementation Relationships Legality - Promotion Legislation, Legitimacy - Legal Rights - Property, Courts and of law and order in Law and Voluntary Governance Civil, Human Litigation social relationships Rules Compliance Controlled and Specialized and Management - of Targets - Corporatist Balanced Rates of Privileged Major Organized Plans Operational Governance Socio- economic Advisory Social Actors Objectives Development Committees Prices - Competition - Regulatory Contracts Resource/Cost Controlling for Market Promotion of Small Boards, and Efficiency and Externalities, Governance and Medium sized Tribunals and Regulations Control Supply and Enterprises Commissions Demand Networks of Subsides and Relationships- Co-Optation of Governmental and Expenditures on Network Promotion of Inter- Collaboration Dissent and Self- Non- Network Governance actor organizational Organization of Governmental Brokerage Activity Social Actors Organizations Activities Source: Modified from Considine (2001) Such distinctions proved useful in distinguishing between two different modes of state-centric governance found in European and Anglo-American systems. However the logic of hybridity is more extensive than put forward by Considine since different combinations of government, civil society and businesses exist and within a combination different sets of 6 actors have different ‗strengths‘. When variations on the strength of each actor in a governance relationship is included, this stretches to at least a dozen types (see Figure 3). Figure 3 – Nuanced Model of Modes of Governance Including Variation by Actor Strength 1,(cid:9)2(cid:9)=(cid:9)Strong(cid:9)and(cid:9)Weak(cid:9)Legal(cid:9)Governance(cid:9) 3,4(cid:9)=(cid:9)Strong(cid:9)and(cid:9)Weak(cid:9)Market(cid:9)Governance(cid:9) 1(cid:9)=(cid:9)HN(cid:9) 5,6(cid:9)=(cid:9)State(cid:9)and(cid:9)Societal(cid:9)Corpora st(cid:9) 2(cid:9)=(cid:9)NH(cid:9) Governance(cid:9) 7,8,9,10(cid:9)=(cid:9)Network(cid:9)Governance(cid:9) 3(cid:9)=(cid:9)MH(cid:9) 11,12(cid:9)–(cid:9)Strong(cid:9)and(cid:9)Weak(cid:9)Private(cid:9)Governance(cid:9) Government(cid:9) 4(cid:9)=(cid:9)HM(cid:9) 5(cid:9)=(cid:9)HMN(cid:9) (Hierarchy)(cid:9) 6(cid:9)=(cid:9)HNM(cid:9) 7(cid:9)=(cid:9)NMH(cid:9) 8(cid:9)=(cid:9)MNH(cid:9) 9(cid:9)=(cid:9)NHM(cid:9) 1(cid:9) 10=(cid:9)MHN(cid:9) 4(cid:9) 11=(cid:9)MN(cid:9) 2(cid:9) 12=(cid:9)NM(cid:9) 3(cid:9) 5,(cid:9)6(cid:9) 7,8,9,10(cid:9) 11,(cid:9)12(cid:9) Civil(cid:9)Society(cid:9) Business(cid:9) (Network)(cid:9) (Market)(cid:9) This added complexity might appear at first glance to make it difficult to assess the nature of success and failure in each mode and this initial insight is largely correct. However, some simplifications can be made. First, as Cashore (2002) noted, pure non-governmental governance arrangements such as those found in certification and private standard setting schemes are not in the public realm and need not always concern governments and those studying governmental or ‗public‘ policy-making (see also Cutler et al 1999). Second, as Capano (2011) argued, in most cases of bilateral or trilateral governance arrangements the presence of hierarchy is overpowering and nuances pertaining to, for example, market and network leadership in bilateral legal and market regimes are less significant than in trilateral 7 arrangements where governments may be dominated by the presence of both market and network actors. That is, network and corporatist forms of governance modes exist as hybrids which are based on different mixes of coordination principles (hierarchy, market and network) (Meuleman 2010). Thus, as Howlett et al (2009) and Tollefson et al (2012) noted, the substance of difference between governance modes lies in their plurilateral or hierarchical nature and state or society-centric nature. How these two factors relate to the main types of public governance relationships described in Figure 3 is set out in Figure 4 below. Figure 4 – Matrix of Typical Public Governance Modes by Central Actor and Mode of Co-ordination Significance of State Role Higher Lower (Hybrid) Corporatist Hierarchical Legal Governance Governance Central mode of Co- ordination of Actors (Hybrid) Market Governance Plurilateral Network Governance As Figure 4 shows, government is not just one of the possible actors in a governance mode but is often the central player either potentially or actually, whether it chooses to play that role or not. Even in the more extreme horizontal or plurilateral arrangements governance modes need to be ‗steered‘ or led towards constructive, positive coordination. That is, the hierarchical government role in any governance arrangement may vary considerably, and change, but remains a core determinant and element of governance, rather than something existing in opposition to, or outside of, it. Government has the inescapable task of defining what governance is, or can be, and may choose to allow a higher degree of freedom to other policy actors with regard to the goals to be pursued and the means to be employed (Capano 2011). 8 Hence the performance of any of the public governance modes set out in Figure 4 – that is, discounting the purely private forms outlines in Figure 1 - is affected by the ―governance capacity‖ that the central actor enjoys: that is, by governments ability to exercise its role in the level of co-ordination that governance mode entails. In those modes of governance which feature an extensive government presence, the nature of that government‘s ability to steer or control governance relationships is a critical component of that mode of governance‘s operation and propensity to fail. The government‘s capacity to make and implement policies in particular is thus a key requisite of the effective operation of any mode of governance (Wu and Ramesh 2014). At the extreme, for example, any effort to hollow out of ―central government‖ weakens the ability of a government to undertake a ―command and control‖ approach to policy-making and hence also undermines the ability of any legal or market form of governance dependent on such instruments to function effectively. Governments (conceived as central political institutions) may find themselves overburdened with economic problems or social demands so that ‗hierarchical governance‘ – that is, a policy framework whereby the most important actors are governments and the state implements policies by ordering and sanctioning – may no longer prove to be an efficient or effective form of governance. Hybrid corporatist and ‗network‘ governance modes in which the size and resources of the state typically greatly outreach those of other network actors and thus can continue to dominate those actors in any governance relationship also encounter such dynamics and public policy performance issues (Osborne 2006). Governance Modes and Their Propensity to Fail: Defining Actor Capabilities and Competences in Policy Capacity The policy capacity of a governance mode is hence a key indicator and requisite of governance success. The term describes the preconditions a government requires in order to 9 make sound policy choices and implement them effectively in achieving its potential to steer a governance mode. This is a broader definition than the widely-used one offered by Painter and Pierre (2006) who focus their attention on capacity for policy formulation rather than both formulation and implementation in their definition of the term as: ―… the ability to marshal the necessary resources to make intelligent collective choices, in particular to set strategic directions, for the allocation of scarce resources to public ends.‖ Theirs is an unduly restrictive definition, as policy capacity is not only about the ability to formulate and make policy choices but also to implement them and evaluate their performance. Policy capacity is at heart a function of three competences or skills which affect the ability of governments in their relationships with other governance actors: analytical ones which allow policy alternatives to be effectively generated and investigated; managerial ones which allow state resources to be effectively brought to bear on policy issues; and political ones which allow policy-makers and managers the room to manouevre and support required to develop and implement their ideas, programs and plans (Wu et al 2010; Tiernan and Wanna 2006; Gleeson et al 2009; Gleeson et al 2011; Fukuyama 2013; Rotberg 2014). These skills or competences are crucial to policy and governance success. However they also rely on the their availability and the availability of adequate resources to allow them to be mobilized. Resources or capabilities must exist at the individual level which allow individual policy workers (Colebatch 2006; Colebatch et al 2011) and managers (Howlett and Walker 2012) to participate in and contribute to designing, deploying, and evaluating policies. It includes not only their ability to analyse but also to learn and adapt to changes as necessary. Resources must also be available at the level of the organization. These are aspects of the structure and make-up of policy-relevant organizations that affect their members‘ ability to perform policy functions come. Organizational features that unduly circumscribe individual decision capabilities or morale among policy workers, for example, can undermine 10