Understanding contemporary policy mixes

Policy design is ‘ubiquitous, necessary and difficult’ but surprisingly little studied and understood (Bobrow, 2006). Over the past three decades it has received some treatment in the existing policy literature, but not as much, or in as much detail, as is necessary (May 1981, 1991 and 2003; Weimer 1992; Bobrow 2006). Within the policy sciences it has been linked to studies of policy implementation and policy instruments (May 2003) and to those of policy ideas and policy formulation (Linder and Peters 1990; James and Jorgensen 2009), but without systematic attention being paid to such basic elements as the definition of key terms and concepts. In addition, it has been a large, if typically implicit, part of a more recent trend towards the study of governance, and even ‘meta-governance’, but again, without the benefit of clear and systematic analysis (Meuleman 2009; 2010).

Despite the vagueness and uncertainties currently associated with its principles and elements, the purpose and expectations of policy design have always been clear. Policy design has been conducted by policy actors in the hope of improving policy-making and policy outcomes through the accurate anticipation of the consequences of government actions (Tinbergen 1958; 1967; Schon 1992). It thus is situated firmly in the ‘rational’ tradition of policy studies, aimed at improving policy outcomes through the application of policy-relevant and policy specific knowledge to policy-making processes, specifically in the crafting of alternative possible courses of action intended to address social, political, economic and other kinds of policy problems (Cahill and Overman 1990; Bobrow 2006).

While somewhat similar in this regard to activities such as planning and strategic management, policy design is much less technocratic in nature than these other efforts at ‘scientific’ government and administration (Forester 1989; Voss et al. 2009). However, it too is oriented towards avoiding many of the inefficiencies and inadequacies apparent in other, less knowledge-informed ways of formulating policy, such as pure political bargaining, ad hocism, or trial and error (Bobrow 2006). In general, it is less specific than planning in developing alternatives rather than detailed ‘plans’, fully acknowledges the uncertainty of the future and the contingent nature of policy outcomes (Voss et al. 2009) and is more open than strategic management to the idea that there are alternative sources of knowledge and design criteria than those residing in, or proposed by, experts (Fischer and Forester 1987; May 1991).

May (2003) argues that rather than treating design as simply a technical activity of finding the best design, it should be seen to involve channelling the energies of disparate actors towards agreement in working towards similar goals. In this sense, policy design contains both a substantive component – a set of alternative arrangements potentially capable of resolving or addressing some aspect of a policy problem, one or more of which is ultimately put into practice – as well as a procedural component – a set of activities related to securing some level of agreement among those charged with formulating, deciding upon, and administering that alternative. It thus overlaps and straddles both policy formulation and policy implementation and involves actors, ideas and interests present at both these stages of the policy process (Howlett, Ramesh and Perl 2009).

Understanding contemporary policy designs

Owing to this purported shift in governance contexts – the presence of more flexible economic and political circumstances than have existed in the past (Lenihan and Alcock 2000) – contemporary policy designs in many advanced countries, it has been argued, have changed. In many cases they are argued to have become more indirect and subtle, and often much less visible than was previously the case (Rhodes 1997).

It has been argued in many circles that in response to the increased complexity of society and the international environment, governments in many countries in Western Europe, in particular, have turned away from the use of a relatively limited number of traditional, more or less command-and-control oriented, ‘substantive’ policy tools such as public enterprises, regulatory agencies, subsidies and exhortation, and begun to increasingly use their organizational resources to support a different set of ‘procedural’ tools (Klijn and Teisman 1991; Peters 1998) such as government reorganizations, reviews and inquiries, government–NGO partnerships and stakeholder consultations. These all act to guide or steer policy processes in the direction government wishes through the manipulation of policy actors and their interrelationships, and constitute a ‘new governance’ system (Bingham et al. 2005).

These processes and theories continue to challenge public administrators, managers and scholars (Peters 1996; Peters and Pierre 1998; Knill 1999) but have generally been found to have less grounding in actual patterns of tool use and the evolution of policy designs than their proponents allege. As the discussion in Chapters 5–8 has shown, even if some sectors have been moving in the direction of network governance, the relationship existing between governance modes and policy tool categories is far from one-to-one and, moreover, other trends exist in other sectors featuring other kinds of governance activities and preferences. And, as the discussion in Chapters 2–4 revealed, the reasons why particular instruments are selected extend beyond macro-factors such as globalization to others associated with expert discourse and historical experience in the process of policy formulation.

Students of public policy-making in many countries have thus begun to move beyond simple macro-level models and theories of policy design and have developed a renewed interest in the meso- and micro-aspects of policy formulation and the investigation of the ways in which governments actually propose and utilize the multiple different types of policy instruments available to them (Goggin et al. 1990; Dunsire 1993). Properly assessing, inventorying, categorizing and modelling procedural policy instrument use and choice has thus become a prerequisite to understanding the evolution of policy designs. And this is just as true for practitioners as it is for theorists and academics.

Effects of internationalization on policy design

In order to assess the actual effects and impact of environmental changes on state behaviour, it is necessary to acknowledge that serious gaps exist in our understanding of the workings and characteristics of globalization and its policy consequences (Hay 2006). Contrary to what is commonly believed and often advocated, in our global era the domestic state remains far from overwhelmed and lacking autonomous decision-making capacity (Weiss 2003; Braithwaite 2008) and the source of many of the changes in the patterns of policy-making and instrument choice found in contemporary society very often lies in the domestic rather than the international arena (Scott et al. 2004; Levi- Faur 2009). Domestic states, be they national or sub-national, do not just react to changes in their international environments but also are very much still involved in the design and implementation of policies expected to achieve their ends (Lynn 1980; Vogel 2001). And, to the extent that global factors have had an impact on domestic policy designs and governance practices, it is often through what can be termed more ‘indirect’ and ‘opportunity’ effects spilling over from trade and other activities, rather than from the ‘direct’ effects that advocates of design alterations typically cite in arguing that state behaviour must change (Howlett and Ramesh 2006).2

That is, there is little doubt that economic, diplomatic, military and aid related relations among nations help shape many choices of policies and policy tools. It is well known, for example, that many countries are aggressive in pressuring other countries to weaken regulations or preferential tax or subsidy treatments that restrict international firms’ business activities and global and regional multilateral agreements are the most direct ways by which extra-territorial factors shape the choice of policy instruments. But these exist only in very few sectors, and often have large areas of exclusion even when they are present. Trade and investment agreements such as the World Trade Organization (WTO) agreements and the North American Free Trade Agreement (NAFTA), for instance, often specify in great detail the measures that governments can or cannot adopt vis-à-vis domestic and international producers. But, powerful as these treaties are, they only prohibit the use of a small number of very specific instruments such as tariffs and quotas to assist domestic producers, and even then they often contain exceptions to those bans. The use of subsidies of various kinds specifically intended to assist domestic producers, for example, is restricted by countervailing duty clauses and other similar measures, but activities in the cultural and agricultural realms are usually excluded from these measures.

Similarly, more general political agreements, whether formal or informal, can also have a constraining direct effect on the choice of policies and policy tools. The European Union is an extreme case of the formal transfer of decisionmaking authority to a supra-national centre, for example, which often severely limits what national governments can do and the instruments they can employ to effect their decisions (Kassim and Le Galès 2010). However even here the restrictions on national governments’ abilities to employ regulatory and fiscal tools on their own are often less significant than often assumed (Halpern 2010). And, even when they are, national governments are often able to craft their own specific solutions to ongoing policy problems with little regard to EU policy through mechanisms such as ‘subsidiarity’ or the ‘Open Method of Coordination’ which allow local states to determine and design their own policy responses to EU-level initiatives (Meuleman 2010; Heidbreder 2010; Lierse 2010; Tholoniat 2010).

Thus, while there is no doubt that the evolution of these kinds of international treaties and arrangements is an important development, in many sectors and areas of government activity they impose only very minimal or no constraints on the choice of policy tools utilized by governments; nowhere near those alleged by both proponents and opponents of globalization-led promarket reforms (see especially Palan and Abbott 1996; Clark 1998; Weiss 1999; Bernhagen 2003). There are few international agreements that specifically require governments to privatize or deregulate, for example, and the sectors which have experienced the deepest deregulation and privatization in recent decades – financial, telecommunications and air transportation services – predate the General Agreement on Trade in Services (GATS) negotiated under the WTO umbrella.

Hence, to date, the actual impact of globalization on domestic state policy designs is much less than often alleged, outside of several well-known sectors and events. The direct international constraints on policy designs typically cited by proponents of the globalization thesis have largely been confined to cross-border economic exchanges and do not cover much of what governments do and how they do it. Traditional command and control instruments of governance, such as regulation by government or independent regulatory commissions, state-owned enterprises and direct taxation and subsidization, are far from antithetical to globalization as is evident in their continued, and at times even increasing, use in policy designs in a variety of national and sectoral settings (Jayasuriya 2001, 2004; Vogel and Kagan 2002; Jordana and Levi-Faur 2004; Ramesh and Howlett 2006).

Effects of networks on policy design: ‘government to governance’

Similarly, the need to shift toward the greater use of network management tools and activities put forward by adherents of the argument that states have moved ‘from government to governance’ as a result of changes in society and the way in which government interacts with it, is also lacking a great deal of empirical evidence (Howlett, Rayner and Tollefson 2009; Schout, Jordan and Twena 2010). While it is clear that the development of modern information and communications technologies have had a serious impact on the way in which individuals and organizations interact and organize themselves in contemporary societies, it is not clear that these developments have had an equally direct effect in altering traditional governance practices or policy designs (Hood 2006; Hood and Margetts 2007b).

Governing involves the establishment of a basic set of relationships between governments and their citizens which can vary from highly structured and controlled to arrangements that are monitored only loosely and informally, if at all. In its broadest sense, ‘governance’ is a term used to describe the mode of coordination exercised by state actors in their interactions with societal actors and organizations (de Bruijn and ten Heuvelhof 1995; Kooiman 1993 and 2000; Rhodes 1996; Klijn and Koppenjan 2000). ‘Governance’ is thus about establishing, promoting and supporting a specific type of relationship between governmental and non-governmental actors in the governing process.

Changes in governance modes entail both alterations in the abilities of various state and non-state actors to prevail in policy formulation disputes and decisions, as well as shifts in the choices of policy instruments used to implement public policy (Scharpf 1991; Weaver and Rockman 1993; March and Olson 1996; Offe 2006). But much is unclear about the application of the concept of governance to considerations of policy design.

Typical management activities related to network modes of governance are those which affect network creation, recognition, capacity-building, and content creation or alteration. Robert Agranoff (Agranoff and McGuire 1999), for example, has observed that in a typical ‘network management’ situation ‘the primary activities of . . . [a] manager involve selecting appropriate actors and resources, shaping the operating context of the network and developing ways to cope with strategic and operational complexity’ (21). Such definitions of network management activities, however, are very vague and their design implications unclear.

Many early proponents of the idea of increased ‘networkization’ simply expected governance arrangements to shift evenly away from sets of formal institutions, coercive power relations and substantive regulatory tools found in hierarchical systems towards more informal institutions, non-coercive relationships of power and a marked preference for procedural instruments and soft law in more plurilateral systems (Dunsire 1993; Kooiman 1993). However, the possible variations in governance types and outcomes are a good deal more complicated once the possibility of sectoral variations is taken into account (van Kersbergen and van Waarden, 2004; Meuleman 2009).

Mark Considine and his colleagues, for example, have investigated these arrangements and linkages and identified four common governance arrangements found in modern liberal-democratic states which they relate to specific policy foci, forms of state-society interactions and overall governance aims (see Table 1.1).

While a change in governance mode involving a shift from hierarchical, imperative state-led coordination to ‘steering’ through reflexive self-organization (‘plurilateralism’) is basic to the idea of a shift from ‘governing’ to ‘governance’, such a shift may be present in only some sectors and also represents only one axis along which different modes of governance can be located (Harrop 1992; Cerny 1993; Pontusson 1995; Daugbjerg 1998; Haas 2004; Zielonka 2007). As observers such as Knill and Lehmkuhl have noted, governance arrangements feature different relative strengths of the public and private actors involved (Knill and Lehmkuhl 2002; Jordan et al. 2005) and vary according to whether these relationships are expressed in formal or informal terms (Treib et al. 2007; Kritzinger and Pulzl 2008).3

These dual dimensions of governance arrangements cover the four modes of governance Considine identified (see Table 1.2) and suggest ‘network governance’ is only one such possible arrangement in any given governing circumstance (Heidbreder 2010). The existence of several possible alternative types or ‘modes’ of governance existing simultaneously at the sectoral level suggests a more complex picture of governance arrangements, instrument choices and policy designs than the one-to-one national level, multi-sectoral shift towards the use of more participatory and less ‘command and control’ tools often proposed by adherents of the ‘government to governance’ thesis (Koppenjan and Klijn 2004; Barnett et al. 2009; Esmark 2009; Hysing 2009; Edelenbos et al. 2010; Hardiman and Scott 2010; Schout et al. 2010).

Moving beyond globalization and networks

The recognition of the continued vitality of the state in a globalized environment along with the existence of multiple modern sectoral modes of governance suggests a more subtle and nuanced account of policy design trends and influences is required than is typically found in the discussions about instrument use and policy design which have flowed from many current studies of globalization and network governance. However, many existing debates about policy design often continue to fixate on the impact of these two processes and, as a result, much of the existing discussion of policy tools and policy design is characterized by misinformation, ideological predilection, and unnecessarily polarized position taking.

Fortunately, other scholarship on policy design and policy instrument choice has offered pathways to a better understanding of instrument selection and policy design by grounding it more carefully in empirical studies and in more nuanced and sophisticated analyses of policy-making practices and activities (Bressers and O’Toole 1998; de Bruijn and Hufen 1998; Van Nispen and Ringeling 1998). As the discussion in this book will show, embedding the discussion of policy design in the Procrustean bed of globalization and network theory is not a useful way to advance thinking on the subject. Understanding contemporary policy design requires an effort to develop a more nuanced understanding of the policy formulation and implementation activities of governments than is provided by adhering to either or both of the ‘globalization’ and ‘government to governance’ hypotheses. A more detailed and systematic understanding of the kinds of policy choice open to governments and of their ability to choose specific combinations of policy tools in their efforts to create and manage public policy-making is needed to advance our understanding of policy design (Ingraham 1987).