States and modes of regulation in the global political economy

This paper develops a comparative perspective on the evolution of states in the global political economy and the emergence of the regulatory states in that context. It sets out to question the utility of the regulatory state model for understanding contemporary states, and does so from two platforms. The first is to consider the broad applicability and relevance of the regulatory state model, and to take issue with the contention that this is a useful framework within which to understand broadly the evolution of states in the different parts of the world. The second is to consider the assumptions on which associated notions of regulation rest, with particular reference to the equation of greater regulation with greater efficiency, effectiveness and transparency. It argues that the study of the political economy of regulation remains excessively dominated by Anglo-American and European – and fundamentally ethnocentric – perspectives, and concludes that these hinder our full understanding of the various modes of regulation that have emerged in recent years. Debates about the evolution of states in the contemporary global political economy have become strikingly tightly bound up with debates about regulation. In both ‘international’ and ‘comparative’ political economy, the so-called ‘regulatory state’ model has become the favoured framework within which to understand the forms of state transformation and adaptation unleashed by the constellation of economic and social processes we have come loosely to term ‘globalisation’. The basic argument is that these processes have acted – and are acting – to reorient the policy making functions of states towards a set of strategies designed to underpin markets, correct market failures, and facilitate market activity, and that these strategies crystallise around the notion of regulation. The aim, in essence, is to redefine the policy making process to the extent that it becomes rules-based and fundamentally technocratic rather than governed by political discretion. This process of reorientation is further conceived as involving a restructuring of the institutional apparatus of the state, and in the European and Anglo-American cases in particular this has involved constructing such institutional conditions as Central Bank independence, a clearer demarcation of bureaucratic responsibilities, and a proliferation of autonomous state agencies charged with administering the various dimensions of a system of regulation. What I want to do in this paper, however, is question the utility of this model for understanding contemporary states, and to do so from two platforms. The first is to consider the broad applicability and relevance of the regulatory state model, and to take issue with the contention that this is a useful framework within which to understand broadly the evolution of states in the different parts of the world. The second is to consider the assumptions on which associated notions of regulation, with particular reference to the equation of greater regulation with greater efficiency, transparency and, moreover, what Peter Burnham (1999) has called a ‘depoliticisation of economic management’. Again, my challenge to these assumptions rests on their evaluation in a wider comparative context. The conclusions from each of these endeavours indicate that the study of the political economy of regulation remains excessively dominated by Anglo-American and European – and fundamentally ethnocentric – perspectives. Before getting to that point, however, our first task is to set out some context by considering the broader debates about the evolution of states under the impact of globalisation, and the place of regulatory states within that conceptual framework. The following section considers the extent to which ideas about emerging regulatory states are accurate or useful in considering the evolution of states in various regions of the world outside the Anglo-American and continental European models of capitalism, and the third considers the ways in which the notion of ‘depoliticisation’ is useful in understanding the aims and the outcomes of a shift towards regulatory governance. Globalisation and the emergence of regulatory states We turn first, then, to debates about the ways in which globalisation has reconfigured the institutional structures of states and their roles and functions. In so doing, we are called upon to navigate a sea of contending perspectives. David Held and his colleagues (1999) have usefully identified three dominant schools of thought – ‘hyperglobalist’, ‘sceptical’ and ‘transformationalist’ – which offer a useful starting point. The gist of the first hyperglobalist argument is well-known and requires only brief re-statement here. It is that the nation-state is poorly adapted to the task of government or governance in the context of globalisation, and that state authority is progressively and decisively undermined by the structural power of markets. Primarily as a result of the ‘structural power’ of global capital (Gill and Law 1989), the autonomy and room for manoeuvre which states previously enjoyed in matters of economic policy has been severely curtailed. Deregulated financial markets and progressively integrated production structures are seen to dissipate national boundaries and render ‘national’ economic strategies increasingly anachronistic. Markets, in short, are now ‘more powerful than the states to whom ultimate political authority over society and economy is supposed to belong’ (Strange 1996: 4). The more extreme statements of the hyperglobalist argument have heralded the ‘end of sovereignty’ (Camilleri and Falk 1992), the ‘end of geography’ (O’Brien 1992), a ‘borderless world’ or the ‘end of the nation state’ (Ohmae 1990, 1995). More nuanced perspectives have perceived the ‘retreat of the state’ (Strange 1996) or the advent of ‘governance without government’ (Rosenau and Czempiel 1992), such analyses being concerned essentially with the withering of state authority as a consequence of progressively transnationalised forms of economic organisation, the heightened mobility of capital, and, moreover, the proliferation of non-state forms of authority ranging from international or supranational organisations to the various forms of ‘private authority’ (see Risse-Kappen 1995; Cutler, Haufler and Porter 1999; Hall and Biersteker 2002). It is important to note that the hyperglobalist terrain is inhabited not just by neoliberals who consider these processes desirable and beneficial, but also by their radical and neo-Marxist rivals which see them as representing the victory of an oppressive and exploitative global capitalism (Held et al 1999: 3-4). But either way, what we might call a ‘market triumphalism’ prevails across the hyperglobalist school of thought, the variation lying in the different receptions of it in the contending ideological camps. It is in this hyperglobalist perspective that we find the strongest validation of the hypothesis of ‘convergence’, which contends, in essence, that ‘all advanced industrial economies tend toward common ways of producing and organising economic life’, explaining any persistent differences in these modes of production and organisation as the result of ‘extramarket’ sources of distortion. (Berger 1996: 1). In its association with liberal globalisation discourses, and based on neoclassical understandings of trade and competition, the hypothesis holds that the universal availability of technological innovation to advanced industrial societies will generate convergent rates of productivity, accompanied inevitably by the emergence of common institutional configurations and patterns of state-society-market interaction (Boyer 1996). Concretely, convergence stems from the process by which high capital mobility renders expansionary fiscal and monetary policy strategies not only ineffective but also counterproductive. In part this is a consequence of a generalised preference for fixed exchange rates in which ideally monetary policy is removed from the range of policy instruments available to governments. In other part this is the result of the significantly more problematic implications of macroeconomic divergence in conditions of high capital mobility. With the globalisation of financial markets, expansionary fiscal and monetary policies generate significant pressures from markets and other governments for a defence or devaluation of the national currency, and consequently foreign reserves come under threat (Andrews 1994). High capital mobility furthermore generates a situation in which the ‘exit options’ of mobile capital asset holders are heightened, along with the ease with which they are exercised, and consequently the interests of these actors are elevated to primacy (Milner and Keohane 1996: 245). The hypothesis thus extends to areas outside these advanced industrial societies in the assumption that the pressures of globalisation and pressures from the international financial institutions (IFIs) will generate convergence on neoliberal forms of economic organisation, and thus on the institutional and social configurations necessary to sustain it. It is in opposition to the convergence hypothesis of the hyperglobalist school that the other two schools of thought have largely crystallised. The ‘sceptical’ school finds its ammunition in empirical evidence, which has been marshalled to advance primarily two arguments. The first is that the existence of globalisation is open to considerable question. Patterns of trade and FDI flows are taken to indicate that globalisation boils down to no more than a continued process of internationalisation, and indeed that the world economy was no more internationalised in the late twentieth century than it was during other historical periods (see Hirst and Thompson 1999). While such perspectives made an important impression on the debate in their insistence on understanding globalisation in historical context, the resulting presentation of globalisation as a ‘myth’ did end up obscuring many important facets of contemporary structural change. The second argument is rather more effective. It is that economic activity remains rooted primarily within the borders of national states and within the institutional order contained therein, thus casting a sceptical light on the prevalent assumptions of convergence. The message, in a nutshell, is that the global does not eliminate the national. Rather, governance structures remain fundamentally the product of national contexts and located within them. As John Zysman (1996: 180-1) puts it persuasively, ‘the national institutional foundations of the several market systems are neither washed away nor compressed into homogeneous convergence by growing international interconnections, or at least not yet’, going on to suggest that ‘increasingly in the years to come, the politics of trade ... will be about reconciling differently structured political economies that express different values’. Thus, consistent with the central arguments of both institutionalist and Marxist strands of the models of capitalism literature, there is no reason to imagine that divergence will become any less pronounced as a result of contemporary ‘globalisation’ than it has been at any other time in the history of social and economic development (see Crouch and Streeck 1997; Hollingsworth and Boyer 1997; Hollingsworth 1998; Coates 2000). The dividing lines between these latter perspectives and the third, ‘transformationalist’, school of thought are rather blurred, and indeed parts of the one melt quite nicely into the other. A transformationalist understanding of globalisation, which Held et al themselves favour, perceives it to be ‘a powerful transformative force which is responsible for a “massive shake-out” of societies, economies, institutions of governance and world order’ (Held et al 1999: 7, citing Giddens 1996). While this understanding might rightly be upbraided for a certain vagueness, it is nevertheless useful inasmuch as it accommodates notions of diversity, unevenness and complexity. In so doing, it insists that ‘the existence of a single global system is not taken as evidence of global convergence or of the arrival of single [sic] world society’ (1999: 7). Even more importantly for our purposes, it sees globalisation as ‘associated with new patterns of global stratification in which some state, societies and communities are becoming increasingly enmeshed in the global order while others are becoming increasingly marginalised’ (1999: 8). It thus provides a framework within which to challenge the notion of convergence on a single (AngloAmerican) model, but also the notion of convergence into a globalising world economy. Such a challenge is further facilitated by the coincidence of the transformationalist school with the ‘middle ground’ that has emerged in studies more specifically of the relationship between globalisation and states (see Boyer and Drache 1996; Amoore et al 1997; Evans 1997; Scholte 1997; Weiss 1998; Smith, Solinger and Topik 1999). In response to the contention that states are uniformly in retreat or decline or, moreover, buckling under the tide of convergence, the middle ground envisages states as undergoing processes of adaptation or transformation. It also restates an established argument (e.g. Helleiner 1994) that globalisation processes are embedded in the structures and actions of states, and conversely that states themselves are structurally integral to the creation and propulsion of globalisation itself. Taking this transformationalist middle ground as a broad framework, then, we can identify three key problems with the hyperglobalist school, which we can here re-cast as the ‘globalisation orthodoxy’. The first is that it assumes uniform outcomes from uniform structural pressures, and consequently a process of convergence on a uniform policy framework and institutional configuration across states and regions. The argument, in short, is an homogenising one which attempts to offer a single, condensed interpretation of how ‘the state’ is affected by globalisation and what the outcomes of that are and will be. The second and related is that it fails to recognise that not all states are engaged by globalisation processes. Contrary to homogenising interpretations of the ‘integrative’ aspects of globalisation processes, many states – particularly those of sub-Saharan Africa, for example – have been excluded from the economic transformations these processes have wrought and thus have become more, rather than less, marginalised from the global economy (Moore 2002). The third problem is that the arguments suggest that states are, generally speaking, in a process of decline. In the strong hyperglobalisation thesis, this might be the result of the dispersion of their authority and the eradication of national boundaries by the supposed globalisation of trade and finance; in the other perspectives outlined under the ‘orthodox’ heading, it is considered to be the result of complex constricting influences on states’ policy autonomy and the parameters of viable development strategies. Either way, trends underlying the contemporary structural context are perceived to exert an intrinsically negative impact on states and state capacities, to the extent that states are perceived now to be engaged in a battle to avoid their own obsolescence under the weight of transnationalising forces. Critical responses to these problems with the globalisation orthodoxy have been oriented primarily to disputing the twin contentions that states’ responses to globalisation are uniform and essentially convergent in their manifestations (including that states are uniformly integrated into patterns of globalising economic activity), and that states are by and large in decline as a result of associated processes. Taking the first of these, recent scholarship has mobilised a good deal of evidence to dispute these orthodox interpretations of the relationship between globalisation and economic policy and homogenising understandings of the sorts of policy landscapes that emerge as a consequence of the changes wrought by globalising processes. Geoffrey Garrett (1998), for instance, has demonstrated using data from the OECD countries that there remains a significant divergence among EU countries in fiscal and tax policies, and patterns of expenditure remain significantly more expansionary than would be anticipated by orthodox understandings of the sources of competitiveness. Moreover, neither the level of variation nor the persistence of expansionary fiscal policies have made these economies any less attractive to global capital, with the implication that conventional arguments in the globalisation debate about the severe circumscription of policy alternatives based on mechanistic understandings of the behaviour of capital are both misplaced and misleading. Other work on Europe, particularly that of Martin Rhodes (1999), has demonstrated in a similar vein that national governments remain able to pursue policy agendas consistent with a range of priorities (such as regulation initiatives and welfare and employment policies) that are not incompatible with the maintenance of an open economy. Similar evidence has been mobilised to demonstrate that ‘old’ forms of state-society compact are not necessarily washed away and that an enormous variety of state-society relationships continues to manifest itself, contrary to contentions that globalisation produces a uniform set of social relations. Rhodes goes on to demonstrate in the European context that ‘not only do open economies seem able to sustain strongly unionised environments, but corporatist bargaining structures remain important in those economies increasingly exposed to international forces’ (1999: 20). Such arguments are forceful and persuasive in the European context, but their conclusions are made as general contentions regarding the nature and implications of globalisation, and have been criticised for this. It has been pointed out, for instance, that it is only in those capitalist systems in which these styles of state-society linkages are already established that such forms of political management are viable: that ‘where social democratic institutions are in place, social democracy is viable; where they are not, neoliberalism is the best that can be done’ (Hay 2000: 146). The suggestion is thus that while globalisation has not generated a uniform pattern of social relations as predicted by the convergence hypothesis, and has not necessarily eroded the corporatist and social democratic characteristics of social relations in Europe, the spaces for innovation in institutional structures or state-society compacts are squeezed by the impacts of globalisation. Perhaps the more important point still is that the implications of globalisation look very different in Europe from in regions like Latin America or Asia, where foreign capital occupies a very different place in economic development processes, issues of credibility and competitiveness present much greater challenges, and other constraining conditions are put in place by the much greater prominence of multilateral agents such as the IMF and World Bank. Yet the challenges to the notion of convergence and uniformity, based on the European case or otherwise, are important here primarily for the ways in which they direct us again to the ‘middle ground’. They indicate that we need to seriously consider the reconfigurations of state forms and state-society linkages propelled by globalisation, at the same time as recognising that these processes of reconfiguration are mediated by existing institutional and social structures. Closely linked with these arguments disputing the contention of uniformity in responses to globalisation are those which dispute the idea that states are necessarily (or indeed uniformly) in decline. This element of the orthodoxy springs in good part from the theoretical licence afforded by its neoliberal foundations – that is, from the ability (and imperative) to distinguish clearly in analytical terms between the public and the private, or the political and the economic, but also to advocate their separation in ideological terms. In other words, the argument that states are generally in decline as a consequence of processes associated with globalisation is equally an argument that they should be in decline. What Peter Evans identifies as the ‘pervasive belief that the institutional centrality of the state is incompatible with globalisation’ (1997: 70) is thus essentially an ideologically-founded contention that this incompatibility is desirable in order to ensure the primacy of markets and maintain the constellation of social forces appropriate to sustaining their free and effective operation. Exposing the ideological underpinnings of the globalisation orthodoxy in this way lends itself to critical endeavours to dispute its central arguments regarding the contemporary evolution of states, especially given that much of the empirical evidence flies in its face. The European case is taken to indicate the ways in which both state corporatism and relatively high levels of social expenditure can be maintained under conditions of high capital mobility, contra the orthodox arguments which posit the retreat of the state from its welfare and spending commitments or from regulatory activity. Asian cases provide grist for the same mill, demonstrating that ‘developmental states’ have been and remain at the centre of the prevailing development model despite shifts towards a greater emphasis on the market in organising economic activity. A number of Latin American states also fit with these arguments in their continuing developmentalist orientations, despite their generally much closer association than in Europe or Asia with neoliberal pressures for fiscal discipline and traditionally much less robust states (particularly in terms of institutional capacity) than in either of these two regions. Such comparative empirical investigation, conducted largely from what we have identified as the transformationalist middle ground, does not necessarily contradict the idea that there might be certain common trends in processes of state adaptation to the demands of development under conditions of globalisation. Nor does it ignore the thorough-going changes and constraints that the contemporary world order poses for states. If there are generalisations to be made about the contemporary evolution of states, they are neatly captured in Peter Evans’ idea that the new ‘stateness’ is a ‘leaner, meaner kind of stateness’ (1997: 85). This for Evans is pursued primarily as a means of addressing the ‘capacity gap’ generated by states’ post-war assumption of unsustainable welfare commitments under the post-war regimes. The ‘social compacts’ between states and societies in which liberalism was embedded and made politically manageable – the core of Ruggie’s (1982) seminal concept of ‘embedded liberalism’ – were specific to the European context, this being the only part of the world in which meaningful welfare states emerged. But the notion of a capacity gap stemming therefrom is equally valid in, for example, the Latin American context of inward-looking industrialisation, in which states assumed very sizeable stakes in their countries’ productive structures, the running of the national economy, and the corporatist management of socio-political relations, bequeathing highly politicised state structures and hugely bloated and inefficient public sectors. Whichever the specific context, the movement towards a ‘leaner, meaner’ form of state is thus a response to the demands of competitiveness in the global economy and a response to the ‘overload’ on states seen to derive from Keynesian experiments in Europe and desarrollista (developmentalist) industrialisation projects in Latin America or, to a lesser extent, the developmental state model in East Asia. What is interesting, however, is that over the course of the 1990s there was a considerable shift in understandings, particularly in the IFIs, of what this leaner meaner stateness involves and how it might be achieved. In the early 1990s, the neoliberal discourse was dominated by the notion of ‘minimalism’ as the byword for an effective, efficient and desirable state. This was driven and characterised, as we have signalled already, by ideological shifts premised on a qualitatively different relationship between state and market from that which had prevailed over much of the post-war period. ‘Effective’ states were conceived in this ideological frame of reference as ‘nightwatchmen’ states, with little or no direct economic or social role. As the 1990s unfolded, however, this neoliberal orthodoxy was jettisoned even by its initial champions, the IFIs, as perceptions mounted of the failures of first generation reforms and, moreover, of the failure of ‘minimalism’ to redress many of the governance problems associated with states in ‘developing areas’ such as Latin America (see inter alia Inter-American Development Bank 1997). Corruption continued to be pervasive across the region, inefficiency and ineffectiveness continued to be characteristic of even the more rationalised bureaucratic structures, transparency and accountability remained in very short supply. Thus in Latin America, in George Philip’s characteristically pithy turn of phrase, ‘the problem is that a nightwatchman state can also be a biased state, and that this bias still matters’ (Philip 1999a: 76). This went together with the (rather churlish) acknowledgement of the ‘East Asian Miracle’, which came to be recognised as contradicting many of the earlier equations of economic growth with a minimal state (World Bank 1993). Around mid-1990s, consequently, the World Bank’s agenda of state reform was pushed around an important corner. The core concept to arise from this shift in thinking was that of ‘good governance’, which we flagged up in the discussion of neoliberalism in Chapter 4. It embodied the new belief in significant parts of the international financial community that economic growth was dependent not solely on an observance of the ‘rationalities’ of markets but also on the quality of governance thus defined – or, put another way, a recognition that a ‘quantitative reduction [of state apparatuses] might end up in qualitative deterioration’ (Welsch and Carrasquero 2000: 34). Perception of actual qualitative deterioration – that quantitative reduction had, in fact, resulted in this deterioration – was of course the key point of departure in this rethinking of the reform agenda. Its essence was consequently an acknowledgement of the centrality of national states and their importance in pushing forward the sorts of reforms deemed conducive to the improved governance of economies and societies. This inclusion of institutions and governance in the Bank’s reform agenda found clear expression in successive titles of its annual World Development Report – from The State in a Changing World in 1997 to Building Institutions for Markets in 2002. In this light the emphasis came to fall on the construction of an ‘effective’ state rather than on a minimal one, effectiveness now understood as involving on the one hand enhanced technical and institutional effectiveness, a special place being afforded here for the issue of judicial independence and the rule of law, and on the other a ‘corresponding liberal public sphere’ characterised by an increasingly empowered civil society (Williams and Youngs 1994: 93-4). Administrative, legal and institutional structures and practices thus had come to be seen as integral parts of the economic reform process, and state reform became oriented to the enhancement of states’ abilities to discharge a range of governance-related functions, understood mostly as the provision of key public goods. The revamped reform agenda, and the measures included under its rubric, have been captured under the general heading of ‘second generation’ reforms. The distinction between the first and second generations lies not only in their substance but also, moreover, in evolving ideas about the appropriate role and structure of states and their appropriate relationship with markets. Put simply, the first generation of structural reforms in Latin America and elsewhere was about the reduction of the size of national states and public sectors, the retraction of their involvement in economic activity, and the ‘rationalisation’ of their administrative and bureaucratic functions – entirely in tune with the ‘anti-statist’ logic of the neoliberal orthodoxy. Second generation reforms were premised on a very different set of objectives captured in notions of ‘rebuilding’ national states, constructing functioning and viable institutions, and strengthening states’ institutional and policy-related capacities. The important point about this transition, however, is that the Bank’s formulation of its governance agenda was predicated on a continuing overriding concern with markets: that is, institutions are of interest purely for their functions in creating the economic conditions in which markets could flourish and investments prosper, and for their potential utility in deflecting political impediments to effective liberalisation and deregulation. The substance of the second generation reform agenda reflects this continuing economism in the Bank’s approach to key development questions, and thus its continues to formulate its prescriptions primarily in technical terms which reflect a set of ‘rational’ criteria for the achievement of ‘effective’ and ‘efficient’ states. On this understanding, the objectives of so-called ‘second generation’ reform fall broadly into three categories (Pastor and Wise 1999): (a) ‘market-completing measures’ which aim to carry forward and/or complete the liberalisation processes initiated by first generation reforms; (b) ‘equity-oriented programmes’ relating to redressing the region’s pervasive distributional problems; and (c) good governancerelated ‘institution-building initiatives’ for the purposes of enhancing state and institutional capacities and drawing civil society into policy making processes. In more general terms, however – and here we (finally) get to the key issue at hand – the primary functions of states became increasingly seen as resting on regulatory activity, and the regulatory state model became deployed as both a useful descriptor of the functions and roles of contemporary states, and an ideological statement about what those functions and roles should be. As McGowan and Wallace point out (1996: 562), regulation is conceptualised by some (such as Majone 1997) as an ‘all-inclusive concept of governance’, but it is most often taken to refer to a particular type of policy making environment oriented to entrenching ‘mechanisms of control’. In this sense, building on our comments in the introductory paragraphs, it refers to a process by which economic management becomes ‘proceduralised’: it is characterised by an increasingly rule-based, juridical approach to the governance of the economic arena and a technocratic bias in decision-making processes, in which there is a greater emphasis on institutional self-regulation. The independence of Central Banks is the most obvious component of this latter dimension (Jayasuriya 2001: 102). The functions of such a state are largely seen to be two-fold: first, to underpin markets and second, to address market failures through the provision of various rights and goods (McGowan and Wallace 1996: 562). As such, the notion of the regulatory state has been developed in order to understand a situation characterised not by complete deregulation but rather also by key areas of re-regulation, particularly of financial markets (Gamble 2000: 114; Burnham 1999: 46). In other words, a disengagement of states from direct administrative responsibility for certain economic functions is accompanied by an increase in their regulatory functions. The classic case in this respect is the United States, in which neoliberal forms of economic organisation are accompanied by an extensive and penetrating web of regulatory provision. The regulatory state is thus fundamentally an enabling – rather than a planning or overtly interventionist – one, but should not be mistaken for a minimal one. Indeed, it is entirely possible that a regulatory state will have interventionist characteristics, but the point is that these are likely to be oriented towards underpinning rather than ‘replacing’ markets (McGowan and Wallace 1996: 563). Burnham’s concept of the ‘depoliticisation of economic management’ shares a good deal of common ground with the regulatory state model, but goes quite a way further towards considering the roots and the implications of such a trend, not just the basic attributes of this form of state. He suggests, in essence, that depoliticised economic management refers to a set of strategies which aim to ‘place at one remove the political character of decision-making’ (1999: 47). This is not to be confused with the elimination of political influence over economic activity (along the lines envisaged in the crudest early 1990s neoliberal rhetoric), nor the ‘wholesale evacuation of politics from policy making (1999: 44). Rather the act of depoliticisation itself is, according to Burnham, highly political, state managers ‘retain[ing] arms-length control over crucial economic processes whilst benefiting from the distancing effect of depoliticisation’ (1999: 47) and the overall upshot being an increase in government control through its consequent protection from the effects of politics on its decisionmaking processes. Thus he envisages a model of depoliticised, or ‘rules-based’ management that has three key pillars (1999: 44) which are worth citing directly as a framework for our discussion in the following pages: 1. The ‘reordering or reassignment of tasks’ from the party in office (operational independence; ‘managerial state’) 2. Increased accountability, transparency and external validation of policy (fiscal codes, IMF Code of Good Practice) 3. Acceptance of binding ‘rules’ limiting government room for manoeuvre (ERM, WTO Dispute Settlement Mechanism) The resulting form of governance, in this sense, is consistent with the mode of regulation envisaged in the regulatory state model – that is, one which is fundamentally oriented to facilitating market operations and limiting states to regulatory activities designed to correct market failures and enhance economic performance. This form of state and mode of economic management are thus sharply differentiated from the post-war welfare states of continental Europe, post-war corporatism in Europe and Latin America, developmental states in Asia, and variations on the developmental theme elsewhere in the world (such as Brazil). Modes of regulatory governance The argument that the regulatory state model provides the most instructive framework for analysing the evolution of states under the pressures of global restructuring is usually made in a spirit of generalised applicability. The ‘regulatory’ conceptualisation has been applied most frequently to European states (McGowan and Wallace 1996; Wilks 1996), the UK (Burnham 1999), increasingly to a number of Asian states (Jayasuriya 2001), and further to the Chilean case (Muñoz 1996, 2000), in addition to the staple concern with regulatory policy in the US public policy literature. In all of these cases, to a greater or lesser extent, there is evidence of most of the processes and strategies associated with the movement towards rules-based, proceduralised and juridical forms of economic governance – including, notably, Central Bank independence, extensive anti-monopoly and competition legislation, financial regulation and banking supervision, and so on. The question that subsequently presents itself, though, concerns the extent to which it can travel outside this relatively small handful of specific contexts, and the conclusion ‘with difficulty’ quickly presents itself. Of course, any model will stumble across exceptions to its criteria, and this is entirely to be expected. An argument that the regulatory state model does not apply in all cases and contexts is rather a facile and simplistic one, not to mention distinctly uninspiring. But a consideration of why it should be applicable only in a narrow set of cases, despite the frequently generalising way that its tenets are presented in the literature, opens up a set of much more interesting questions which, I suggest, invite a more careful consideration of key elements of the debate on the evolution of contemporary states in the context of globalisation. Stated baldly, the argument here is that regulatory states can only emerge in contexts characterised by an existing level of state capacity and a type of existing institutional configuration amenable to the development of these strategies, and thus that the study of regulation needs to pay significantly more attention to the interface of regulatory strategies considered ‘desirable’ for achieving the aims outlined above and the political and institutional environments in which they are intended to take root. At the very least, an independent and effective judicial system is a clear prerequisite, and the development of regulatory capacity involves a process of institutional reorganisation that involves extensive institutional innovation. In systems not possessed of such a judicial system, and not equipped (politically or technically) to engage in appropriate processes of state reform, an emphasis on regulation as a key function of states in the current structural context is very often meaningless. Let us try to put some flesh on these arguments by turning to the Latin American context. The best way into this is perhaps through an observation of experiences with the ‘good governance’-related state reform agenda, itself geared primarily to creating the relevant conditions for effective regulatory governance. In a nutshell, there has been a manifest and problematic breach between good governance-related visions of a desirable state and the historically entrenched characteristics of Latin American states, and consequently the externally-propelled state reform agenda has travelled into this regional setting with distinct difficulty. In one sense, this is because the broad institutional requirements for the successful implementation of governance reforms as formulated by the Bank are simply not in existence in the vast majority of Latin American states. This is not necessarily a function of the ‘under’-development of state institutions, although in some cases this is manifestly a problem. The real issue lies in a central inconsistency in the good governance agenda: that its success depends on a level of institutional and political capacity that frequently is only available to precisely the sorts of state that the agenda aims to dismantle or reconfigure. Most Latin American states have shown themselves to lack the necessary levels of institutional and political capacity for successful judicial reform, for example, or labour reform, in contrast with the patterns of capacity deriving in Asia from forms of ‘embedded autonomy’ (Evans 1995). At the same time, ‘state bias’, to recall Philip’s phrase, persists as their dominant and pervasive trait. Historically this bias has been essentially a ‘pro-elite’ one (see Philip 2003), which both reflects and stems from the other dominant characteristics of states in the region – clientelism as the organising principle of the institutional and political system, corruption, patronage politics, patrimonialism, personalism and presidentialism. Apart from their consequences for fiscal resources and overall macroeconomic stability, these traits – which, it should be said, are consciously generalised for the moment – have consistently obstructed the construction of democratically accountable, technocratic states which conform to the criteria of effectiveness and efficiency contained in the World Bank’s good governance agenda, and indeed to the types of regulatory state that are assumed to be of generalised applicability. They translate into the frequent absence of a technocratic reforming elite possessed of both the political capacity and the political will necessary to advance an agenda of this sort. The pro-elite bias, in other words, removes many of the incentives that would need to be present in order to there to take place a radical reorientation of the prevalent political culture – from a clientelistic, personalist and biased one to a technocratic, meritocratic and intrinsically accountable one. This feeds into the second key inconsistency in the good governance agenda: the sorts of state which have enjoyed higher degrees of reformist capacity in Latin America have usually been those possessed of a range of other attributes considered ‘undesirable’ in dominant understandings of what ‘good’ governance looks like, such as the very significant centralisation of power in the executive branch, corruption, clientelism, quasiauthoritarian styles of governance and the relegation of state institutions. In such systems the legislature is commonly marginalised, judicial independence is rare (as is Central Bank independence), and the development of transparent, accountable bureaucracies is significantly impeded. Put another way, any ‘effectiveness’ – understood as the capacity for the consistent implementation of reform initiatives – has traditionally derived precisely from conditions of institutional weakness. We stumble in this sense across an interesting anomaly: while the fragmentation of states may be useful in constructing governments’ capacities to implement a number of first generation reforms, at the same time it significantly impedes the meaningful reconstruction of the state and the development of the institutional capacity necessary for second generation reform. The historically important result has been something of a trade-off in Latin America between a state capable of reform and a democratically accountable one. While state effectiveness and accountability might be conjoined in theory, and indeed are so in the good governance agenda, they are by no means unequivocally linked in the Latin American setting (see Philip 1999b, 2003). In such a context, regulation has been a distinctly absent feature of Latin American states and their adaptation to the demands of globalisation. The regulatory state model has something to offer in light of its closeness of fit with neoliberalism: the emergence of new regulatory regimes has been encouraged by the policies of expenditure reduction allied to neoliberal preferences for market solutions (Wilks 1996: 541), and expenditure reduction has been vigorously pursued in the vast majority of neoliberal economies in the region. It is also somewhat helpful in light of the contention that the policy constraints of globalisation make regulation a more attractive policy framework than other forms of intervention (McGowan and Wallace 1996: 564), and rhetoric in Latin America has indeed emphasised regulatory governance as an key aspiration. The problem, though, is that the implementation of neoliberal strategies in South American countries has both preceded and outstripped the development of regulatory regimes and capacities, reflecting pervasive institutional weakness. Privatisation generally took place without previous preparation of the state to assume the role of regulator of competition, and consequently the politics of post-privatisation constitute less an example of a decisive shift to regulatory governance (as in the European and UK cases) than an example of the ‘chronic inability of Latin American states to fulfil a regulatory function’ (Fleury 1999: 76). This applies virtually across the board, with the possible exception of Chile. In Argentina, for example, mechanisms for the regulation of competition among privatised enterprises emerged only after the state had divested itself, in a notably rushed an haphazard manner, of its assets. Those regulatory bodies that did come into being only after privatisation had taken place were consequently burdened from their inception with myriad legal problems and litigation (Manzetti 2002: 8). There are a couple of instances of fairly effective and transparent regulatory bodies – notably the gas and electricity commissions – but the overall pattern has been one in which regulatory bodies are financed from sources laden with political connections, subject to intense lobbying by large firms, and with little autonomy from the executive branch, the latter intervening and trespassing frequently on the exclusive authority of regulatory agencies (Pastor and Wise 1998: 16; Manzetti 2002: 8-9). There have been some cases of more successful or ‘pre-emptive regulation, such as in the Uruguayan case, but here these have been largely as a substitute for privatisation due to widespread political opposition to eliminating public ownership. Governments in the 1990s sought instead to emphasise greater competition in the public sector, the budget law of 2001, for instance, providing for the demonopolisation of telecommunications and insurance, except for worker’s compensation insurance and basic telephony, and establishing regulatory bodies in the telecommunications and electricity sectors (U.S. Department of State 2002). Yet overall there was little serious political impetus from the government to significant reform of the state and public sector. And in general, serious regulation through competition policy has been either non-existent (as in the smallest, most bureaucratically underdeveloped states such as Paraguay and Ecuador) or largely ineffective (as in Argentina). With the exception of Chile, Central Banks have not been accorded significant independence. The institutional realignments and ‘operational independence’ associated with the depoliticisation of policy management thus find minimal expression. But the elements connected with external mechanisms of policy validation and the acceptance of binding rules for limiting government room for manoeuvre (see Burnham 1999: 49). Such mechanisms most obviously include agreements with multilateral and financial institutions. While these are important to establishing credibility as well as necessary financing, though, it should be noted that these mechanisms of external validation remain perceived more as signs of weakness than as signs of economic health or as manifestations of an overall ‘depoliticisation’ of economic governance. Mechanisms of external validation which have found rather firmer ground relate to the domestic implantation of a rules-based policy-making environment. The elaboration in Argentina of a code of fiscal responsibility (which its architects chose to call ‘fiscal convertibility’) was agreed in mid-1999, which legally established limits to the fiscal deficit (1.5% of GDP for 2000; 1% for 2001); stipulated that public spending may not exceed real GDP growth; set ceilings on public debt; and established a ‘Fiscal Stabilisation Fund’ to cover the contingencies of severe international or internal crisis. Similarly, in the Brazilian case following the devaluation, Cardoso declared that the policy priorities lay in the fiscal area, and proposals included plans for a fiscal responsibility law which would impose criminal penalties for over-spending and prohibit the federal government from bailing out states and cities in financial difficulty. Similar mechanisms have been put in place in Mexico, Chile and elsewhere, but are certainly not universal in Latin America. Apart from being politically charged, of course, the implementation of such laws of fiscal responsibility is complicated by the aforementioned levels of institutional weaknesses: at the very least, the successful installation of such policies requires a strengthening of audit tribunals (and their increased independence), as well as further reforms in the areas of pensions and taxation (The Economist, 6 May 2000), all of which have been especially problematic throughout the region, except – yet again, but even so only up to a point – in Chile where pensions reform has been held up a neoliberal beacon of best practice. The fact that the Chilean case is exceptional in so many ways relates primarily to the exceptional (in a comparative sense) nature of its state, and this brings us back to the key argument outlined above. Despite the significant concentration of power in the executive branch, consistent with most Latin American political systems, various bureaucratic arrangements in the Chilean system have afforded it a notable degree of institutional strength, cohesion and coherence, and inter-agency cooperation, which set the Chilean state apart from the other cases in the region. This is largely due to the strength of the General Secretariat of the Presidency which has been the principal mechanism of coordination within the executive branch and between the branches of government, largely through its Division of Inter-Ministerial Cooperation, and was granted ministerial status early in the 1990s. It also helps that there is a considerable degree of ideological consensus on the market model in the structures of the Chilean state, despite political problems in the 1990s in areas such as labour and tax reform. Despite the persistence of high levels of politicisation in the judicial system, furthermore, the judiciary is constitutionally and technically independent of both the executive and the legislature, and on matters of regulatory enforcement has been reasonably – although not fully – effective. The Chilean case has thus been characterised by a much more extensive and effective system of regulation than elsewhere in the region, maintaining, for example, a system of financial regulation and banking supervision from the time of the banking crisis in the early 1990s, a structure of capital controls from 1991 onwards, and a fairly effective regulation of competition in the private and public sectors. The point about the centrality of the institutional and political context is equally salient in other regions. In China, for example, mechanisms of external validation and constraint have been gradually adopted – most notably in this case through WTO membership, which has created considerable pressure for more adequate regulation in services and financial sectors. Yet the prospects for the elaboration of this sort of regulation are extremely thin given the strength of local governments and their political autonomy from (and defiance of) central government. The structure of local monopolies is extensive (at a very basic level making it difficult for producers to sell products in other regions of the country), local economies are utilised largely as the fiefdoms of local governments, and state banks are used by the latter as key sources of what has been called ‘parafiscal investment’ (see Breslin 2003). Thus pressure from the WTO for increased regulation in specific sectors of key concern runs immediately into fierce political opposition from the provinces, and moreover into an absence of a national and local institutional infrastructures appropriate for sustaining it. Central bank independence has not been a major issue in Chinese discussions about reform, and neither have privatisation nor the development of a judicial system capable of providing a basis for juridical forms of economic governance. In most of Africa, in addition, it makes very little sense to talk about regulation in the first place – across a swathe of the region states have either a very fragile and contested existence, or else are ‘failed’ or ‘collapsed’, or else have minimal levels of institutional capacity or bureaucratic development. South Africa is clearly the principal exception to this generalisation, but here again to talk of a functioning ‘regulatory state’ – or indeed the adoption of the regulatory state model – has very little foundation. In other parts of Asia, however, the model has found a somewhat firmer footing, and Kanishka Jayasuriya’s (1998, 2001) work has emphasised the movements towards Central Bank independence in such cases as Japan and Korea. Yet his work focuses almost entirely on Central Banks, and it would be a considerable stretch to claim that, even in the cases just mentioned, the other dimensions of the ‘regulatory state’ have 1 In other matters, such as human rights and military-related affairs, its track record has been considerably worse. 2 These comments are based on conversations with Shaun Breslin, and I am grateful for his ideas and insights. been put in place. Again, mechanisms of external validation have become important in the East Asian economies – notably binding agreements with the IMF, conformity with WTO regulations, and regional integration initiatives designed to entrench rules and procedures as the basis of economic management. Yet financial regulation and banking supervision remain weak, as demonstrated by the financial crises of the late 1990s, and most East Asian systems share with the Latin American region a prevalence of clientelism and presidentialism as the key features of their political systems, contrary to the stipulations and the aims of the regulatory state model. These (very) brief sketches indicate, then, significant problems with thinking about ‘a’ regulatory state model on the one hand, but on the other significant problems with deploying the model as a general framework within which to analyse the contemporary evolution of states in the context of global restructuring. Accountability, effectiveness and depoliticisation Similar reservations present themselves when we turn to core assumptions about the outcomes of a shift towards regulatory governance. In a nutshell, these are that a depoliticisation of economic management along the lines described by Burnham generates greater effectiveness, efficiency, accountability and transparency in policy making systems. The depoliticisation framework does not, as we have seen, connote some sort of benign process by which attributes such as transparency – by common consensus desirable ones, one would hazard – are acquired by a state and an economic system, in that the regulatory state is fundamentally about insulation from the ‘unpalatable consequences of politics’ (Burnham 1999: 44). But nevertheless the assumption is that rules-based economic management is an intrinsically more efficient system which does afford greater accountability and greater state effectiveness than other forms of governance. The ideological substance of such a claim is manifest. The point I wish to make here, however, is that while government and state agencies might be shielded from the ‘unpalatable consequences of politics’, these are not generally accompanied by a shielding of systems of regulation from other forms of political influence. That is, two situations have commonly prevailed in states outside the Anglo-American and European contexts which have moved towards some degree of regulation in economic management. The first is that the elaboration of regulatory policies has proceeded without the necessary institutional changes or institutional innovation, and thus regulatory strategies have remained governed by government and other non-independent agencies, as opposed to the sorts of independent, nongovernmental state agencies or para-statal bodies which have been associated with regulation elsewhere. The second, alternatively, is that regulatory bodies have been developed to oversee regulatory activity, but that these have remained highly vulnerable to political ‘capture’, especially by the executive branch, and highly politicised in their operations. Either way, it has frequently been the case that greater levels of regulation have not been accompanied by greater levels of accountability and transparency, and overall politicisation has meant that technical and technocratic efficiency levels have not been significantly increased. [Section to be expanded with empirical detail]