Theorising Privatisation: Policy, Network Analysis, and Class


Fran M. Collyer
[email protected]


In the last two decades of the twentieth century many governments across the world have displayed an unprecedented attraction toward policies which encourage and allow the marketisation of the public sector and the privatisation of public assets and services. While the financial and social impact of this policy shift has been empirically documented, theoretical development regarding possible underlying causal processes remains in its infancy. This paper explores current theories of the policy process of privatisation from a sociological perspective, and proposes a potentially useful direction for future policy analysis.

In the 1980s Australian policy followed other nations (particularly Britain) in participating in what Hood described as ‘an international policy boom’ (1994:4) and reversed previous commitments to a large and healthy public sector. Within the space of a few years, policy shifted from a reliance on bureaucratic regulation and the use of public enterprise as mechanisms for national development, to a new focus on cutting government budgets, self-regulation, de-regulation, competition, privatisation and the marketisation of the public sector.

Evidence of this policy reversal in the Australian context is mounting, with studies documenting the impact on wages, jobs and conditions (eg. Herd, 1998); on citizenship (eg. Salvaris, 1995), on accountability and the provision of information to the consumer (eg. Carter, 1998:17); on the fiscal health of the state (eg. Quiggin, 1996; Walker and Walker, 2000); on welfare (eg. Saunders, 1994; Nevile, 2000); and differentially across the spectrum of social groups (eg. Langmore ,1987).

Policy Studies and the Policy Process

There have been however, far fewer attempts to analyse this policy reversal, to define its characteristics, explain its origins, and theorise its global spread. In part, this lacuna is the consequence of the comparatively recent emergence of the trend, for although privatisation, commercialisation, and other state-private sector relationships are not new, they have seldom previously been entered into with such vigour nor been so widespread (see Collyer et al., 2001). However this theoretical ‘black hole’ may also stem from problems within policy studies itself. Despite the central importance of policy to the everyday life of the citizen, to the structure and functioning of institutions, and to the exchange relations between countries, the theory of policy has been a peculiarly undeveloped area of scholarship. In particular, sociology, like anthropology, has in many ways failed to contribute to the study of policy making and policy processes (for a discussion of the potential role for anthropology, see Shore and Wright, 1997), though it must be acknowledged that sociology has invested heavily in exploring the impact of both social and public policy, particularly on vulnerable social groups.

This sociological deficit in policy studies is reflected in the way privatisation has become defined in public discourse. Though there remains some contention over the concept (Thynne, 1995:3-4), the prevailing view is of privatisation as an economic or fiscal technique of government, fundamentally concerned with the transfer of assets or activities from the public to the private sector (eg. Domberger and Piggott, 1986:145). While there is much truth to this definition, it remains an extremely narrow and one-dimensional view, with an exclusive focus on financial transactions and exchanges. More pertinent, at this point, is that this mainstream definition assumes a particular kind of policy process in operation, in which privatisation has been a rational, intentional, coherent, policy response to a fiscal problem. In the typology devised by Dudley and Vidovich, this conception of policy would be akin to the ‘rational comprehensive model’ in which the best course of action and the public interest is determined through a rational, linear, neutral process of collective decision making (1995:16-18).

The rational perspective also presents policy making as a process which relies upon, and is the application of, rational techniques such as the gathering of empirical data, the systematic clarification of aims and objectives, cost-benefit analysis, economic modelling, and impact simulation. In the policy literature, many documents offer ‘flow charts’ and diagrams setting out the ‘stages’ of privatisation, often beginning with feasibility studies and options papers and moving through to legislative and organisational changes before the sale process begins (for an example, see Walshe and Daffern, 1990:90-3). Such analyses often confuse the ideal case with the actual processes, and assume policy makers will, and have, learned from past experiences. For example Domberger states that ‘Governments now have sufficient experience in the sale of public sector assets to avoid unintentional underpricing’ (1995: 44). There is also little evidence that policy makers use rational techniques or existing research evidence when making policy choices (Jansson, 2000:47) particularly in the case of privatisation (Kay and Thompson, 1986:31; Collyer, 1997; Collyer et al., 2001).

This rational perspective on policy is highly problematic. It is steeped within a positivist epistemology, proceeding on the premise that there is a readily identifiable problem which can be addressed through policy (Bacci, 1999:17). Yet this is how privatisation is widely presented and understood. For instance, at a time when the claims by pro-privatisation advocates were at their most articulate in Britain, the Financial Secretary to the Treasury stated ‘The privatisation program is coherent, and well thought out’ (Moore, 1986:93). Claims continue to be made that privatisation will address a very diverse range of ‘problems’ such as economic inefficiency and lack of innovation in the public sector, political ‘interference’ in the management of enterprises, unemployment, trade union power, welfare dependency, ‘big government’, and the high cost of public services – to name just a few. Within the rationalist perspective on policy, the basis of these claims are rarely questioned. Indeed policy is often presented as a rational response to ‘market forces’ or ‘market problems’ (eg. Bradley and Neja, 1989:12). Sociologically of course, social and economic ‘problems’ are not fixed properties of society but socially constructed, and there is a concern with how only specific issues are recognised as a ‘problem’ and placed on the political agenda (Edelman, 1987).

The Politics of Privatisation

Although the orthodox, economic definition of privatisation is virtually hegemonic, other discipline areas have sought to develop alternative understandings of privatisation. For example Feigenbaum and Henig (1994), Gormley (1994), and Chamberlin and Jackson (1987) see privatisation as fundamentally a political phenomenon rather than economic, administrative, or fiscal. From this perspective, privatisation is a policy strategy used to shift power between various constituencies, alter the structural capacity of government and its responsibilities, realign and restructure decision making and institutions, and/or create new interest groups and classes (cf. Henig et al., 1988; Samson, 1994:90).

These analyses throw light on the political aspects of the policy process, and generally avoid many of the pitfalls of the rationalist approach by acknowledging the diverse aims of government and the range of interest groups that have input into the policy process. In particular, they avoid the fallacy of assuming policy to be the consequence of a normative consensus between key individuals, and expressing a shared set of societal goals to which all, particularly the state, aspire. This normative perspective of the policy process is one that has come to us from TH Marshall’s (1981) conception of the welfare system, in which policy is the end product of an harmonious interaction between market economy, government and welfare institutions, each defined by their distinctive value-realm. This normative focus is extremely limited (Mishra, 1982:106). Without a theory of interest group struggle and competition, Marshall cannot explain the dominance of some value positions, and it problematically assigns values and goals to organisations and institutions. Despite these faults, the normative approach is prevalent, with many studies presuming that privatisation policy is selected in order to achieve key ‘goals’ such as equity, efficiency, reliability and/or consumer choice (see for example, Wanasinghe, 1991; Waters, 1987; Gormley, 1994:230).

As already stated, an alternative to the consensus approach is to probe deeply into the nature of the nation state and the policy process. Here it is proposed that policy is a fundamentally political activity, that politics are not confined to the final, implementation stage of the policy cycle (Dalton et al., 1996:106,118). Furthermore, the state is itself an arena of political activity. It is not conceived as a neutral mechanism through which the will of the people is expressed. Instead government often fails to act in the interests of the people as a whole, may act in its own interests, and policy outcomes generally accord with the interests of the more powerful members of society. This allows us to understand why privatisation policies might be adopted by governments even where rational economic theory suggests it is not always the best financial or economic solution; why privatisation is used even though the proceeds of a sale may not cover more than a few years of dividends from the public enterprise; why it is used even when the newly private firm demands a costly indemnity against previously incurred risk; and why it is used where the costs of regulating the new company clearly outweigh tax revenues returned to government.

This political approach acknowledges that power and resources are not equally distributed across society, and that competition between these groups does not in itself prevent the concentration of power within segments of society (cf. Bachrach and Baratz, 1977). Instead power is unevenly distributed, and, despite the democratic processes of government, well organised groups are able to influence public policy according to their interests (eg. see Dahl, 1984). When used in empirical studies of privatisation, this theoretical approach is able to reveal the multitude of interest groups which shape the policy process. For instance Kay and Thompson found in their study of privatisation in the UK that the most influential interest group in the decision making process is the senior management of the potentially privatised enterprise (1986:29), and Gupta reveals that those who gain most from privatisation are the political leaders, bureaucrats and the newly emergent capitalists (1996:46). Not surprisingly perhaps, this group of studies tends to be highly critical of government rhetoric and pro-privatisation discourse, disputing claims of greater cost efficiency, reduced administrative costs, or a reduction in government outlays (eg. Gupta, 1996:46-7).

Policy and Pluralism

Despite its strengths, this approach to policy via the pluralist perspective has its (well documented) shortcomings. One of these is the way it places an emphasis on policy as the outcome of the action of key individuals. Here the underlying causes of privatisation can be located in the differing motivations of key actors, each of whom has a specific goal and the means to make policy through attracting supporters or by strategically altering the ‘rules of the game’ so that other actors must reassess their interests and strategies (eg. Feigenbaum and Henig, 1994). Where it is used simplistically, pluralism may portray ‘governments’ as having human characteristics such as ‘desires’ and ‘objectives’. For example:

‘Since the government was eager to privatize … [and] fearful of being accused of “selling out” to the private sector …’ (Campo-Flores, 1994:241).

‘Although governments in developing countries are eager to increase the efficient use of the resources that they have at hand …’ (Vernon, 1987:7).

More sophisticated uses of pluralism generally avoid such reductionism, though they do have the tendency to conflate the various concepts of ‘motivation’, ‘intentions’, ‘objectives’ and ‘interests’, and fail to maintain a distinction between ’objectives’, which are publicly announced by the government in power, with ’objectives’ that are analytic constructions imposed by the theorist (eg. Ernst, 1994:81,83; Whitfield, 1992:129). Pluralist studies also remain open to the charge of offering a ‘one dimensional’ perspective on power. This is the very criticism levelled by Lukes (1974), who reminds us that policy rarely conforms to the intentions of particular individuals, but results from collective forces and social arrangements. Power is multifaceted, yet pluralism focuses only on observable acts of power, ignoring the way in which action can be indirectly manipulated by the creation of a set of circumstances that favour one set of preferences over another. Pluralist perspectives ultimately ignore the impact of underlying social arrangements upon which the struggle for dominance is played out. Applying Lukes’ exposé on power to privatisation policy, Samson argues pluralism focuses only on policy decisions that are on the political agenda and fails to investigate how agendas are narrowed to prevent the consideration of alternative proposals (1994:81). Thus we can see that a pluralist conception of the policy process fails to offer an adequate analysis of privatisation because it reduces power to the interaction between public figures, revealing only the ‘public face’ of power. As such it is in danger of over-rationalising the policy process, presenting the outcome as the consequence of rational, intentional, and conscious choices of individuals and groups.

Policy and Functionalism

Interestingly, when we seek to theorise policy change (ie. the rise in popularity of privatisation), rather than explain the policy process itself, privatisation research is often implicitly functionalist rather than pluralist in orientation. Here specific policy initiatives are thought to result not from particular group interests and pressures but as a consequence of becoming ‘functionally necessary’ for societies that have reached a ‘certain stage in their development’ (Hall et al., 1975:6). Here privatisation policy is conceived as an inevitable response to a societal ‘need’ or prerequisite. Although it is rarely explained how this ‘need’ came to be recognised (or by whom), the policy literature is replete with functionalist examples. Some imply that privatisation has become functionally necessary as a consequence of normative or cultural prerequisites, while for others it has been an ‘inevitable’ response to new economic circumstances. For instance, Vernon proposes that privatisation occurs when the ‘time is ripe’ (1987:2); and Foster contends that privatisation is a response to ‘the need to reduce public expenditure and public borrowing in a period of severe economic crisis’ (1994:493; see also Gupta, 1996:42; Flora and Heidenheimer, 1984).

Privatisation as a result of economic ‘need’ is the more common of the functionalist explanations. This is perhaps understandable given the prevailing emphasis on privatisation as an economic or financial transaction. The idea here is that as a consequence of the global and national economic downturn after the oil crisis of the early 1970s, governments found it increasingly difficult to continue their support for an extensive welfare state and a large public sector: and were thus ‘driven’ to privatisation and other market oriented ‘reforms’ (eg. Franzway et al., 1989:24). This version assumes a) that policy is a function of economic circumstances, b) that all countries were facing the same ‘problems’, and c) that they all adopted privatisation as the ‘solution’. Yet the national differences are vast. For example, where British public enterprises were restrained from capital expansion, Australian enterprises were relatively efficient and endowed with flexible borrowing regimes (cf. Steel and Heald, 1982:337-4; Productivity Commission, 1998). Thus while some argument might be made that Britain was forced to privatise, this ‘need’ was not part of the Australian experience. Moreover, such a proposition assumes privatisation to be a mechanism for reducing government outlays and is thus the ‘best’ policy option in a time of scarcity. As yet there is no conclusive evidence that privatisation has decreased government outlays, or that privatisation is a more cost efficient means to provide essential services. This argument is discussed more extensively below, but it is necessary at this point to note the tautological problem inherent within this perspective: that is, the idea that we can argue a particular social practice occurred because of the outcome that it subsequently produced (or claimed to produce).

Although functionalism has been criticised for this logical fault in regards to many areas of social life, it appears to have been rarely tackled in social policy: possibly there is an assumption that policy practices are different from other social practices, enabling us draw an immediate and inevitable causal link between intention, strategy and outcome. Sociologically this proposition cannot be sustained. Policy practice, like other social practices, needs to be understood as historically produced and socially contingent. Thus an answer to the puzzle of privatisation cannot be founded upon financial or other system imperatives, nor upon claims based upon the subsequent outcomes of policy initiatives.

Policy and Ideational Theories of Change

Among the alternatives to pluralist and functionalist theories of privatisation, are ideational theories of policy change and policy spread. Here we find suggestions that privatisation results from an intellectual disillusionment with the Keynesian approach (eg. Hall, 1993; Bell, 1997:25), a response to the fall of the command economies such as the USSR (eg. Gupta, 1996:39), the cross-national adoption of ‘policy fashions’ (eg. Ikenberry, 1990), and the rise of doctrines of economic rationalism and neo-classical economics (eg. Ernst, 1998; Self, 1990:19, 21).

There is much to be said for a theory of privatisation and policy change which depend, at least in part, on the role of ideas, values and beliefs. A new ‘paradigm’ has indeed emerged in which privatisation is portrayed as a politically neutral process of economic exchange. Within this paradigm there is a new moral imperative to remove public and collective forms of ownership, and to seek social progress and efficiency through market mechanisms. This package of ideas and beliefs, like all ideas and beliefs, are influential because ‘they structure the terms of political discourse’ (Bell, 1997:24). However, it is difficult to find empirical support for an ideational theory of change: as even when there was strong presidential support for ‘small government’ during the 1980s in the USA, there were few asset sales (see Henig et al., 1988:477,450). It is also difficult to build a theory of privatisation around the role of ideas. An ideational theory of policy change offers an explanation only in terms of a policy ‘fashion’, implying an innocent or accidental pattern, as if other policy initiatives might equally have emerged. Missing from such a conception is a sense of the social context that shapes the ‘reservoir of ideas’ from which policy makers draw, and within which problems and opportunities for action occur. Thus in order to explain the systematic emergence of privatisation in countries that are economically, culturally, and politically diverse, ideas must be linked theoretically to social practices and social structures.

Policy, Ideology, and Discourse

Two concepts which potentially link the two realms are ‘ideology’ and ‘discourse’, with the latter tending to be a fashionable replacement for the former. A handful of studies have emerged recently which imply that privatisation is the result of new (or revived) discourses or ideologies concerning the role of the state and the relative merits of the private and public sectors. For example Michael Pusey suggests there has been a ‘fundamental change of orientation towards civil society and culture’ (1991:224) and elsewhere it is suggested that privatisation is the result of the adoption of ‘fashionable’ ideologies (eg. Walker and Walker, 2000; Mansfield, 1987). These observations are not without foundation. It has been noted that rather than a rational and careful consideration of the evidence, many studies of privatisation merely assert the greater efficiency of the private sector, and fail to offer compelling evidence to support their generalisations (cf. Mansfield, 1987:75; Heald, 1983:308). It is not difficult to find evidence of ‘simplistic assumptions’ and ‘assertions’ of this nature within the literature. For instance:

‘A state entity cannot be as efficient as a private entity in the production of the same output. We know that’ (Waters, 1987:55).

‘The generalization that private is more efficient than public is a reasonable enough assumption …’ (Vernon, 1987:7).

‘… government-owned enterprises perform poorly, provide an inadequate service or product, and serve as a burden on the taxpayer’ (Campo-Flores, 1994:235-6).

These and similar claims are endlessly repeated by advocates of privatisation, often with little evaluation of the evidence for the relative efficiencies of the two sectors. Although there are methodological difficulties with the comparative evaluation of such claims (see Collyer 1996), well designed and independent studies show that neither sector is inherently more efficient (eg. Martin and Parker, 1997; Goodman and Loveman, 1991:35). What most comparative studies of the two sectors generally fail to emphasise however, is the different functions of the private and public sectors, and the failure of the private sector to perform all previous functions of the divested public sector unit, leaving the state the least profitable areas to assign to other units within the bureaucracy (and without the capacity to fund these activities through cross-subsidisation). Such an analysis should render the private/public comparison invalid. However the new pro-privatisation paradigm has been so effective in its message that any current defence of the public sector is considered illegitimate, ‘out of date’, and utterly misplaced. Rationally produced evidence which stands contrary to the paradigm is rejected and it is clear that a process is operating in which the truth, or the possibility of discovering the truth, is being systematically obscured.

For some, this process is regarded as ‘ideology’, for others it is ‘discourse’. Unfortunately, these concepts are rarely explicitly defined, and often embedded within an ideational concept of social change where they are assumed to refer to a set of ideas or beliefs, or to the communication of these ideas and beliefs. Indeed the words are often used inter-changeably with ‘rhetoric’ or ‘doctrine’ (eg. Shore and Wright, 1997:18; Agger, 1998:8,19,23). These definitions propose that discourses and ideologies have no material or observable existence: there is only the text itself. This prioritising of ideational phenomena makes it difficult to adequately theorise the relationship between language and lived experience, and moreover, social change cannot be explained. Only within Marxism is ideology defined differently. Here it is a set of beliefs which serves the interests of its proponents at the expense of other social groups (see for example, Hughes et al., 1995:38). This definition separates ideologies from other common or prevalent sets of ideas, because they are fundamentally connected to the activities of elites and interest groups as these seek to secure their own interests (cf. Marx and Engels, 1976:67).

Policy and Marxism

There are only isolated examples of research on privatisation which use the concept of ideology in its fully theorised, Marxist, and specifically Gramscian sense, where its definition includes the customs, practices and behaviours which reflect, consciously or unconsciously, ideas about society and the nature of reality (cf. Navarro, 1982:46). Here, ideology is the ‘organic cement’ which does not simply express the reality of the economic base, but is constituted by both social practices and institutional arrangements, and thus bridges both structure and superstructure (Daly, 1999:69). One of these studies is Samson’s (1994), where this conceptual framework is used to critique the prevailing, economic definition of privatisation. Sampson regards this definition as part of an ideology of privatisation, in which people are coerced into accepting a particular set of policy strategies. Although these policies are ‘harmful to some social groups’ this facet of privatisation is conveniently obscured by another: a view that privatisation is a politically neutral, economic transaction. Beyond this however, privatisation is more than the mere sale of public assets or the contracting-out of services, it also involves public sector and welfare cuts, trends toward intensive market commodification, a new emphasis on self-provisioning, and the increased domestic burden being placed on women and families. Samson argues that these aspects of privatisation are kept hidden from the public, as a broader and more political definition of privatisation would have ‘politically negative consequences’, revealing the full range of likely effects and exposing the yawning gap between the beneficiaries of the policy and those who are harmed by its introduction (1994:88-94). Thus for Samson, privatisation is an ideology which is not just about ideas but is fully realised in social practices and institutions.

It is only when the Gramscian version of ideology, or the Foucaultian version of discourse, are used, that we can successfully theorise policy change. If either is simply a system of, or communication of, ideas, then there is no explanation for how these ideas change over time. It is only when we use these concepts to refer to relationships between ideas, practices and structures, that we can see that ideas and policies change because they are always part of an ongoing struggle: that there is always conflict and dissent and a struggle for meaning and dominance. And this social activity produces change – or perhaps rather, it is the process of change in action.

Despite its appeal, few studies have examined policy change from this perspective: that is, as a consequence of class conflict and dissent. Indeed the concept of class – once a major tool of sociology – has currently fallen out of favour. Not only has it become unfashionable to use class, but there is a particular reluctance to use it as an analytic rather than descriptive tool in policy analysis (Mooney, 2000:157,161-2).

Policy and Capitalism

As a consequence of the demise of class analysis, the more common ‘conflict’ approach is to examine policy actors in terms of interest groups, often referring to them as ‘stakeholders’ (eg. Gupta, 1996; Chamberlin and Jackson, 1987; Henig et al., 1988) or as ‘players’ (eg. Kerr and Savelsberg, 2001:25). In most cases, these studies employ a pluralist framework in which ‘class’ denotes the characteristics of a group of similar individuals: they rarely refer to class as an element of a social system. This means they are unable to explain possible links between the interest groups and other developments in the social system (such as globalisation). Policy is presented as an adhoc process of struggle between diverse interest groups and taken up by the many national policy communities as a new policy ‘fashion’ and coincidently imitating the policies of other, apparently, successful nations. They do not show the possibility of privatisation as a ‘class project’, in which government restraints on the market system are systematically eased or removed, and where government activities are depoliticised by transferring them to the market. Nor do they show how privatisation policy might alter the power balance between classes: such as in Britain where privatisation was directed at curtailing the power of groups such as the medical profession, coal miners and steel workers. This is because the balance of power between classes is not altered by changes in power between different interest groups. Thus they are clearly unable to demonstrate how privatisation maintains a system of class inequality, by taking from the majority of the population their collective access to, and ownership of, public resources, while simultaneously enriching already privileged groups.

Where class is used in an analytic or relational sense, these problems do not emerge. Notable here is the analysis of privatisation by Samson (1994). Samson sidesteps the typological problem presented by pluralism, where privatisation is a class policy only where there has been an ‘intention’ to alter class boundaries and where these projects are ‘undertaken with little thought to ideology’ (Feigenbaum and Henig, 1994:195, 200). Samson proposes instead that privatisation is always about class. He defines privatisation as ‘a cultural and political mission’, in which a package of ideas about individual liberty and consumer choice is intertwined with strategies and techniques for the management of the national economy. This package of ideas and practices alters established relationships between both groups and, importantly, between classes, and as such it can be understood as an ‘hegemonic project’ aimed at altering consciousness and power relationships (Samson, 1994: 79,90).

Samson’s thesis on the class basis of privatisation rests on an interpretation of Marxism in which the motivations of individual policy actors, and the characteristics of these actors (eg. social background, level of education or past career choices) play little part. Instead this approach concentrates on identifying key groups according to the ‘objective effects of their actions’ (eg. Therborn, 1982:229), and emphasises the structural and institutional restraints placed on these policy actors rather than on their capacity for agency (see also Deacon and Mann, 1999). It is a level of analysis which explains ‘macro-level factors that can play a major role in shaping the environment in which individuals behave and in which policy choices are made’ (Bell, 1997:41).

This approach is however, problematic in the way it entirely collapses the individuals and groups within the policy making community into the institution of the state. It assumes that politicians choose privatisation because the state has a commitment to the maintenance and reproduction of capitalist social relations: that the interests of the policy makers and the state are the same (Mooney, 2000:165). The reductionism of this approach has been extensively discussed elsewhere (see Hay, 1999:164-71, on the Miliband-Poulantzas Debate), and does not need to be revisited here. Nevertheless some further points need to be made, given that the recent global shift toward privatisation can only be explained within a coherent theory of the policy process, and specifically, the policy process within capitalist societies.

From an orthodox Marxist position the state is merely an ‘agent’ of capital, and policy is fully determined by class interests. This is the Althusserian interpretation of Marx, and it relies on the view that capitalism has a hidden structure that generates, in a mechanistic way, its forms and surface movements (Jessop, 2001b:3). It is a concept of ‘structural causality’ wherein the economic base of the society determines the political system, and thus the realm of policy. In its favour, this framework offers a strong challenge to pluralism with its plethora of unrelated stakeholder groups each acting to secure their own interests, where policy can be reduced to individual intentions and motivations, and where policy outcomes show little overall patterning or systemic bias. It also offers a challenge to neo-statism, where the state is presumed to be an autonomous entity, fully able to pursue its own interests (eg. Dandeker, 1990; Skocpol, 1985); and to autopoietic theories of the state, in which the state is portrayed as a self-producing, self-referential, self-controlling political system (cf: Jessop, 2001a:9). Nevertheless, this orthodox Marxist position is untenable because of the way it limits the role of policy makers to ‘agents of capitalism’, collapses their actions into the logic of the whole, and theorises the policy process as a direct outcome of the logic of capitalist accumulation.

Fortunately there is an alternative Marxist interpretation of the policy process. Here there is a rejection of economic reductionism, and a re-focusing on the ‘complex interplay of structure and agency’ (Mooney, 2000:162). This is an interpretation of that within Marxism which is, Daly argues, ‘already postmodern’ (1999:65). Marx tells us in the Grundrisse that everything ‘that has a fixed form’ appears merely as a ‘vanishing moment’ in the movement of society:

‘The conditions and objectifications of the process … are themselves equally moments of it, and its only subjects are individuals, but individuals in mutual relationships, which they equally reproduce and produce anew’ (1973:712).

This passage highlights the idea of agency as fundamental to any notion of social structure, and points toward a solution in the way it integrates structure and superstructure, economy, state, culture and ideology into a relational, rather than just a causal, schema (Daly, 1999:69). This interpretation of Marxism shares some similarities with the approach to the state taken by Gramsci (1971). Gramsci provided an analysis of the institutional form of the state, how this influences class struggle, and how this political struggle in turn shapes the institutional form of the state. This method of analysis places emphasis on the state as an arena of action, but not one which is separate from capital or other social relations, nor one which is fully determined by capitalism. Instead the emphasis is on the reciprocal nature of the relationship between base and superstructure (Jessop, 2001b:4). The Gramscian capitalist state is not necessary, but contingent, and its actions are political and hegemonic, not determined or economistic – even though it advances the interests of capital (in Marsh, 1999:327-8). This reciprocity of state and capital is important because it allows for the possibility of a capitalist economic system connecting with a policy community, in which policy is often, but not always, capitalist in character or impact. Policy decisions may be made in the interests of policy makers themselves, may assist specific capitalists, and sometimes may assist the working class by protecting employment conditions or negotiating public access to privately owned goods (cf. Collyer et al., 2001). Thus policy is produced through a process of social struggle. The policy system is historical, unstable, contingent and contested.

What then, is the specific relationship between policy and capitalism? How is it that, as Gramsci argued, the economic and political orders of society are ‘coupled’ into an ‘historic bloc’? (cf: Jessop, 2001a). For both Jessop and Harvey, the answer is found in the construction of capitalism itself. As an historically developed system of social relations, capitalism comprises both a system of economic relations and a system of political relations. The state has a crucial role in the construction of economic forms of capital, and in how capital is organised and circulated (Harvey, 1982:281-2). Or, as Jessop argues, the capital relation is incomplete and indeterminate, and relies, in a rather unstable way, on extra-economic conditions (2001b:8). And because ‘capitalism cannot reproduce itself purely through market forces’, the system of political relations is a part of the constitution and regularisation of capital accumulation (Jessop, 2001c:9).

Policy Actors

It is with this framework that we can better understand the relations between policy and capitalism. In a capitalist social system, policy reflects capitalist relations not because policy is determined and unfolding according to a teleological plan or logic of necessity, but because policy is created as a result of human behaviour and interaction in the complex interplay between the political and the economic systems. And it cannot be created without policy actors.

Throughout the 1970s and 80s, sociological analysis of policy focused on the structural qualities of the social system, and was largely silent on the role of actors. Perhaps this might have been, in part at least, a reaction to the increasing dominance of a neo-liberal discourse in which the victim was blamed for their own plight. More recently however, there has also been a rejection of over-arching structural categories and a developing interest in the role of actors and active human subjects in the policy process (Deacon and Mann, 1999:417; Mann, 1992). This has rejuvenated the search for theoretical frameworks that can explain the role of policy makers in the policy process. Past attempts at linking agency with structural levels of analysis through locating specific groups within the class structure have been theoretically problematic: as Narravo (1978:86-7) and Johnson (1972) discovered when they sought to empirically demonstrate the class location of the professions. They found they could only assert the connection as a ‘truth’ through theoretical analysis, because at the empirical level, the task involved conflating occupational or professional groups of individuals with an abstract notion of class (eg. Willis, 1983:13).

The same reasoning applies if instead of investigating the class location of the professions we seek to empirically examine the class position or role of policy makers. Again the problem would be that policy makers do not constitute a class. This is a diverse group: policy (in its widest sense) is produced by individuals and groups from many occupations and institutions: media owners and employees, bureaucrats and politicians, owners and managers of private firms, philanthropic organisations, community representatives etc. Class locations are mixed: employees, owners of capital, petite bourgeois. Consequently, using an orthodox Marxist analysis we cannot empirically demonstrate the class basis of privatisation without conflating the concept of the group with the concept of class.

What is needed then, is a theoretical framework which bridges the empirical world of policy making with the conceptual and analytical structures of the state, class and capitalism. This framework must not be reductionist, it must not seek complete closure or full determinism, and, most importantly, it must link macro and micro levels of analysis to make sense of the relationship between the activities of the individual (and the group) and structural formations. One answer to this problem is the policy network approach.

Policy Networks

There are a variety of approaches to policy network analysis. Some analyses tend to collapse the notion of a policy network with that of a policy community or policy culture, and as such emphasise a sharing of values and collective goals (eg. Marsh and Rhodes, 1992; Putnam, 1995). For example, Pross (1992:119) defines a policy network as a group of individuals with similar outlooks and values and a common approach to policy. As such, it excludes actors with alternative perspectives or different policy agendas. Moreover, although values and norms are an important concern within the policy making process, a sole focus on this aspect of social life results in a lack of attention being paid to the impact of the state (and other institutions) on the policy process (Maloney et al., 2000). It also fails to include the role of groups and organisations in the policy process, as it is problematic to assign a ‘mindset’ or ‘worldview’ to a group or organisation, and, importantly, it overlooks the possibility that policy outcomes may be an amalgam or compromise of various positions, views, and values.

Coleman and Skogstad take a different view. They argue that a policy network is ‘a concept reserved for describing the properties that characterise the relationships among the particular set of actors that form around an issue of importance to the policy community’ (Coleman and Skogstad, 1990:26). The focus of the policy network approach is thus the relationships among the actors involved in the policy making process, rather than the values or characteristics of the actors.

This reading of the policy network avoids the problem of trying to ascertain the intent or motivation of participants in the policy process, and enables a greater focus on relationships, particular relations of power, within the network. This form of network analysis has been used in conjunction with a pluralist perspective, where there has been an examination of the role of interest groups in policy process. Saward (1990) for example, examines the nature of resources held or captured by different interest groups, and how these influence their capacity to shape policy; and Whiteley and Winyard (1987) propose a typology in which policy actors can be characterised according to their lobbying strategies.

There is no inherent reason however, why pluralism must be the only, or dominant theoretical approach to network analysis. Pluralism doesn’t offer an adequate explanation for the way policy initiatives, such as privatisation, both reflect existing inequalities and become structured into the operation of social institutions, thus ensuring a policy regime will become entrenched and resistant to further change. One attempt at a solution for this problem has been the ‘path dependency’ approach (eg. Berman, 1998; Goldstone, 1998; Wilsford 1994; Greener 2002a, 2002b). This approach assumes that policy reflects established regulatory regimes, entrenched institutional positions, established industrial structures, and past government-industry relationships. This entrenchment of institutional patterns ensures policy becomes fixed and difficult to change, so that policy development is likely to be incremental rather than innovative. It argues that historical legacies determine future policy trajectories. As Löfgren and Benner (2003:27) suggest, ‘path dependency perspectives emphasise the ‘stickiness’ of institutional arrangements: once a nation has developed a particular institutional set-up, it is difficult to change tack’.

The path dependency approach provides an explanation for the apparent stability of a policy regime: that it is the unique configuration of institutional relationships in each national setting which determine the policy outcome. Unfortunately, this approach doesn’t account very well for policy innovation: and this is important if we are to theorise the recent radical shift toward privatisation. Empirically we know that policy designs are not created and then simply ‘fixed’ in time, but continue to be contested and remain malleable (Alexander, 2001). And although institutional legacies can restrict change, social and institutional resources are also combined in creative ways to produce new policy solutions and directions (Neilson, Jessop and Hausner, 1995:3-46). Moreover, cross-national comparisons have found that path dependency is not a good predictor of policy direction, particularly when countries, each with very different institutional histories, are facing profound pressures for radical change (Löfgren and Benner, 2003:27). Even the idea of ‘conjunctures’, offered by Wilsford (1994) to explain policy change, is not of assistance here. Wilsford argues that structural forces curtail policy processes, keeping policy to established pathways, while conjunctures enable deviations and radical innovations. The established pathways may occur initially ‘as the result of quite random variables’ (Wilsford, 1994:253), but then, like the QWERTY keyboard, ‘channel present and further policy along certain paths’ (ibid:256). Conjunctures, on the other hand, ‘are the fleeting coming together of a number of diverse elements into a new, single combination’ (ibid:257), providing incentives for radical change.

The main difficulty with this approach is that Wilsford has no theoretical framework to assist with distinguishing between factors or conditions that are likely to inhibit change, and those which encourage innovation. His selection of ‘conjunctures’ is ultimately adhoc, and this produces dispute from others. For instance Greener (2002a:170-1) contests Wilsford’s selection of critical conjunctures to explain radical policy change for Britain’s National Health Service, arguing that, amongst other factors, Wilsford places too much emphasis on Margaret Thatcher’s authority, and on the disorganisation of the medical profession during the salient period. As Greener argues, the conjuncture argument ‘seems to place an excessive burden onto fate’ and ‘does not give us a coherent theoretical underpinning’ for how actors are constrained or break free of structures (2002b:615).

The solution then, is to disregard both the pluralist and the path-dependency approaches in favour of a combination of class and network analysis within the parameters set out in the above discussion. This means employing a combination of the strengths of network analysis, with its emphasis on the relationships within policy networks, and those of class analysis, with its capacity to make sense of the continuing, asymmetric relations between actors in the policy making process. With this combination, we can reveal the way that policy networks, and their outcomes, are shaped by the class system at both the national and international levels.

What would such a combined approach look like? We might take the example of the Commonwealth Employment Service (CES). This was an agency established in 1947, initially corporatised, and then fully privatised in 1998. Over its fifty years of operation there were several attempts to privatise it, most notably perhaps by the McMahon government, which sought to allow private employment agencies entry to the job placement system. Its demise began with the publication of the controversial Norgard Report (1977) which recommended systematic changes in the operation of the CES, and the successful elimination of the Professional Employment Service by the Fraser Coalition government. The end of the organisation was secured with the passage of legislation in 1994 under the Hawke-Keating Labour government, to allow the introduction of private sector competition. These changes prepared the way for the later full privatisation and dismantling of the CES under the Howard Coalition government. Announced in August 1996, the CES ceased to function in its old form in November 1997, interim arrangements were put into place, and the new Job Network emerged in May 1998. Heralded under the rhetoric of enhancing consumer ‘choice’ and introducing the ‘efficiencies’ of market competition, the new service (estimated at the time to be worth approximately $AUS1.7 billion), offered via tender, contracts in which agencies would be paid by government to place individuals into jobs. Over 300 contracts were signed with private (for-profit) enterprises, community organisations, and one public organisation, Employment National. The privatisation of the CES is a typical example of the emergent form of privatisation for policy delivery, in which private operators distribute funds or perform services that were once the responsibility of government.

. network analysis would reveal the characteristics of, and relationships between constituents of the network, showing for example, their degree of autonomy or inter-dependence (as it does in the work of Ornstein (2003) on Canadian corporate networks, or the work of Burt (1983) on the market relationships between various industries). In the case of the CES, network analysis would also indicate the prominent role of religious organisations in the new Job Network, and the formation of alliances between agencies (particularly in the early period, where many were without adequate skilled staff or resources to fulfill the terms of their contracts). Class analysis on the other hand, would highlight the asymmetry in the relationships of power between capital and labour, showing how privatisation shifted the balance to favour employers and capital. For example, with the demise of the CES we see the undermining of union power, as 5000 unionised, permanent public servants lost their jobs and were shifted into casualised positions in the private sector, many working on contracts offering poorer and less secure working conditions, and most no longer represented by a union.

A combined network/class analysis however, would reveal the way the policy networks are shaped by the class system. To focus once again on the new Job Network that emerged in May 1998, a combined approach shows, first of all, the way in which the state works with capital to initiate and support the privatisation process. Evidence of this is found in the awarding of contracts to favour private businesses, particularly those in electoral areas held by a member of government. Indeed private job agency sites granted to safe government seats were double the number of those in safe Labor seats (Haslem, 2000c). Evidence is also provided in the stated intentions of government. In discussing the privatisation of the job network, the discourse focuses explicitly on the need to form a ‘social coalition’ between businesses, community organisations and government. This coalition would enable the private sector to expand their workforces, develop new business opportunities, and, in return for delivering welfare services, enable these entities to ‘put something back into their communities’ (Shanahan, 1999).

Secondly, the effects of the class system on the network itself are evident in the alliances formed to place pressure on government. Some of these are large and formal alliances, such as Job Futures, a consortium of 40 agencies; National Employment Services, representing over 60 agencies; and Jobs Australia, representing 300 delivery organisations. In 1998 these actors lobbied for large funding increases of over $AUS1b on the grounds that 80 businesses were ‘about to go to the wall’ (Gratton, 1998; Dore, 1998). These alliances have enabled particular agencies to monopolise resources, shape the rules and processes for inclusion and exclusion, and ensure that the flow of benefits is toward private capital. As the powerful alliances swung into action, they demanded not only large increases in their fees, but a role in the policy process. Their demands included a role in designing the terms of subsequent tender rounds, to have more areas of the job network placed under tender (rather than publicly provided), the granting of ‘base’ grants not tied to the fee-for-service system, and a widening of the eligibility criterion to ensure agencies will be paid to service other groups of job seekers (Green, 1998; Gratton, 1998). This heavy lobbying from placement agencies had its intended effect, with several subsequent ‘overhauls’ of the infant Job Network, including an additional $AUS55m proffered just prior to the election and a further $AUS100m in grants and higher payments prior to the second round of contracts (Dore, 1998; Dore et al, 1998).

Clearly missing from these alliances was Employment National, the one public provider of services in the Job Network. This organisation was expected to operate as a corporate entity on the same basis as its private sector competitors. The birth of this organisation was a difficult one, given that it was the offspring of the previously much larger CES, and began its new life with only 10% of its workforce and staff morale at a low ebb (Peake, 1998). Moreover, it came into operation in an environment that was fiercely pro-privatisation and opposed to government involvement in employment services. A Morgan and Banks opinion poll of 3,700 businesses in 1997 indicated the business community was behind the privatisation of employment services, with more than half stating they believed privatisation would make the system more efficient (Sunday Telegraph, 1997).

While Employment National was initially provided with a 40% share of government contracts (Haslem, 2000a), press coverage of its first year of operation was generally negative, with constant demands for its removal from the network via privatisation. The criticism had the intended affect, and in the second round, announced at the end of 1999, Employment National was awarded only 1% of the contracts (McGregor, 1999). This left the Salvation Army as the largest provider, with contracts worth over $AUS280m, and ‘reflect[ed] the Howard Government’s push to deliver welfare services through community-based organisations, particularly the churches … [which now] deliver nearly 25% of $AUS700 million worth of employment services’ (Haslem, 2000b). Other reasons given for the demise of the agency concerned problems with its tendering strategy, problematic performance, bias and politicisation of the tender process (Kelly, 2000; Dore, 1999), and, more pointedly, the ‘deliberate’ destruction of its capacity to function effectively by the government (Haslem, 2000a; Haslem and Marris, 1999). The agency continues to operate in a limited way with government financial support, although the threat of privatisation remains, most of its staff have taken redundancy packages, and many of its offices have closed.

Thirdly, the operation of the network is clearly capitalist in the way it functions to channel public funds to the private sector. Despite the stated intentions, these public funds are not necessarily given to providers. Ten percent of the businesses that succeeded in obtaining contracts perform no employment services, but merely subcontract these to other agencies (cf: Peake, 1998). Interestingly, this sub-contracting is done with the explicit permission of government (Taylor, 1998). This case of the ‘capture’ of public funds by the private sector offers support for Jamrozik’s thesis that the welfare system, and social expenditure more generally, has many beneficiaries other than the ‘poor’. Jamrozik (2001:51) argues that social expenditure also ‘enhances the social functioning of the service provider’ through the provision of wages or salary for middle class occupational groups such as doctors, teachers and social workers. Broadening the debate, Jamrozik (2001:49) argues that although not usually classified as ‘welfare’, public expenditure is also provided to business and industry in the form of tax concessions. Extending from this, it is not difficult to see how the private sector has, in this and many other cases, managed to obtain a significant level of public funds out of the welfare budget. Indeed there has been a massive shift of public funds between the private and public sectors in Australia in recent decades, with the Productivity Commission estimating that ‘budget handouts’ to industry in the form of business welfare is now greater than $AUS14b, an increase of 15% in the past five years (Marris, 2002). While it may not be necessary to support Samson’s (1994:79,90) concept of privatisation as an ‘hegemonic project’ – with its underlying notion of a class conspiracy or at least ruling class co-operative activity – such cases of privatisation lend support to Taylor Gooby’s (1997:184) view that the pursuit and implementation of privatisation policies have enabled the ruling class to ‘articulate its interests more effectively’.

The example of the CES shows that a combined network and class approach throws new light on the theory of privatisation. O’Connor (1993) has previously argued that state expenditure on services and infrastructure is essential to lower production and reproduction costs and to ensure the expansion of private industry. Moreover, O’Connor argued that one of the problems of late capitalism is the tendency for state expenditure on the social expenses of production to increase beyond the state’s financial capacity, thus increasing the risk of a legitimation crisis. Although the capacity of privatisation to reduce state expenditure is difficult to ascertain (in part because of the claim that data is commercially sensitive, cf: Wright, 1998; Taylor, 1998; Kelly, 2000) and is certainly much disputed (cf: Collyer, 1996; Collyer and White, 1997, 2001), nevertheless it can be concluded that privatisation operates at two levels. At one level privatisation is ideological. Privatisation is used by the state to make it appear as if there is a sharing of social expenditure between the private and public sectors. At the level of practice and structure, privatisation is not just a ‘fashionable policy trend’, but a unique development in late capitalism which enables capital to secure new sources of funds, expand business opportunities into areas previously prohibited to private capital (such as welfare services), and to expand the commodification process.


This paper has traversed much territory and it is time to meld its many strands. We have seen how the policy process of privatisation has been inadequately characterised within a functionalist framework, leading to a distortion of policy as if it were an inevitable response to societal and economic pre-requisites; within a pluralist framework, where the policy process is fundamentally political but is theorised, problematically, as arising from individual motivations and accidental groupings; within an ideational framework, where policy is reduced to ‘fashion’ and untethered to material life; and within orthodox Marxism, where policy is merely derivative of structure. Against these conceptions, this paper has argued that the policy process is fundamentally social, historically contingent, and structured by relations of power – as is found in all other areas of social life.

As noted at the start of this paper, policies of privatisation have become commonplace in almost all countries over the past few decades. This policy trend, which began in the capitalist countries and was largely driven by their agendas, is best explicated in terms of policy networks. Such networks are embedded within structures of power relations, where networks operate within a framework of institutions and a capitalist global market system. The paper has shown how particular groupings of policy actors form networks that monopolise the policy process and shape the rules and processes for inclusion and exclusion. A combination of network and class analysis was used to explain the privatisation ‘boom’ (the recent intensification of privatisation and other pro-market policies), not as a policy ‘fashion’ but as a structural process within which actors agitate for the removal of all government restrictions, for the creation of new opportunities and for further privatisation. Thus the process of privatisation itself assists with the building of new, powerful alliances demanding greater ‘reform’, and thus drives further privatisation. In this way we can understand policy change, not as an outcome of logical structural imperatives, nor as an adhoc, discursive process, but as a social process undertaken by social actors and in accordance with a capitalist social structure.


Agger, B. (1998) Critical Social Theories: An Introduction Colorado: Westview Press.

Alexander, G. (2001) ‘Institutions, Path Dependence, and Democratic Consolidation’ Journal of Theoretical Politics 13(3):249(22).

Bacci, C.L. (1999) Women, Policy and Politics London: Sage Publications.

Badie, B. and Birnbaum, B. (1983) The Sociology of the State Chicago: University of Chicago Press.

Bachrach, P. and Baratz, M. (1970) Power and Poverty New York: Oxford University Press.

Baumgartner, F. and Jones, B. (1993) Agendas and Instability in American Politics Chicago: University of Chicago Press.

Bell, S. (1997) Ungoverning The Economy Melbourne: Oxford University Press.

Benner, T.; Reinicke, W. and Witte, J. (2003) ‘Global Public Policy Networks’ The Brookings Review Washington: Spring 21 (2) 4 pgs.

Berman, S. (1998) ‘Path Dependency and Political Action: Re-examining the Responses to the Depression’ Comparative Politics 30:379-400.

Bradley, K. and Nejad, A. (1989) Managing Owners: The National Freight Consortium in Perspective Sydney: Cambridge University Press.

Burt, R. (1983) Corporate Profits and Copation Vancouver, University of British Columbia Press.

Campo-Flores, F. (1994) ‘Lessons From Privatization of Argentina’s National Telephone Company’ Policy Studies Review 13(3/4):235-48.

Carter, M. (1998) ‘Case Studies Of Public Hospital Privatisation’ Health Issues 56:15-17.

Chamberlin, J. and Jackson, J. (1987) ‘Privatization as Institutional Choice’ Journal of Policy Analysis and Management 6(4):586-604.

Cohen, P. (1968) Modern Social Theory London: Heinemann.

Collyer, FM. (1996) ‘Measuring Privatisation: An Overview of Methodological Approaches’ Just Policy Number 7, August. pp. 24-34.

Collyer, F.M. (1997) ‘The Port Macquarie Base Hospital’ Just Policy 10:27-39.

Collyer, F.M; McMaster, J. & Wettenhall, R.W. (2001) Case Studies in Australian Public Enterprise Divestment Fiji: University of South Pacific Press.

Collyer, FM. & White, KN. (1997) ’Enter The Market: Competition, Regulation and Hospital Funding in Australia’ Australian and New Zealand Journal of Sociology 33 (3) November, pp. 344-363.

Collyer, FM. and White, KN. (2001) Corporate Control of Healthcare in Australia Discussion Paper No. 42, Australia Institute, Australian National University.

Coleman, W. and Skogstad, G. (1990) ‘Policy Communities and Policy Networks’ in Coleman, W. and Skogstad, G. (eds) Policy Communities and Public Policy in Canada: A Structural Approach Toronto: Copp Clark Pittman.

Dalton, T; Draper, M; Weeks, W. and Wiseman, J. (1996) (eds) Making Social Policy in Australia Sydney: Allen and Unwin.

Daly, G. (1999) ‘Marxism and Postmodernity’, in A. Gamble; Marsh, D. and Tant, T. (eds) Marxism and Social Science London: Macmillan, pp. 61-84.

Dandeker, C. (1990) Surveillance, Power and Modernity: Bureaucracy and Discipline From 1700 to the Present Day Cambridge: Polity.

Deacon, A. and Mann, K. (1999) ‘Agency, Modernity and Social Policy’ Journal of Social Policy 28(3):413-35.

Domberger, S. (1995) ‘What Does Privatisation Achieve? – A Comment on Quiggin’ The Australian Economic Review 2nd Quarter:43-47.

Domberger, S. and Piggott, J. (1986) ‘Privatization Policies and Public Enterprise: A Survey’ Economic Record 62 (177):145-62.

Dorner, D. (2002) ‘The Essential Services Policy Network: Organizational Influence in Canada’s Information Highway Policy Development Process’ Library Quarterly 72(1):27-84.

Dore, C. (1998) ‘Howard’s $100m Gift For Job Agencies’ The Australian 16th December.

Dore, C. (1999) ‘Job Network Agency Losses $12m’ The Australian 9-10 January.

Dore, C.; Green, P. and Lunn, S. (1998) ‘$55m Rescue For Ailing Job Network’ The Australian 26 August.

Dowding, K. (1995) ‘Model or Metaphor? A Critical Review of the Policy Network Approach’ Political Studies XLIII(1):136-58.

Dudley, J. and Vidovich, L. (1995) ‘The Politics of Education’ Australian Educational Review No. 36 Melbourne: The Australian Council for Educational Research.

Edelman, M. (1987) Constructing The Political Spectacle Chicago: University of Chicago Press.

Ernst, J. (1994) Whose Utility? Buckingham: Open University Press.

Ernst, J. (1998) ‘Privatization’ in Shafritz, JM. (ed) International Encyclopedia of Public Policy and Administration Boulder, Colorado: Westview Press.

Feigenbaum, H.R. and Henig, J.R. (1994) ‘The Political Underpinnings Of Privatisation: A Typology’ World Politics 46 (1):185-208.

Flora, P. and Heidenheimer, A. (1984) The Development of Welfare States in Europe and America New Jersey: Transaction Books.

Franzway, S. Court, D. and Connell, R.W. (1989) Staking A Claim Sydney: Allen and Unwin.

Gains, F. (2003) ‘Executive Agencies in Government: The Impact of Bureaucratic Networks on Policy Outcomes’ Journal of Public Policy 23(1):55-79.

Goldstone, J. (1998) ‘Initial Conditions, General Laws, Path-Dependence, and Explanation in Historical Sociology’ American Journal of Sociology 104:829-45.

Goodman, J.B. and Loveman, G.W. (1991) ‘Does Privatisation Serve The Public Interest?’ Harvard Business Review 69 (6):26-38.

Gormley, W.T. (1994) ‘Privatization Revisited’ Policy Studies Review Autumn/Winter 13(3/4):215-34.

Gramsci, A. (1971) Selections From The Prison Notebooks London: Lawrence and Wishart.

Grattan, M. (1998) ‘Job Providers Want Some Urgent Changes’ The Australian Financial Review July 24th.

Green, P. (1998) ‘Job Agencies Demand A ‘Fix’ The Australian 8-9 August.

Greener, I. (2002a) ‘Understanding NHS Reform: The Policy-Transfer, Social Learning, and Path-Dependency Perspectives’ Governance: An International Journal of Policy, Administration, and Institutions 15(2):161-83.

Greener, I. (2002b) ‘Theorising Path-Dependency: How Does History Come to Matter in Organisations?’ Journal of Management History 40(6):614-19.

Gupta, A. (1996) ‘Privatisation’ Journal of Institute of Public Enterprise (India) 19 (1-2):38-51.

Hall, P.A. (1993) ‘Policy Paradigms, Social Learning and the State’ Comparative Politics 25:275-96.

Hall, P.; Land, H.; Parker, R. and Webb, A. (1975) Change, Choice and Conflict in Social Policy London: Heinemann.

Harvey, D. (1982) The Limits of Capital Oxford: Blackwell.

Haslem, B. (2000a) ‘Job Agency Dumps Long Contacts’ The Australian 7th Feb.

Haslem, B. (2000b) ‘Army Steals March on Job Providers’ The Australian March 25-6, p.5.

Haslem, B. (2000c) ‘Job Sites Disparity Queried’ The Australian 23 December, p. 4.

Haslem, B. and Marris, S. (1999) ‘Tax Error Cost Job Agency Contracts’ The Australian December 20th.

Hay, C. (1999) ‘Marxism and the State’ in Gamble, A.; Marsh, D. and Tant, T. (eds) Marxism and Social Science London: Macmillan, pp. 152-74.

Heald, D. (1983) Public Expenditure Oxford: Martin Robertson.

Henig, J. Hamnett, C. and Feigenbaum, H.B. (1998) ‘The Politics Of Privatization’. Governance 1 (4):442-68.

Herd, B. (1998) ‘Deconstructing The Current Account Balance Of Payments’ in M. Alexander; S. Harding; P. Harrison; G. Kendall; Z. Skrbis, and J. Western (eds) Refashioning Sociology Brisbane: TASA Conference Proceedings.

Hood, C. (1994) Explaining Economic Policy Reversals Buckingham: Open University Press.

Ikenberry, G.J. (1990) ‘The International Spread Of Privatization Policies’ in Suleiman, E.N. and Waterbury, J. (eds) The Political Economy of Public Sector Reform and Privatization Oxford: Westview.

Jamrozik, A. (2001) Social Policy in the Post-Welfare State Longman, Frenchs Forest.

Jansson, B. (2000) ‘Policy Analysis’ in Midgley, J.; Tracy, M. and Livermore, M. (eds) The Handbook of Social Policy London: Sage Publications, pp. 41-52.

Jessop, B. (1982) The Capitalist State Oxford: Martin Robertson.

Jessop, B. (1999) ‘Narrating The Future of the National Economy and the National State? Remarks on Remapping Regulation and Reinventing Governance’ in Steinmetz, G. (ed) State/Culture: State Formation After The Cultural Turn Ithaca, Cornell Press, pp. 378-405.

Jessop, B. (2001a) ‘Bringing The State Back In (Yet Again): Reviews, Revisions, Rejections, and Redirections’ published by the Department of Sociology, Lancaster University, at

Jessop, B. (2001b) ‘Capitalism, the Regulation Approach, and Critical Realism’ in Brown, A.; Fleetwood, S. and Roberts, J. (eds) Critical Realism and Marxism London, Routledge.

Jessop, B. (2001c) ‘On The Limits of Limits to Capital’ published by the Department of Sociology, Lancaster University, at

Johnson, T. (1972) Professions and Power London: Macmillan Press.

Kay, J.A. and Thompson, D.J. (1986) ‘Privatisation’ The Economic Journal 96 (March):18-32.

Kelly, P. (2000) ‘Jobs Agency Bungled Tender Bid’ The Australian 14th Feb.

Kerr, L. and Savelsberg, H. (2001) ‘The Community Service Sector in the Era of the Market Model’ Just Policy 23:22-32.

Langmore, J. (1987) ‘Privatisation’ in Abelson, P. (ed.) Privatisation: An Australian Perspective Sydney: Australian Professional Publications.

Löfgren, H. and Benner, M. (2003) ‘Biotechnology and Governance in Australia and Sweden: Path Dependency or Institutional Convergence?’ Australian Journal of Political Science 38(1):25-43.

Lukes, S. (1974) Power: A Radical View London: Macmillan.

McGregor, R. (1999) ‘Job Agent Put Out of Work’ The Australian 4-5th Dec.

Maloney, W. Smith, G. and Stoker, G. (2000) ‘Social Capital and Urban Governance’ Political Studies 48:802-20.

Mann, K. (1992) The Making of an English ‘Underclass’? The Social Divisions of Welfare and Labour Bucks: Open University Press.

Mansfield, B. (1987) ‘The Public Sector’ in Abelson, P. (ed.) Privatisation: An Australian Perspective Sydney: Australian Professional Publications, pp. 70-95.

Marris, S. (2002) ‘Business Welfare Budget Costs Taxpayers $14bn’ The Australian 23rd December.

Marsh, D. (1999) ‘Resurrecting Marxism’ in Gamble, A.; Marsh, D. and Tant, T. (eds) Marxism and Social Science London : Macmillan, pp. 320-40.

Marsh, D. and Rhodes, R. (1992) Policy Networks in British Government Oxford: Clarendon Press.

Marsh, D. and Smith, M.J. (2000) ‘Understanding Policy Networks: Towards a Dialectical Approach’ Political Studies 48(1):4-21.

Marshall, T.H. (1981) The Right To Welfare and Other Essays Heinemann.

Martin, S. and Parker, D. (1997) The Impact of Privatisation London: Routledge.

Marx, K. (1973) Grundrisse Harmondsworth: Penguin.

Marx, K. and Engels, F. (1976) German Ideology [1845-6] Moscow: Progress Publishers.

Melossi, D. (1990) The State of Social Control: A Sociological Study of Concepts of State and Social Control in the Making of Democracy Cambridge: Polity.

Mishra, R. (1982) Book Review of TH Marshall (1981) ‘The Right To Welfare and Other Essays’ Critical Social Policy 2 (1) Summer:105-7.

Mooney, G. (2000) ‘Class and Social Policy’ in Lewis, G.; Gewirtz, S. and Clarke, J. (eds) Rethinking Social Policy London : Sage, pp. 156-70.

Moore, J. (1986) ‘Why Privatise?’ in Kay, J.; Mayer, C. and Thompson, D. (eds) Privatisation and Regulation Oxford: Clarendon Press.

Navarro, V. (1982) ‘The Crisis of the International Capitalist Order and Its Implications For the Welfare State’ Critical Social Policy 2 (1) Summer:43-62.

Navarro, V. (1978) Class Struggle, The State and Medicine New York: Prodist.

Neilsen, K.; Jessop, B. and Hausner, J. (1995) (eds) Strategic Choice and Path-Dependency in Post Socialism. Institutional Dynamics in the Transformation Process Aldershot: Edward Elgar.

Nevile, A. (2000) ‘The Implementation of Competitive Tendering in the Australian Social Welfare Sector’ Just Policy 18:15-23.

Norgard, JD. (1977) Review of the Commonwealth Employment Service AGPS, Canberra.

O’Connor, J. (1993) The Fiscal Crisis of the State London, St Martins Press.

Ornstein, M. (2003) ‘The Canadian Corporate Network in Comparative Perspective’ Comparative Sociology 2(1):197-213.

Peake, R. (1998) ‘Jobs at Stake – But Whose?’ Canberra Times May 2nd.

Poulantzas, N. (1979) State, Power, Socialism London: Verso.

Productivity Commission (1998) Performance of Government Trading Enterprises 1991-92 to 1996-97 Canberra: AusInfo.

Pross, P. (1992) Group Politics and Public Policy second edition, Toronto: Oxford University Press.

Pusey, M. (1991) Economic Rationalism in Canberra Melbourne: Cambridge University Press.

Putnam, R. (1995) ‘Bowling Alone’ Journal of Democracy 6 (1):65-78.

Quiggin, J. (1996) Great Expectations St Leonards: Allen and Unwin.

Salvaris, M. (1995) ‘Privatisation, Citizenship and the Public Interest’ Just Policy 4:37-43.

Samson, C. (1994) ‘The Three Faces Of Privatisation’ Sociology 28 (1):79-97.

Saunders, P. (1994) Welfare and Inequality Melbourne: Cambridge University Press.

Saward, M. (1990) ‘Co-option and Power’ Political Studies 38:588-602.

Self, P. (1990) ‘What’s Wrong With Government?’ Canberra Bulletin of Public Administration 60:17-21.

Shanahan, D. (1999) ‘Abbott Given To Charity’ The Australian 4-5 December.

Shore, C. and Wright, S. (1997) ‘Policy’ in Shore, C. and Wright, S. (eds) Anthropology of Policy London: Routledge, pp. 3-39.

Skocpol, T. (1985) ‘Bringing The State Back In: Strategies of Analysis in Current Research’ in Evans, P.B. et al (eds) Bringing The State Back In Cambridge; Cambridge University Press, p. 3-37.

Steel, D. and Heald, D. (1982) ‘Privatising Public Enterprise’ Political Quarterly 53 (3):333-49.

Sunday Telegraph (1997) ‘Firms Back Private CES’ 10 August.

Taylor, M. (1998) ‘Subcontractors’ Job Fees Go Unchecked’ The Canberra Times 11th June.

Taylor-Gooby, P. (1997) ‘In Defence of Second-Best Theory: State, Class and Capital in Social Policy’ Journal of Social Policy 26(2):171-92.

Therborn, G. (1982) ‘What Does The Ruling Class Do When It Rules?’ in Giddens, A. and Held, D. (eds) Classes Power and Conflict London: Macmillan, pp. 224-48.

Thynne, I. (1995) ‘Basic Concepts and Issues’ in Thynne, I. (ed.) Corporatization, Divestment and the Public-Private Mix Hong Kong: AJPA and IASIA, pp. 1-9.

Toke, D. and Marsh, D. (2003) ‘Policy Networks and the GM Crops Issue: Assessing the Utility of a Dialectical Model of Policy Networks’ Public Administration 81(2):229-51.

Vernon, R. (1987) ‘Economic Aspects of Privatization Programs’. The Economic Development Institute and the World Bank March, pp. 1-14.

Walker, B. and Walker B. (2000) Privatisation: Sell Off or Sell Out? Sydney: ABC Books.

Walshe, G. and Daffern, P. (1990) Managing Cost-Benefit Analysis London: Macmillan.

Wanasinghe, S. (1991) Privatisation’ in Gouri, G. (ed.) Privatisation and Public Enterprise New Delhi: Oxford and IBH Publishing, pp. 123-37.

Waters, A.R. (1987) ‘The Techniques of Privatization’ Economic Impact 61:55-9.

Whiteley, P.F. and Winyard, S.J. (1987) Pressure For the Poor: the Poverty Lobby and Policy Making London: Menthuen.

Whitfield, D. (1992) The Welfare State London: Pluto Press.

Willis, E. (1983) Medical Dominance Australia: George Allen and Unwin.

Wilsford, D. (1994) ‘Path Dependency, or Why History Makes It Difficult But Not Impossible To Reform Health Care Systems in a Big Way’ Journal of Public Policy 14 (3):251-83.

Wright, L. (1998) ‘“Unflattering” Job Figures Withheld’ Canberra Times 6th August.

© Electronic Journal of Sociology