Abstract
Industrial policy has become one of the buzzwords of state interventionism in the political economy of states. Continuity and change in social contracts, the increasing complexity of multi-level governance in the world economy, and the need for greater degrees of co-ordination and greater amounts of capital in key industrial sectors all require a reinvigorated understanding of the state and of state power. This article seeks to develop such an understanding of one state form: the Developmental State. Through an analysis of the Argentine development experience under Nestor Kirchner (2003-07), this article will reveal the need to develop sophisticated understandings of the state and associated concepts of capacity and autonomy to fully grasp its role in the development process, with a particular focus on industrial policy. Such examination will draw attention to the changing nature of industrial policy, and how this is linked to changes in the state as a result of pressures from below (changing constellation of social forces) and from above (changing nature of global forces).
Never has there been a greater need for development; never have there been larger obstacles to it. From climate change to global pandemics, socio-economic division to systemic economic crisis, the challenges to a sustainable model of growth and catch-up industrialisation are manifold. The power of markets to meet this challenge should not be underestimated, but nor should they be over emphasised. The historical record has demonstrated that one key actor must play a pivotal role: the state. An understanding of the state and its key role in the development process has evolved in recent years, moving beyond a reductionist understanding of globalisation facilitating the retreat of the state (Ohmae, 1995) to one of an ‘active’ state embedded within both societies and markets at national and global levels (Bremmer, 2008; Kurlantzick, 2016).
One historical state function that has received much attention has been that of industrial policy. The hegemony of a neoliberal paradigm in development thinking, and its policy manifestation in the developing world in the form of the Washington Consensus, has led to the dismantling of many state capacities. However, this renewed interest in states as a key actor in the development process has led to more sophisticated understandings of state interventionism in general and its role in rapid catch-up industrialisation in particular. Of particular interest has been to refocus away from specific functions and associated policies and more into state form; its associated social forces (of production) and the attendant manifestations of development paradigms.
This article will attempt to give the same treatment to industrial policy, which has enjoyed a resurgence in both academic literatures and in the practice of state’s attempts to engage in rapid catch-up industrialisation. In fact, Industrial policy never went out of fashion. The shift towards embracing industrial policy is therefore more a reversion to what has always been known: rapid catch-up industrialisation requires the seducing and cajoling of markets by the state. What has evolved is the form of this intervention, both in the context of changing international structures (globalisation) and how states are inserted into the subsequent global division of labour. This has led to contributions that identify a transition away from ‘inward looking industrial policy’ towards ‘open market industrial policy’ (Bulfone, 2022: 7), or from ‘old’ industrial policy to ‘new’.
The central argument of this article is that Bulfone (2022) represents a framework for analysing the capacity and autonomy of states to enact these ‘new’ forms of industrial policy. In-so-doing, it will contribute to two sets of literatures. First, the neo-developmentalism literature will be critiqued and rationalised though a systemic mapping of the core relevant concepts associated with (neo)developmentalism: from the absolute autonomy of Evans (1995) to the relative autonomy of Jessop (2016). Second, the literature on new industrial policy. An analysis of Bulfone (2022) will show how his categories of ‘logic of intervention’, ‘protagonists’, ‘policy instruments’, and ‘constraints’ in the context of industrial policy map nicely onto key lessons distilled from analysis of the state through the framework of autonomy and capacity (Berwick and Christia, 2018; Cingolani, 2013). Namely, how a focus on class relations – both domestic and international – and their interaction with processes of state formation (or institutionalisation) are key to understanding the nature of public policy profiles associated with new industrial policy. In so doing, the article contributes to a tradition of understanding state formation as the product of a set of mutually dependent underpinnings understood through state capacity and autonomy (Fine et al., 2013; Fishwick, 2019; Fishwick and Selwyn, 2018; Mann, 1984; Tilly, 1990) rather than as independent or isolated dimensions of state capacity (Brambor et al., 2020; Cole, 2015; Lee and Zhang, 2017).
The intention is to then synthesise these literatures so as to show the conceptual linkages between neo-developmentalism and ‘new’ industrial policy. Such an endeavour will facilitate an ability to discern state features that lead to growing or declining autonomy: demonstrating that the form of state institutions understood in relational terms help us gauge the likelihood of success of key aspects of twenty-first century industrial policy. Domestically this means analysis of the social contract and how different class formations lead to different processes of institutionalisation, which subsequently impact on state form (Clark and Dear, 2021; Jessop, 2016). Internationally this means examination of the key aspects of contemporary IPE, how they interact with domestic social contract formation, and the subsequent ‘political economy of the possible’ (Santiso, 2006) that twenty-first century states must operate within: both generally and broadly, but for our purposes specifically in the context of industrial policy.
These theoretical insights will be empirically contextualised and tested through an examination of Argentina; more specifically, the post-crisis administration of Nestor Kirchner 2003–07. This historical-structural approach is embedded in a broader comparative political economy agenda (Haggard, 1990, 2004; Naseemullah, 2023; Nem Singh, 2023; Rodrik, 2023) that seeks to empirically induce mid-level theorisation that helps us understand broader social transformations such as state-driven developmental policies. Argentina as a country case study was chosen for its place as a middle-income country with a history of state intervention in its catch-up industrialisation processes. It has oscillated between (neo)liberalism and interventionism in different guises across the twentieth-century. More importantly, it experienced a series of systemic crises that culminated into the 2001–2 collapse of its convertibility regime. The scale of the subsequent economic collapse, combined with a crisis in political legitimacy, represented a watershed moment in national and regional history. The widely held desire at the time in Argentina to replace the political, legal, and economic establishment as well as the neoliberal model, were articulated with profound effects (Levey et al., 2014: 2). Analysis of the first significant post-crisis regime – that of Nestor Kirchner, 2003-07 – represents an opportunity to study a country under significant political, economic, and institutional transformation. Crisis has a profound impact on pre-existing patterns of political economy (MacIntyre et al., 2008: 2), and study of the Argentine case facilitates a testing of theoretical frameworks under empirical conditions of change. Yet at the same time, the crisis and post-crisis periods reveal a surprising number of continuities with the pre-crisis panorama. Any framework for analysing such phenomena must therefore be robust enough to capture the elements of continuity as well as change in the post-crisis period in Argentina and therefore represent a good litmus test of the proposed model.
To achieve these contributions, this article is split into three sections. The first of these will set out how state autonomy and state capacity represent key concepts for understanding state formation in the twenty-first century. This will necessarily encompass an integration of the role of global power through a consideration of the contributions of the IPE literature, as well as more domestic concerns involving continuity and change in social contracts and concomitant processes of institutionalisation of state capacities and autonomies. A second section will then articulate how Bulfone’s (2022) framework for understanding ‘new’ industrial policy can be used to analyse the changing capacities and autonomies of states. This will generate more sophisticated and sharpened understandings of dynamics that lead to success or failure of these policies. In the third section, these theoretical contributions will be distilled empirically into an examination of post-2001 Argentina. Examining the role industrial policy played in the post-crisis administration of Nestor Kirchner will represent an opportunity to empirically test the framework under changing and sometimes profound political, economic, and social conditions.
Capacity and autonomy in state form
The type of state that this article is interested in is the one that has the authority, power, and capability to strike a relationship with the market that is not one of domination, but one that is able to ‘govern the market’ (Wade, 1990). The literature has described this kind of formation as a Developmental State, and is often concerned with the necessary authority, power, and capability to enact appropriate market governance, which distils into the concepts of autonomy and capacity. Capacity can be defined as the ability of a state to implement goals, that is, the institutional forms of different state apparatus. It follows that in order to have capacity a state must also possess autonomy, defined as when it is able to ‘formulate and pursue goals that are not simply reflective of the demands or interests of social groups, classes, or society’ (Evans et al., 1985: 9). Government may have to enact unpopular or even harsh policies in the name of development. To effectively guide economic development a state must enjoy the power to direct society and lead it through traumatic changes. Bureaucrats must be able to draft policies that promote national development, not the advancement of private lobbyists. This analytical distinction is useful because developmental structures may exist within a state, but it requires their active use to achieve developmental ends. A state may have the capacity to be developmental, but not the autonomy.
State capacity initially focused on the role of bureaucrats and bureaucracy (see, for example, Johnson, 1982; Evans, 1995). This literature highlighted industrial policy as a key variable in understanding success in terms of rapid catch-up industrialisation. For example, Johnson discusses the importance of a pilot agency in the Japanese Developmental State (MITI – the Ministry of International Trade and Industry) as a key co-ordinator of other agencies in structuring that country’s contemporary industrial policy (Johnson, 1982). The contributions of Ha-Joon Chang (see, for example, 1999) advanced our understanding further in this area. Chang identified that successful industrial policy in Developmental States was grounded in government directives that were broad-based in scope and left operational detail to individual firms (Chang, 1999; Madjd-Sadjadi and Karagiannis, 2007: 244). Furthermore, due to the need for complementary industries or infrastructure to ensure success of an investment, there needs to be a centralised co-ordination of investment plans (Chang, 1999: 192). Finally, infant industry arguments can be deployed (Rapley, 2008: 143). Unlike the Structuralists of the 1970s, these infant industry arguments are used to build export industries in order to foster new comparative advantages based on dynamic rather than static efficiencies.
State autonomy began its intellectual journey in the form of ‘absolute autonomy’: the state was seen as separate from society in the form of a classic authoritarian state (see, for example, Amsden, 1989). Peter Evans (1995) developed this early material, emerging with the concept of ‘embedded autonomy’. This is where, following Weber, an autonomous, professional bureaucracy is embedded in elements of society that allow it to respond to market signals. In his words (Evans, 1995: 12), bureaucrats are ‘embedded in a concrete set of social ties that binds the state to society and provides institutionalised channels for the continual negotiation and renegotiation of goals and policies’. It is this embedded autonomy that facilitates the kind of strategic industrial policy witnessed in the East Asian Newly Industrialised Countries (NICs), rather than that seen during the Latin American Import Substituting Industrialisation (ISI) period. This very much underpins Rodrik’s thinking (2010: 2) when he suggests that contemporary industrial policy’s practitioners, that is, state bureaucrats, need to bear in mind that it aims to serve society at large, not the businesses that receive the incentives.
Whilst analytically powerful in the context of the rise of the NICs in the second half of the twentieth century, Evans’ traditional, neo-Weberian reading of autonomy obscured the role of social relations. The agent-centred institutionalism of (for example) Sikkink (1991) or Thurbon (2016) focused too much on how ideas shape and change institutions, and thus state capacity. Neo-Marxists (see, for example, Jessop, 2016) critiqued this as it requires politics to be viewed as analytically separate from economics; the state as autonomous from class forces. In other words, a focus on the developmental structures (Vu, 2007: 38), or organisational features, of the state – in Evans’ case bureaucracies and their links with industrial capital – generates a procrustean bed upon which all Developmental State’s must fit. In considering social relations, Jessop (2016) shows that what is needed is a greater understanding of structure and the ways in which it constrains agency. This helps us determine why certain sets of ideas (or ideologies) are adopted over others in the context of state-driven development.
These critiques distil into a further development of the autonomy debate: that of ‘relative autonomy’. Understanding the state as a social relation (see, for example, Bonefeld, 2012) means not seeing the state as a mere reflection of class interests – the economic committee of the bourgeoisie – rather the state should be seen as possessing ‘relative autonomy’ from different class interests as it advocates the interests of capitalism rather than capitalists. Class domination of the state does not simply mean that its form – the institutionalisation of a suite of polices in the form of a public policy profile and the ideological justification underpinning that profile – is a simple reflection of the interests of the dominant class. Rather, the very existence of this domination is made possible by the relative autonomy of the state. That is, state power must be seen in relational terms, that is, as founded on an unstable equilibrium of compromise among class forces rather than as a monopoly of one class (fraction) (Poulantzas, 1973: 191–3).
This overcomes a key weakness of Developmental State literature as the state no longer has an a priori national interest in favour of developmentalism; instead, the right agents in the right structural conditions, which set the strategically selective limits of agency, must coalesce within a particular social formation at a given time. The concept of relative autonomy thus helps overcome traditional Developmental State theory’s reductionist and constricting approach by ‘bringing class back in’. The state fulfils an adjudicatory role between contradictory interests and is therefore simultaneously autonomous from and embedded in social relations. It is autonomous because the state needs to pursue the ‘true’ interests of the people, rather than any distinct class (fraction), and it is embedded as it is itself part of the social formation of society. In the words of Ben Fine (2013: 23): ‘economic development is a complex amalgam of processes and outcomes derived from capital accumulation, where state and market – and their interaction – are themselves attached to the economic and political relations and interests which act upon them’.
It must always be remembered that this process happens at multiple spatial levels of structuration (Hay, 2014). Globalisation is the structural condition that dominates current study of the political economy of development. It is not though a new phenomenon; rather, it is the ‘intensification of an old process, the continuing internationalisation of commodity production and capital accumulation’ (Soloman and Rupert, 2002: 284). This instantiates itself through the agency of different political actors at different levels of spatial analysis. In other words, these actors can come from ‘above’ the state in the form of regional or global institutions, (transnational) corporations, or global markets and their attendant social relations that produce World Order (Cox, 1996: 98); or ‘below’ the state with more national or local actors and, at least for Foucauldians, the governance of minds and bodies through the concept of ‘governmentality’ (Foucault, 2008[2004]).
Governance then can be said to operate at different spatial levels of analysis, with complex mutually re-constitutive interactions between them. These different spatial levels of analysis interact with each other differently across those very spatial contexts, as well as changing over time. In other words, ‘...interdependence is hardly likely to prove politically neutral’ (Hay, 2014: 33). Relationships between actors at different spatial levels of analysis can be characterised by either interdependence or dependence.
One analytical solution to the problem of differentiated power across spatial levels is called structuration. This is a complex process that would benefit from extended definition: ‘...a dynamic system in which the conduct of actors and the context in which they find themselves are intimately and iteratively interlinked... we conceive of political-economic processes as operative at distinct spatial levels within each of which actors orient themselves strategically to the context in which they perceive themselves to be located and seek to realise specific goals and intentions’ (Hay, 2014: 38).
Governance then shapes and is shaped by a complex process of structuration across multiple spatial levels. The state has a key role as a site of institutional integration of power relations and social domination; structuration clearly shows that it simultaneously exists in a complex, heterogeneous, and multi-level network. On the one hand, the state is but one spatial level in this model, but on the other, the state can be seen as more than this. What structuration gives us is not a state that is of equal ontological weight but analytical weight. In other words, the state is clearly not a sovereign authority in a single hierarchical command structure, but rather it is an institution that is primus inter pares (Jessop, 2016: 185).
It thus becomes imperative to understand the precise relational nature of any given state formation through the analytical tools of both capacity and autonomy if the suite of policies that are generated from that constellation are to be fully comprehended. The analytical separation of roles from structures reveals the role of material interests, previously obscured through an integration of state purpose and institutions when analysing state capacity. These are present at distinct levels of spatial analysis – national, regional, and global – and interact in complex ways to shape and limit any given state’s (developmental) autonomy, although not to the extent of being absolutely constraining. When understood in these terms, the developmental state can move away from the lionising of narrow bureaucratic embeddedness and authoritarianism. Instead, broader suites of socio-economic relationships can be examined, which are the product of complex, mutually re-constitutive relationships between material interests and ideas at different levels of structuration.
The historical emergence of society and associate social entities cannot be separated from the evolution of the state. For example, development often stalls where a state has no relative autonomy from society – with sub-Saharan Africa being a clear example of this as states have often been captured by discrete interests. The state centralises and concentrates social power so that it can be exercised, executed, and directed. Policy is therefore the application of social power, and the policies of the state must be understood in the context of this application. These conclusions are also echoed in the work of Fishwick (2019) and Fishwick and Selwyn (2018) when they develop the concept of ‘labour centred development’ to critique the manifestation of developmentalism in Latin America; one that led to the failure of early twenty-first century ‘pink tide’ states to overcome deep structural problems of social and economic exclusion. State-led development strategies in this era ‘deepened relations of exploitation, blocked and co-opted social movements seeking to represent the most marginalised, and made little progress in dismantling the neoliberal macroeconomic consensus’ (Fishwick and Selwyn, 2018: 235). Another example of this analysis would be Lombardozzi and Pitts (2019) where they analyse Universal Basic Income, concluding that this policy may not generate positive outcomes in terms of reducing social and economic exclusion through analysis of the historically specific capitalist forms of labour, money, and the state.
Analysis of capacity and (relative) autonomy to underpin our narratives of state formation clearly bears much fruit. It is the intention of the rest of this article to connect these theoretical strands together into a consideration of one suite of emblematic state policies: industrial policy. The categories of ‘logic of intervention’, ‘protagonists’, ‘policy instruments’, and ‘constraints’ as defined by Bulfone (2022) serve to capture the relational aspects of developmental tendencies in industrial policy. The next section will take each of these categories in turn and leverage the taxonomy into an analysis of the nature of how different social relations generate different and complex institutional milieu. In the process, a state’s autonomy and capacity are collectively shaped. Such an exercise leverages up these concepts into a heuristic device to understand how different social relations facilitate more successful developmental outcomes than others.
‘Old’ to ‘New’ industrial policy
Industrial policy can be defined as ‘any type of intervention or government policy that attempts to improve the business environment or to alter the structure of economic activity towards sectors, technologies, or tasks that are expected to offer better prospects for economic growth or societal welfare that would occur in the absence of such intervention’ (Warwick, 2013: 16). Industrial policy can be conducted using a variety of direct and indirect instruments including grants, subsidies, tax credits, low-interest loans, public guarantees, procurements, credit provisions, and equity investments (Bulfone, 2022: 2).
Industrial policy, in the developing world at least, has a chequered history. Mid-twentieth century experiments with State Owned Enterprises (SOEs) and a model of Import Substituting Industrialisation (ISI) ended in economic and social torpor at best and dislocation at worst in many more cases than those that evidenced any successes. The neoliberal turn from the 1980s onwards not only meant industrial policy dropped out of fashion but also led to a fundamental restructuring of the nature of the global economy into what has become known as the post-Fordist production regime. Such restructuring was in favour of liberalisation, market integrations, and privatisations that combined into a collection of policies that became known as the Washington Consensus (Williamson, 1990).
This triggered a re-orientation of state intervention that has coalesced into what the literature has termed ‘new industrial policy’ or ‘open-market industrial policy’ (Bulfone, 2022: 6). The goal of state intervention shifted from sheltering domestic markets from foreign competition, to managing the integration of domestic firms, industries, and sectors in regional and global capital chains in ways that benefited certain actors over others (Bulfone, 2022: 6). This was accompanied by a concomitant shift in the policy instruments needed to achieve such a shift in goals.
The purpose of industrial policy can vary, but the chief goal that this paper is interested in is the one of rapid catch-up industrialisation. Given the institutional engineering required for such interventions (Cimoli et al., 2009), the sophisticated understanding of the processes of institutionalisation at the state level given in the previous section becomes essential. It is the purpose of the rest of this section to map out the changing nature of industrial policy and its associated shift from ‘old’ to ‘new’ using a framework developed by Bulfone (2022). The categories of ‘logic of intervention’, ‘protagonists’, ‘policy instruments’, and ‘constraints’ (Bulfone, 2022: 7) will be used to capture the nature of those changes. This will facilitate a mapping of the contours of state capacity and autonomy due to the model’s ability to capture the relational aspects of these changes. Consideration of state capacity and (relative) autonomy in this manner generates an ability to offer tentative conclusions with regard to the underpinnings of successful catch-up development policies as opposed to unsuccessful ones. This exercise will be completed in a third section, which will apply this theoretical framework onto the empirical case of Argentina; outlining the nature of the post-crisis social relations that generated such successful (albeit temporary) developmental outcomes.
Logic of intervention
This widening of both policy instruments and the constellation of actors required to achieve the aims of new industrial policy has led Dani Rodrik to suggest that ‘industrial policy is a state of mind, not a list of policies’ (Rodrik, 2010: 1). Analysis of this ‘state of mind’ as well as its associated policies, actors, and institutions requires the tools of state capacity and state autonomy. This is because, whilst successful policies require the necessary state capacity, the motivation to implement them is very much predicated on what Pempel (1999: 171) termed a ‘hegemonic project’. Traditional Developmental State literature suggests this comes from absolute autonomy, or authoritarianism (Johnson, 1982: 52). However, through embracing the concept of relative autonomy as analysed in the previous section, it becomes clear that a hegemonic project of rapid, catch-up industrialisation can come from a multitude of different constellations of socio-economic alliances (Haggard, 2004: 71), with the state acting as a key institutional site for the mediation of the different interests of those forces through relational power struggles and subsequent institutionalisation of those interactions into public policy.
As a result, attention in the literature has shifted away from old industrial policy associated with the sheltering of domestic companies from foreign competition through protectionist measures, improvements to infrastructure associated with specific industrial sectors, and the provision of other basic services by the state (Bulfone, 2022: 7). Instead, new industrial policy in the context of both broader societal bases of legitimacy and changing international structures consists more of policies that facilitate the adaptation of core sectors to structural shifts in the global economy, the creation of employment domestically, and the carving out of niches of international competitiveness. This has led to a necessary blending of previously separate policy realms: ‘traditional’ (sectoral) industrial policy and macroeconomic policy, most notably exchange rate policy (Bresser-Pereira and Rugitsky, 2018). Such policies represent an industrial policy fit for the twenty-first century, and will be analysed empirically in the context of Argentina under Nestor Kirchner in the next section.
Protagonists
The key actors involved with these new suites of policies have also evolved. Under old industrial policy, at least for Developmental States, analysis has traditionally centred on state bureaucrats and their relationships with industrial elites, particularly domestic bourgeoisie (Evans, 1995). The core logic of successful industrial policy under the traditional developmental state rested on the combination of bureaucratic autonomy and a subsequently high and sophisticated level of public-private cooperation. This led to the central insight that it was the degree of government-business cooperation and consensus of national goals that facilitated successful rapid catch-up industrialisation. However, ‘bureaucratic autonomy and mandarinate competence… are thin reeds on which to rest a strategy of economic development’ (Pempel, 1999: 147). This is echoed by Woo-Cumings (1999: 31): ‘…trimming some bureaucratic fat off the developmental state does not mean the end of the developmental state; rather, it is a requirement for survival’.
Embracing the concepts of capacity and autonomy as outlined in the previous section suggests that the new industrial policy of Developmental States must concern a wider array of actors. Successful capacity for new industrial policy therefore relies upon a ‘very broad range of state-society ties necessary for the effective delivery of capability expanding services’ (Evans, 2008: 18). Institutions help capture this concept of capacity within the Developmental State literature. Therefore, at the state level, these can be defined as the formal organisations of the state and society: government institutions, economic institutions, and those that connect government, economy, and society (Pempel, 1998: 23–6). For Bulfone (2022: 7), these include institutions such as national development banks, regional, and local authorities, as well as central government. Government institutions include areas such as the civil service, courts, parliaments, cabinets, the military, and the police. Economic institutions broadly express how the state interacts with the market, such as the central bank as well as specialist development banks. Institutions that connect the government, economy, and society include areas such as electoral systems, interest associations, party systems, and economic policy networks (Pempel, 1998).
Institutionalism becomes an essential concept for understanding the evolution of state capacities and autonomies in this context, as it stresses the importance of the interaction between economic activities and different actors that constitute society. Institutions regularise the organisation of decision-making through which actors’ channel and manifest their power. Subsequent institutionalisation of actor’s actions around a certain suite of policies that constitute new industrial policy is essential to ensure longevity beyond any single administration is also key. As suggested by Kohli (2004: 16–19), institutions are social patterns that gel only over time; and once gelled, they often endure beyond the forces that brought them into being. In the words of Schumpeter (2003[1942]: 12), ‘they are coins that do not readily melt’.
Policy instruments
The policy instruments of old industrial policy in the developing world revolve around the traditional development models of Import Substituting Industrialisation (ISI) or Export-Oriented Industrialisation (EOI). With differences in emphasis regarding external orientation and the ultimate destination of goods, these models had much in common. Industrial policy was essentially inward-looking, ‘as the ultimate goal of state intervention was to protect the domestic economy from foreign interferences’ (Bulfone, 2022: 5). State intervention mainly led to the fostering of a manufacturing base in sectors such as steel, car-making, and chemicals (Ibid; see also, Pianta, 2014). In other words, strategic sectors were cherry picked by the state and promoted through a variety of instruments such as tariff barriers, special credit lines, legal monopolies, and merger control. The general idea was to nurture a domestic industrial bourgeoisie that could lead national development (Rivera-QuiÑones, 2018: 47).
Whilst EOI was largely successful, ISI – albeit with some important exceptions – was largely unsuccessful; although the reasons for this remain hotly contested to this day. For the purposes of this article, the important consideration is the politics that underpinned the facilitation of these models. This is because ISI and its associated suite of economic policies were adopted in Latin America largely by ‘leftist-populist’ governments between the 1950s and 1970s. Peron and the Peronist party in Argentina are often portrayed as the quintessential populist movement, facilitating mobilising support within organised labour and lower middle-class groups, obtaining complementary backing for domestically orientated business, and politically isolating the rural oligarchy and foreign enterprises (Kaufman and Stallings, 1991: 16). EOI, on the other hand, was characterised more by embedded autonomy (Evans, 1995) and state bureaucratic links with large domestic bourgeoisie. What is clear, is that specific constellations of socio-economic relations underpin different models of political economy, which in turn generate different levels of success.
New industrial policy is characterised by a different set of policy instruments. The tools of economic diplomacy, selective market opening, taxation policy, and R&D grants can all be used as part to further contemporary industrial policy (Bulfone, 2022: 7). An understanding of the central government actors/institutions that facilitate these policies must embrace a broadened understanding of the state as a social relation, so as to sharpen our understanding of how different constellations of social forces generate the necessary capacity and autonomy to pursue developmental goals. In other words, the power of social forces as they manifest in social relations and expressed through modes of social domination constitutes the capacity and autonomy of any given regime. These modes of social domination are grounded in, for example, class, ethnicity, religion, party, status, patriarchy, and nationalism. It is the task of the next section to map out the contours of these relationships in the context of Argentine political economy under Kirchnerismo.
Constraints
Consideration of global political economy through the lens of structuration proves highly effective in complementing any analysis of twenty-first century industrial policy. Indeed, it sharpens any analysis of industrial policy. Globally, the international regime that emerged after World War 2 was founded on a principle of what Ruggie (1983) termed ‘embedded liberalism’. In its essence, a reinvigorated class compromise between capital and labour that at its core was representative of a classic compromise: unlike the economic nationalism of the thirties, the international regime would be multilateral in character. However, unlike the multilateralism of classical liberalism in the form of the gold standard and free trade, it would be predicated on domestic interventionism (Ruggie, 1983: 209). Movement towards greater openness in the international economy was to be coupled with measures designed to cushion the domestic economy from external disruptions. State intervention through industrial policy combined with movements to set standards for the social wage by constructing a variety of welfare systems (Harvey, 2005: 10). Embedded liberalism thus formed the broader international regime backdrop for the role and nature of (‘old’) industrial policy at this time, constraining not in an absolute sense but in a complex web of governance structures present at multiple sites of spatial analysis.
By the end of the 1960s, this model began to breakdown, and led to a number of responses by different states (for various reasons) that can be broadly identified by what Williamson (1990) termed the Washington Consensus. This consisted of a series of technocratic prescriptions that emerged from the International Financial Institutions (IFIs), the US Treasury, and the Federal Reserve. Such technocratic prescriptions can be summarised as ‘macroeconomic prudence, outward orientation, and domestic liberalisation’ (Williamson, 1990: 1). This was a set of neoliberal ideas, insisting upon macroeconomic stability through cutting government spending, deregulating domestic markets, privatisation, and opening both capital and current accounts to foreign finance and trade. Such structural constraints crowded out the appetite for systemic state interventionism in the form of industrial policy and its use, particularly in the developed world, fell away.
Emerging markets that desired to engage in such behaviour found their agency conditioned under the constraints of this international regime, facilitating ‘new’ industrial policy responses. Such constraints are an important consideration in the next section when analysing the nature of how this broader international regime manifested in the particular development experience of Argentina under the Nestor Kirchner. Analysis of the Nestor Kirchner regime will proceed through examination of contemporary state forms: institutions that emerge as crystallisations of social relations that are layered onto previous manifestations of such interactions. Tracing these social relations alongside institutional development, for this article in the context of industrial policy, will offer us insights into the developmental process, at least in the context of Argentina 2003–07. In so doing, it will tell us that the transition to a ‘new industrial policy’ was taking place in this country in the political space opened up by crisis; that economic recovery and some modicum of success was achieved in this process; but that this occurred alongside strong elements of policy continuity from the neoliberal period themselves the result of enduring institutionalised structures underpinned by social relations of production.
In turn, this allows us to offer some tentative conclusions with regard to social conditions under which state capacity and state autonomy are to combine in a way that could facilitate success in developmental terms. This will occur in the context of an analysis of how social relations underpin different policy profiles, subsequent processes of institutionalisation, and eventual developmental outcomes. It will show how a social contract that considered a broader range of interests generated greater focus on labour, employment, and welfare; facilitating better outcomes in terms of reducing sources of exclusion than under neoliberalism as manifested in the Washington Consensus. At the same time, it will demonstrate how both historical forms of political economy domestically and enduring international structures that remain grounded in a neoliberal global political economy served to restrain and constrain the actions of the Argentine state; allowing us to learn the lesson that social relations matter, both domestically and globally.
‘New’ industrial policy in Argentina: the regime of Nestor Kirchner, 2003–07
There was a real and important shift in Argentine economic policy post-2001 in the direction of interventionism. This shift is best understood in terms of what is currently known with regard to the nascent new industrial policy literature as outlined in the previous sections. This section will seek to demonstrate the efficacy of considering both state autonomy and capacity when analysing ‘new’ industrial policy in Argentina. This will reveal a sophisticated understanding of the nature of institutional change and the political conditions under which such change happens in the context of overarching development models and the ability of states to forge their own varieties of capitalism in the midst of an international regime characterised by neoliberalism.
Nestor Kirchner was elected to the Presidency of Argentina in 2003. He took over from an administration run by Eduardo Duhalde, who had taken over in an emergency capacity after the collapse of the government in an economic, political, financial, and social crisis in 2001. His administration ran to 2007, attempting to facilitate a new development model dubbed neodesarrollismo (Katz, 2006) that sought to refashion the relationship between the state and the people. In terms of industrial policy, Nestor Kirchner’s neodesarrollismo was centred on what Heidrich (2005) called a ‘competitive change’: a shift of incentives to encourage production (or industrial sectors), rather than historical incentives under the Presidency of Carlos Menem in the 1990s that were directed towards the financial sector in the form of banking and capital markets (Ferrer, 2005: 370). A clear logic of intervention grounded in neodesarrollismo facilitated a strategic move away from a neoliberalism that failed so spectacularly in 2001–02, towards one of adaptation to structural shifts in the global economy that came in the form of the fostering of an export-oriented manufacturing base that stimulated employment. This was made possible through a re-orientation of social relations in the wake of the crisis of 2001–02. The traditional pueblo of unionised workers, lower-middle classes, and the urban poor associated with Peronism – Nestor Kirchner’s base – was supplemented by middle classes who longed for financial stability and a macroeconomic environment more conducive to stable consumption and investment.
This particular ensemble of social relations generated a suite of policy instruments that chimed with neodesarrollismo. In a world of multi-level governance, organised through processes of structuration, consideration of regional and international actors and their interactions with the state is crucial. Especially when these interactions underwent fundamental change in the period under discussion. In the context of a crisis that had led to dislocation and then isolation from international capital markets Argentina was able to restructure its defaulted debt with the IMF on very favourable terms, changed its relationship with foreign investment, and began the process of renegotiating its position in the international division of labour through an active industrial policy based on export promotion. Significant domestic demonstrations against the IMF (who was blamed for precipitating the events of December 2001/January 2002), combined with their support for the currency peg that had ultimately facilitated the crisis led to a weakening of their position vis-a-viz the Argentine state. The outcome was a large haircut on the defaulted amount, a 43.5% nominal reduction (Damill et al., 2007: 22), and restructuring of the remainder on more favourable terms. This deal, or megacanje, was accepted beyond the IMF and included all private debt holders (or 76.2% of them at least).
Argentina’s relationship with foreign investment was also different to the past, both the neoliberal past of Menem whose policies favoured the financial sectors of the economy and the structuralist past of Peronism that was outright hostile to Trans National Corporations (TNCs). Instead, Nestor Kirchner engaged and accommodated foreign business that already had a presence in Argentina, but was more hostile foreign capital looking to initiate interest in the country. He maintained links with international finance and banking to court investment to fund expanded consumption, but did not actively promote their interests like under more neoliberal times (Wylde, 2011: 448).
The policies associated with industrial policy represent a blending of what has traditionally been demarcated in the literature: industrial policy and macroeconomic policy (Bresser-Pereira and Rugitsky, 2018). Therefore, a programme of public works and infrastructure projects, selective protective tariffs, funding for small and medium enterprises (SMEs or PyMES in Spanish), and promotion of exports of a manufacturing origin were all pursued. Public works and infrastructure projects were more significant in the provinces and leaned heavily towards extraction, but nevertheless also included road building and bridge construction. This process was further limited by lack of capacity, with a relative absence of active policies aimed at strategic reconstitution of production chains (Bunga and Porta, 2008: 41). Nevertheless, industrial policy was used to help promote private levels of economic activity in areas such as mining, extraction, and agricultural products; as well as significant growth in mechanical engineering, electric machinery and equipment, and automobiles – which were all the fastest growers across the period (Kosacoff, 2008: 40). These products were often focused on export markets: with automobiles often exported to Brazil – Argentina’s largest trading partner –, electric machinery to the USA, or mechanical engineering products going to Europe (Bezchinsky, 2007: 27).
Crucially, this was in combination with an exchange rate policy dubbed the ‘SCRER’ (Stable and Competitive Exchange Rate Policy) that supported and underpinned those policies that were more sectoral in nature (Frenkel and Rapetti, 2008; Ortiz and Schorr, 2009: 3). Briefly, the SCRER was a policy of exchange rate manipulation to keep the Argentine peso artificially weak, to stimulate exports. This also facilitated broader macroeconomic stability as the sterilisation of the influx of dollars led to healthy balance sheets for banks, which had been rocked by the financial crisis of 2001–2, and a positive impact on national debt to GDP ratios (Ibid; see also, CEPAL, 2007). This complemented the rest of Kirchner’s suite of industrial policies through export stimulation, as well as further injections of funds for other welfare programmes, debt reduction, and investment.
Internal active state spending on infrastructure combined with stimulation of private investment into sectors that benefited from a low exchange rate, itself partly a product of active state policy, facilitated growth of an export driven economy that had climbed the value-added food chain in comparison to previous eras of Argentine catch-up industrialisation (Bezchinsky, 2007; Chudnovsky, 2007). In addition, there was significant revenue for the Argentine state (Heymann, 2006). This was spent on employment programmes (Beccaria et al., 2007), welfare programmes (Grugel and Riggirozzi, 2007), and other areas of political economy that generated modest outcomes in terms of health, education, and other development indicators. This occurred under twin surpluses in the fiscal and current accounts (CEPAL, 2007), helping facilitate a broader macro-economic stabilisation. In the context of the recent crisis of 2001–02, this was a very welcome phenomenon. Poverty declined, as did inequality (Kosacoff, 2008: 43).
This was a clear movement away from policies linked to traditional industrial policy, and politically in Argentina associated with Peronism. An awareness of the inefficiencies associated with many forms of state intervention and state ownership, the acceptance of the primacy of the market in setting prices, the abandonment of economic protectionism in favour of relative economic opening and regional integration, and a general welcoming of foreign investment were all present in Argentina under Nestor Kirchner (Panizza, 2005: 12). He proclaimed the need for ‘capitalismo en serio’ (serious capitalism), which led to an understanding of the need for sound fiscal policy and the importance of low inflation. This manifested in targeted social welfare as opposed to widespread and indiscriminate social programmes (Grugel and Riggirozzi, 2007: 98), a new form of segmented neocorporatism (Etchemendy and Collier, 2007) rather than old Peronist-style corporatism, prioritising inflation control over wage increases, and limited attempts to address poverty and indigence in the country.
Continuing with Bulfone’s framework, protagonists came in the form of central government and powerful provincial governors, who facilitated key infrastructure developments in order to stimulate complementary private sector investment. These governments possessed the relative autonomy to enact the policies outlined above due to the nature of neodesarrollismo as a mode of political economy and its concomitant social relations. As previously stated, the social relations that underpinned this, or facilitated a relative autonomy for the state to engage in such behaviour, were an interesting expansion of the pueblo to include consuming middle-classes, and investing domestic and international bourgeoisie. In terms of industrial policy, these were collectively interesting policy manifestations of change and continuity. Prioritisation of sound money, combined with fiscal surplus, very neoliberal concepts, were combined with welfare promotion and industrial policy of a more interventionist and developmentalist nature.
In sum, we see all the elements of Bulfone’s (2022: 7) framework for understanding new industrial policy present. These were made possible by a judicious blend of social relations that generated a relative autonomy grounded in neodesarrollismo; one part of which was changing industrial policy. Neodesarrollismo then represented a historical condensation of social relations that expanded the pueblo in Argentina during the Presidency of Nestor Kirchner, fusing elements of labour and elements of capital that generated positive developmental outcomes. In more concrete terms, an expansion to a broader social contract that created more inclusive state forms that significantly reduced poverty (Riggirozzi, 2009: 98), inequality (Riggirozzi, 2009: 99), and facilitated areas of industrialisation in value-added production – particularly in the export market (Bezchinsky, 2007). The possibility for more radical solutions to Argentina’s post-crisis structural issues was blunted by both historical forms of neoliberal political economy combined with the ongoing characterisation of global political economy as neoliberal. The need to integrate key elements of international capital such as the IMF, international investors, and transnational bourgeoisie means that the potential for new industrial policy to more fundamentally address sources of exclusion and inequality in society remained spatially and temporally limited.
Conclusions
This article has attempted to show how the fusing of Bulfone’s taxonomy (2022) with analysis of a state’s capacity and (relative) autonomy generates powerful understandings of contemporary, or ‘new’ industrial policy. What becomes clear from this analysis is that state capacities matter. Efficient bureaucracies that implement appropriately crafted policy instruments are key. Simultaneously, these capacities are grounded in a logic of intervention that facilitates a relative autonomy in social relations that generate a national interest in the direction of developmentalism. The contours of these social relations thus also become key to understanding the success or failure of such industrial policy through application of the concepts of state autonomy and state capacity. Empirical analysis of states in relational terms reveals the equilibrium of compromise among class forces, which gives us a basis for understanding different criteria for success in terms of twenty-first century industrial policy. By looking for the presence of a series of policy instruments enacted by certain protagonists through a specific logic of intervention, we can discern the presence of the right capacities and autonomies that may generate positive developmental outcomes.
In addition, this intersects with the global political economy through a process of structuration, creating political economies of the possible (Santiso, 2006) in the context of powerful agents above the state. Understanding the relative autonomy of a state with these power relationships integrated into the analysis becomes crucial. Limitations are inevitably placed on states by international agents, which in the contemporary era are grounded in neoliberal principles. The room for manoeuvre however changes: crisis and dislocation from international capital markets, strategic insertion into the evolving international division of labour, and relationships with foreign investors all shape the ultimate nature of a state’s relative autonomy to pursue an independent development path.
The case of post-crisis Argentina under Nestor Kirchner tells us that broad-based social contracts that fuse old and new social relations under the right international conditions can crystallise into modest developmental outcomes. More specifically, a logic of intervention grounded in neodesarrollismo that emerged in post-crisis Argentina was the result of expanding the traditional Argentine social pact based on the urban working class, lower-middle classes, and urban poor through integration of consuming middle classes, national bourgeoisie, and limited input from international capitalist classes. A relative autonomy that fused elements of labour and elements of capital in a judicious manner facilitated a new industrial policy that demonstrated impressive results, albeit for a limited time. The reasons for this limitation lie beyond the scope of this paper, but analysis of changing relative autonomies and state capacities through consideration of logics of intervention, protagonists, and changing policy instruments – sharpened with analysis of the constraints placed by the changing contours of global political economy – would represent a starting point for such an endeavour.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
