Rethinking Growth Model Analysis: A Life Cycle and State-Centric Relational Approach

Since the publication of ‘Rethinking Comparative Political Economy: The Growth Model Perspective’ in this journal in 2016, growth model analysis (GMA) has become the subject of a sustained research programme. This article offers a critique of both the overall focus and key substantive arguments associated with GMA and, in doing so, develops an alternative perspective on how growth models function over time, especially in relation to the politics shaping growth models.

Our critique centres upon two arguments. First, GMA downplays the role of the state in the politics and governance of growth models and, conversely, overstates the significance of producer coalitions and business actors. Second, current GMA analysis offers a rather static account of growth models, with Stockhammer arguing that GMA is ‘by design focussed on national comparisons at a given point in time.’ ¹ We introduce a more dynamic account of growth models: a life cycle account. We argue that current GMA focuses almost exclusively upon the functioning of existing and established growth models, neglecting two other stages in their life cycle relating to their initial establishment and eventual decline or termination. Our two critiques are closely connected. States, we argue, are very likely to play a leading and decisive role in the creation and termination of growth models. Hence the neglect of these two stages leads to a relative neglect of the state. This does not, however, mean that the state, during the mature, middle, stage of their life cycle plays little or no role in the maintenance of an existing growth model. Drawing on an established literature on the ‘relational’ power of the state, we argue that states often cooperate with business interests to advance and protect an existing growth model and that they do so not because they are necessarily coerced or beholden to those interests but because they independently share a favourable assessment of the growth model's utility.

In developing our alternative ‘state-centric relational’ account, we draw upon a ‘statist’ literature associated with the work of, among others, Theda Skocpol, which, in seeking to ‘bring the state back in’ to political science, highlighted the distinctive interests, capacities and authority of the state. ² In analysing relations between states and business and producer interests we also draw on a distinction we have previously drawn in our work on business politics between coercive business power and noncoercive business influence in relation to state actors. ³ We also employ the concept of ‘structural prominence’: a term developed by Kevin Young in his useful sectoral analysis of the variable nature of the structural power of business but which we argue can also be used to help identify which sectoral interests are likely to be able to exert influence in relation to the state. ⁴

Armed with these concepts, the article's central argument is that states typically play an authoritative role in relation to business and producer interests across the life cycle of a growth model, albeit featuring differing patterns of state and producer group relations at different stages of the growth model life cycle. It is usually the case, we argue, that states play a dominant, authoritative role during the initiation and termination of a growth model. However, during the middle or maturation stage, relations might be more variable, perhaps featuring business and producer group power in relation to the state. Yet, more often, we argue, business and producer interests will seek to influence authoritative state policymakers in a context where producer groups and the state share a degree of mutualism and a commitment to the success of the prevailing growth model.

We illustrate our argument about the salient role of the state in developing, maintaining, and terminating growth models through an analysis of the so-called Australian Settlement. This was an economic growth model based primarily on manufacturing industry development mainly through protectionist tariff policy linked to various types of labour protections. This was the dominant growth model that persisted during the first six decades of the twentieth century in Australia. In the post–World War II era, the model matured further in becoming the productive foundation of a postwar Fordist growth model that saw rising productivity strongly linked to rising wages underpinned by growing trade union strength and a centralised wage fixing system. An important justification for using this case is that it perfectly illustrates our key argument that states usually play an authoritative role in relation to business and producer interests across the life cycle of a given growth model. As we show, the role of the state was central in growth model initiation and termination, whilst during the maturation phase, business and producer interests engaged in cooperative mutualism with the state because of largely shared goals and agendas in supporting and further developing the growth model.

We also briefly examine some further cases of growth model dynamics from a range of time periods and countries which also highlight aspects of our state-centric relational GMA. These include the development of Germany's export-based growth model and the unintended evolution of Britain's ‘Anglo-Liberal’ growth model centred on finance and housing-led economic growth, the power exercised by the American financial industry following the 2008 crisis, the role of the state within the East Asian developmental state model, the political impotence of the British finance industry during the Brexit negotiations and the decline of the British coal industry. These cases support our state-centric relational account although the contrast between the power of the state in the British finance case during Brexit compared to the power of finance in the US case after the 2008 crisis illustrates our argument about the potential for more variable patterns of business-state relations during the mature stage of a growth model.

We proceed as follows. In the next section we identify the key claims made in relation to the existence and functioning of growth models and track important developments in GMA since the publication of Baccaro and Pontusson's original article. Here, we focus on a debate about the how state policies apparently come to be aligned with the functional economic requirements of growth models. In particular, we focus on Baccaro and Pontusson's argument—one subsequently endorsed and developed by, amongst others, Rothstein, and Frank, Arthur, and Friel—that ‘producer coalitions’ representing the sectoral interests of businesses within a growth model can effectively shape state policy. ⁵ In the third section we develop our alternative state-centric relational account, arguing that the state can and does operate with a degree of relative autonomy from and authority over societal forces and is capable of independently developing its own views as to which economic sectors are best positioned to contribute to future growth: views which may not necessarily align with those being expressed by business or producer interests. The three remaining sections of the article then develop and illustrate this argument in relation to the establishment, maintenance, and termination of growth models: returning, in each section, to a discussion of the Australian Settlement.

Growth Model Analysis

GMA, in its most recent incarnation (see below), was initially advanced by Baccaro and Pontusson as a critique of and an alternative to the then dominant ‘Varieties of Capitalism’ framework developed by Hall and Soskice. ⁶ Baccaro and Pontusson argue that in order to properly understand the nature and dynamics of capitalist economies our attention should focus not just on firm-level institutions and the existence of institutional complementarities, as in the VOC approach, but more directly upon economic growth and the channels through which aggregate demand expands (or fails to expand). They argue that, in the decades following the conclusion of World War II, growth was sustained in Western economies through a dominant ‘Fordist’ growth model characterised by a combination of rising wages and rising productivity facilitated through the institutionalisation of collective bargaining and an associated rise in trade union power. That model began to break down in the 1970s and had, they suggest, almost entirely disappeared by the mid-1990s.

This leads Baccaro and Pontusson to a set of related questions which have since defined the GMA research programme: ‘How have advanced capitalist political economies addressed the problem of finding a replacement for wage-driven growth? Have they all followed the same path? Have some countries been more successful than others in transitioning to another growth model?’ ⁷ Their answer to these questions is that, at a national level, capitalist economies are characterised by the existence of a relatively small and stable number of growth models, distinguished by their reliance upon consumption-led growth facilitated by credit expansion (exemplified by the United Kingdom), export-led growth (exemplified by Germany), or a more ‘balanced,’ hybrid, model characterised by high levels of investment, exports and rising wages (exemplified by Sweden). ⁸ Finally, Baccaro and Pontusson argue that some countries (Italy, for example) have failed to find an alternative to the Fordist model and so have struggled to generate sustained economic growth.

Baccaro and Pontusson are careful to acknowledge that their work ‘resonates’ with and ‘draws’ upon earlier work on the nature and functional requirements of growth within Neo-Kaleckian macroeconomics and the French regulation school. ⁹ Indeed Amable, in a sometimes-coruscating review, argues that Baccaro and Pontusson add very little which is new to this existing work on growth models whilst neglecting many of the key insights within it. ¹⁰ Nevertheless, it is now clear that GMA, in its most recent incarnation, led by Baccaro and Pontusson, has achieved a salient place within Comparative Political Economy. We focus upon their work and that debate in this article.

As the research programme associated with GMA has developed new economic indicators, including private borrowing and public spending, social protection, fiscal policy, and the role of the ‘supermultiplier’ have been added. At the same time, new institutions and structural elements, including central banks and financialisation, have been incorporated within GMA and new levels of regional and city analysis explored. This work has resulted in the identification of further types, or subtypes, of growth models; a gendered analysis of growth models; and efforts to reconcile the key insights of GMA with that of the Varieties of Capitalism approach. ¹¹

Our focus in this article, however, is on a specific debate within GMA about the political economy of growth models. In an essay on ‘making sense of populist nationalism’ contained within a special issue of New Political Economy on ‘blind spots’ in political economy, Andrew Gamble noted that whilst the term ‘political economy’ might suggest that the political has equal weight with the economic in explaining how societies and economies evolve, in practice political economists have tended to give primacy to the economic. This has meant that the political has often been understood only in terms of political arrangements appropriate for a particular form of economy, rather than in broader analytical terms relating to the way in which the economic is also constituted by the full gamut of political and power relationships and ideas. ¹² GMA is, we argue, illustrative of this tendency.

In their original article Baccaro and Pontusson recognise, in passing, that a growth model, if it is to function effectively and deliver growth, must be supported through an appropriate range of state policies. Yet how can we know whether the ‘right’ set of policies will be selected by the state? How, in other words, can we be sure that the economics and politics of growth models will align? Baccaro and Pontusson's initial answer to this question is that an alignment will be achieved because democratically elected governments have every reason to prioritise economic growth, and presumably the interests of key producer groups, in order to secure reelection. Amable suggests that this argument exhibits ‘a somewhat basic economic functionalism by subordinating politics to a single economic imperative, namely growth.’ ¹³

Having recognised, in their 2016 article, that ‘politics needs to be introduced into our analytical framework in a more systematic fashion than we have managed to do so far,’ Baccaro and Pontusson return to the issue of how the economics and politics of growth models become aligned in a later article devoted to ‘the politics of growth models.’ ¹⁴ Drawing on earlier work within Comparative and International Political Economy on ‘producer coalitions’ involving business and perhaps labour interests, they argue that ‘growth models are sustained by a coalition of organized interests that enjoy privileged access to the policy-making sphere and ensure that government policies, especially macroeconomic policies, reflect its interests and the functional requirements of the growth model.’ ¹⁵ They argue that ‘the key policy foundations of a growth model are shielded from electoral competition,’ that ‘governments of different political orientations internalise the functional requirements of the respective growth models,’ that ‘growth coalitions successfully project their key interests as the “national interest,”’ that ‘established parties converge on the policies that are crucial for the growth model,’ and that ‘in most times and circumstances, producer-group politics prevails.’ ¹⁶ This argument has subsequently been endorsed by Rothstein, who argues that ‘producer groups constitute social blocs, whereby sectors with complementary interests band together to shape economic policy,’ with ‘the growth models perspective focused on business as the primary actor in a social bloc.’ Similarly, Frank, Arthur, and Friel argue that growth models ‘depend on dominant growth coalitions … drawn primarily from the ranks of producer groups.’ ¹⁷

But why are growth coalitions featuring business and producer groups apparently able to shape government policy so effectively? Here, Baccaro and Pontusson follow Culpepper in arguing that because economic policy is a highly ‘technical domain,’ it ‘lends itself more to the quiet politics of technical agencies and committees, in which well-defined sectoral interests are more likely to be heard.’ ¹⁸ Baccaro and Pontusson also argue at one point that ‘key sectors have structural power’ and that ‘governing parties (politicians) have strong electoral incentives to cater to the interests of these sectors.’ ¹⁹ Rothstein similarly argues that ‘social blocs have been conceived as exercising a type of power that is structural and historically anchored: systemically relevant sectors, those that contribute to GDP and employment, are likely to see economic policy meet their interests.’ ²⁰

It may indeed be true that producer groups can at times exert either influence or power in relation to states in shaping growth model politics. Yet the problems within the current GM literature are that, first, the prevailing arguments about producer group dominance are only very partially explored and are often simply asserted; second, they contain, as noted, strong functionalist elements; and third, the accounts accord virtually no detailed or active role for the state. This last is important because arguments about influence or power in this context must always be focused on the detailed dynamics of relations between producer interests or growth coalitions and states.

Currently within the GM literature there are only a few hints of a more nuanced approach in this regard. One comes from Frank et al., who comment that ‘there is often a startling degree of collaboration between mainstream political parties with respect to the operation and maintenance of GMs.’ ²¹ True, this could reflect functionalism and producer interests but at least the role of political parties is acknowledged. Another comes from Baccaro and Pontusson, who suggest that in moments of crisis within an existing growth model ‘new political entrepreneurs, within or outside mainstream parties, may attempt to intervene.’ ²² An even more rounded account is provided by Bohle and Regan, who recognise ‘the power that specific societal groups [or sectors] exercise,’ but opt for a model of mutual interdependency between producer groups and the state, arguing that growth models ‘are the product of those strategies that key actors—government, producer groups, and business elites—pursue.’ They further argue that their account ‘stresses the reciprocal and mutual dependency between business and state interest and treats business power as a variable.’ ²³ This is indeed an important insight, one which this article is designed to further elaborate and sustain.

Typically, as we argue, such mutualism would see key producer groups working with governments to sustain sectoral growth models. At other stages in a growth model's life cycle, especially during initiation and termination, we argue that states play a more prominent role. This illustrates that there needs to be space in a theory of growth model politics for authoritative governments and state agencies. Otherwise it is hard to account for how growth models change, especially over their life cycle, when key interests, though in varying degrees, both within and beyond the state, play a role in the initiation, development, and termination of growth models.

A State-Centric Relational Account

One of our major arguments is that Baccaro and Pontusson's account, and more broadly the current GM literature, significantly downplay the capacity of government and state actors to act with a degree of relative autonomy and authority in relation to societal interests, and offers, for this reason, a defective account of how and why states operate in the governance of growth models.

Our alternative ‘state-centric relational’ account is derived from the earlier work of ‘statist’ theorists such as Krasner, Skocpol, Nordlinger, Tilly, and Mann, who all sought to ‘bring the state back in’ as a reaction to its near elimination by an earlier generation of political scientists who either saw the state as a cipher for other societal forces, most obviously capital, or viewed the concept of the state as a ‘conceptual morass.’ ²⁴ The core of this statist position was that states, by virtue of being states, have a recognisable set of distinctive interests, resources, and capacities, which means they are almost always capable of exercising a measure of relative autonomy or authority in relation to societal forces.

Statists argue that states have an enduring primary interest in securing their own survival: an interest which requires them to satisfy the material demands of their citizens, repel internal or external threats to their autonomy and sovereignty, and, in doing so, to secure their legitimacy. ²⁵ In most economic, political, and social contexts, this also gives states—and not simply politicians eager to secure reelection—a secondary but salient interest in achieving economic growth as all of the above primary objectives are more likely to be achieved with economic growth. As such, states and state actors, we argue, will want to develop a successful growth model if one does not currently exist; will want to work with and support business interests within an existing growth model if they believe that this growth model can continue to generate growth; will have an interest in resisting the demands of business interests (whether inside or outside of the existing growth model) if they believe that they are advocating policies which constitute rent-seeking that might jeopardise future growth; and, finally, will have an interest in terminating an existing growth model if they believe that it is no longer delivering growth or if they believe a superior alternative is available to it.

What is a growth model? Baccaro and Pontusson argue that growth models are answers to the question of how best to secure economic growth. They also suggest, however, that growth models are ‘descriptions’ of the current workings on an economy which can be derived from data on ‘the decomposition of GDP growth.’ ²⁶ On such an account, growth models can be objectively inferred and verified. Against this we argue that growth models can also be partly understood as ideational constructions defined by key interests of how an economy is functioning and of how future growth might be achieved.

Business interests operating within a growth coalition are likely to possess strong views on how growth is being achieved, and which state policies will best sustain it. But states are not simply ciphers of societal forces. State actors, drawing on their own ideas and the expertise of an organised bureaucracy—a key source of state capacity in statist accounts—can also engage in ‘economic reasoning’ about how best to secure future economic growth and in doing so can develop and, over time, revise and change their own subjective accounts of growth models. ²⁷ States thus operate with a relative degree of autonomy in this regard. As James recognises, government ‘policymakers are embedded within institutional structures that provide valuable political and financial resources, which can be used to accumulate technical knowledge and expertise.’ Business, James adds, might, in an exercise of ‘instrumental power,’ seek to use its own sources of expertise to lobby government, but that does not necessarily mean its arguments will always be considered credible. ²⁸ As Crouch argues, ‘There is always an area for discretion in the extent to which governments take seriously industry's [views] or complaints.’ ²⁹ In an article on postcrisis banking regulation in the United Kingdom, for example, Bell and Hindmoor show how the economic expertise of the independent Vickers Commission and of the Bank of England contributed to the eventual acceptance of proposals relating to increased bank capital and to the ‘ring-fencing’ of the capital assets of the commercial and investment arms of the largest banks against strong opposition from the banking industry. ³⁰

Business interests do not, however, simply have to rely upon their claimed expertise to influence policymakers to help secure their goals. They may attempt to exert structural power over the state through threats to reduce their future levels of investment in ways which will negatively impact upon economic growth. The concept of structural power in relation to business interests was initially developed by Charles Lindblom and Fred Block, with Block in particular pointing, even in this initial formulation, to the potential variability of structural power depending on the context. ³¹ Subsequently, in one of many more recent articles which have sought to clarify the nature, sources, extent and variability of structural power, Young advocates the use of a sectoral approach focusing on ‘differentiations within the business community as a way of assessing which economic entities actually matter.’ ³² He then develops a two-step analysis arguing that the structural power of business is about a hypothesised relationship between cause ‘(state dependence on business investment etc.) and effect (what governments or states do in response).’ For Young, the key variable here is the economic importance, or, what he terms the ‘structural prominence’ of a firm or sector. He thus proposes that

to improve the language we use to discuss structural power we call the things that we think might engender structural power ‘structural prominence,’ which is different from structural power itself. The reason for this differentiation is it allows us to conduct empirical evaluations of the structural power of business as an empirical conjecture, rather than to conflate hypotheses about cause [structural prominence] and effect [the actual exercise of structural power]. ³³

Young does not suggest that there is a linear and deterministic relationship between structural prominence and structural power. He does, however, argue that ‘there is evidence that structurally prominent firms may be catered to more than less prominent firms.’ ³⁴

What makes a firm or sector structurally prominent? Young points to the significance of a sector's size, value-added or employment. Similarly, Fairfield suggests that ‘structural power will tend to be stronger where that sector contributes a large share of GDP, generates significant employment, or has multiple linkages to other sectors that amplify the economic impact of its investment decisions.’ ³⁵ Massoc also concludes that ‘a firm or economic sector is deemed structurally powerful when, due purely to their position in the economy, weakening them would also result in weakening an essential dimension of the whole political economy.’ ³⁶

We do not necessarily disagree with Young's analysis here. Yet we need to return to our previous point that growth models are not simply descriptions of how growth has occurred but are also ideational constructions of how an economy functions, why growth occurs, and what the future may hold. Here we would argue that a conviction, on the part of key state actors, that a particular sector and the businesses within it which are particularly well-placed to generate future growth, will result in that industry being viewed as structurally prominent.

If a particular economic sector is viewed by state actors as being important for future growth and as being structurally prominent, then businesses within that sector, and their organised representatives, may be well placed to exercise structural power threats in order to achieve their policies. Yet, even in such cases, state actors retain a measure of autonomy and authority insofar as they must decide whether a threat to reduce investment levels or to move to another jurisdiction is plausible as well as potentially damaging. Bell has argued that the views and ideas of state actors in this regard can be crucial in shaping the perceived degree or impact or any structural power moves by business. ³⁷ Relatedly, in an article on the introduction of an Australian ‘super profits’ mining tax, Bell and Hindmoor show how the federal state initially dismissed the objections of the mining industry because it considered the investment threats the mining industry was making were hollow. ³⁸ Similarly, and in relation to postcrisis financial reregulation proposals, the British state dismissed threats by the large, global bank HSBC to move its headquarters to Hong Kong. ³⁹ Bohle and Regan, for their part, note that whilst multinational firms may be in a strong position to bargain with states before deciding where to invest, ‘most global firms are more dependent on national governments than the structural power perspective would often assume’ because initial investments ‘create sunk costs, making the firm less mobile, particularly in high-tech manufacturing.’ ⁴⁰

We have so far focused upon the ability of organised business interests to achieve their goals through lobbying (instrumental power) or threats to reduce investment (structural power). Yet whilst important, this focus is also too narrow for it ignores other forms the relationship between government and business interests can take which do not involve conflict and power but instead entail business's noncoercive influence and a broader notion of cooperation or mutualism between states and business.

One aspect of this is when business finds itself in a ‘systematically lucky’ position due to an alignment of its interests with the state. ⁴¹ As Dowding argues, if business gets what it wants, this may not be due to power but instead because business interests can often be ‘systematically lucky’ in the sense that ‘what is in the interests of government is also, by and large, in their [business] interests as well.’ ⁴² In the context of this article, firms or sectors that are regarded by the state as being structurally prominent because they are part of a viable growth model may find their interests converging with the state and vice versa, thus being favoured without having to attempt to exert power.

Such interaction usually occurs in the arena of ‘quiet politics.’ Yet in contrast to Culpepper's view, where quiet politics is seen as an arena of business power, our argument here is that quiet politics is often an arena where business is well placed to exert noncoercive influence, trading on firsthand information and business or sectoral expertise. ⁴³ This important distinction between coercive power and noncoercive influence is explained and illustrated by Bell and Hindmoor, with the latter influence strategy typically being favoured as a first option by business interests, especially given the potential costs and risks associated with more coercive strategies. ⁴⁴ Noncoercive influence entails efforts by business lobbyists to persuade and shape the views of state actors regarding appropriate policies in the absence of any conditional or coercive threats or offers. As one senior corporate executive in government relations in Australia argued, ‘in the majority of cases’ our aim is to ‘exert influence,’ and that only in ‘extreme circumstances’ would there be an attempt to be ‘coercive’ in relation to government. This respondent was also adamant that power-based coercive ‘tactics are the exception rather than the norm.’ ⁴⁵

From the state's perspective, this pattern of noncoercive business-state interaction arises when states develop and deploy what Mann, a key theorist in the revival of the statist position, refers to as their ‘infrastructural power.’ Mann argues that within modern liberal states the deployment of despotic, coercive force has decreased and that the key source of state capacity, one whose use has ‘increased enormously,’ is its infrastructural power, that is, its capacity to penetrate civil society and, through the formation of coalitions with societal actors, achieve its goals. ⁴⁶ In other words, the state derives its capacity not, as Skocpol suggests, by separating itself from society, but by collaboratively integrating with it. This state-centric relational perspective suggests that in place of a zero-sum conception of business-state interaction, both business and state actors possess resources and seek to adjust to each other's position. As Marsh and Chesters argue, ‘The relationship between government and business is an exchange relationship; one which is not asymmetric,’ with Woll adding that ‘even when business appears to lead the dance, it is more promising to look at resource distributions and the interdependence of both sides, instead of assuming the domination of business power over policy outcomes.’ ⁴⁷ On such an account, it is the ‘relational’ power of the state, one mobilised through the development of close working relationships in ‘quiet’ political arenas governed by the existence of mutually recognised protocols or ‘rules of the game’ which constitute a key source of state capacity. ⁴⁸

In his work on the governance of the East Asian development state, Evans has, in this way, emphasised the embedded autonomy of the state. ⁴⁹ Doner points to a ‘complex bargaining process between state and the private sector’ in the development of East Asian export industries after World War II. ⁵⁰ Booth similarly argues that policymakers in Southeast Asian states have ‘always been ready to listen to influential business lobby groups, who in their turn have been catalysts of policy reform, and active in promoting structural adjustment measures.’ ⁵¹ In a detailed account of the development of the South Korean export-led industrialisation strategy, Chibber pushes back against accounts which present the state as having ‘impose[d] a new development strategy’ and instead argues that the state pursued its new development strategy ‘in alliance, with domestic business, not over it.’ ⁵² Yet this does not necessarily mean that the state and business engage in such situations as equals. In her account of the East Asian development model Weiss argues that, far from being ‘subservient,’ the state consistently exercised a ‘definite capacity to set broad goals and to coordinate policies and resources to that end.’ ⁵³ In the Korean case analysed by Chibber, for example, the state, having decided to pursue an export-led strategy, imposed strict performance conditions upon business in return for its offer of financial and other support.

Having outlined the basic terms of our ‘state-centric relational’ account of growth model politics, we now need to turn to the question how such politics and the pattern of business-state relations vary across the three phases of the growth model life cycle. We argue that states are most likely to act unilaterally and decisively during the initiation and termination of growth models, whilst in the middle or mature stage, when a growth model has already been established, the relationship will be more variable and that business interests will be in a stronger position to exert influence or perhaps even power, sometimes successfully, though sometimes not.

Growth Model Initiation

Baccaro and Pontusson's argument about the key role played by growth coalitions in shaping government policy crucially assumes that a growth model is already in existence and that business interests have mobilised to shape and support it. But growth models do not simply exist; they must be established. As Bohle and Regan remark in their discussion of the development of the Irish and Hungarian growth models, ‘growth models do not emerge from thin air.’ ⁵⁴

Where, then, do they come from? It is entirely possible that they can sometimes emerge from the interplay of market forces and private capital in ways which reflect the unanticipated consequences of existing state policy or regulation, but which were not the express purpose of such interventions. This, arguably, was the case with the ‘Anglo-liberal,’ consumption-based, growth model in the United Kingdom. As both Hay and Barnes show, this developed as a result of the rapid growth of the City of London following the collapse of the manufacturing sector in the early 1980s, the ‘Big Bang’ financial reforms of the mid-1980s, and the combination of low interest rates and rising house prices (the latter the result of the collapse of new social housing construction which made owning and trading home ownership an attractive form of ‘asset based’ wealth). ⁵⁵ The resulting growth model, which was eventually recognised and then supported by the incoming Labour government in 1997, was a ‘largely unanticipated and unsought consequence rather than the product of a more conscious plan or strategy.’ ⁵⁶

Yet, historically, the development of new growth models has often been consciously led by states precisely in order to escape from an existing market or developmental trajectory, often at moments of acute economic and political challenge. Here, state actors, drawing upon their empirical understandings of the global and national economic context, assessments of any existing growth model and its viability and future prospects, theoretical assumptions and understandings of how key economic variables interact, and institutional understandings of policy capacities and the operation of policy instruments, will seek to identify a future growth model through an act of what Beckert calls economic ‘imagination’ or, to use an alternative phrase recently employed by Weiss and Thurbon, one of ‘economic statecraft.’ ⁵⁷

The development of the German export model after World War II offers an example of the state-led initiation of a new export-led growth model, one centred upon the automobile, chemical, and machine engineering sectors. It can be dated back to the establishment of the sovereign independence of the new German Federal Republic in 1949 and its desperate need to fund postwar reconstruction efforts whilst achieving a balance of payments surplus. ⁵⁸ Its development was led by state actors in the Economics Ministry and the Christian Democratic Union party, including, crucially, Economics Minister Ludwig Erhard, and was supported through the provision of subsidised loans, export financing, credit insurance and the relaxation of American-imposed anticartel rules.

The state-led development of a new growth model also occurred in Australia at the start of the twentieth century. In the nineteenth century, colonial state governments prior to Australia's federation in 1901 had sought to achieve growth through a model based mainly on state investment in infrastructure and the market-based export of commodities such as wool and gold. The state's infrastructure investment, largely financed by Britain, was important, with Butlin describing it as a system of ‘colonial socialism’ whereby private investment did not exceed state investment until the 1930s. ⁵⁹ However, the depression of the 1890s, during which commodity prices fell, highlighted Australia's vulnerability to international economic shocks in commodities markets. ⁶⁰ That economic crisis and associated class conflict saw a strengthening of the labour movement, and the formation of a federal state system in 1901. This resulted in a substantial increase in state capacity and the development of a new growth model based on a conviction that, as Australia's second prime minister, Alfred Deakin, argued, ‘no nation ever claimed national greatness which relied on primary industry alone.’ The prevailing view was also that a narrow, commodities-based economy would never support the population Australia was seeking and that Australia needed to follow Britain, Germany, and America in developing a ‘modern’ industrial economy. ⁶¹

The key aims of the new growth model—which has variously been referred to as ‘Domestic Defence,’ ‘New Protectionism,’ or the ‘Australian Settlement’—were to restructure the economy, transfer income to the urban sector, better shield Australia from the volatility of international markets, support labour interests and Australia's high-wage economy, and spur much-needed population growth based on new urban industries. ⁶² The resulting policy framework, as it was developed by the federal state, consisted of tariff protectionism and other subsidies for manufacturing industries, compulsory state arbitration of industrial disputes, a mandated wages floor, and a highly restrictive ‘white’ Australia immigration policy designed to protect domestic labour interests. ⁶³ The strong link here between industry protection and labour protections was fundamental to the new model. Over the first six decades of the twentieth century, as Murphy argues, ‘tariff protection would grow into a fundamental institution of public policy, interwoven with industrial arbitration because high wages were only possible behind the barrier that shielded Australian workers from the world market.’ ⁶⁴

By the end of the first decade of the twentieth century, the new protectionist growth model had been established. The key role played by the state during the model's initiation resulted in the structural demotion of the previously dominant, rurally based agricultural and mining industries in favour of an embryonic industrial sector. Initially, the economic costs of the new model were borne by exporters and by rural industries who were not under the protectionist blanket, who were exposed to competitive foreign markets, and who suffered domestic cost increases because of protectionism. To compensate for this, and in response to the concerns of farmers in particular, the federal and state governments ultimately provided a range of statist supports, including rural assistance in the form of access to cheap loans, the establishment of state marketing board for rural products, dam projects and subsidised water for irrigators, and transport assistance. The resulting system, one which became known as ‘protection all round,’ became something of a national religion, one which generated positive returns for politicians and key relevant producer groups. Australia's first major tariff inquiry in 1929, the Brigden Report, only saw upward momentum in the statist protections and warned that it saw ‘no influence’ within the Australian tariff system ‘to counteract the indiscriminate and indefinite extension of the tariff.’ ⁶⁵ As the economist J. B. Brigden, who chaired the inquiry, noted, this new growth model saw ‘the protection of manufacturers and of labour march in one indissoluble unity . . . the two lions of employer and employed laying down at the same feast.’ ⁶⁶ In contrast to the European small economy model of ‘domestic compensation’ and adjustment to international market forces outlined by Katzenstein, Australia's model was one of ‘domestic defence’ especially for urban industry and labour. ⁶⁷ As Castles argues,

Almost every section of early twentieth century Australian society was protected in some way: the exporters and pastoralists by the very profitability of the export trade in staple commodities [as well as state subsidies]; the manufacturers by tariffs; and the working class by a minimum wage and a mechanism of controlling labour supply. ⁶⁸

Based on these foundations, the Australian Settlement achieved widespread political support and was widely seen as an important nation-building exercise. By early in the first decade of the twentieth century the majority of politicians in the new federal parliament were protectionist. ⁶⁹

Maintaining Growth Models

Once a growth model has been established, we can expect, as Baccaro and Pontusson suggest, a growth coalition, composed of key business and perhaps labour interests, to emerge and argue for the continuation of favourable state policies in support of the growth model. At this point, and as long as the state retains a commitment to the continued survival of that growth model, business and producer interests are likely to be regarded by state actors as being structurally prominent and will be in a strong position to secure their goals. As we have argued above, the resulting interactions may then reflect the power of business and producer groups, or alternatively and probably more typically, degrees of mutualism in which business interests seek to exercise influence rather than power in relation to the state.

In the United States, following the 2008 financial crisis, the largest banks, which had previously been able to exercise considerable influence in persuading policymakers to pursue deregulation of their industry suddenly found themselves in a new, precarious, political position in which they were being blamed for having caused the crisis through the reckless pursuit of short-term profit. The executive branch and elected politicians in Congress, under intense pressure to be seen to be taking punitive action against the banks, proposed or supported legislation to reform banking and to cap the maximum size of any single bank as one way of ameliorating the ‘too big to fail’ problem which had necessitated highly unpopular bank bailouts. Young et al. argue that the Obama administration was nevertheless forced to dilute its reform agenda when the largest banks which had, at this point, lost virtually all of their earlier credibility and influence, instead threatened, in a classic expression of structural power, to reduce the amount of funding available for lending if the reforms were rolled out in full. At a time of credit scarcity this was a potent weapon. ⁷⁰

Yet even in those cases where a business sector is regarded by state actors as being a key part of an existing growth model and, for this reason, structurally prominent, it may not always be able to achieve its goals. In the run-up to the 2016 Brexit referendum, the finance industry (with some exceptions) emerged as a strong public advocate for the Remain campaign. After a narrow majority of the public had voted to leave, the banking industry then switched its focus to a demand that the government negotiate a ‘soft’ Brexit, one in which the banks retained their ‘passporting’ rights allowing them to operate seamlessly in the European Single Market. In doing so, the chief executive of the British Bankers Association warned ministers that his members were ‘quivering over the relocate button.’ ⁷¹ The sector's efforts to exert power over the government were, however, entirely unsuccessful. True, the banking industry was still regarded as a key economic actor, one whose revenues accounted for over 7 percent of GDP, contributed £60 billion in tax revenue and employed over two million people. ⁷² Yet, after the referendum, in a new and highly charged political environment in which the Conservative Party was fragmenting and the new prime minister, Theresa May, was convinced that the British state's continuing legitimacy and her party's electoral future required it to negotiate a ‘hard’ Brexit, the interests of the City of London became, at most, of secondary importance, as politics, for a time, eclipsed economics. ⁷³ This shift in priorities was perhaps most succinctly expressed by Boris Johnson at a diplomatic reception in London in June 2018 at a point when he was still foreign secretary but plotting to displace May. Asked by the Belgian ambassador, Rudolf Huygelen, about business opposition to Brexit, he reportedly replied, ‘Fuck business.’ As James and Quaglia argue, ‘Brexit illustrates the highly contingent nature of business power, which can change suddenly and unexpectedly in response to electoral developments.’ ⁷⁴

In contrast, the politics of the Australian Settlement during its mature stage, which lasted from around 1910 to the late 1960s, was marked not by conflict between producer groups and the state but by high degrees of mutualism and the exercise of producer group influence rather than power. State leaders—led by dominant political figures of the postwar era such as Sir John McEwen, the minister for trade and deputy prime minister—as well as industrial interests, all benefited from what the Bridgen Report had earlier warned of as the ‘indiscriminate and indefinite extension of the tariff.’ ⁷⁵ McEwen was the dominant figure in postwar economic policy and was a major supporter manufacturing industry. As leader of the Country Party in the face of a declining rural electorate, he was strongly encouraged to secure a new political support base from producer interests in the manufacturing sector. ⁷⁶ Australia ended up with one of the most protected industrial sectors in the world, propelling the manufacturing sector from a small base at the turn of the century to almost a third of GDP by the late 1960s.

According to Robert Menzies, Australia's prime minister between 1949 and 1966, this was a world in which the boundaries between government and industry had ‘disappeared under the pressure of modern complexities,’ with ‘politics and industry deeply involved with each other, acting and reacting on each other.’ ⁷⁷ The resulting political economy was one in which business groups and trade unions representing manufacturing industry were ‘systematically lucky’ insofar as they could achieve their goals without trying, largely because the state's interests were aligned with theirs and producers could exercise influence without having to make threats or act coercively. The general view in politics and society during this era celebrated the growth of manufacturing, with the economist F. G. Davidson writing in the late 1960s that the ‘story of Australian manufacturing was an encouraging even an inspiring one.’ ⁷⁸ The regime of ‘protection all round’ had generated a wide cross-sectoral coalition that supported the long-running Liberal-Country party government of the postwar era. As the administrative body responsible for the tariff regime, the Tariff Board, admitted in 1969, ‘the tariff making procedures are thus . . . in response to requests from interested parties seeking greater protection.’ ⁷⁹

Given the shock of the 1890s, followed by crises of first half of the twentieth century, with two world wars interspersed by the Great Depression, McLean argues that Australia's protectionism was defensible and not necessarily growth retarding during this period. ⁸⁰ However, the main Achilles’ heel was that the generous protectionism was unaccompanied by any significant attempt to impose conditions such as performance or efficiency criteria on recipients. This reflected Australia's unusual combination of statism at the broad structural level combined with a kind of laissez-faire approach at the microeconomic level. Indeed, as the Brigden Report had warned in 1929, the ‘most disquieting feature of the tariff has been the stimulus it has given to the demands for government assistance of all kinds with the consequent demoralising effect upon self-reliant efficiency throughout all forms of production.’ ⁸¹ Australian protectionism thus came without strings, and without microeconomic oversight. ⁸² The manufacturing industry in particular ‘was left largely to its own devices.’ ⁸³ During the postwar economic boom and a period of rising prosperity, the costs of such inefficiency were conveniently neglected. But by the 1970s, the system had helped produce a privileged, but oversized and uncompetitive, manufacturing sector that was ill-suited to looming structural challenges.

Terminating Growth Models

GMA is predisposed toward accounting for the enduring nature of growth models in terms of the existence of entrenched growth coalitions. Baccaro and Pontusson acknowledge, in their original article, that ‘all growth models are fundamentally precarious,’ yet they have also emphasised, in their more recent work, that, when growth models come under pressure, they can be stabilised through exercises in ‘coalitional engineering’ in which policies are adjusted and new political actors incorporated in an effort to secure their political legitimacy and longevity. ⁸⁴

Yet we also know that growth models do disappear. It was, after all, the failure of the Fordist growth model which provided the intellectual starting-point for Baccaro and Pontusson's initial article. More recently, as Bondy and Maggor have documented, Israel, Ireland, and Brazil have transitioned from ‘export-led growth to a more balanced growth model, simultaneously driven by both net exports and wage-based private consumption.’ ⁸⁵ Growth models which once seemed eternal can be undermined by changes in context, including technological changes, shifts in the international economic environment, ideological changes, the emergence of new rival industries, and changes in consumption patterns. Moreover, and whatever the background context, a growth model can be terminated if key state actors conclude that the existing growth model it had once supported is no longer capable of delivering future economic growth and so no longer regards the industries within it as being structurally prominent.

The coal industry was an important part of the postwar British growth model which had employed over 700,000 people in the late 1950s. In the 1960s, however, it became increasingly dependent upon extensive state subsidies and market privileges, including a de facto monopoly on sales to electricity generators. Successive Labour governments were nevertheless willing to support the industry and to countenance a further £600 million investment in it in the early 1970s to significantly increase production, in part because the coal industry, based largely in Scotland, Wales, and the North of England, was seen as being key to future regional economic development and to labour's electoral prospects. ⁸⁶ This changed when Margaret Thatcher's Conservative Party was elected in 1979, underlining the fact that an important shift in the growth model stemmed from a political and electoral change. The new government expressed its determination to reduce public debt, operate the nationalised industries on market-confirming principles, and eviscerate the politically powerful National Union of Mineworkers. The eventual result was a yearlong strike which the government won, based in part on authoritarian tactics and the prior secret stockpiling of substantial coal supplies by the government. This resulted in the closure of one-third of the remaining pits in 1985–86, a switch to the large-scale importation of coal from Eastern Europe and South Africa and, eventually, the privatisation of the industry. ⁸⁷

The fate of the Australian Settlement is somewhat similar in that it reflects ideological change by political leaders and underlying structural shifts in the economy. Starting in the late 1960s, the position of the manufacturing sector initially came under intellectual attack from the Tariff Board and its chairman, Alf Rattigan. Rattigan had been appointed by the trade minister, John McEwen, in the expectation that he would continue to support ‘protection all round.’ Instead, Rattigan used novel economic research on the economy-wide effects of tariff protection pioneered by economists such as Max Cordon and in effect became one of Australia's first major neoliberal policy reformers. Rattigan was quickly supported by other economists and by the wider neoliberal intellectual shift that was occurring by the 1970s, one increasingly supported by the press. Through strident advocacy, the Tariff Board was able to show that Australia's high and unconditional levels of protectionism had significantly increased costs for the rural, mining, and consumer sectors and was acting as a drag on economic growth. ⁸⁸ Moreover, in the late 1960s and the 1970s the Australian economy was undergoing profound structural change. A key aspect of this was the rapid growth of the mining sector which helped to erode support for further import-substitution manufacturing development as well as the rapid development of low-cost manufacturing in East Asia. This put the competitiveness of local manufacturing in the spotlight.

The Tariff Board's conclusions were ferociously attacked by John McEwen. Yet the Tariff Board was clearly winning the intellectual and policy debate, and because it was a statutory authority it had a measure of political protection. By 1971 McEwen had resigned. The initial attack on the growth model that had favoured manufacturing therefore came from within the state from a public official. Rattigan's economic reasoning became the basis for government policy when the Australian Labor Party, which had been out of power for twenty-three years, was elected in 1972 and soon implemented an across-the-board tariff cut of 25 percent. The new prime minister, Gough Whitlam, shocked an audience of manufacturing interests in a speech in 1974 in which he welcomed them to a more competitive environment and the ‘first free enterprise government in twenty-three years.’ ⁸⁹ Manufacturers reacted angrily. The head of the Associated Chamber of Manufactures of Australia stated that ‘manufacturers are getting sick and tired of being pushed around . . .  it is not a threat but merely a statement of fact to suggest that the present Government would be out of power in quick time if it monkeyed about too much with tariff levels.’ ⁹⁰ Yet Whitlam's change in policy direction came at a point when, even with the levels of protection they enjoyed, Australian manufacturers were still regularly being undercut by the East Asian export industries and public support for protectionism was waning. All this placed the interests that had benefited from protectionism in an increasingly weak position. Manufacturers made repeated threats to reduce investment and employment, but policymakers reacted by welcoming manufacturing disinvestment. As Rattigan argued, since ‘Australia was short of labour and capital and its economic growth was adversely affected by rising costs and inflationary pressures, the deferment of large investments in manufacturing activities which required high protection would be a benefit, not a disadvantage.’ ⁹¹

This battle continued into the 1980s, led by the Hawke-Keating Labor governments initially elected in 1983. The government embraced globalisation and the retrenchment of the protectionist Australian Settlement. By the late 1980s and early 1990s the Settlement, one which had featured ‘no-strings-attached’ protectionism and a highly centralised wages system, had largely been dismantled. The final dramatic blows came in the early 1990s when the Labor government pushed ahead with the final major cuts in protection amidst the deep early 1990s recession.

Australia's replacement growth model from the 1980s was underpinned by a stronger embrace of economic globalization and a booming commodities sector based on rural and mining exports. This was supported by a range of government subsidies and a light-touch resources rent taxation regime. A second key source of growth was associated with the liberalization of finance in the 1980s which by the 1990s was spurring surging mortgage and property markets accompanied by steeply rising asset prices. As in the United Kingdom, this model was not initially planned, but by the 1990s the conservative Liberal-National coalition government saw an electoral advantage by giving property owners favourable capital gains taxation and supportive negative gearing arrangements for investors. As Frank, Arthur, and Friel write, this model ‘is maintained by an influential coalition of interests from the financial and insurance services sector, the construction industry, increasingly leveraged voters who own housing assets, and centre-left and centre-right parties.’ ⁹²

The Australian Settlement was the defining growth model for the Australian economy for the first six decades of the twentieth century. It featured a state-protected system of trade relations, sectoral privileges, especially for manufacturing, and labour protections, all of which were widely supported politically. In the period after World War II this was supported at the macroeconomic level by wage-driven growth especially in the protected sector. Our analysis of this model's initiation and termination supports our general argument that growth model politics often features the state as a weighty actor. Yet the state's role was also apparent during the long maturation stage when government and state leaders acted in an authoritative position in dealing cooperatively with key beneficiary sectors in industry and labour.

Conclusion

This article has made two contributions. First, it has sought to rebalance the debate about the politics of growth models, arguing that the existing literature places too much weight on business and producer group interests in shaping the dynamics of growth models and too little weight on the role of governments and the state. Second, the article introduced the concept of a growth model life cycle involving stages of initiation, maturity, and termination. The article showed that different patterns of politics and of producer group–state relations are apparent in each of these different stages. To explore such variable relations, the article utilised two further concepts: the structural prominence of a sector and the distinction between business or producer group power and influence. The article argued that the structural prominence of producer interests and their associated influence or power are constituted not just by the economic but also by the political, wherein the ideas and the power and authority of state actors often loom large.

Armed with such concepts, the article argued that growth models follow a life cycle and that the relative power and capacities of states and producer groups depends on the stage of the cycle. The influence of business and producer groups is more prominent in a mature growth model and less prominent in the period when the model is being formulated and when it is in decline. Hence, during growth model initiation, the state is likely to be a key player in looking forward, assessing possibilities, and steering the development of a new growth model. During the maturation stage, the relationship between business and producer interests and the state becomes more variable, largely because producer groups are likely to have a higher level of structural prominence implying greater state reliance on such interests and greater influence or power for producer groups. If conflict situations arise at this stage, this is likely to see attempts by producer interests to convert structural prominence into instrumental and/or structural power. However, as argued above, a more likely outcome in a mature growth model is a reasonably high degree of state-business mutualism and codependence, implying a stronger role for business influence rather than conflictual attempts to exert power. Finally, the termination stage is likely to be associated with a decline in the structural prominence of once favoured sectors and the outbreak of conflict and attempts by producer groups to exert influence or instrumental and/or structural power in order to retain their former privileged position. Yet at this point, and precisely because their structural prominence has declined, the state will be well positioned to resist such efforts.