The history of industry policy in Australia

For much of its history, Australian industry policy has been characterised by ‘ideas in good currency’ in government. From the time of Federation until the early 1970s, the ruling idea was that Australia needed manufacturing for employment and for the overall development of the economy. The mechanism of choice was tariff protection. This emphasis reflected both the importance of trade and the prominence of the state in the evolution of the economy (Jones Reference Jones2016).

However, protectionism was not well-favoured among many economists, and had the potential (by increasing the costs of inputs) to detract from the economy’s main export earners, agriculture, minerals, and energy production. From 1921, these factors were acknowledged through the operations of the Tariff Board, an independent body which advised on appropriate levels of assistance in response to industry or government requests.

The Whitlam government (1972–75) overturned this arrangement. Its historic 1973 25% reduction in tariffs, undertaken without consultation at a time when the dollar was appreciating, signalled the beginning of the end for many types of manufacturing. In 1974, the Tariff Board was replaced by a new body, the Industries Assistance Commission (IAC). In the 50 years since then, any sense that governments might productively alter the structure of the Australian economy (other than by winding back assistance) – has generally not been well-supported in Australia.

Increasingly from the 1970s, Australian policy elites, from both sides of politics, and including the powerful Treasury bureaucracy, endorsed the theory of comparative advantage, which meant open markets and minimal government involvement. During this time (and still), theory permitted only one rationale for assistance, and this was to counteract market-based externalities. In practice, this meant support for incentives designed to promote research and development (R&D) in business and little else.

Nevertheless, a change to the structure of assistance came slowly. Interestingly, it was the Labor governments of Bob Hawke (1983–1991) and Paul Keating (1991–1996), which did most to dismantle the tariff regime, with progressive reductions in protection from 1983 onwards. With tariffs effectively removed as a policy tool, and a disinclination to commit to direct support, the theme from the Hawke Labor government to the Howard years (1996–2010) and beyond, has been one of managed decline of Australian manufacturing. In the 1980s and 1990s, structural adjustment schemes lessened some of the pain for heavy engineering, steel, motor vehicles, textiles, clothing, and footwear (Capling and Galligan Reference Capling and Galligan1994).

The government of Liberal Prime Minister Tony Abbott (2013–2016) delivered the coup de grace to what was left of the vehicle manufacturing industry, as well as eliminating support to emerging initiatives to combat climate change. Abbott also commenced what was to become an embarrassing period of submarine-mania, himself making a Japanese choice for the Australian navy, later changed by Prime Minister Turnbull to favour the French, then changed again by Scott Morrison in favour of the US and UK in a deal usually known by its acronym, AUKUS. AUKUS highlighted the politicisation and lack of an independent perspective in Australian defence thinking. Less noticed was Abbott’s defunding of a number of promising science initiatives, including in 2016, the National Institute of Information and Communications Technology, a cross-university and the Commonwealth Scientific and Industrial Research Organisation (CSIRO) entity which at its peak employed 700 researchers and promised to be the foundation of an enduring Australian presence in the emerging field of artificial intelligence.

Climate change remained on the agenda, but in keeping with the general political and media indifference to industry issues, none of these events caused much comment at the time, although this was later to change as the scale of AUKUS started to become apparent. So, it came as something of a surprise when, in late 2024, the Albanese government, through its FMIA programme, made a bid for supporting new industries. The policy, legislated later that year, provided some serious funding ($7 billion over 10 years) for uniting two important policy ideas – the science, technology, and manufacturing needed for a new era, including notably net-zero emissions, while at the same time making an explicit pitch for Australian involvement in these activities.

As matters turned out, FMIA, although it certainly did not harm the government’s chances, was not a prominent theme in the 2025 federal election campaign. Voters were concerned about ‘bread and butter’ economic issues, the cost of living, health care, and childcare. The Opposition’s attempt at its own version of FMIA – the founding of a domestic nuclear power industry – was almost universally decried.

Whether FMIA will yield results remains to be seen. Opinion at the time of its release was either neutral or critical. I argue that some scepticism is in order, not because the ideas themselves are inadequate, but because the implementation networks and expertise required to make them work will take time to establish. Apart from defence and health-related research, there has been so little positive government involvement in science, technology, and industry that it is difficult for any government to find a credible way forward. It is just not possible to switch industry policy on, then off, and then on again.

The international policy agenda

Industry or industrial policy refers to government efforts to alter economic structure in productive ways (Millot and Rawdanowicz Reference Millot and Rawdanowicz2024). Empirically, there is evidence that, provided national circumstances are correctly understood, governments can accelerate economic development through targeted assistance to industry. Trade-related policies have been central in this respect, tailored to orientate demand towards domestic production, while also exploiting export opportunities (Higgott and Robison Reference Robison and Higgott1985; Zysman Reference Zysman1983). The US and Germany in the nineteenth century, and in the post-World War II period, Japan, Hong Kong, Taiwan, South Korea, and, more recently, developing countries such as China and Vietnam, have all employed these mechanisms.

But trade-related policies formed only one part of the picture. In 1990, Michael E Porter described the importance of highly competitive nations of firm clusters and complementary government policies, particularly in relation to the support of innovation (Porter Reference Porter1990). Clearly, governments did not always get it right. But equally, if governments did nothing, they risked getting left behind by those that were even only partially successful.

Little of this work proved persuasive in Australia, where, as Michael Pusey noted in his pathbreaking work of 1991, economic rationalism held sway over many areas of public policy, including industry in the fullest sense of the term. The key public advisory body, the IAC, transformed into the Industry Commission (1989), later the Productivity Commission, continued to see policy options through the lens of neoclassical economics, and took little interest in a growing body of empirical work, particularly in relation to innovation, that suggested a more complex story (Dalitz Reference Dalitz2016).

There were dissenting voices, but they had to come from outside mainstream economics. From a policy studies perspective (Dodgson Reference Dodgson1989; Stewart Reference Stewart1994), put forward arguments for innovation support and for various forms of post-tariff industry policy. From a political science perspective, Tsokhas (Reference Tsokhas1994) and Bell (Reference Bell1997) outlined the uniqueness of the Australian state in its international context. Policy analysts Elizabeth van Acker and Tom Conley (Reference Conley and van Acker2011) argued for an industry policy that followed a ‘middle course between free trade and interventionism’. Stewart (Reference Stewart2016) also showed how the effects of path dependence reduced opportunities for useful action.

Economists working in the political economy tradition had long argued for the importance of understanding the particular qualities and dependencies of Australian capitalism, which tended to fragment the manufacturing sector (Crough and Wheelwright Reference Crough and Wheelwright1982). Later work, drawing, in particular, on the work of Mazzucato (Reference Mazzucato2021) and on innovation studies, argued for the revitalisation of Australian manufacturing (Dean and Jackson Reference Dean and Jackson2024; Dean et al Reference Dean, Rainnie, Stanford and Nahum2021). As well, Australia’s Asian trading partners had shown what could be done to support competitive manufacturing. The work of Australian-based scholars such as Linda Weiss, John Mathews, and Elizabeth Thurbon highlighted the continued importance of the developmental state in East Asian economies (see, for example, Weiss (Reference Weiss1998), Mathews et al (Reference Mathews, Thurbon, Kim and Tan2023), and Thurbon (Reference Thurbon2016).

While Australia continued its conventional policy approaches, there was a significant change in policy-related thinking, particularly in the US. From around 2010, there was a major surge in industry-policy writing (Millot and Rawdanovicz Reference Millot and Rawdanowicz2024), so much so that the term ‘the new industrial policy’ has been used to describe it. The newer policies were more likely to feature explicit coordination measures between government and business, rather than ‘picking winners’ or identifying national champions (MacNeil Reference MacNeil2016). These approaches were underpinned by work on the dynamics of learning required to enter and to prosper in new fields (Stiglitz and Lin Reference Stiglitz and Lin2013).

At the same time, trends in the capitalism of the 2000s onwards shifted opinion away from unquestioning support for globalisation. China’s opening to the world from the days of Deng Xiaoping had aroused keen Western interest in penetrating the new markets, even while China exacted a skilful price for the market access it gave the eager Westerners. After an initial period of euphoria in the West, the success of China, by far the largest economy in world history to industrialise in such a short period, began to give pause to the conventional notion that the more trade, the better.

For the US, in particular, the political consequences of de-industrialisation formed a large part of the impetus behind Donald Trump’s electoral success. But disillusionment with the prospects of open trade with China pre-dated President Trump. Worried by China’s strategic ascendancy, Barack Obama had earlier introduced restrictions on exports of key technologies, steps that were reinforced by Joe Biden in the early 2020s.

Concern had also grown that the US was itself falling behind in advanced manufacturing globally. In October 2014, the Obama administration announced incentives for investment, networking, training, and science. The Biden administration went further with the Inflation Reduction Act (2020), which introduced a wide range of measures designed to strengthen American science and technology (Dean and Jackson Reference Dean and Jackson2024).

Over the same period, there was a resurgence of academic interest. Factors that were not, strictly speaking, trade-related were to have a profound impact. The COVID-19 pandemic, which began in early 2020, showed many countries the extent of their vulnerability to disruption of supply chains. Russia’s invasion of Ukraine in 2022 had similar effects. The opportunities and challenges posed by climate change brought forth a surge of interest in green industrial policies. These policies took their lead from the need to proactively transform economies in the interests of ultimately achieving net-zero carbon emissions.

Rationales for industry policy

From a mainstream economics perspective, the rationale for industry policy is to act when and where markets are likely to fail. Potentially, this gives a rather large remit, as markets fail almost all the time. Spillovers (unpriced externalities) provide the overarching rationale for intervention. Traditionally, attention was given to the so-called ‘infant industry’ argument, which relied on learning by doing and economies of scale. More recently, analysts have tended to focus on positive externalities (spillovers) associated with research, development, and (to some extent) commercialisation (Millot and Rawdanovicz Reference Millot and Rawdanowicz2024; Productivity Commission 2021). Ideally, ex ante cost-benefit analysis supports the identification of priority sectors for assistance, but because future market conditions are unknown, this technique is difficult to apply and, in practice, subject to considerable political distortion (Flyvberg and Bester Reference Flyvberg and Bester2021).

Evaluations of industrial policies are also difficult to undertake, as complex cross-cutting and spillover effects are almost always present. Firm-level effects (for example, as a result of regional subsidies) are easier to identify than the costs and benefits of types of policy. The OECD has reported positive impacts of European Union (EU) structural funds on output and employment in specific instances, but failed to find demonstrable overall economic benefits from the broader range of industry-related incentives (Millot and Rawdanovicz Reference Millot and Rawdanowicz2024, Annex D).

Another way of resolving the problem of choosing effective and apt policies is to identify broadly the kinds of objectives that industry policies might serve, thus allowing for a broader range of social and political factors to be considered. Millot and Rawdanovicz (Reference Millot and Rawdanowicz2024, 42) identify such objectives as pursued by OECD members: green industrial policies (which rely essentially on market failure arguments); the creation (or retention) of regional employment (social equity); and national security, arguably a form of risk management where nations may be cut off from supply in the event of war, or disruption of trade due to political action, or to a pandemic.

In practice, industry policy continues to present a dilemma for policymakers, particularly in countries with less-developed, more specialised or smaller economies. How to respond in these situations is much more difficult than where, for example, there is a large domestic market, an abundant supply of relatively cheap, skilled labour, and equally importantly, market access for exports.

The prospects for Future Made in Australia (FMIA)

As Treasury noted in its supporting paper of FMIA (Australian Government The Treasury 2024), identification of industries to receive support would be determined by a ‘national interest framework’, based on five key factors: ‘Australia’s grounds for lasting competitiveness, the role the industry will play in securing an orderly path to net zero, the role the industry will play in building Australia’s economic resilience and security, whether the industry will build key capabilities, and whether the barriers to private investment can be resolved through public investment in a way that delivers compelling public value.’

It is hard to argue with these objectives. Worthwhile firms and activities will no doubt benefit. However, without an appropriate administrative strategy, there are reasons for scepticism as to the prospects of success. How, in practice, are firms to be chosen and activities monitored? Recent Australian commentary has focused on the importance of goal-setting and priority-setting as the foundation of the new industrial policy (Wood Reference Wood, Reeve and Suckling2022; Phillips and Koh Reference Phillips and Koh2024).

However, there is a danger that without a firmer grounding in Australian experience and a more realistic perspective, the implementation of FMIA will founder. In particular, if it leads to governments ignoring signals that are already there, in favour of ‘blue sky’ initiatives, it may make matters worse. A report in June 2025 that the scrapping and recycling of a redundant oil platform would be undertaken by an American firm based in Denmark, rather than either of two Australian contenders, suggests this is a real and continuing risk (Quail Reference Quail2025, 2). In this case, price was considered the decisive factor, although there have been cases in high technology fields where more expensive foreign tenders have been preferred to those put forward by local suppliers considered to be higher-risk. FMIA’s emphasis on technology raises an underlying problem that needs to be discussed if policy is to evolve.

Australia’s cultural cringe, at least as it affects the arts, is no longer as strong as it once was. Australian science, especially in the health field, is acknowledged as world-class. Technology, however, is a different matter. Indeed, the technological cringe, the belief that Australian firms are not really up to the challenges involved, may be worse than ever. If these attitudes are to be addressed, policy communities must walk a fine line. They need to be realistic about Australian capacities, so as to avoid over-reach. But if the stance is too negative, opportunities are missed, and confidence fails to develop.

Getting the balance right between trying to do too much and not doing enough is not easy, particularly where, as in many fields in which Australia might wish to succeed, the competition from other advanced economies is intense. Policy-makers are, understandably, risk-averse. Most commentators continue to favour tax and regulatory reforms that will reduce costs and enhance competition, rather than the more direct approach implied by industry policy.

Identifying a middle way appears to be the way forward. As Zhang et al (Reference Zhang, Green and Dodgson2024) put it, in relation to science and innovation.

Without a domestic industry, though, there is a risk that good ideas will simply go offshore. Nurturing firms that have the potential to commercialise research is vital. The problem is that, while it is conceptually easy to identify activities that might fit the bill, governments are understandably wary of ‘picking winners’ at the firm level.

One way, insufficiently explored in both the Australian and international literature, is to use activities governments are already involved in for social, environmental, security, or even political reasons, as information signals to mitigate risk and to increase responsiveness (Stewart Reference Stewart1994; Reference Stewart2016). In the past, Australian governments have intervened directly in situations of natural monopoly or where sufficient private sector investment was not likely to occur. Two of Australia’s most successful firms (Telstra and CSL) started life as publicly-owned entities. NBN Co has continued this tradition (Australian National Audit Office 2024; Moyal Reference Moyal1984). Where the public sector presence remains substantial, a good deal of market-congruent and demand-oriented intervention is possible. Defence procurement, medical expenditures, and even the provision of public transport all offer opportunities for the domestic industry to be involved in ways that, over time, support the growth of firms that offer a wide range of technology-intensive jobs and have the capacity to take part in research partnerships.

The politics of Australian industry policy

The problem with not having an explicit industry policy is that governments end up with an implicit one, so that other factors, some intended, others not, shape economic activity. In such a political environment, if there is no space for discussing the type of economy people want, a version of it will emerge, determined by whatever interests, bureaucratic pressures, and ideologies happen to be strongest at the time. At times, policies themselves generate competing priorities that evade easy compromise, as Australians have seen with the effects of rising power prices on energy-intensive domestic manufacturing, such as the production of aluminium.

Budgetary decision-making can also be significant. In Australia’s case, a de facto industry policy has emerged from the chronic underfunding of universities. This has led to universities recruiting more and more international students in order to gain revenue and has produced, indirectly, a booming export industry. But there have been significant costs. Many universities, it has been reported, have neglected investment in the next generation of academics, leading to significant casualisation of employment and an emphasis on student numbers at the expense of quality education for all students (The Senate 2025, chs 4–5). Overall, a type of hybrid industry has been produced, publicly-owned, but lacking any clear commitment to the public good (Turner Reference Turner2025).

At the same time, manufacturing has reached an all-time low, comprising only 5% of GDP in 2023 (Ai Group 2024). We know that a relative decline in manufacturing has occurred in all advanced economies. However, in Australia’s case, the decline has been precipitous, from roughly 25% of GDP at the end of the 1960s, to 14% by the mid-1990s (Commonwealth Treasury 2001), then reducing still further from the turn of the century. Australian energy resources, notably gas, which had previously been seen as an asset to be exported and also used domestically, have, in the age of renewables, come to be seen as liabilities. In the absence of a carbon tax, climate change policies have been pursued without regard to cost. Even the production of critical minerals and rare earths faces major difficulties (Australian Government Department of Industry, Science and Resources 2023). All this has been happening while the US gathers as much strategic industrial power as possible to itself and China pursues its hard-won advantages relentlessly.

Structural change is insidious. During the COVID-19 pandemic, shortfalls of goods, such as hospital saline, caused many citizens to think – surely we could make that here? But such concerns are usually fleeting. Industries that are otherwise well-established and efficient reduce production in Australia because it is easier to switch to importing, or to go offshore completely. In turn, imports become progressively more expensive, as supply chains become more dependent upon just a few producers. In theory, price signals should lead to a return to domestic production, but few are willing to take the risk.

Where activities are no longer undertaken locally, there are obvious implications, not just for employment, but for the types of jobs that are on offer. The manufacturing decline has affected a wide range of skills. In an economy that lacks firms in the relevant fields, there is structurally less need for business-based research and development (Davis and Tunny Reference Davis and Tunnyc2005). A comprehensive report from Science and Technology Australia (2023) highlighted problems in maintaining a viable STEM workforce in an environment where it is difficult to support continuing careers. The STA report identified a ‘leaky pipeline’ in engineering, leading to a progressive loss of skilled people over the course of their careers (Science and Technology Australia 2023, 42). Future inaction would mean more of the same: jobs for trade-skilled workers, baristas, and healthcare workers, but not so many for engineers, scientists, and production workers, with employers finding it difficult to recruit local workers for the higher-skilled jobs that are on offer.

There is an even more fundamental issue, which goes to the question of what is valued in the economy. Industry policy asks questions about how nations earn their living, but also about their place in a changing world. What does this mean? As some are now arguing, it means, in Australia’s case, re-engaging with questions of national independence that have long been subsumed by the American alliance. President Trump may be a passing phenomenon, but the politics he represents are not likely to disappear.

Australia’s response to the great power rivalry of China and the US requires subtlety and even-handedness in the definition of the national interest. An industry policy that builds on national strengths (such as critical minerals) gives more leverage in this context, but the key question remains: how is the leverage to be used? The mid-2020s have brought a sharp reminder that a continuing congruence of policy values, let alone interests, with the United States, cannot be assumed. Doing something about this does not mean Australia has to go it alone, far from it. But it has become increasingly evident that societies need room to make choices, and where needed, to draw economic and social boundary lines. In turn, this means considering values – of creativity, inclusion, and social equity – and developing activities that will strengthen ambition and performance in intelligent ways.

What should Australia (and other small to medium-sized nations) do?

I want to end on a prescriptive note, largely because so many arguments about industry policy reflect a failure to think realistically and creatively about the policy agenda and its implications, and the challenges of implementation. Implementation, often neglected in policy studies, is not dictated by policy design, but rather requires its own impetus and vision.

The importance of events in creating a positive environment for industry policy has been noted by many commentators. But events are changeable and changing. If industry policy is to be sustained on the institutional agenda, it requires bipartisan political support. It would help, in this context, if there were to be some acknowledgment from both major parties that there is a problem with Australia’s policies towards industry,

As the voter constituency for a more innovative and wide-ranging approach is not strongly evident, it is important for those in policy circles, including within the bureaucracy, to continue to argue for change. Even if nothing is done, the consequences of past neglect should be acknowledged. Relying on market forces to do the work is attractive for some policy-makers, but it means that options that require more detailed attention begin steadily to contract. Good people continue to go overseas because the jobs they want to do are not here. Industries that add value to Australian resources continue to go offshore.

The role of information in policy is critical. Successful industry policy builds on information, on developing systemic complementarities between those who make and do, and those with access to knowledge, incentives, and buying power. If we think of industry policy in this connected way, the task of policy and of policymakers is to listen and evaluate, and then to implement linkages.

The starting point has potential. Years of neglect have resulted in Australian manufacturing and related industries that are resilient. To survive, as they have done for many years, in a policy world that largely ignores them, is a major achievement. To dismiss their views as those of rent-seekers ignores real-world experience.

In turn, forms of engagement with industry are shaped by the confidence and capacity that skilled and experienced public servants bring to policy discussion within government and outside it. Unfortunately, the subservience of the ‘doers’ to the policy mandarins of Treasury, Prime Minister, and Cabinet has produced a bias in policy thinking that undervalues implementation, evaluation, and a sense of emerging strategy. Good programmes are lost in a welter of small-scale initiatives (Toner et al Reference Toner, Agarwal, Li, Bajada, Paul, Phan, Pugalia and Green2023). Rectifying this fragmentation requires re-orienting the roles of public servants towards adaptation, learning, and the creation of public value, rather than the service of short-term politics.

The next areas to consider are the vast amounts of research that governments fund, and the organisations that both do much of this work and train those who will contribute to it in the future. Governments and universities, even governments and schools, find little common ground in Australia. Educational institutions complain about a lack of money, but the underlying problems run deeper. Neither side is listening properly to the other. As any academic can attest, the last 20 years have produced a tertiary sector that has lost its way (Hil Reference Hil2012). International students have tended to become cash cows when they should, together with domestic students, form part of a teaching and research system embracing both Australian and international activity.

While the universities have over-reached, other levels of the education system, in Australia’s case, controlled by state governments, have struggled to maintain the performance of former years. For example, Australian education needs a vocational sector that restores its connection with Australian business and skills. It needs high school systems that, while continuing to support disadvantaged learners, do not fail their best students. As the 2022 PISA data show, the performance levels of the top group in reading and in mathematical and scientific literacy are significantly below those of twenty years ago (De Bortoli et al Reference De Bortoli, Underwood and Thomson2023).

Much publicly funded research is both fragmented and wasteful (Universities Australia 2025). Long-term competitors in many rounds of research funded by the Australian Research Council can testify to the cumulative effects of missing out, and at least in the social sciences, a tendency for reviewers to choose the latest ‘hot topic’ without any sense of a discipline-based agenda. In a small, chronically underfunded system, there is too much counterproductive competition when people should be working together. Collaboration, both nationally and internationally, is key, so funders should be supporting those who are doing the most important work, as long as they are taking others with them.

This is not an argument for governments to direct research or to set nominated national priorities. They need to know what is going on, but they also need to resist their tendency to politicise and to undermine what is working well. An entity like the national government-owned strategic research body, the CSIRO, is a case in point. After a long period in which CSIRO developed well-established links with industry, from the early 2000s, CSIRO was required to orientate itself towards ‘flagship’ programmes that stressed broad-brush national outcomes, such as climate change (Upstill et al Reference Upstill, Spurling, Healy and Simpson2023). I would argue that the assumption that Australian industry no longer needed CSIRO’s creative ideas overlooked the importance of public research and development in national innovation systems.

In the 2020s, events, particularly the actions of China, have focused attention on the interactions between industry and strategic power. China uses its industrial power to further its international ambitions. For Australia, the strategic dimension does not distort the underlying rationale of connection, but informs it in particular ways. In order to work, sovereign defence capability has to be part of a broader purchasing-oriented stance in industry policy. Such a stance recognises the need for Australian-based firms to work across a range of fields in order to remain viable.

It is difficult to discuss issues of national sovereignty without talking about values. There is a sense that values of freedom, self-determination, and ecological stewardship are more important than ever. In order to defend and promote these values through the policy realm, market signals are not enough. Some degree of self-reliance is required. But this need not mean that lists of ‘strategic’ industries are compiled, because this kind of priority-setting ignores the interconnections that need to be nourished, the firms we do not know about, the ideas that could matter most.

For countries with very large domestic markets, the power of demand, harnessed by tariffs, has proved paramount in building a basic industrial structure. But most nations are not large; they are small and medium-sized. If countries in this group are to develop as independent, middle powers, that perception has implications for all that is done. Path dependence shows that it is not possible to turn science, technology, and industry policy on and off. It takes time and patience to build networks, to change the culture, and to enlarge the perspective.