In the mid-1960s, manufacturing employed close to one in four Australian workers; by 2024 its share of employment had fallen below 7 per cent, and the sector contributed barely 6 per cent of gross value added (Australian Bureau of Statistics 2024). The closure of Holden’s Elizabeth plant in Adelaide’s north in October 2017, weeks after Toyota’s Altona line fell silent, ended more than nine decades of Australian passenger vehicle assembly and confirmed, for many observers, that globalisation had hollowed out the nation’s industrial base. Yet the meaning of that decline remains deeply contested. One reading treats it as the textbook operation of comparative advantage in an open economy, a necessary shedding of activities Australia was never suited to perform. Another treats it as an avoidable surrender of strategic capability that governments are now scrambling, through the Future Made in Australia agenda, to reverse.
This essay evaluates both propositions. It argues that the relative decline of Australian manufacturing was largely inevitable once the post-war protection model collapsed under the weight of its own costs, but that the mining boom pushed the contraction well beyond what comparative advantage alone required, destroying capabilities that market forces will not automatically restore. It follows that the current revival agenda is defensible in its narrow form, targeted at genuine emerging advantages and demonstrable security externalities, and economically unsound wherever it shades into a nostalgic attempt to rebuild manufacturing breadth for its own sake. The argument proceeds from the historical arc of reform, through the economic logic of decline, to an assessment of the reshoring turn.
From Protection to Openness: The Long Arc of Decline
Manufacturing’s rise and fall cannot be separated from tariff policy. From Federation, protection was a load-bearing pillar of what Kelly (1992) describes as the Australian Settlement: high tariff walls raised a manufacturing sector in the suburbs of Melbourne, Sydney and Adelaide that, by the 1960s, employed a quarter of the workforce while remaining subscale, inward-looking and dependent on continuing assistance. The unravelling began with the Whitlam government’s 25 per cent across-the-board tariff cut of 1973 and the creation of the Industries Assistance Commission, which for the first time forced the economy-wide costs of protection into public view.
The decisive shift came under Hawke and Keating. The dollar was floated in 1983, general tariff phase-downs were announced in 1988 and 1991, and the effective rate of assistance to manufacturing, which had exceeded 35 per cent in the early 1970s, fell to around 5 per cent by the turn of the century and sits at negligible levels today (Productivity Commission 2023). The automotive industry, the emblem of protected manufacturing, was managed down rather than simply abandoned: the 1984 Button plan traded gradual tariff reduction for model rationalisation and export facilitation, an attempt to build a leaner, export-capable industry rather than to preserve the old one indefinitely (Kelly 1992).
The endgame nonetheless arrived. Nominal tariffs on passenger motor vehicles, still 57.5 per cent in the mid-1980s, reached 5 per cent by 2010, and combined assistance of roughly A$30 billion between 1997 and 2012 could not overcome the arithmetic of a fragmented domestic market producing subscale volumes at high cost (Productivity Commission 2014). By the time the Commission concluded that further assistance would merely delay an inevitable exit, Ford, Holden and Toyota had already announced their departures, completed between 2016 and 2017. Importantly, the direction of change was not unique to Australia: manufacturing’s share of employment has fallen across every advanced economy since the 1970s, including in Germany and Japan, as productivity growth and shifting demand move labour into services (Rodrik 2016). Australia’s distinctiveness lies in the depth of the fall, not in its existence.
Comparative Advantage and the Logic of Letting Go
The orthodox account explains that depth without embarrassment. Australia’s factor endowments, abundant land, minerals and energy alongside scarce and expensive labour in a small market distant from dense industrial regions, generate comparative advantage in resources, agriculture and services exports such as education, and comparative disadvantage in scale-intensive assembly manufacturing. Protection did not repeal this logic; it merely taxed exporters and consumers to conceal it. On this view, tariff reform did not destroy manufacturing so much as stop compelling the rest of the economy to subsidise it, releasing labour and capital toward higher-value uses during precisely the decades in which Chinese industrialisation was transforming the returns to Australian resources.
The gains from that reallocation were real and widely shared: substantially cheaper vehicles, clothing and appliances for households, and a terms-of-trade windfall that lifted national income. Successive Trade and Assistance Reviews document how far the assistance burden has fallen and how concentrated its remnants remain (Productivity Commission 2023). Judged by aggregate welfare, letting go was sound policy, and the counterfactual of continued high protection implies a poorer and less flexible economy.
Yet the orthodox account is incomplete on its own terms. Comparative advantage explains specialisation; it does not guarantee that the resulting production structure carries the knowledge spillovers, export sophistication and adaptive capabilities that sustain long-run growth. On the economic complexity measures developed by Hausmann et al. (2014), Australia ranks strikingly low for a high-income country, its export basket dominated by unprocessed minerals and energy in a profile more typical of far poorer economies. A nation can be rich and simple while the ore carriers keep sailing; the strategic question is what remains when they slow.
The Mining Boom as Accelerant: Dutch Disease
The second complication is that much of the contraction occurred not as smooth structural adjustment but as a violent exchange-rate shock. Gregory (1976) warned, before the term Dutch disease had entered circulation, that growth in mineral exports would squeeze Australia’s other traded industries through the currency; Corden and Neary (1982) later formalised the mechanism as resource-movement and spending effects that pull factors into the booming sector and non-tradables while crushing the lagging tradables sector. The resources boom of 2003-2013 ran this experiment at national scale. The terms of trade roughly doubled, and the Australian dollar held above parity with the US dollar for much of 2010-2013, precisely the window in which all three remaining car makers announced their exits.
Reserve Bank modelling estimates that the boom raised real per capita household income by about 13 per cent by 2013 while leaving manufacturing output in the order of 5 per cent below its counterfactual level (Downes, Hanslow & Tulip 2014). The geography of these effects was equally stark: the income gains accrued disproportionately to the resource states and to shareholders, while the output losses concentrated in the manufacturing suburbs of Melbourne, Geelong and Adelaide that had least capacity to absorb them.
The analytical significance of Dutch disease is that it severs any easy equation of decline with efficiency. A temporarily overvalued currency destroys firms that would have been viable at the long-run exchange rate, and the destruction is not symmetric: supplier networks, tacit engineering skills and accumulated process knowledge did not reconstitute when the dollar subsequently fell toward US$0.65. Hysteresis of this kind means that part of Australia’s manufacturing loss reflects an unmanaged macroeconomic shock rather than revealed comparative disadvantage. Norway sterilised a comparable windfall through sovereign wealth arrangements; Australia largely spent its boom and let the exchange rate perform the adjustment. Decline was inevitable in direction, but the boom made it deeper and more abrupt than efficiency required.
Strategic Capability: What the Market Logic Misses
A third strand of argument holds that manufacturing matters beyond its measured output because it anchors innovation, complexity and national resilience. Manufacturing performs a disproportionate share of business research and development, and complex manufactured exports embody capabilities that diffuse across the wider economy (Hausmann et al. 2014). If such spillovers are real, purely private decisions will underinvest in manufacturing capability, and some public support is efficient rather than sentimental.
Recent shocks converted this abstract case into practical political economy. COVID-19 exposed Australia’s dependence on imported personal protective equipment, medicines and vaccines; domestic fuel refining has contracted to two plants; and the AUKUS partnership presumes a defence-industrial base that decades of deindustrialisation have thinned. The language of sovereign capability that now saturates Canberra reflects a judgement that market prices do not internalise the value of resilience against supply-chain disruption or economic coercion. That judgement is defensible, but its logic licenses targeted intervention in identified vulnerabilities, not reindustrialisation at large.
A Future Made in Australia? Evaluating the Revival Agenda
The Future Made in Australia agenda is the most ambitious Commonwealth industry policy since the Button era. Anchored by the Future Made in Australia Act 2024 (Cth), the 2024-25 Budget committed A$22.7 billion over a decade to the package, alongside the A$15 billion National Reconstruction Fund legislated in 2023, spanning Hydrogen Headstart, production tax incentives for renewable hydrogen and processed critical minerals, and support for solar manufacturing (Commonwealth of Australia 2024). Treasury’s National Interest Framework supplies the agenda’s intellectual discipline, admitting support through only two gates: net zero transformation sectors in which Australia can plausibly attain durable comparative advantage, and economic resilience sectors in which security concerns justify paying a premium (Commonwealth of Australia 2024).
The strongest case for the agenda rests on dynamic comparative advantage. Garnaut (2019) argues that the endowments which once condemned Australian manufacturing now invert the calculus: in a decarbonising world, world-class solar and wind resources co-located with iron ore, bauxite and critical minerals make Australia a natural site for energy-intensive processing, because it is cheaper to embody renewable energy in green iron, aluminium and ammonia than to ship that energy abroad. Grattan Institute modelling similarly finds that a green steel industry could sustain substantial employment in the coal regions most exposed to transition (Wood & Ha 2020). Support framed in these terms is not a defiance of comparative advantage but an anticipation of it, accelerating investment where learning curves, coordination failures and the massive subsidy programs of competitors, most obviously the United States Inflation Reduction Act, would otherwise delay or divert it.
The sceptical case is equally serious. Australia’s assistance history suggests that subsidised sectors assemble lobbies faster than they assemble capabilities, and that governments struggle to collect what Banks (2008) calls the exit fee when support fails to produce competitiveness. In an economy near full employment, subsidised projects bid workers and capital away from unsubsidised ones, so the true test is not whether funded factories open but whether they outperform what the same resources would otherwise have produced. Scale scepticism also persists: Australia accounts for a very small fraction of global manufacturing output, and activities such as solar module assembly confront Chinese cost structures that no plausible subsidy will offset. The soundness of the agenda therefore turns on governance rather than aspiration: rigorous sector assessments under the National Interest Framework, sunset clauses, and a demonstrated willingness to let failing ventures fail. Where those disciplines hold, the policy is a reasonable hedge on a plausible future; where they slip, it risks reconstructing the protected mediocrity that the reforms of the 1980s dismantled.
Counterargument and Rebuttal
Against the position advanced here, a substantial school contends that the decline was never inevitable but was chosen. On this account, small high-wage economies such as Sweden, Switzerland and Germany retained sophisticated manufacturing through patient capital, deep vocational training systems and relentless export orientation, and Australia could have done likewise had governments treated the sector as a capability to be cultivated rather than a distortion to be unwound. The 2013 decision to withdraw comparatively modest automotive assistance, on this reading, extinguished an industry whose annual cost was small relative to the design, engineering and supplier capabilities it anchored, and the abruptness of liberalisation, compounded by an unmanaged exchange rate, converted an orderly adjustment into a rout.
The rebuttal must be partly concessive. The Dutch disease evidence reviewed above supports the narrower version of this claim: the pace and depth of decline were policy-contingent, and a Norwegian-style stabilisation architecture would probably have preserved more capability through the boom. But the stronger claim, that a broad advanced-manufacturing base could have been sustained, fails on the evidence. Decades of assistance at rates far exceeding anything contemplated today produced neither export competitiveness nor innovation leadership in the assisted industries (Productivity Commission 2014). The comparator economies owe their positions to century-deep institutional complementarities, adjacency to dense industrial markets and firm-level scale that Australian policy could not conjure at will, and even their manufacturing employment shares have fallen steadily (Rodrik 2016). The realistic counterfactual for a protected Australian manufacturing sector was never Bavaria; it was a larger and costlier version of the industry that actually existed, subscale, inward-looking and perpetually assisted. Inevitability, properly stated, attaches to the direction and rough magnitude of the decline, not to every plant closure along the way.
Conclusion
The decline of Australian manufacturing was overdetermined. Endowment-driven comparative disadvantage, the universal drift of advanced economies toward services and the dismantling of an indefensible protection regime together guaranteed a shrinking manufacturing share; in that sense the decline was inevitable, and the reforms that permitted it enriched the nation. What was not inevitable was the severity of the loss. An unmanaged mining boom pushed the exchange rate to levels that destroyed viable capability alongside unviable capacity, and the failure to stabilise that windfall stands as the genuine policy error of the era. The revival agenda should be judged against this diagnosis. Where Future Made in Australia follows emerging comparative advantage into green iron, refined critical minerals and energy-intensive processing, disciplined by Treasury’s national interest tests, it is sound policy for a world of subsidised competitors and fragile supply chains. Where it drifts toward recreating manufacturing breadth as an end in itself, it repeats the tariff century’s central mistake. The lesson of that century is that protection cannot manufacture competitiveness; the lesson of the boom is that markets left entirely alone can destroy more than efficiency requires. Sound policy for Australian industry lives, uncomfortably but necessarily, between the two.
References
Australian Bureau of Statistics 2024, Labour Force, Australia, Detailed, ABS, Canberra.
Banks, G 2008, ‘Industry policy for a productive Australia’, Colin Clark Memorial Lecture, Productivity Commission, Melbourne.
Commonwealth of Australia 2024, Future Made in Australia: National Interest Framework, The Treasury, Canberra.
Corden, WM & Neary, JP 1982, ‘Booming sector and de-industrialisation in a small open economy’, The Economic Journal, vol. 92, no. 368, pp. 825-848.
Downes, P, Hanslow, K & Tulip, P 2014, The Effect of the Mining Boom on the Australian Economy, Research Discussion Paper 2014-08, Reserve Bank of Australia, Sydney.
Garnaut, R 2019, Superpower: Australia’s Low-Carbon Opportunity, La Trobe University Press, Melbourne.
Gregory, RG 1976, ‘Some implications of the growth of the mineral sector’, Australian Journal of Agricultural Economics, vol. 20, no. 2, pp. 71-91.
Hausmann, R, Hidalgo, CA, Bustos, S, Coscia, M, Simoes, A & Yildirim, MA 2014, The Atlas of Economic Complexity: Mapping Paths to Prosperity, MIT Press, Cambridge, MA.
Kelly, P 1992, The End of Certainty: The Story of the 1980s, Allen & Unwin, Sydney.
Productivity Commission 2014, Australia’s Automotive Manufacturing Industry, Inquiry Report no. 70, Productivity Commission, Canberra.
Productivity Commission 2023, Trade and Assistance Review 2021-22, Productivity Commission, Canberra.
Rodrik, D 2016, ‘Premature deindustrialization’, Journal of Economic Growth, vol. 21, no. 1, pp. 1-33.
Wood, T & Ha, J 2020, Start with Steel: A Practical Plan to Support Carbon Workers and Cut Emissions, Grattan Institute, Melbourne.