Continuing to browse our website indicates your consent to our use of cookies. For more information, see our Privacy policy.

How to use it

The pitfalls of protectionism: Import substitution vs. export-oriented industrial policy


Published 14 May 2024

After decades of dismissal, industrial policies are experiencing a revival. Were past industrial strategies truly ineffective? What insights can we gain from them in order to inform present industrial policies? The IMF Working Paper from Reda Cherif and Fuad Hasanov analyzes successes and failures of the past in order to draw powerful conclusions about today’s policies.

Here’s how to use the report entitled The Pitfalls of Protectionism: Import Substitution vs. Export-Oriented Industrial Policy.

Why is the report important?

As many economies are currently adopting or considering industrial policies to spur economic and industrial growth, it is important to understand what policies worked or failed in the past, and how to mitigate risks of industrial policies. This report undertakes exactly the analysis necessary to isolate the factors that led to success and failure during the wave of industrial policies undertaken by developing economies in the 1960s and 1970s. This analysis both reminds policymakers of the risks of industrial policies and makes a strong case for continued open trade to drive competition and innovation for sustained, high-quality economic growth.

What’s in the report?

The report includes seven principal sections:

How to find the key insights

The industrial policy debate

  1. Industrial policy narrative has made a comeback since the 2008 financial crisis, with increasing focus on income inequality, deindustrialization, and the hollowing out of the middle class; since the 1980s and its ‘Washington Consensus’ on free market reforms, industrial policy has been considered a fringe idea; now it has returned to the public debate. (p. 3)
  2. Return of industrial policy culminated in an acceleration of public announcements of major industrial policy measures by the EU, US, and China; in 2015, China announced ‘Made in China 2025’; starting from 2018, the EU announced a series of industrial strategy plans to achieve autonomy and leadership in clean technologies manufacturing and green jobs; the US has launched multiple programs on a scale not seen for decades. (p. 3)
  3. In the new debate about industrial policies the interpretation of past experiences plays a crucial role and represents the main intellectual battlefield; those in favor cite the example of Asian miracle economies (i.e., Korea and Taiwan); those against cite the experience of developing countries trying to create new industries beyond their capabilities; these two types of experiences and their associated strategies rely on fundamentally different principles: inward-looking or import substitution (IS) and export-oriented (EO) policies. (p. 4)
  4. Export oriented policies are associated with industrial policy success, achieving five key objectives such as (i) market signals and feedback free of distortions, (ii) fierce competition on foreign markets, (iii) spillovers –deeper value chain integration resulting in positive spillovers to other firms, (iv) market size – large enough to take advantage of economies of scale and scope, and (v) accountability – incentives mitigating cronyism and corruption. (p. 4)
  5. Import substitution policies cannot take advantage of these five objectives; deprivation of market signals is particularly challenging and can result in a lack of sufficient investment in innovation or production of large productivity gains; this system produces firms that can only survive under protection against both foreign and domestic competition; in the absence of state support domestic producers quickly go bankrupt. (p. 5)
  1. Export orientation could be the secret ingredient of a successful industrial policy; a large domestic market providing economies of scale may not be sufficient to avoid the pitfalls of import substitution policies; analyzing the tradeoffs of the IS and EO strategies in the context of new industrial policies, the authors find that although protectionism and high tariffs were used in the past, they may not be necessary now as there are alternatives, and may even be counterproductive for large economies. (p. 5)

Industrial policy always failed in the past…because a “True” one was barely tried

  1. A standard argument against state intervention or industrial policy to spur growth of new industries are the many past failures; enterprises created in earlier waves of industrial policies ended up in financial disarray, wasting resources. (pp. 5-6)
  2. In fact there were very few examples of “true” industrial policy, i.e., focused on export orientation in technologically sophisticated industries; many industrial policies focused on non-tradeable and non-sophisticated services or the creation of productive capacities in inward-looking heavy industries or examples of import substitution industrialization (ISI). (p. 6)
  3. The golden age of import substitution (1965-1980) yielded a sizable increase in manufacturing production although not to the level of Asian miracle countries; many developing countries achieved relatively high growth rates in manufacturing spurred by IS policies; when IS polices were rolled back in most developing economies from 1980 to 2010 average growth rates in manufacturing dropped significantly and manufacturing stagnated in many economies. (pp. 6-7, Figure 1)
    Figure 1: The rise and fall of IS-led growth
  4. Among developing economies, very few pursued an export-oriented industrialization policy on a large scale as in the Asian miracle economies, with the Asian miracle economies clearly outliers over the 1970-1990 period, especially with respect to export performance and sustained growth in their production of manufacturing. (pp. 7-10, Figures 2-4)
  1. The level of manufacturing export intensity of an economy by its market share in world manufacturing exports is a helpful way of measuring whether an economy is making a serious foray in to world markets in relation to its population size; countries like Korea, Malaysia and Thailand grew fast both in terms of manufacturing output and manufacturing export intensity; other countries known for import substitution polices like Egypt, Nigeria, and India had relatively high growth in manufacturing output with a negative change in their manufacturing export intensity; besides Korea and some other countries, developing economies did not pursue simultaneously a rapid industrialization and an export-oriented policy. (pp. 10-12, Figure 5)

The mid-20th century import substitution policies in India: The state as a micromanager

  1. India presents a good case study of industrial policy failures; though it has economies of scale, it demonstrates that lack of export orientation leads to a lack of productivity improvements, innovation, and dynamism and how import substitution policies are accompanied by distortions and perverse incentives. (p. 12)
  2. India had an ambitious import substitution policy until the late 1970s with the goal of producing domestically a high proportion of the manufactured goods it consumed; India suffered regularly from hard currency shortages and a low growth rate in per capita terms; import substitution goals were achieved through intricate rules and regulations, in particular a licensing system; domestic firms were completely shielded from international competition through heavy restrictions and even bans on imports; domestic competition was also heavily curtailed; profits for firms depended more on the ability to secure adequate licenses than on competition through improved productivity or innovation; exports rarely exceeded the minimum requirement due to high costs, inability to meet order deadlines and volumes, and low profits. (pp. 12-13)
  3. India’s IS policy contradicts the key principles of “true” industrial policy except for interventions to create domestic capabilities, which it did with relative success; IS gave the illusion of self-sufficiency while leaving India vulnerable to external shocks as it suffered from currency shortages and still relied on imports of raw materials and key industrial equipment. (p. 13)

How to apply the insights

  • This section provides a case study for policymakers of how industrial policies can go wrong, even when there are economies of scale at play and incentives to export.

  • Without the objective of competing in export markets and with heavy regulatory burdens to protect domestic firms, India could not achieve efficiency, innovation, true incentives, or the capability to export.

  • Nor could India shake reliance on imports for key inputs.

  • These are key issues for policymakers to bear in mind in designing industrial policies today.

Export orientation, not tariffs, is the secret ingredient

  1. Export orientation has been a critical ingredient for the “true” industrial policy of the Asian miracle economies and represented a major departure from the IS policies of other developing economies, though both policies used tariffs to protect domestic markets and subsidies to support domestic champions; typical developing economies applied tariffs without EO policies, successfully increasing domestic production and improving capabilities, though over time, protection from international competition meant little investment in R&D and innovation and some dependence on imports, particularly critical technologies. Even if protected firms were well-managed, they would perpetually depend on protection from international and domestic competition and on receiving state subsidies and would be vulnerable to currency devaluation and elimination of tariffs. (pp. 13-14)
  2. In contrast, in Asian miracle economies, export promotion was the main objective from the beginning and tariffs were only one tool among many to ensure a minimum rent for domestic producers while facing fierce competition domestically and internationally; other tools included production and investment subsidies, cheap financing, support from public research institutes, government procurement, and tax incentives; firms had to invest heavily in R&D and innovate to compete internationally and to create economies of scale and efficient operations to take advantage of low wage environments in their economies; firms could still compete when protections were lifted and benefit from devaluation; outcomes were different from economies with IS policies because export orientation acted as an organizing framework for all state policies and the extent of support with clear market signals as feedback. (pp. 14-15)
  3. The state gave strong incentives to export in specific sectors and provided support but ultimately private firms were in the lead, including in innovation, international partnerships, and marketing; while a lot of support was given, the firms had sufficient autonomy to shield them from undue interference or abrupt policy changes; export orientation prevents policymakers from falling into the illusion of economic “independence” and tariffs are neither necessary nor sufficient to grow industries; the development of Korea’s Hyundai and Malaysia’s Proton automakers illustrates these points. (pp. 15-16)
  4. What matters is export orientation in a broad sense, including both vertical specialization industrialization, which focuses on exporting intermediate good and linkages with global value chains and regional networks, and traditional export orientation focused on final goods in advanced markets. (p. 16)

How to apply the insights

  • This section makes a powerful but elegant case not only for the importance of export orientation in policy making, but the factors that make export orientation the secret ingredient and allow firms to sustain competitiveness even when protections are removed.

  • These are key lessons for policy makers designing industrial policies now.

The limits of laissez-faire

  1. The pursuit of unfettered free market policies adopted by the developing world in the 1980-90s, can produce undesirable outcomes; average growth rates in developing economies in the 1980s-90s were lower than in the IS policy years of the 1960s-70s; laissez faire policies did not tackle market failures in sophisticated products and allowed market forces to focus on commodities, low-skilled, and non-tradeable sectors resulting in low productivity growth. (pp. 16-17)
  1. Chile, which pursued a laissez-faire policy until the early 1980s with disappointing results, witnessed a sizable loss in manufacturing exports market share; the 1982 crisis triggered a roll back of Chile’s laissez-faire policies when it raised tariffs on imports and re-introduced some state intervention to spur growth; despite some success and relatively steady real GDP growth and macroeconomic stability, Chile’s productivity growth has lagged its peers. (p. 17)

Risk tradeoffs in industrial policy strategies

  1. The main considerations for the development of a new industry are the local production’s distance to the technological frontier, the level of domestic ownership, and the macroeconomic costs in terms of the current account deficit and possible fiscal deficits. Each strategy in developing a new industry – import substitution industrialization, export orientation, or a mix of the two – entails risks. (p. 18, Figure 6)
    Figure 6: Industrial policy strategies and risk tradeoffs
  2. Risks for import substitution industrialization: this strategy relies on the belief that by closing a domestic market to competition, a sector can become self-sufficient. (pp. 18-19)
  3. There are three categories of ISI, and each has its risks:
    1. quasi autarky, where local producers dominate the market and fully own firms protected by trade restrictions; also characterized by minimal imports and no exports, with a minor trade deficit and large rents to producers (India). Minimizing links with MNCs hinders technology transfer, resulting in outdated, low-quality products with little innovation – the industry stays below the technological frontier, eventually risking loss of the entire industry when exposed to competition. (pp. 19-20)
    2. a strategy of retaining full ownership domestically while relying more on foreign partnerships allowing for an automotive trade deficit and creating a sizable consumer cost (Malaysia’s automotive industry). (p. 19)
    3. economies that negotiate access to their protected local market in exchange for FDI to produce locally, can involve forced technology transfer in exchange for access to a captive market resulting in lower quality/higher cost offerings for the domestic consumer (Brazil’s automotive industry). (p. 19)
    4. For both b) and c) there is an increased risk of the appearance of a more technologically advanced industry that is but nevertheless still relatively far from the frontier, and with an internationally uncompetitive cost structure. (pp. 19-21)
  4. Though there is evidence that the use of ISI may have helped jumpstart industries, especially in the “easy stages,” the strategy is not sustainable in the long run; learning, innovation, and competition are lacking, jeopardizing long-term stability of the industry (pp. 19-20)
  5. ISI is often unsustainable, sacrificing welfare of consumers by restricting their choice and making them pay higher prices without developing an internationally competitive (p. 21)
  6. Risks for an export-oriented (EO) approach: Asian miracle economies combined elements of protectionism and export orientation, succeeding at developing competitive firms in sophisticated industries while managing macroeconomic and political economy risks and using support measures and incentives beyond trade protectionism to make them competitive and innovative globally, allowing learning and firm growth; the focus on export orientation dwarfed protectionist policies that entailed risks; providing incentives for domestic firms to compete at the outset without a domestic captive market to fall back on makes firms take state support seriously to do their best, putting them on the path to the technological frontier. (pp. 21-22)
  7. An EO approach requires a much stronger coordination of policy tools and availability of long-term financing in order to support newly created firms until they can become competitive in the absence of protectionism; Asian miracle economies pursued an optimal combination of EO and ISI policies but the strategy required sizable support, entailing the risk of sliding into a traditional ISI strategy; in a globalized world dominated by GVCs, a combination of ISI and EO may not be feasible or desirable requiring strong incentives and incurring high monitoring costs; given the level of trade liberalization and rise of GVCs, an ISI/EO strategy is harder to implement and less desirable. (p. 22)
  8. Industrial policies should aim to create sustainable, innovative, and efficient industries in sophisticated products, the development of which is riddled with market failures; state intervention in these industries without rigid pre-requirements could bring success, but raises the question of what kind of intervention is the most effective; tariffs and local content requirements might not be sufficient and could be counterproductive. (pp. 22-23)
  9. An EO industrial policy, or “true” industrial policy, is flexible enough to mitigate the risks in the 21st century; this policy should provide different types of support and incentives to firms producing locally while encouraging them to export, without imposing local content requirements or a specific production structure or technology; adoption of tariffs or trade restrictions could provoke retaliation and be fatal in small economies lacking scale. (p. 23)
  10. Several factors require focus as part of an EO strategy: 1) complexity of production is far greater than in the 1960s-70s, requiring firms to build partnerships with the most advanced firms; tariffs and trade restrictions will make establishing partnerships and participating in GVCs more difficult and costly; 2) the EO policy’s key advantage of reliable market signals from export markets mitigates risks of capture and avoids the illusion of self-sufficiency be helping different stakeholders and investors to realize when the need to restructure arises; 3) the incentive to export combined with adequate support and appropriate institutions provides flexibility and incentives to achieve a competitive cost structure while encouraging innovation. (p. 23)
  11. An EO industrial policy can achieve better results at increasing local content and domestic value-added content without imposing such requirements before the local industry is ready with cost-competitive and comparable quality substitutes. (pp. 23-24)
  1. Is an EO policy viable in the post-pandemic world with US-China technological competition and national security and economic resilience concerns, causing a slide toward geoeconomic fragmentation?; despite signs of fragmentation into trading blocs, non-aligned countries play a larger role in trade than in the past and empirical evidence does not show large dislocations in trade patterns; demographic trends and the establishment of large regional trade areas imply that new opportunities for trade will emerge. (p. 24)
  2. Although there are risks to EO policies in times of geopolitical competition, the world is more integrated than in the past, with a freer flow of information because of digital technologies; these gains may not be easily reversed; international cooperation to set mutually beneficial rules for trade in the current context would be optimal for all. (pp. 24-25)

The chips must flow: Is protectionism back?

  1. From the 19th century until after WWII, the US and other of today’s advanced economies used state intervention – particularly tariffs – to protect and grow their manufacturing industries; drawing parallels with current industrial policy would be misleading, as during that era, protectionism was only one tool used among a wide array of policy support, the economies quickly became exporters of manufactured goods, and the distance to the technological frontier was much shorter. (pp. 25-26)
  2. Protectionist policies started fading after WWII and by the mid-1980s, free trade and liberalization were believed to best support development and growth; after the 2008 financial crisis, the adoption of China’s Made in China 2025 policy to focus on high-tech industrial advancement, the Covid pandemic, and all of the challenges to economic growth these experiences brought to light, policies to support domestic industries in the US were resurrected. (p. 26)
  3. In the 2010s and early 2020s, new policies adopted went beyond the goal of economic development to include national security, ensuring resilience, and climate change mitigation and adaptation, though at their heart, the economic justifications for these policies are the same as for any other kind of industrial policy: the use of government intervention to acquire new industrial capabilities in a sustainable way which would otherwise not take place because of market failures; though no major economy has gone back to IS to accomplish these new goals, most countries have included local content requirements (LCRs) and other industrial policy tools to achieve resilience which amount to indirect restriction on competition and trade, which may not strike the right balance. (pp. 26-27)
  4. US Inflation Reduction Act (IRA) provisions as an illustration of arguments in this paper: the IRA is a long-term commitment to a subsidized financing program to build green infrastructure while ensuring firms receiving support pay adequate wages; the IRA differs from ISI policies in that its main goal, financing green infrastructure, is not intended to substitute suppliers of existing tradable product but to create the supply chain of a new non-tradable product: clean power. (pp. 27-28)
  5. The IRA’s electric vehicle (EV) subsidy is closer to an ISI policy, with the goal of the tax credit to product a tradable product in North America and precluding unfettered competition with imported EVs and batteries; however, the incentive creates a new demand and invites both domestic firm investment and FDI as well as new opportunities to export components to the US; although use of LCRs could result in higher prices and/or lower quality of domestic EV production, subsidies should cause a price hike similar to the Korean strategy; it is important to make sure competition and innovation among domestic producers stays high and exports are encouraged to mitigate the risks of this policy. Temporary subsidies should provide an incentive for domestic producers to invest for the future rather than become complacent with protectionism. (p. 28)
  6. The current structure of the investment and production tax credit for clean energy allows for both domestic and international competition but risks protectionist measures that could reduce innovation; a combination of high demand and low competition for some low-tech inputs could reduce the effort to create technology and new capabilities. (p. 28)
  7. Without explicit export orientation policies in the IRA, in order to maximize benefits to the US economy, it will be critical to adopt policies and take actions to ensure that industrial capabilities are developed around green technologies and reduce barriers hindering innovation and competition; LCRs may not be sufficient to help industries become cost competitive and first developing capabilities in key; in the design of industrial policy packages like the IRA and the EU’s Green Deal, policies to develop local capabilities are much more important than protectionist measures. (p. 29)
  8. The key difference between the success of industrial policy in Asian miracle economies and the failure of IS policies in developing economies is export orientation; the fact that these economies combined export orientation with protectionist measures is not inconsistent with this argument; the Asian miracle economies were able to pursue both policies without retaliation but today would have to pick one policy or the other; export orientation should be the absolute priority; IS or laissez-faire policies would not necessarily produce sustained growth; currently, the race for semiconductors and renewable technologies is tempting policymakers to pursue protectionism but to achieve success, creating a race to the top could bring technology costs down and benefit all economies. (pp. 29-30)

How to apply the insights

  • This section evaluates both the historical US use of industrial policy and today’s US industrial policy against the thesis developed in this paper, clearly defining the differences between historical industrial policies and the US IRA provisions today, and giving policymakers both a clear sense of the risks of today’s provisions as well as ideas for how to mitigate those risks.

  • The conclusion importantly spells out the role of protectionist measures in combination with export oriented policies historically and now, and makes a strong case for avoiding risks that would dampen international competition and forces that drive innovation.

Conclusion

The report provides a persuasive analysis of how export oriented industrial policies led to the remarkable successes of the Asian miracle economies of the 1960s and 70s, as well as why the import substitution oriented industrial policies of other developing economies failed, and the lessons these histories hold for development of industrial policies now.  This report is a must-read for any policymaker formulating industrial policies today.

Download report

Complementary reports and analysis

Hinrich Foundation

External Resources

© The Hinrich Foundation. See our website Terms and conditions for our copyright and reprint policy. All statements of fact and the views, conclusions and recommendations expressed in this publication are the sole responsibility of the author(s).