Industrial policy is a term that is often interpreted differently depending on the audience. The fact that it cuts through a variety of economic policy tools, ranging from innovation programmes and tax policy to trade and Foreign Direct Investment (FDI), makes the matter even more complex. As such, it lends itself to easy misinterpretation. At its core, it refers to “any type of selective intervention or government policy that attempts to alter the structure of production toward sectors that are expected to offer better prospects for economic growth than would occur in the absence of such intervention” (Pack and Saggi, 2006, p. 268). This might or might not include sectors/technologies where leadership might have geopolitical, security, and military implications. The concept is built on two fundamental principles: (i) production in some sectors is more desirable than in others (Hausmann et al., 2007), and because of this, (ii) governments should make an active effort in nudging the production structure in that direction.
It is important to note that, irrespective of simplistic characterisations (e.g. capitalism vs socialism) or standard economic modelling of markets under perfect conditions, practically all countries engage in various forms of industrial policy, and always have, including the US, UK, France, Singapore, China, Taiwan, and South Korea, just to name a few (Rodrik, 2009; Terzi et al., 2023, 2022). Notoriously, Japan has a long history of using industrial policy to foster rapid industrialisation and growth, at least since the end of WWII (Okazaki, 2017).
Historical cycles in industrial policy
Over the past few decades there has been a general reduction in the use of industrial policy, which moved to the margins of mainstream economics. As this principle got engrained in policymakers’ minds, it was framed under the narrative that governments cannot pick winners in a market economy, and in fact they are at a relatively high risk of being captured by interest groups (Rodrik, 2014). And of course, narratives eventually shape policies (Shiller, 2019).
Industrial policy was also seen as harmful to the pursuit of a more globalised world economy, which to some extent became a leading objective in and of itself (Rodrik, 2011). In the service of a rule-based global trade order, richer nations of the West sided against a preponderant government role in altering production in a certain direction, preferring comparative advantages to manifest themselves freely.
The idea that industrial policy was unhelpful was to some extent codified in the so-called ‘Washington Consensus’, or else the idea that a small government (together with open current and capital accounts) was instrumental in achieving rapid development. As such, the resistance to industrial policy was exported to the Global South, in particular during macroeconomic adjustment programmes (Irwin, 2022). In Japan’s case, it was the Industrial Policy Dialogue with the US in 1983/84, that led to a scaling down of standard industrial policy practices, which were substituted with a more broad-based push for structural reforms (Okazaki, 2017).
Over the last few years, however, we have been living through a return of what has facetiously been called “the policy that shall not be named” to the main stage of economic policy (Cherif and Hasanov, 2019). There are several reasons for this. First, and most prominently, is the fact that China has been making extensive use of it, and has been experiencing meteoric growth in the process. Moreover, the fact that the country’s accession to the WTO in 2001 did not encourage it to abandon such practices effectively invalidated the argument for others to favour the safeguarding of a level playing field at global level. Second, COVID-19 required a large degree of government intervention in the economy, including for the stockpiling and provision of Personal Protective Equipment or the fast development and production of vaccines. And, of course, in the moment in which global supply chains came to a grinding halt due to COVID-induced restrictions, there was a sudden realisation of the central role that microchips play in today’s economy, from cars to military applications. Lack of access to them could be weaponised, as AI and the digital economy will play a crucial role in defining military supremacy in the 21st century (Terzi, 2023).
Finally, the returning appeal of industrial policy is due to climate change, defined as the greatest market failure the world has ever seen (Stern, 2006), which itself questions the narrative that market forces should be left largely unfettered (Terzi, 2022).
Guiding principles for an effective industrial policy
After reviewing the most relevant industrial policy interventions of the past in Europe, the US, China and Japan, including both successes and failures, and crosschecking these findings with the relevant recent literature, Terzi et al (2023) conclude that six basic design characteristics should be maintained when crafting a successful industrial policy in order to maximise its impact and minimise risks. These include:
- I.Future-oriented. Industrial policy has proved most effective when focused on the future, not the present (or the past). This means one cannot identify strategic sectors by simply looking at the home country’s current competitive advantage. Industrial policy must be future-, and hence innovation-oriented. Its aim is to kick-start sectors where market failures are preventing a desirable equilibrium, not lean against the wind of global structural change.
- II.Sector- and Technology-driven. Focus should be on areas/technologies and not companies. One should refrain from designing a policy around specifically ‘picking winners’, i.e., single companies that are considered to be a strategic asset, as this is likely to lead to perverse incentives, decrease the innovativeness of the ‘champion’, and in the end counteract the very objective of industrial policy i.e., fostering long-term competitiveness and sustained growth on top of harming consumers. On the other hand, the objective of an effective industrial policy must be built around sectors or technologies where advanced capabilities are desirable.
- III.Competition is a strength. An effective industrial policy should not be about making incumbents bigger. Clearly, having companies that grow to become global leaders would be a sign of competitiveness in a specific sector. However, artificially fostering scale to increase the number of a nation’s corporations in global rankings is similar to treating the symptom rather than the root cause of an illness.
- IV.Top-down, but also bottom-up. By definition, industrial policy is equivalent to setting a top-down direction of economic development. However, in any form it takes, it should not transform itself into economic planning, or it risks leading to decreased productivity growth. In other words, it should encourage experimentation and bottom-up innovation and creativity. This is why the policy goal must be defined in a balanced way that is tangible enough to make it concrete, but broad enough to allow for creativity in achieving it. It should therefore not undermine the entrepreneurial ecosystem. As such, it complements and perhaps even reinforces the crucial role of strict competition policy enforcement discussed in point III above.
- V.Accountable, non-partisan and adaptable. Transparency is key to avoiding or reducing the political economy risk of capture by interest groups. Moreover, politicians should avoid the tendency of tying their political success (or failure) to the success of a specific domestic company. Failure to do so creates a strong incentive to keep funds flowing, even when an objective assessment would suggest otherwise. An effective industrial policy should therefore remain as independent as possible, be assessed against clear targets, allowing for adaptability along the way. This implies that policies, outcomes, and assumptions must be constantly monitored, questioned, and quickly adapted if need be.
- VI.Holistic approach. Supply and demand considerations, together with offensive and defensive tools, must be designed consistently, shaped in tandem with a supportive regulatory environment. A successful industrial policy should not only foster the development of innovation in critical technology areas (e.g., through targeted R&D), but also use tools to ensure increasing demand for those sectors.
In the era of a resurgence in industrial policy, Japan stands out with several advantageous factors. Firstly, the country possesses a wealth of experience in employing industrial policy tools, supported by a competent bureaucracy at METI and beyond. This expertise provides a solid foundation for effective policy implementation. Secondly, Japan's strong access to financial markets gives it a significant edge on the global stage, enabling the mobilization of substantial financial resources that are crucial for the success of industrial policy initiatives. Additionally, Japan benefits from a robust industrial base and a high level of technological advancement in various fields, including cutting-edge digital applications. This advantageous starting point allows Japan to leverage existing dynamic ecosystems, rather than having to start from scratch, when fostering key industries of the future.
On the downside, however, industrial policy in the 21st century strongly relies on an integration of economic and security considerations. Japan’s pursuit of enomic progress post-WWII has largely detached from geopolitical or military rivalry, benefitting from a rules-based open global trade order. As that order comes to an end, adjusting economic policymaking, including industrial policy, for a world of increasing geopolitical tensions will represent a challenge for Japan. In many ways, this challenge is shared with the European Union, whose policymaking was similarly designed based on peaceful principles. A fruitful exchange of best practices can therefore be envisioned going forward, as both Europe and Japan, in collaboration with likeminded partners, readjust and design new industrial policies that are fit for the 21st century.