IEA-METI-RIETI Conference

New Thinking on Industrial Policy (Summary)

Information

  • Time and Date: 12:00pm-3:00pm (GMT) / 9:00pm-12:00am (JST) / 7:00am-10:00am (CDT) / 8:00am-11:00am (EDT), Friday, June 10, 2022
  • Hosts: Research Institute of Economy, Trade and Industry (RIETI) / Ministry of Economy, Trade and Industry (METI) / International Economic Association (IEA)

Summary

Key Takeaways

There is a general consensus that industry policy is once again gaining momentum today. Industrial policies have been in place regardless of whether they are branded as industrial policy. Participants identified factors such as underinvestment and low productivity in the past, necessity of missionoriented policy targeted to address today’s pressing social problems, and importance of stable institutional framework for micro-policy as the drivers behind industrial policy today (Mr. Hirai, Ms. Leong).

Comparing industrial policies today with those in the past (first generation industrial policy or “grandpa’s industrial policy), participants noted the risk of policies such as subsidies and carbon tariffs falling into the path of the first generation industrial policy which was based on competitiveness concerns rather than externality concerns (Professor Goolsbee). Government participants (Mr. Hirai, Dr. Steinberg) pointed to transnational coordination as a key in mitigating this problem.

Scholars identified information asymmetry and rent seeking as the source of general criticism on industrial policy. In overcoming these problems, participants highlighted the importance of embedding industrial policy in society through shared financial and implementational responsibility between public and private sector (Professor Lerner, Professor Stiglitz, Professor Tomiura) and engagement of civil society and regional actors (Professor Hanson, Ms. Leong, Professor Sabel, Professor Stiglitz), thereby bringing the best out of the private sector in achieving new social and development objectives (Dr. Criscuolo). On the degree of government intervention and selection of appropriate instruments in policy implementation, frameworks such as the US National Science Foundation, DARPA (Professor Goolsbee), Israel’s Yozma scheme (Professor Lerner), and Peru’s Mesas Ejecutivas policy (Dr. Ghezzi) were highlighted as pertinent examples.

There was also a discussion on targeted policy. Government participants cited the unintended consequences or emissions arising as a result of a focus on horizontal policy in the past (Ms. Leong), and the necessity to promote newly defined set of sectors (e.g. green industry, resilience-related industry) (Mr. Hirai) as parts of rationale behind today’s targeted policy. Several scholars (Professor Goolsbee, Professor Lerner) cautioned against governments making too specific targets ex-ante, and the importance of learning-by-doing and a feedback loop among stakeholders is highlighted in this regard (Dr. Ghezzi, Professor Hamaguchi, Professor Hanson, Professor Sabel).

Participants also stressed the necessity of (i) capacity-building and trust-making among policy designers and stakeholders (Dr. Ghezzi, Professor Hamaguchi, Mr. Hirai), and (ii) data collection, data sharing and evidence-based policy making (Dr. Corrado, Professor Tomiura) for effective industrial policy.

1. Opening Remarks

Speakers:
  • Mr. HAGIUDA Koichi (Minister of Economy, Trade and Industry, METI, Japan) (pre-recorded)
  • Dr. Dani RODRIK (President, IEA)

H.E. Mr. Hagiuda, Minister of Economy, Trade and Industry of Japan welcomed participants, and stressed Japan’s willingness to lead the global economic and social reforms amid increasing changes in geopolitics and rising concerns over the future of the global economy. In light of this, the minister revealed the New Direction of Economic and Industrial Policies, a policy initiative by METI to achieve a virtuous cycle of growth and wealth distribution through mission-oriented industrial policies and reconstruction of economic and social systems. Minister Hagiuda concluded his remark by calling on the participants to lead the way in identifying new industrial policies that will best serve the world going forward.

Echoing Minister Hagiuda's remarks, Professor Rodrik noted that industrial policy is back in fashion as many countries and governments seek new forms of capitalism and industrial policy. Professor Rodrik further elaborated that while industrial policy has always been practiced, there is an opportunity to make it better by being more conscious and systematic about the policy. He also pointed to two traditional sources of skepticism about industrial policy: (i) the lack of information among policymakers and (ii) the political capture that can arise from industrial policy, and stressed the importance of mutual learning and cooperation between policymakers and economists.

2. Keynote Speech: The Role of Industrial Policy in the New Era

Speakers:
  • Dr. Joseph STIGLITZ (Columbia University, US)

Professor Stiglitz noted a confluence of factors, including financial crisis, climate crisis, inequality, Covid-19, and the war in Ukraine, behind industrial policy taking a center stage once again. Professor Stiglitz stressed that market alone do not suffice systematic failures, and governments can improve matters by government-designed economic strategy, and cited that the systemic short-sightedness to risk is evidenced most strikingly by an European country which made itself significantly dependent on Russian gas.

Professor Stiglitz recalled oppositions from many advanced countries and from within the World Bank on his project on the East Asian miracle in the last 1980s’-early 1990s’, a World Bank project with support from Japan. His observation is that what was behind these opposition was unwillingness to seek for an alternative to the Washington Consensus, and industrial policy has discredited not by experience or theory, but by ideological hostility toward industrial policy, a view that is widely rejected today.

Professor Stiglitz raised following four reflections on industrial policy.

  1. 1)All countries do have industrial policy in reality. Support for expansion of derivatives, policies embedded in the Defense Department, policies that encouraged de-industrialization are the industrial policy in the U.S. Policies, expenditures, and taxes all shape the economy, and being conscious on every government action helps shape the economy to have a democratic discussion about where the economy and the society is going.
  2. 2)We need to keep in mind both social objectives (e.g. climate change, equality and resilience) and market failures (e.g. failures in capital market, widespread racial and gender discrimination in many societies). Borders do matter as well; while we have created a view after WWII that we should be working and striving for a borderless world, and economic policy was often based on the notion that we were about to achieve a borderless world, we began to realize that the borders do matter (e.g. actions by President Trump, vaccine nationalism, export restraints under food crisis).
  3. 3)An industrial policy program has to focus on addressing double or triple duties. With limited resources and instruments, policies need to address as many of social objectives as possible. It is pleasing that the Biden administration is pursuing industrial policy with discrimination inequality and green transition in consideration.
  4. 4)Opposition to industrial policy is not based on economics but is based more on political economy and imperfections of information. Governments have made a difference (e.g. US agricultural programs in the 19th century, DARPA and the Internet in the 20th century). The East Asian miracle showed unambiguously that the government industrial policies have made a difference. In terms of institutional design that reduces the likelihood of failures, one distinct source of success by East Asian economies is access to credit versus giving away free money. Transparency, peer review, engagement of civil society can mitigate political economy risks.

3. Latest Development on Industrial Policies (Presentations and Discussions)

Government officials to introduce development on industrial policies in respective countries, and distinguished scholars to share latest findings on industrial policy such as theoretical background and challenges, followed by discussions on each presentation.

Key issues for consideration on presentations from government officials:

  • Priorities, challenges, and backgrounds of the recent development on industrial policy

Issues to be explored through discussions:

  • Why the effectiveness of industrial policy has been “overlooked” until recently
  • Challenges which market cannot solely address such as climate change, income and spatial inequality, and supply chain resilience
  • Successful implementation of industrial policy
  • Concerns regarding today’s industrial policy

3-A. Session1: Presentations from Government Officials

Chair:
  • Dr. Dani RODRIK (President, IEA)
Speakers:
  • Mr. HIRAI Hirohide (METI, Japan)
  • Ms. Donna LEONG (BEIS, UK)
Discussants:
  • Dr. Gordon HANSON (Harvard University, US)
  • Dr. Charles SABEL (Columbia University, US)

Presentations from Speakers

Mr. Hirai elaborated the concept of METI’s New Direction of Industrial Policy and its rationale. Acknowledging that Japan has underinvested in the fields such as green technology, digital technology, and human capital in the last 30 years of Japan’s low economic growth, Mr. Hirai reiterated the idea of inducing private investment with a mission-oriented approach in targeted areas, and providing large-scale, long-term and well-planned governmental support.
He identified six pillars of mission-oriented industrial policies, with the 15 billion dollars Green Innovation Fund and support for Japan Advanced Semiconductor Manufacturing as examples of mission-oriented industrial policies already in place. Regarding socioeconomic system reform, Mr. Hirai also stressed the importance of addressing low and declining investment in human capital through promotion of human capital management, improving flexibility of labor market, and diversification of education from elementary school to doctoral levels.

In line with Mr. Hirai, Ms. Leong also linked the starting point of modern UK industrial policy to its relatively weak productivity performance. She presented her analysis that UK policies in last 20 years tended to have too much focus on horizontal policies and lacked sufficient consideration for (i) the effects of policy across various sector, (ii) stable policy institutions for sound micro-policy, and (iii) importance of place, and local productivity.
Ms. Leong emphasized the UK has been addressing the issues inherent in past industrial policy by (i) developing a framework to prioritize sectors, (ii) setting up long-term micro policy institutions such as the National Infrastructure Commission and the Productivity Institute, and (iii) improving social and institutional productivity and implementing place based policy decisions through the initiatives identified in the recently published Leveling Up Strategy.
She concluded her presentation by identifying challenges for the future (i.e. the uncertainty with regard to the ultimate shape of a post-Covid new normal, path dependency effects in advancing sector and place based policies, the unprecedented challenges of ‘big transitions’ such as automation, digital, and net zero).

Discussions

Following the presentations from speakers, Professor Hanson made following points on purposes and implementation of industrial policy.

  1. 1)Purposes of industrial policy: Given the externalities of capitalism, we need to be mindful against the belief that one policy fits all, and it is imperative to have separate, yet complementary, set of policies to achieve different objectives.
    -In pursuing decarbonization of the economy and reduction of economic disparities concurrently, there is a risk that industrial policy is highjacked to achieve the former without sufficient consideration about how to achieve the latter.
    -The combination design of the four components of US industrial policy (i.e. (i) workforce development, (ii) technical assistance to firms, (iii) tax incentives to firms, (iv) necessary infrastructure) will depend on the social objectives. Promotion of economic growth will favor STEM education/training, whereas decarbonization will require a much narrower part of STEM, and addressing economic disparities will call for vocational and technical training in non-fouryear college institutions.
  2. 2)Successful implementation of industrial policy: It is imperative to involve firms, workers, and civic actors early in the process. Studies on place-based policies in the US by Professor Rodrik and Professor Hanson reveals that the economic development practice in the U.S. has begun to resemble the practices in emerging economies, with the successful policies involving non-governmental actors, industry groups and local policymakers early in the process of problem identification and policy design. We also need to be mindful about local context in workforce development.

Professor Sabel echoed with previous speakers on the point that decades-long ideological disposition to industrial policy has disappeared. He characterized challenges for industrial policy today as follows.
-With many countries committing substantial degree of financial resources on industrial policies, the concern today is not impediment by ideology, but rather the information problem for authorities.
-A number of successful industrial policies in the past were designed and implemented by relatively small number of sophisticated firms, experts and academics in developed economies (e.g. DARPA). Today’s agenda (e.g. green transition, addressing accumulated poverty) will require (i) local actors to adapt to new types of policy planning, gaining new set of competence to work with national/regional actors, and (ii) workers to acquire new skills. This task is common to both advanced and developing economies (e.g. the UK Levelling Up Program). Local, place-based, participatory approaches will be required in implementing Next Generation EU as well.

On the relationship between policy objectives and policy instruments, Mr. Hirai mentioned that while each of the six missions under New Direction of Industrial Policy is a different socio-economic problem, he is seeing growth opportunities in addressing these problems given the underlying enormous demands. Mr. Hirai also touched upon the importance of a well-functioning multilateral trading system and dispute settlement system, along with the importance of seeking alternative mechanisms as necessary.

Ms. Leong acknowledged the tradeoffs among policy objectives, and noted innovation will provide one answer in escaping from tradeoffs. On the issue around local actors, she stressed that what is at the heart of the Levelling Up Strategy is the recognition the local economy relies significantly on local decision makers and it is a “rediscovery” of necessity to ensure that local decision makers take decisions that are appropriate for local areas.

As for the relationship between place-based and sector/industry-based policies, Ms. Leong described them as intertwined in the case of the UK’s levelling up policy. Mr. Hirai also elaborated that while Japan is not focusing on industry in conventional terms (e.g. heavy industry, light industry), Japan is seeking to prioritize new set of industry groups (e.g. green transformation industry, resilience-related industry).

3-B. Session2: Presentations from Academic Researchers

Chair:
  • Mr. WATANABE Tetsuya (RIETI, Japan)
Speakers:
  • Dr. Austan GOOLSBEE (The University of Chicago, US)
  • Dr. Josh LERNER (Harvard University, US)
Discussants:
  • Dr. TOMIURA Eiichi (Hitotsubashi University, Japan)
  • Dr. Philipp STEINBERG (BMWK, Germany)

Presentations from Speakers

Professor Goolsbee echoed the observations in previous discussions that all policies are industrial policies, regulatory captures and information-based failures were behind the negative views by economists on the first generation of industrial policy, and that the new generation of industrial policy needs to fix externalities and market failures. He made following arguments on the forms and challenges regarding today’s industrial policy.

  1. 1)New policies to fix externalities: Successful new policies often are seen in areas that are largely non-political. While the recent rise of industrial policy typically comes from various crisis (e.g. concerns on climate change leading to subsidies and regulations, supply chain crisis pushing for vaccine productions, national security concerns on semiconductors), the competitiveness concern behind them entails a risk of going back to old-fashioned industrial policy. Even if carbon intensity tariffs are intended to address global externalities, such tariffs could lead to other countries claiming for national security interests or tariffs based not on externality.
  2. 2)Short-run orientation arising from information failure: Early in the Covid-19 crisis was a huge attention to ventilator, but in a few months, the importance of having domestic manufacturing base for ventilator in dealing with the pandemic was denied. This resembles an information failure which haunted old industrial policy.
  3. 3)The degree of government intervention: A source of success behind the U.S. National Science Foundation, National Institute of Health, DARPA, and funding system for US universities is that policy designers did not design where the money should go. In a sense, immigration, attracting human capital, is one of the ultimate new types of industrial policy.
  4. 4)Further caution on old-fashioned industrial policy: The US industrial policy in last five years was about old-fashioned protection of industry rather than about externality (e.g. large assistance to airline, cruise and other politically connected industries). We should also be cautious about hijacking antitrust policies with the intention of blocking foreign goods.

Professor Lerner revealed his observations on entrepreneurship and stated that private entrepreneurial finance can help mitigate information and capture issues surrounding public programs.
-Professor Lerner noted the benefits of private entrepreneurial finance from the governments’ perspective, which include (i) mitigating informational asymmetries in selecting new ventures, (ii) venture capitalists’ ability to make “tough” decisions, and (iii) different types of compensations for private financiers and governments (e.g. financial returns, addressing externalities).
-Using data on 755 programs worldwide, Professor Lerner’s empirical study identified several factors behind involvement of private capital investors in public programs: (i) effective governments, (ii) programs targeting earlier-stage firms, and (iii) the presence of existing private venture activity. Public entrepreneurial finance programs tend to be associated with higher subsequent growth in innovation.

Discussions

Professor Tomiura provided his reflections on presentations from Professor Goolsbee and Professor Lerner regarding unintended policy consequences by citing the widened productivity gap that has resulted from the industry relocation program in Japan. He also echoed Professor Lerner’s argument on private-public collaboration by noting that cluster policy in Japan was successful only when local banks were involved as the main bank.
Professor Tomiura further highlighted the importance of forward-looking design of data collection in evaluating industrial policies, as the objectives and policy tools for the new industrial policy today are more complex than in the past while the big data accumulated in private sector and recent development on place based policies are providing opportunities in advancing policy evaluation. In this context, he revealed a recent initiative by RIETI on evidence based policy making (EBPM).

Dr. Steinberg noted that modern interpretation on market failures has contributed to the growing momentum of industrial policy in the last few years. His argument stretched the rationale for industrial policy from externalities to planetary boundaries that is motivating decarbonization efforts. He also cited positive spillover effects of promoting digital economy. Dr. Steinberg further made following arguments.
-Capital market failure and the importance of private sector engagement: The German economic stabilization fund is an example of German government’s effort for close collaboration with private investors.
-Path dependency: The Gaia-X project is trying to help businesses to collaborate to overcome path dependency.
-New rationales for industrial policy: Strategic autonomy and resilience are gaining attention as the new theoretical backbones of industrial policy, resulting in increased support for semiconductor industries and value chain relocations in Europe and the U.S. There is both a point and a danger to it.
-Policy scope and policy instruments: While Germany has rather narrow scope on its industrial policy, it is important to consider criteria and boundary of industrial policy, given the political nature of the issues which industrial policy is expected to address. The main challenge from a practitioner’s perspective is how to determine and calibrate right instruments.

Professor Lerner touched upon the establishment of Yozma in Israel as an example of successful involvement of private sector under which the government distance itself from entrepreneurs by providing funding to intermediary venture capital funds rather than directly to entrepreneurs, and matching validation to the public funding made possible by co-fundings from venture capitalist and government. Professor Lerner also argued for setting a broad set of parameters for entrepreneurship rather than trying to dictate where technologies should be going.

Professor Goolsbee echoed Professor Lerner’s argument on the importance of involving intermediaries; in response to Covid-19, the US policy was designed for swift provision of financial resources from Fed (Federal Reserve System) to small business by using banks as intermediaries, but this approach has faced conflicting objectives as banks tended to provide finance first to their most favored clients (e.g. high income entrepreneurs). All policy decisions are not free from politics, and this makes industry policy prone to the risks of going back to the old-fashioned policy. He also raised the issue of whether autonomy can be considered as a new rationale for industrial policy, if autonomy is fueled by competitiveness concerns and desires to do things at home.

Dr. Stenberg responded to Professor Goolsbee’s comment on the danger of repeating the oldfashioned industrial policy by stating that pure nationalization cannot be the solution today, and cited the Important Projects of Common European Interest as an example of policy coordination effort within the EU to mitigate the problem of subsidy race.

4. Actions for Achieving the Purpose of the Forum (Panel Discussion)

Chair:
  • Dr. Ufuk AKCIGIT (IEA)
Panelists:
  • Dr. Carol CORRADO (The Conference Board, US)
  • Dr. Chiara CRISCUOLO (OECD)
  • Dr. Piero GHEZZI (Former Minister of Production, Peru)
  • Dr. HAMAGUCHI Nobuaki (Kobe University, Japan)
  • Mr. HIRAI Hirohide (METI, Japan)

Participants from academia, government and international organization jointly discussed how to achieve the purpose of the forum: building a better industrial policy and promoting a better understanding of new thoughts on industrial policy.

Topics raised by the moderator:

  • What should be the new ways to think about industrial policy
  • How to make industrial policy more effective

Dr. Corrado explored the roles and features of intangible economy in today’s digitalized economy, noting the rise of proprietary data as a vital factor for competition in the intangible economy. She elaborated that even when digital products and services themselves are not a secret, the proprietary data that trains algorithms cannot be copied, and this characteristic has a significant impact on both productivity and competition among companies. She concluded that frameworks for new industrial policy need to incorporate this aspect of modern competition.

Dr. Criscuolo noted that objectives of industrial policy are changing in response to the new socioeconomic and geopolitical challenges, and industrial policy itself is also evolving in parallel. Referring to the OECD's recent work on the SDGs and industrial policy, she emphasized that industrial policy has an important role to play in bringing the best out of the private sector in achieving the new social objectives such as green transition, resilience of global value chains, and inclusive growth. She also stressed the importance of relying on strategies that combine industrial policy tools with the wider range of complementary instruments to address different market failures.

Dr. Ghezzi presented three important points for implementing effective industrial policies based on his experience with the Mesas Ejecutivas (Executive Roundtable), which he implemented to improve the productivity of the Peruvian economy when he was the Minister of Production in Peru. The first point he stated is that collaboration among the main public and private stakeholders is the key to improving productive diversification. He explained that coordination failures between the public and private sectors as well as within compartmentalized government entities of different levels and within the private sector are the major cause of low productivity. Since these failures tend to be very specific to the sector, industry, value chain and often territories, starting a process of public-private cooperation is key to identifying and solving the main problems affecting them.
As the second point, he explained that the Mesas Ejecutivas are neither purely top down nor bottoms up. They include actors on the ground that know the problems and potential solutions. But, occasionally, when ground actors can’t solve problems, these must be bumped up to highlevel authorities (with more capabilities and resources).
Finally, he emphasized that the focus on starting small and implementing policies quickly and learning during execution is crucial because political cycles are much shorter than that required for industrial policies to come to fruition. This has allowed the methodology to work at solving problems but also learn to solve them, generating significant capabilities both in the public and private sector and in their capacity to coordinate with one another.

Professor Hamaguchi shared his observation on the capital flight to risk-free assets in the last two decades in Japan due to uncertainties arising from structural changes (e.g. aging population, technological paradigm change toward carbon neutrality and digital transformation). He called for a holistic approach in industrial policy to reduce such uncertainties and to promote investment in more productive assets. As for practical implementations of such a holistic approach, he highlighted the importance of (i) competent executive institutions composed of staff with high analytical and communication skills, and (ii) a functioning feedback loop among stakeholders, including politics, to ensure transparency in operations.

Mr. Hirai stressed the importance of policy coordination and shared narratives among like-minded countries in achieving objectives such as carbon neutrality and secure supply chains, inducing large private investments, while avoiding conflicts of interests among countries in their pursuit of industrial policy.

Professor Akcigit pointed out the monotonous application of policy instruments among countries as a cause of policy ineffectiveness, and stressed the importance of countries using their own data to analyze and understand the particular problems of each country.

Dr. Ghezzi noted that country’s capabilities for industrial policy can be developed in the course of policy implementation by fostering trust between public and private sectors. On the other hand, he pointed out that since Mesas Ejecutivas is intensive in government capabilities – which are not abundant – and promoting stakeholder interaction at a nationwide level is difficult, policymakers often seek for easier alternatives such as tax reduction and protection of specific industries. In addition, he reiterated that building strong industrial policies, and more so solving long term or structural problems, require perseverance and policy continuation through administrations over time.

Dr. Criscuolo pointed out the dilemma inherent in designing both politically acceptable and socially effective policies; politicians may seek to create efficient policies that satisfy everyone while policy implementation inevitably creates winners and losers, and this conflict results in policy exceptions that make policies ineffective. She also stated that it is important to embed downscaling and exit strategies into policy evaluation cycles because the tendency of politicians illustrated above makes it difficult to abandon existing policies, ending up with a myriad of similar policies.

Answering a question from Professor Akcigit on whether governments have adequate human resources and dynamism to fully comprehend and formulate appropriate policies in response to the impact of digital economy on employment and competition, Dr. Corrado enumerated two points.

  1. 1)The lack of a proper framework for digital business cause misunderstanding of cost and rent structure of digital economies, leading to misguided discussions.
  2. 2)The rise of proprietary data is hindering competition.
    She stated that policies to promote data sharing among companies may help address these issues, citing practice of credit scoring companies as an example, and that we should understand how data is being used in each industry to address the problem of entry barriers.

Mr. Hirai presented several reflections from his work on the New Direction of Economic and Industrial Policies; (i) the importance of international cooperation, as the free flow of data, goods, and services is becoming increasingly difficult amid the changing geopolitical situation, and (ii) need for talented human resources and functional institutions in keeping up with rapid technological change.

Professor Hamaguchi pointed to two types of abilities that are essential for government agencies in charge of industrial policy; (i) the ability to incorporate social issues (e.g. employment, equality) into specific policy programs, and (ii) communication capability. He also noted the value of technology and knowledge transfers through international cooperation.

Finally, Professor Akcigit reiterated that a myriad of similar policies distorts policy outcomes, and that there is a lack of proper evaluation of the effectiveness of subsidies and other programs. Dr. Ghezzi commented that the lack of coordination between key stakeholders rather than failing to provide the right subsidy is the real and deep-rooted problem. Following this, Dr. Criscuolo emphasized that an ecosystem in which universities, producers, customers, and competitors play their respective roles and interact with each other is key to the new industrial policy.

5. Closing Remark

Speakers:
  • Dr. YANO Makoto (Chairman, RIETI, Japan)

Chairman Yano wrapped up the conference by emphasizing that the new industrial policy is indispensable to tackle major issues we are facing today, and thanking the participants for their valuable insights and discussions which will serve as the basis for future work on industrial policy.