(www.oecd.org/sti) develops evidence-based policy advice on the contribution of science, technology and industry to wellbeing and economic growth. STI Policy Papers cover a broad range of topics, including industry and globalisation, innovation and entrepreneurship, scientific R&D and emerging technologies. These reports are officially declassified by an OECD Committee.
ABSTRACTThis paper reviews the evidence on emerging thinking and new trends in the sphere of industrial policy. The paper adopts a broad and inclusive definition of industrial policy, and proposes a new typology based on the orientation of policy and the policy domain. Looking at a typology according to the policy domain, the paper proposes a framework based on growth accounting, which parallels the evolution of thinking about the rationale for industrial policy interventions, which has moved from a traditional approach based largely on product market interventions (production subsidies, state ownership, tariff protection), through market failure-correcting taxes and subsidies operating mainly on factor markets (R&D incentives, training subsidies, investment allowances, help with access to finance) to a focus on interventions that help build systems, create networks, develop institutions and align strategic priorities.Recent developments in both the theory and practice of industrial policy suggest that it is possible to find a theoretical rationale for a government role. However, there are important practical difficulties, not least the risk of government failure, the risk of rent-seeking behaviour and the potential use of industrial policies for protectionist goals. Unless these problems are addressed, traditional approaches to industrial policy seem destined to fail, and the historical record has not been kind. 'Horizontal' measures are still the preferred way forward but there may be some aspects of policy where strategic choices need to be made. The risks associated with selective-strategic industrial policy can be minimised through a 'soft' form of industrial policy, based on a more facilitative, co-ordinating role for government, consistent with the systems approach described in this paper. While not immune to the dangers of government failure, such an approach, if carefully designed and implemented, has a much higher chance of success than the costly and distortionary selective-defensive industrial policy interventions of the past. One clear message to emerge from the paper is the need for much better monitoring and evaluation of industrial policy initiatives.