Industrial Development Think Tank: Policy Brief 12[1]
Antonio Andreoni and Fiona Tregenna
Industrial policymaking is a complex process as it entails the management of multiple interactive measures and instruments (Andreoni, 2016). In his account of the lessons learned from East Asia, Stiglitz (1996) emphasises how the successful industrialisation of these countries can only be understood by analysing their “packages of interactive measures” whereby companies were exposed to different types of internal and external competitive pressures.
This policy option is also stressed by Chang (2011, 100) when he writes, “In East Asia, free trade, export promotion (which is, of course, not free trade), and infant industry protection were organically integrated, both in cross-section terms (so there always will be some industries subject to each category of policy, sometimes more than one at the same time) and over time (so, the same industry may be subject to more than one of the three over time).”
Industrial policy makers in middle-income countries like South Africa needs to focus on three set of issues:
selecting appropriate policy instruments targeting specific production, technological and organisational challenges;
integrating and coordinating these instruments in coherent industrial policy packages;
addressing the governance challenges that middle-income countries will face in managing these policy instruments.
Industrial policy instruments: selection
Table 1 provides a list of industrial policy instruments used by several middle-income countries. They are here organised around five key policy areas:
Production, technological and organisational capabilities building
Innovation and technological change
GVCs integration, Local Production System (LPS) development and industrial restructuring
Demand and trade
Industrial finance
These five policy areas correspond to the critical industrialisation challenges faced by the majority of middle-income countries. In some cases, like South Africa, these middle-income countries have been also showing signs of premature deindustrialisation (Andreoni and Tregenna, 2018).
Industrial policy matrix: integration and coordination
The identification of a mix of policy instruments is only the first step. Indeed, these instruments must be aligned, coordinated and synchronised over time. Policy matrixes are useful tools in addressing the challenges industrial policy makers face in aligning and coordinating packages of interactive measures across different policy targets and areas.
First, a policy matrix allow for mapping out the different policy instruments a given government is implementing. In doing so, policy matrices provide a good basis for reflecting upon the degree of targeting of each policy instrument and how it is governed. Secondly, the policy matrix helps in identifying the potential interactions linking the different policy instruments adopted by different MDAs and implemented across different policy areas. Finally, by revealing the presence of policy interactions within the overall policy package, through the policy matrix it is possible to identify potential policy misalignment or trade-offs which would remain unnoticed otherwise.
Table 1: An industrial policy toolbox for middle-income countries
Source: Andreoni and Tregenna, 2018
These misalignments might also be related to lack of coordination or duplication among MDAs, as well as the fact that the instruments adopted by one government are not synchronised with those left by previous governments.
First, a policy matrix allow for mapping out the different policy instruments a given government is implementing. In doing so, policy matrices provide a good basis for reflecting upon the degree of targeting of each policy instrument and how it is governed. Secondly, the policy matrix helps in identifying the potential interactions linking the different policy instruments adopted by different MDAs and implemented across different policy areas. Finally, by revealing the presence of policy interactions within the overall policy package, through the policy matrix it is possible to identify potential policy misalignment or trade-offs which would remain unnoticed otherwise. These misalignments might also be related to lack of coordination or duplication among MDAs, as well as the fact that the instruments adopted by one government are not synchronised with those left by previous governments.
Figure 1 provides an example of these types of policy matrixes (Andreoni, 2016) and shows the importance of integrating and coordinating different set of policy instruments (as detailed in Table 1). The industrial policy matrix framework takes the following three dimensions into consideration:
(i) Industrial policy governance model.
This axes defined the extent to which policies are implemented by MDAs at the regional/state level or at the national/federal levels. Of course, as in the case of regional agreements such as the East Africa Community, there is also a supranational level of policy-making.
(ii) Industrial policy targets and areas.
Each industrial policy instrument targets a specific set of goals, which can be clustered in various policy areas. For example R&D credit, standardisation policy and public technology intermediaries are all instruments/measures/ institutions targeting the ‘Innovation and Technology Infrastructure’ policy area. While industrial policy generally relies on supply side instruments, there are also clusters of policy areas including demand-side type of instruments such as procurement policy and external market development policies.
(iii) Industrial policy levels of intervention.
Each industrial policy instrument can be more or less selective. Some policy instruments are tailored as sector specific measures, and can also target specific firms within those sectors (SMEs in the food supply chain for example). There are then policies that are focused on the manufacturing system as a whole, in particular those targeting export promotion or development of technology platforms that are critical for manufacturing development (e.g. capabilities in machine tools). Some industrial policy instruments can be more openly focusing on cross-sectoral targets in the industrial system, for example, those targeting better integration between agriculture and manufacturing industries.
Figure 1: Policy matrix for industrial policy package analysis
Source: Andreoni, 2016
Finally, there are policies which are more macroeconomic in nature, such as interest rate and exchange rate policies. Despite the fact that they will affect the overall economy, this does not mean that they will affect all sectors of the economy in the same way. A certain interest rate policy will affect sectors with different degrees of capital intensity differently.
Countries can adopt different packages of industrial policy measures and can coordinate different policy instruments, to either have a combined effect on the same target or to manage potential trade-offs among different goals. For example, education policies can be aligned to labour market reforms to improve workers’ conditions. Technology policies can also be aligned to trade policies or public procurement measures supporting domestic industrial sectors development. Potential trade-offs arising between economic growth and increasing pressure on natural resources can be counterbalanced with aligning sectoral policies and technology policy, in particular green technologies development and deployment over time.
Effectiveness of industrial policies and governance challenges
The effectiveness of a single policy measure depends on its linkages with other policy measures acting upon the same companies, sectors and specific institutions. This implies that the policy effectiveness of a certain instrument might be improved by both/either using the instruments more effectively and/or by changing or introducing other complementary instruments. The combined effect of different policy instruments tends to be different from the one that the government can achieve by the independent implementation in time of the same policy measures.
The governance of industrial policy as packages of interactive measures present several challenges that are particularly severe in middle-income countries like South Africa. Industrial policy governance challenges are due to multiple factors and dynamics that operate and unfold in three areas: (i) the public sector, (ii) the interaction between public MDAs and private sector constituencies, and (iii) the private sector among different powerful groups and interests.
These challenges often result in policy instruments misalignments, duplication of mandates, institutions and misallocation of scarce resources, ultimately lack of coordination among different MDAs and ineffective implementation and enforcement of policies. Moreover, these governance challenges within the public sector are often intertwined with political economy factors and dynamics involving both the public and private sectors.
Understanding the roots of these governance challenges and political economy dynamics is a crucial step in identifying effective strategies for their mitigation and better coordination in implementation and policy enforcement.
Learning to industrialise
By focusing on the experience of three major middle-income countries – Brazil, China and Malaysia – a number of industrial policy packages can be identified. These cases also points to a number of policy recommendations for South Africa.
Intermediate institutions: The Embrapa case study in Brazil
Over the last thirty years, Brazil has been among the most active countries in terms of its use of policies designed to expand natural resource-processing industries and food production. The results of these transformative policies are reflected in the remarkable results Brazil has achieved in manufacturing its agrarian change. At the centre of the transformative policy package implemented in Brazil is a network of intermediate institutes – e.g. Embrapa, which have fostered technological change, diversification and upgrading in agriculture and farming.
Embrapa was founded in 1972 as a response to the main weaknesses of DNPEA (the National Agricultural Research and Experiment Department). During its first two decades, Embrapa created a network of national commodity centres and regional centres that focused on major cropping and animal production systems, as well as on eco-regional and national themes. It also increased its internal capabilities by signing partnerships with American universities such as Purdue and Wisconsin, which allowed Embrapa’s staff to receive postgraduate training. In 1993, the establishment of the Embrapa Planning System (SEP) for the first time introduced a systems approach to R&D planning. This allowed a redefinition and reintegration of the centre’s mission, objectives, programmes, human resources, infrastructural needs and priorities.
From the late 1980s, agricultural research became even more cross-pollinated by research in advanced manufacturing. A good example of this is the satellite monitoring services for the acquisition and processing of remote sensor images and field data. The Satellite Monitoring Centre was created in 1989 in an area of 20 000 m2 in Campinas (Sao Paulo state), assigned by the Brazilian Army to Embrapa for the development of a special unit focused on territorial management systems and electronic networks for modern agriculture.
Throughout the 1990s, Embrapa was involved in a wide range of activities related to agricultural research and technology including plant breeding, pest management, food safety, satellite monitoring, sustainable agricultural development, and hunger relief. Soybean breeding and pest management activities are headquartered at the Embrapa facility in Londrina in the state of Paraná, but crop research activities are carried out at locations around the country to develop crops and varieties that are suited for local conditions.
The trend started in the 1990s continued during the next decade, in particular in 2005 and 2006, when Embrapa made a serious effort to improve and renovate its infrastructure. Included among these investments, at the interface between agriculture, biotechnologies and advanced manufacturing, were:
Facilities for quality improvement in the meat production chain;
An aquaculture laboratory prioritising water quality control, fish feeding and health;
A new Oenology Laboratory to boost wine production in the North-eastern Semi-Arid Region;
The construction of one of the world’s first National Agribusiness Nanotechnology Laboratories focused on the development of sensors and biosensors for food quality control, certification and traceability. The Laboratory was also dedicated to the synthesis of new materials, such as polymers and nanostructured materials or thin films and surface-to-manufacture smart packages.
The building on six new walk-in freezers to increase the storage and preservation capacity of the Embrapa Germplasm Bank (from 120 to 240 thousand seeds).
Probably the most recent remarkable achievement of Embrapa has been the claiming of the Cerrado (the Brazilian savannah) for modern agriculture. Today, Embrapa includes 47 research centres (15 National ‘Thematic’ Centers, 16 National ‘Commodity’ Centers and 16 Regional ‘Resource’ Centers) throughout the country. It hosts 9 284 employees and has an annual budget of over US$1 billion. Similar to Fraunhofers in Germany, Embrapa plays a critical intermediary role between agricultural and manufacturing R&D, education, markets, and in-farm agricultural production. It also bridges and transfers knowledge, technical solutions and innovations across different sectors, thus facilitating various forms of inter-sectoral learning.
Technology and R&D financing policies: The Innofund case study from China
Since the 1980s, China has adopted several technology and R&D financing policies. In 1986, the National High Tech Development Plan (also known as the 863 Plan) introduced the first articulated national technology strategy targeting biotechnology, space, information technology, laser technology, automation, energy, and new materials.
The Innofund was set up in 1999 as a special government R&D programme. It aims to “facilitate and encourage the innovation activities of small and medium-technology based enterprises (SMTEs) and commercialization of research by way of financing, trying to bring along and attract outside financing for corporate R&D investment of SMTEs” (Innofund institutional website). The Innofund has precisely defined eligibility criteria and provides different types of financial support targeting companies at different stages of development, from loan interest subsidies to equity investment.
To be eligible companies must meet the following criteria: i) the project should comply with national industrial technology policies; ii) it should exhibit high potential for social and economic impact; iii) it should be competitive in the market. Companies interested in applying for financing should be a business corporation with no more than 500 employees, 30% of which must have received higher education. Moreover, the company must also show to have an annual R&D investment of at least 3% of total sales, and the number of employees dedicated to R&D should be at least equal to 10%. If the company is operating in an industrial sector with high economies of scale potential, it must also exhibit good economic performances.
Among qualifying companies, the following projects are given priority:
Projects with advanced technology content or independent intellectual property rights and high value added
Projects established by researchers or oversees returnees interested in commercialising their scientific achievements;
Innovation projects jointly initiated by firms, universities and other research institutions;
Projects that utilise new and advanced technologies to drive industrial restructuring in traditional sectors (especially stock assets revive) and jobs creation.
Companies can apply for three different forms of financing. This is a key feature of this programme as it allows for targeted industrial financing acknowledging different company financial needs. Innofund offers several forms of financing:
Appropriation
Partial subsidies
Interest-free loans
Equity investments
The governance structure of such articulated programme relies on two levels of governance agencies. The Innofund Administration Center (IAC) is the central authority under the Ministry of Science and Technology (MOST) responsible for: i) the identification of the preferred application technology fields and industries and develop the application guidelines; ii) the ex-ante screening and evaluation of the applications; iii) releasing the contract with the selected firms; iv) post-investment project assessment. At the local level, Innofund offices are distributed in each province and operates under the Provincial Science and Technology Committee. After 2005 the province offices received much more power from the centre and were put in charge of performing many of the functions originally performed by the central IAC.
Between 1999 and 2011, Innofund provided more than 19 billion RMB to more than 30,000 projects. A large majority of those were supported through appropriation (around 27,000) while all the others received several types of financing support, subsidies credit, bank loan insurance and equity investments. According to official reports, Innofund has a 1:11 multiplier effect in terms of inducing external financing and has succeeded in the incubation of today’s world leading companies, including the ICT giant Huawei. After only ten years of operation, 82 of the 273 companies publicly listed in the China’s SME Stock Exchange were former Innofund supported companies.
The programme has also had a significant impact on job creation, with estimated more than half a million new jobs and 3.4 billion RMB in export. More recently, in 2007, the Ministries of Sciences and Finance supplemented this scheme with the Venture Guiding Fund, a program aimed at investing directly in VC funds, with co-equity investment, grants and risk subsidies compensations.
Diversification of the local production system: the case of palm oil in Malaysia
Among the early industries picked by Malaysia’s government in the 1960s, the palm oil industry represented an example of an initially risky industry involving a non-native crop that eventually grew to achieve phenomenal international success. Malaysia accounts today for roughly half of the world’s production of palm oil.
A number of selective interventions were undertaken to grow internal capabilities and expanding the coverage of oil palm trees.
First, to encourage landowners of rubber to switch to palm oil production, replanting grants were offered to finance the replanting of old rubber trees with oil palm from 1962. Such grants greatly lessened the financial burden on smallholders and estates arising from switching crops. Second, foreign-owned oil palm estates were acquired and developed by the state. Such initiatives dramatically altered the composition of Malaysia’s output.
The government also provided strong institutional support and facilitated coordination between the public and private sectors by setting up the Palm Oil Registration and Licensing Authority (PORLA), the Palm Oil Research Institute of Malaysia (PORIM), and the Malaysian Palm Oil Promotion Council (MPOPC). PORLA, PORIM and MPOPC were responsible for regulation and licensing, specialised training and public sector R&D, and export promotion respectively. In 2000, to harness synergies between related functions, PORLA and PORIM were merged to form the Malaysian Palm Oil Board (MPOB).
Not merely content with the direct economic contributions of palm oil, the government actively sought to develop targeted downstream industries, such as the palm oil processing, oleochemicals, biotechnology, biodiesel and biomass industries (see Malaysia’s industrial master plans – MITI 1986, 1996 and 2006). Defying earlier sentiment that Malaysia lacked a comparative advantage in palm oil processing, the government undertook a slew of targeted measures, including fiscal incentives and price distortions, to grow downstream capabilities.
First, fiscal incentives were used to attract investments in strategic areas related to palm oil. Under the 1968 Investment Incentives Act, qualifying oil palm firms enjoyed two years of (renewable) corporate tax exemptions, and eight years of excess profit and development tax exemptions. Pioneer status awards, offered before 1974, granted palm oil refineries tax exemptions for seven years. Tax exemptions were also offered on the basis of export performance and capital investments.
Second, higher duties on crude palm oil exports and tax exemptions on processed palm oil exports greatly skewed producers’ incentives towards the latter. Following from this, the 1970s were characterised by significant price distortions in palm oil exports.
Third, bleaching earth, a key ingredient in the palm oil-processing industry, was initially subjected to tariffs and import quotas until internal production capabilities were built. However, to cap costs during the import-substitution phase, subsidies were provided such that the price of bleaching earth purchased by domestic industries was similar to world prices.
Fourth, various policies dovetailed to promote the development of downstream industries. For instance, the 2006 Malaysian Biofuel Policy aimed to facilitate the gradual substitution of diesel fuel with palm oil. To promote the development of the palm oil industry, international restrictions on industrial policy were also occasionally circumvented. For instance, the Palm Oil Credit and Payment Arrangement (POCPA) scheme was introduced in 1992 to provide a two-year credit facility for countries purchasing palm oil from Malaysia.
Lessons for South Africa
Over the last years, South Africa has relied on several ambitious rounds of industrial policy, including sectoral targeting as in the case of the automotive industry. The international industrial policy experiences of countries like Brazil, China and Malaysia offer insights into the challenges associated with the implementation of different industrial policy tools and their coordination.
The Brazilian case highlights the importance of promoting the development of public technology intermediaries supporting the absorption, adaptation and diffusion of technologies. The Embrapa case study is particularly important for middle-income countries, as it suggests opportunities for technological development and value addition at the agricultural-manufacturing sectoral interface.
The historical experience of China stresses inter alia the importance of long-term commitment and a staged approach to industrial transformation, especially with respect to technological advancement in the manufacturing sector. The Innofund is presented as a successful case of a funding scheme for technology policy. This specific case also shows the importance of supporting industrial policy implementation with responsive multi-governance institutions.The Malaysia case study points to the importance of a diversification strategy. Product and sectoral diversification is a key driver of sustained industrialisation and the development of a well-integrated domestic production system. The Malaysian experiences in the electronics industry and in palm oil emphasise the importance of developing production and technological linkages in the domestic economy, while integrating into global production networks.
The lessons learned from the country cases reviewed in the paper point to three important policy recommendations.
First, there are significant opportunities for value addition and technological development in agro-business value chains. The industrialisation of agriculture opens new venues for increases in productivity, upgrading in global markets and diversification. However, in order to capture these opportunities, public technology intermediaries must provide key technology and product services to reach product quality standards, and transform agricultural activities into highly productive industrial processes.
Second, the promotion of technological upgrading in manufacturing industries cannot be done simply by jumping to frontier technologies – i.e. the so-called fourth industrial revolution. The China case study, in particular, emphasises how the building of a solid productive and technological capability foundation over several years is a precondition for innovation across several industrial fields. The other lesson is that institutions promoting technological innovation will have to change over time to respond to the changing nature of the innovation challenges firms face in the fast-changing global landscape.
A third policy lesson is that, to increase value addition and value capturing in the domestic economy, countries need to target the development of their local production system. The building of integrated supply chains in the domestic economy gives countries sustained industrial and productivity growth. Therefore, industrial policy focusing on the attraction of a multinational or the setting up of an export-promotion zone must be fully coordinated with other policy measures that induce the real industrialisation process.
Reversing premature deindustrialisation in South Africa will not result from the use of one of these instruments. Instead, it will depend on the coordination of a feasible set of interventions reinforcing each other in a coordinated manner.
References
Andreoni, Antonio. 2016. “Varieties of Industrial Policy: Models, Packages and Transformation Cycles.” In Efficiency, Finance and Varieties of Industrial Policy, edited by Akbar Noman and Joseph Stiglitz, 245-305.New York: Columbia University Press.
Andreoni, Antonio and Fiona Tregenna. 2018. “Stuck in the Middle: Premature deindustrialisation and industrial policy”. IDTT Working paper. Johannesburg: CCRED and South African Research Chair in Industrial Development.
Chang, H.-J. (2011). Industrial Policy: Can We Go Beyond an Unproductive Confrontation? In J. Y. Lin, & B. Pleskovic (Eds.), Annual World Bank Conference on Development Economics 2010, Global: Lessons from East Asia and the Global Financial Crisis (pp. 83-109). Washington, DC: World Bank.
Stiglitz, J. E. (1996). Some Lessons from the East Asian Miracle. The World Bank ResearchObserver, 11(2), 151-177.
[1] The Industrial Development Think Tank at UJ is housed in the Centre for Competition, Regulation and Economic Development, in conjunction with the SARChI Chair in Industrial Development, and supported by the DTI which is gratefully acknowledged. This paper reflects the views of the authors alone and not of the DTI or any other party.