«© 2013 The International Institute for Sustainable Development © 2013 The International Institute for Sustainable Development ...»
Developing countries often seek to attract FDI in order to facilitate technology transfer to domestic firms operating in the same sector, hoping to encourage the development of indigenous firms. Foreign investors usually lack incentives to actively engage in technology transfer with their own domestic competitors. And unlike China and Korea earlier, IISD REPORT JUNE 2013 2013 The International Institute for Sustainable Development © Industrial Policy for a Green Economy many developing countries today may not have enough leverage to force such transfers (through copying designs, engaging in reverse engineering, etc.) without risking a massive exodus of foreign firms (Altenburg, 2009). Hence, incentives and tax breaks to multinational investors could end up crowding out domestic firms and transferring rents from domestic taxpayers to foreign investors (Pack & Saggi, 2006).
On the other hand, there is ample empirical backing for the conjecture that FDI might be beneficial to domestic firms through vertical spillovers, that is, positive externalities stemming from the relationship of foreign enterprises with domestic suppliers and customers (Harrison & Rodríguez-Clare, 2009). Evidence on forward spillovers—the supply of inputs embodying new technologies or processes—is scant. However, the existence of backward spillover effects— productivity gains in foreign firms’ domestic suppliers—has been confirmed in various studies for a number of countries (see, for example, Gorodnichenko, Svejnar & Terrell, 2007, and Javorcik, 2004).
These findings rationalize industrial policy that sets local content requirements, requiring foreign firms to source a certain percentage of their inputs from local suppliers. Sutton (2004) explains the mechanism by which local content
requirements have helped suppliers in India and China:
“From the early 90s onwards, a wave of multinational firms entered both markets. In both countries, these entrants were required to achieve a high level of domestic content within a specified period (typically, 70 percent within 3 years). For at least some of the new entrants, this was seen as an unreasonable target, as domestic suppliers could not meet the price and quality requirements of the carmakers. Achieving the 70 percent target required the car makers to switch rapidly from a reliance on imported components to sourcing from local vendors; and this in turn gave the car makers a strong incentive to work closely with (first-tier) suppliers, to ensure that quality standards were met, within an acceptable price.” (p. 1) But while the imposition of such criteria might actually have positive benefits, it may as well scare off investors in the first place. Such concerns could be avoided, for example through subsidizing domestic inputs instead (Harrison & Rodríguez-Clare, 2009).
Notwithstanding these findings, many developing countries have not been able to reap such gains from FDI, as FDI has often been directed to extractive natural resource sectors that have only few linkages with the rest of the economy (Belloc & Di Maio, 2012). In general, FDI policies seem to have been the most successful when they were part of a broader industrial policy strategy, including efforts at improving the supply of skilled workers in targeted industries, improving regulation and infrastructure and export promotion (Harrison & Rodríguez-Clare, 2009).
Industrial Policy Targets We have up to now reviewed and assessed a number of more-or-less successful industrial policies in the hopes of giving a broad overview of instruments used, and we have stressed a few factors that do or do not contribute to their success. However, on a more general level, we have not really discussed what industries countries’ industrial policies should target. Keeping in mind the goal of achieving structural transformation of the economy to achieve long-term growth, have any lessons been learned on how radical industrial promotion away from the current structure really can be? For example, having noticed that the automobile sector appears to have many backward linkages with various suppliers, does this mean that all countries should start developing automobile industries? Obviously, countries have differential capacities regarding what they are good at producing, which is exactly why we see so much international IISD REPORT JUNE 2013 2013 The International Institute for Sustainable Development © Industrial Policy for a Green Economy trade. While a thorough discussion that would do justice to the complexity of this topic is beyond the scope of this paper, we will briefly present the main ideas around this debate, building on the notion of comparative advantage as discussed in the introduction to Chapter 2.
The term comparative advantage as used in economic theory is essentially an analytical device. Comparative advantage arises from between-country productivity differences, which are not directly observable (Costinot, 2009). Moreover, the sources of productivity differences vary, and they include a disparate set of factors such as human and physical capital, natural resources, institutional quality, capital-to-labour ratios, financial market development, and labour market characteristics (Kowalski, 2011). Outcome-based measures of comparative advantage such as Balassa’s (1965) index of revealed comparative advantage based on trade shares suffer from various real-world market distortions and thus have to be viewed with caution. Nevertheless, comparative advantage remains probably the most powerful benchmark in assessing trade flows and hence global production patterns.
This being said, industrial policy faces a certain dilemma: on the one hand, moving away from current comparative advantage is the very raison d’être of industrial policy. On the other hand, one would expect the success of a country’s industrial policy to diminish the further it strays from its comparative advantage. This view seems to have recently been adopted by the World Bank, and it is worth pausing a moment to have a closer look at what motivates this finding. An earlier study by former World Bank Chief Economist Justin Yifu Lin and a colleague (Lin & Liu, 2004) serves as the basis for his recent book New Structural Economics: A Framework for Rethinking Development and Policy (Lin, 2012). The book makes a strong case for government involvement in the industrialization process, but advocates for industrial policy to be consistent with a country’s comparative advantage as determined by its current endowment structure. Lin views economic growth as a constant process of technological upgrading, but emphasizes that such upgrading—which can be facilitated by the state—needs to be consistent with comparative advantage. While empirical studies generally struggle with finding indicators that appropriately measure and summarize countries’ often disparate sets of industrial policies, Lin and Liu (2004) find an elegant—albeit certainly imperfect—way to circumvent such measurement issues.
They construct the following technology choice indicator (TCI), where subscripts i and t indicate that each observation
is per country i and per year t:
share of manufacturing in GDPi,t TCIi,t = share of labour in manufacturingi,t The lower a country’s comparative advantage in capital-intensive manufacturing activities, the more support domestic firms will need in order to be viable. Hence, the numerator will increase with government support to manufacturing firms. An increase in the share of manufacturing alone would not in itself be an indicator of industrial policy, but could simply indicate a structural transformation of a country regardless of whether the government has intervened or not. The authors claim that what makes the TCI an appropriate indicator for intensity of industrial policy that defies comparative advantage is the presence of the denominator: directed investments in priority manufacturing sectors are expected to be more capital intensive than would otherwise be the case and hence absorb less labour than would occur if the labour absorption were a result of purely market forces. That is, the pace of hiring in the sector will not keep up with the pace of capital injections as supported by governments. Hence, a country that actively pursues a comparative advantage–defying industrial policy strategy will have a relatively higher TCI, resulting from an increasing numerator and a relatively decreasing denominator.
IISD REPORT JUNE 2013 2013 The International Institute for Sustainable Development © Industrial Policy for a Green Economy Using the TCI index in various regressions, Lin and Liu (2004) go on to test several hypotheses and find that countries with high TCI levels generally have slow long-term economic growth, unequal income distributions and significant market distortions as well as bad economic and political institutions. Therefore, the central theme of Lin (2012) and recent World Bank research more generally is not to reject industrial policy per se, but to caution countries not to stray too far away from where their comparative advantage lies. A useful guide for policy-makers to identify where exactly their country’s comparative advantage lies is provided in the same volume. The “Framework for Growth Identification and Facilitation” basically suggests determining which of a country’s industries are worth promoting by drawing up a list of tradable goods that have been produced over the last 20 years in countries with similar endowment structures and twice the country’s own per capita GDP.
Nevertheless, a number of scholars have criticized this approach to industrial policy, pointing out that the very purpose of industrial policy is to move away from a country’s current comparative advantage and calling for more radical industrial policy (see, for example, Chang, 2002; Mattoo & Subramanian, 2009; or Singh, 2011). Justin Lin’s focus on the importance of countries’ endowment structures assumes that countries with similar endowment structures will have similar technology-industry structures, implying similar levels of GDP per capita. Singh (2011) points out that many Asian countries successfully moved into industries in which they clearly did not have a comparative advantage at that time. For example, Japan and then Korea started producing steel when their per capita income was only 2.5 per cent that of the United States.
A wider strand of industrial policy literature relates countries’ export or production baskets to their levels of GDP.
Hausmann et al. (2007) developed a measure of sophistication of a country’s exports, evaluating each good according to the GDP per capita associated with each exporter of that good worldwide. The authors then calculated, using the export mixes of different countries, an implied level of GDP and compared it with the actual GDP per capita. They found that a number of countries have export baskets that substantially overpredict their actual GDP per capita. Using this technique in a case study on China, Rodrik (2006) identifies a large gap between implied and actual GDP and ascribes it to dirigiste government industrial policy that is directing the economy toward higher export sophistication, as the gap would be too large to have occurred naturally.
A third recent strand of literature might offer an alternative to the traditional comparative advantage–centered approach. Hidalgo, Klinger, Barabási and Hausmann (2007) introduce the idea of the global product space, which illustrates the extent to which certain products are connected to other products. As products are interconnected, countries that produce more “complex” products with more linkages tend to have more possibilities to also produce “nearby” products, allowing them to move up the industrial structure with greater ease. Calculating the proximity between products based on the likelihoods of various product pairs actually being produced as evidenced by trade data has the advantage of not having to rely on a priori assessments of countries’ business environments and product relatedness. According to Hidalgo et al. (2007), their measure of proximity is “based on the idea that if two goods are related, because they require similar institutions, infrastructure, physical factors, technology, or some combination thereof, then they will tend to be produced in tandem, whereas highly dissimilar goods are less likely to be produced together” (p. 2).
IISD REPORT JUNE 2013 2013 The International Institute for Sustainable Development © Industrial Policy for a Green Economy
FIGURE 6. THE GLOBAL PRODUCT SPACEReprinted with permission from Hausmann et al. (2011).
Hidalgo et al. (2007) find evidence that some countries have indeed hit a dead end, specializing in sectors that are relatively secluded from other products in the global product space. Other, more connected countries have historically been able to gradually upgrade their economies. Their research has enormous implications for industrial policy, in that it might help to identify promising sectors with higher growth prospects from government intervention, for example through overcoming coordination problems. Such strategic government policy may help countries accelerate their structural upgrading, or even jump to higher production structures.
IISD REPORT JUNE 2013 2013 The International Institute for Sustainable Development © Industrial Policy for a Green Economy A more comprehensive overview of the latest research on determining promising targets of industrial policy can be found in Reis and Farole (2012).
Best Practice The literature reviewed so far suggests the following best practices for optimizing the benefits of industrial policy.
Industrial policies should be only a subset of a broader industrial agenda. The successful deployment of industrial policies requires a sound enabling policy framework that is conducive to improving the business climate and economic competitiveness more generally. A number of complementary policies are needed to overcome potential bottlenecks.
These include education policies to increase the supply of relevant skills within the domestic workforce, R&D support and provision of adequate infrastructure. A sound macroeconomic environment, including a competitive exchange rate regime, has also been found to be an important driver for industry growth. Broader industrial agendas also help to lock in commitments in that they create incentives for consistent policies over successive administrations with sometimes conflicting policy objectives (Devlin & Moguillansky, 2012).