© The Author(s) 2018
Murat A. YülekHow Nations Succeed: Manufacturing, Trade, Industrial Policy, and Economic Developmenthttps://doi.org/10.1007/978-981-13-0568-9_12

12. The ‘How ’ of Manufacturing: Industrial Policy

Murat A. Yülek1  
(1)
Istanbul Commerce University, Istanbul, Turkey
 
 
Murat A. Yülek

This chapter discusses why and what kind of industrial policy is needed at what stages along the industrialization process? It also discusses the sequencing between industrial policy on the one hand and science, technology, and innovation (STI ) policies on the other. Should they be used concurrently? Can STI policies be effective in a less industrialized country? Finally, the chapter proposes a methodology on how strategic sectors can be selected by policy makers. Do all manufacturing subsectors have the same developmental impact? Or, are some subsectors different from others? If so, how can these strategic sectors be identified?

12.1 Why Industrial Policy in Developing Economies? The Middle-Income Trap

In developed economies, industrial policy is back in the policy agenda. In the EU, for example, after retreating between 1990 and early 2000s, industrial policy reclaimed an official position in the policy agenda in 2002 with the objectives of reviving productivity and growth and increasing competitiveness.1 Following the global financial crises, the EU’s industrial policy effort intensified as growth further diminished. In the USA, in addition to defence-related industrial policies, policies to support technology and advanced manufacturing were adopted during Clinton and Obama administrations to enhance the country’s competitiveness.

In developing countries which need growth to catch up, unnecessary imports cause slowdown, as discussed in Chap. 6. The costs of slowdown are higher in developing countries than in developed ones, as it impedes the needed catch-up process. Worse, if the import-led slowdown is systemic and sustained, then the developing country risks falling into the middle-income trap, which is a critical impediment to sustained growth.

The persistence of the gap between the per capita income levels of developed and developing countries has attracted the attention of economists.2 In many developing economies, the growth of per capita GDP has been quite volatile and its average value in the long run remains relatively low.3

As empirical studies have demonstrated, a typical growth path for relatively successful low-income countries helps them reach the middle-income threshold, but then they slow down, keeping the country in the middle-income levels for protracted periods of time.4 This is symbolized in Fig. 12.1 by the fall in the steepness of the growth trend when the economy reaches the middle-income level.
../images/463538_1_En_12_Chapter/463538_1_En_12_Fig1_HTML.png
Fig. 12.1

Industrial policy and the middle-income trap

In fact, this so-called middle-income trap has been successfully avoided by very few countries in the world such as Finland and Sweden in Europe and Japan, South Korea, Taiwan, Hong Kong, and Singapore in East Asia. In other words, avoiding the middle-income trap has proven to be the exception.

Economists Eichengreen et al. (2012, 2013) made some empirical estimations and reached the conclusion that there are two middle-income traps at $15,000–16,000 and $10,000–11,000 (in constant 2005 prices) levels. They argue that the growth of the per capita GDP in the middle-income countries and in many of the developing countries gradually fall after these thresholds. Calculations show that countries may stay in the trap for quite long periods; for example, Bulgaria, Costa Rica, and Turkey have remained there for 50 years.5

A limited number of studies examined the possible key factors helping countries avoid the middle-income trap.6 Their results suggest that some basic factors such as the level of education in the society, physical infrastructure, and macroeconomic framework may play a role in getting out of the middle-income trap or never falling into it. More importantly, efficient transfer of labour from agriculture to knowledge-based manufacturing sectors, R&D infrastructure, and industrial policies are also important factors.7

The results of an empirical study by Hausmann et al. (2005) can be interpreted to suggest that typical market reforms consisting of liberalization of financial markets and trade are not a significant determinant of growth accelerations that might help countries get out of the middle-income trap. On the other hand, macroeconomic (consisting basically of monetary and fiscal) policies have a short time span and their primary scope of action is to eliminate the output gap rather than adding to the long-term growth performance of the economy.

In contrast, the industrial policy, as other structural policies, is designed and implemented in order to improve the long-term growth performance of the economy. In particular, it can help countries surmount the middle-income trap (Fig. 12.1) by raising the growth performance over the long term due to certain characteristics of the manufacturing sector. This is based on the growth-friendly nature of the manufacturing sector (Kaldor 1966, 1967).

The ‘traditional’ (sectoral) industrial policies played a significant role in the developmental performance of Japan between the 1953 and 1973 high-growth periods.8 On the other hand, in the 1960s and 1970s European countries resorted to national champions and sectoral industrial policies (in sectors where European firms were far from the technological frontier) were employed in order to catch up with the US economy.9

12.2 Is a Country Certain to ‘Naturally’ Industrialize in the Full Range? General and Sectoral Industrial Policy Along the Industrialization Process

Industrialization, as a Kaldorian engine of growth, assisted the developed countries of today to increase their productivity and income levels. It may also assist today’s developing countries, some of which face the risk of premature de-industrialization.

De-industrialization is primarily a problem of developed economies and shows itself in the form of falling shares of manufacturing in total employment or in GDP. However, there is evidence that lower-income countries also face the risk of a ‘premature de-industrialization,’ comprising a shrinking manufacturing sector before it reaches the relative levels where the advanced economies began to de-industrialize.10

If industrialization can help reaccelerate the growth rates and take the developing country out of the middle-income trap, a natural question is whether or not specific policy, in particular industrial policy, would be warranted to trigger industrialization.

Earlier stages of industrialization are easier to undertake than later stages. Manufacturing goods without using capital goods is nowadays limited to very a few societies in remote parts of the world. In today’s world, thus, Stages I and II (Fig. 12.2) are relatively easy and even natural for countries where international manufacturers of machinery offer their products to the international market and with abundant financing opportunities; developing countries have ample opportunities to import machinery and increase capital deepening along Stages I and II. Moreover, transition from Stage I to Stage II mostly involves relatively simple learning processes and is a matter of the passage of adequate time for labour and firms to work with machinery.
../images/463538_1_En_12_Chapter/463538_1_En_12_Fig2_HTML.png
Fig. 12.2

Industrial and STI policies along the industrialization process

The difficulty arises in the transition from Stage II to III and III to IV. This is because that transition requires not only larger amounts of physical and financial capital and competition with international incumbents, but also capacity building consisting of larger human capital investments and a build-up of TCs ties at the firm and national levels. Thus, as empirical evidence based on country experiences show, countries are not assured naturally to reach Stage III or Stage IV. This can be called the ‘non-industrialization trap,’ which is likely to coincide with the middle-income trap. Moreover, countries are prone to losing certain manufacturing industries over time due to losing cost competitiveness—the de-industrialization.

Thus, there is a role to be played by the industrial policy along all of the Stages I–IV; however, it is particularly important in facilitating passage from Stage II to III and III to IV. The reasons are that several factors provide convenient conditions for these earlier stages of industrialization: (i) Stages I and II require relatively less policy support, as the supply of low cost labour is available in newly industrializing countries, (ii) these countries have opportunities to join global value chains with their competitive labour costs, (iii) the aforementioned availability of supplier loans financing the machinery imports, and (iv) foreign direct investment (FDI) inflows.

Starting with Stage III, technical capabilities start to become critical. There is probably a bi-directional relationship between technological progress and industrialization. Technological progress is driven by factors that are endogenous to the manufacturing activity (such as by LbD) as well as by exogenous factors such as demand (e.g. tendency of public and private procurement agents to resort to foreign rather than locally manufactured goods), private and non-private R&D activity, and the general education system in the country.

In Stages III and IV, exogenous technological progress drivers become critical and the need for an industrial policy to address these issues becomes more prevalent. It is theoretically possible and empirically evident that a country may fail to raise its technical capabilities beyond critical levels. As a result, it may fall into a ‘low-technology trap’ or a ‘middle-technology trap.’ In turn, the low- or middle-technology trap may be a key reason why a country may fall into the ‘non-industrialization trap.’ These can be considered as market failures that necessitate industrial policy for the achievement of the second best.

In addressing the above challenges, it can be said that industrial policy becomes a more important driver of the industrialization process starting with Stage III. Its targets in terms of technical capabilities and sectoral focus should be appropriately adjusted along the industrialization process; sectoral focus is particularly necessary if the country wants to move beyond Stage II. Moreover, industrial policy and STI policies should be appropriately sequenced in order to address the relevant needs of the industrial sector (Fig. 12.2).

Industrial policy primarily aims at changing the production structure of the economy in favour of the manufacturing industry by channelling the government’s selected budgetary and non-budgetary resources and by channelling labour towards the manufacturing sector. In fact, such a definition would more correctly referred to as a general ‘industrial(ization) policy’ (Fig. 12.3), which would comprise a set of incentives that are expected to render private fixed capital investment into the manufacturing sector as a whole (i.e., without targeting specific industrial subsectors), attractive to entrepreneurs and businesses, or at least to compensate for some of the disadvantages that manufacturing investments have over the non-manufacturing sectors.
../images/463538_1_En_12_Chapter/463538_1_En_12_Fig3_HTML.png
Fig. 12.3

The layers of industrial policy

The decisions of the industrial entrepreneur are not made in a vacuum. The industrial entrepreneur would decide to invest (and take risks) in a sector if non-economic obstacles were removed to a sufficient degree.11 The non-economic obstacles to investment carry costs for the entrepreneur. In a feasibility study, they would appear as negative factors, reducing the overall returns to the investing firm and thus jeopardizing the feasibility of the investment. General industrial incentives which have been employed in many countries have been supposed to raise the attractiveness of industrial investments by raising the overall private returns. Industrial policies, in general, thus, would aim at reducing the costs of non-economic barriers to manufacturing investment and increasing the attractiveness of manufacturing investments to private investors. Thus, they provide a more conducive environment to the industrial entrepreneur and the industrial layer.

However, the primary reason to employ industrial policy is not only to change the composition of the overall GDP in favour of the manufacturing sector, but also to change the subsectoral composition of the industrial (manufacturing) sector in order to increase the value added generated by the existing set of resources. Thus, the ‘traditional,’ ‘sectoral,’ or ‘focused’ industrial policies consist of ‘targeting’ certain subsectors or products of the manufacturing sector and supporting them by various tools. These more focused industrial policies have also been referred to as ‘picking the winner.’ By this, limited budgetary resources can be used more efficiently and effectively.

If the targeted sectors are selected ‘properly’, positive externalities (which can be called ‘extra-sector externalities’) can be generated while building capacity within the realm of the targeted subsectors. Such externalities increase the societal returns reaped from the industrial policy. They would also reduce the costs of ‘wrong’ (sub)sector selections as extra-sector spill overs would generate social returns in other (sub)sectors. As an example of such positive extra-sector externalities, US defence and space policies (which are industrial policies in disguise, as mentioned before) seemingly do not generate economic benefits directly to the society but provide technological externalities to other sectors.

Though there are many critics of picking-the-winner policies for well-known reasons,12 the USA and some European and East Asian countries, particularly Japan (in the second half of the nineteenth century and the third quarter of the twentieth century) and Korea (between 1962 and 1997), have designed and implemented them successfully. Their policy experiences led to a drastic change in the overall economic structure and economic performance, including (i) high GDP growth rates, (ii) a significantly higher share of industrial sector in the overall GDP and in exports, (iii) a substantially different subsectoral structure of the industrial sector and composition of exports,13 and (iv) increased international competitiveness of the country’s manufacturing firms.

A related set of tools comprises the STI policies whereby governments allocate budgetary funds to assist private firms’ R&D activities in order to develop innovative products, processes, and technologies . These are generally regarded as cross-cutting (‘horizontal’) policies without a sector preference. However, technologies which are supported and those which are not nevertheless relate to some manufacturing sectors more than others. Thus, they can be considered to be also somewhat related to the industrial policies.

12.3 How to Pick the Winner? Strategic Manufacturing Sectors

How can a policy maker select the ‘strategic’ sectors or products? While targeting sectors have been both praised and criticized, this key question has not been adequately dealt with in the economic literature. Strategic sectors may, for example, be defined based on national security concerns. For example, in the USA, defence and related sectors are considered strategic.14 However, strategicness of a sector based on national security considerations may not coincide with its strategicness in economic terms. According to some, a technologically sophisticated sector is automatically a strategic one. However, that is also quite vague and inadequate in assessing the overall economic merits of a sector.

Based on the foregoing discussion on the industrialization process and technical progress, the design of industrial policy should take into consideration a number of important factors, including the capability-building aspects. These factors can be quantified to reach an overall ‘sector strategicness index’ in economic terms that may be used to build sectoral rankings or make comparisons. Sub(sector) targeting (picking the winner) can then be based on such rankings in order to achieve maximum long-term benefits for the economy.

Here a framework is proposed based on four factors:
  1. (i)

    Value added potential which directly feeds into per capita income or its growth rate in the overall economy,

     
  2. (ii)

    backward linkages,

     
  3. (iii)

    depth (potential) of LbD, and

     
  4. (iv)

    depth (potential) of technological learning.

     
Each can be scored as a standardized index (e.g. ranging from 0 to 100) and combined under a simple functional form in order to determine empirically the importance of the sector or product:
$$ {\mathrm{S}}_i={\beta}_{\mathrm{EVAP}}{\mathrm{EVAP}}_i\times {\beta}_L{\mathrm{Li}}_i\times {\beta}_{\mathrm{LbD}}{\mathrm{LbD}}_i\times {\beta}_{\mathrm{TD}}{\mathrm{TD}}_i $$
where Si is the estimated strategic importance of sector/product i, EVAPi represents the economic value added, Lii represents the ‘linkages,’ LbDi represents the ‘learning potential’, and TDi represents the technological depth of the sector or product. The coefficients β represent the weight given to the specific factor by the policy maker; they should add up to 1.

The reason for the multiplicative formulation is straightforward. Hypothetically, a product may have one or more factors with zero or very low indices (e.g. a very low EVAP). That should diminish or even eliminate the overall strategic importance of the sector for the decision maker. In other words, a zero or close to zero multiplicative strategicness for a sector should lead to its elimination from the list sectors of strategic priority.

An alternative formulation to eliminate the risk of an unwarranted overlooking of an otherwise strategic sector or product which has a zero (or very low) value for one of the four indices but has very high levels of strategic importance in terms of the remaining three factors can be considered as:
$$ {\mathrm{S}}_i={\mathrm{EVAP}}_i+{\mathrm{Li}}_i+{\mathrm{LbD}}_i+{\mathrm{TD}}_i $$
or
$$ {\mathrm{S}}_i=\alpha \left({\mathrm{EVAP}}_i+{\mathrm{Li}}_i+{\mathrm{LbD}}_i+{\mathrm{TD}}_i\right)+{\mathrm{EVAP}}_i\times {\mathrm{Li}}_i\times {\mathrm{LbD}}_i\times {\mathrm{TD}}_i $$
where α is a standardization coefficient between the additive and multiplicative combinations of the indices of appropriate size.

In what remains of this section, each of the four factors will be discussed and will be linked to the main question, that is, how the policy maker can select the strategic sector and base itself in assigning index values to each strategic factor.

Economic Value Added Potential

A key variable determining the strategic importance of a sector/product is the potential of economic value generation it offers to the manufacturer and the economy. Different firms and sectors take roles in different parts of the value chain. Ultimately, the structure of the value chain defines both the final price of the product (to the end-user) and the value-added contributed by each firm and sector in the chain. The same product in different countries may be subject to different value chain structures and two firms which take a role in a similar part of the value chain in two different countries may create different levels of value added. By putting a cap on the final product price, the competition in the final product market can also significantly affect value generated by firms in the value chain.

The metric which measures the value generation of a firm/sector i is the ratio of the total value generated (value added) to the total sales of the firm/sector; it can be called the value generation rate (VGRi). The higher this ratio, the higher is the potential of generation of total value added by the production of more product(s). That, in turn, would directly translate into a higher GDP and GDP growth. Nevertheless, there are obviously upper boundaries to value generation; for example, if the level of production starts to depress final product prices, then higher production may cause decreasing returns to scale in value creation.

Table 12.1 presents VGRs for the largest 1000 industrial firms in Turkey, a middle-income country that can be considered to be in Stage II. Clearly, the VGR s vary significantly across sectors. VGRs are relatively low in sectors such as paper products, food products, and beverages and relatively higher in oil refining, computer and electronics products, and transport equipment (Table 12.2).
Table 12.1

Value generation rates in large Turkish industrial companiesa

Sectorb

Median VGR (%)c

Number of firmsd

Mining of coal and lignite

35.3

2/4

Extraction of crude petroleum and natural gas

39.4

1/1

Mining of metal ores

67.3

8/10

Other mining and quarrying

29.4

2/6

Manufacture of food products

12.0

110/208

Manufacture of beverages

20.0

7/12

Manufacture of tobacco products

712.4

1/4

Manufacture of textiles

25.2

62/114

Manufacture of wearing apparel

21.9

14/38

Manufacture of leather and related products

21.1

2/2

Manufacture of wood and of products of wood and cork, except furniture; manufacture of articles of straw and plaiting materials

24.9

6/12

Manufacture of paper and paper products

18.5

8/31

Printing and reproduction of recorded media

32.3

4/9

Manufacture of coke and refined petroleum products

34.1

5/10

Manufacture of chemicals and chemical products

19.5

29/60

Manufacture of basic pharmaceutical products and pharmaceutical preparations

32.3

6/14

Manufacture of rubber and plastic products

19.6

20/42

Manufacture of other non-metallic mineral products

35.5

38/76

Manufacture of basic metals

15.7

52/111

Manufacture of fabricated metal products, except machinery and equipment

23.5

14/33

Manufacture of computer, electronic, and optical products

32.2

6/7

Manufacture of electrical equipment

28.0

20/55

Manufacture of machinery and equipment

28.4

13/23

Manufacture of motor vehicles, trailers, and semi-trailers

27.9

32/67

Manufacture of other transport equipment

46.8

3/10

Manufacture of furniture

27.1

7/12

Other manufacturing

3.8

6/9

Electricity, gas, steam, and air conditioning supply

15.5

8/20

Notes: aThe data relate to largest 1000 industrial companies in Turkey

bNace Revision 2 classification

cMedian value of the ratio of value-added 2. Total net sales for each firm in the sector

dThe first number is the number of firms with data. Second number is the total number of firms in the sector

Source: Istanbul Chamber of Industry (2015a, b) and author’s calculations

Table 12.2

Backward linkages of two selected manufacturing industry subsectors

Manufacturing industry subsectors

Beverage bottling plant

Shipyard

Sub-sectors of backward linkages

Glass manufacturing firms

Iron and steel firms

Sweetener and dye firms

Casting and Forging

 

Electronics, electronic equipment and IT firms

 

Glass manufacturing firms

 

Various mechanical equipment firms

 

Power generation equipment firms

 

Engine manufacturers

 

Composite material manufacturers

 

Petrochemical firms

 

Propeller manufacturers

 

Internal furniture firms

The potential market size of the product/sector is also important as a source of value generation; if the overall domestic or international market potential is too small for a product/sector, even high VGRs might not warrant strategic importance. The market potential for a product/sector i (MPi) is a function of not only the current market size but also its future trends. Consequently, a simple formulation for a product’s (or sector’s) potential of value generation can look like the following:
$$ {\mathrm{EVAP}}_i={\mathrm{VGR}}_i\times {\mathrm{MP}}_i $$

Altogether, VGRs for the sector/product and its market potential in the country as well as export markets can be used to assign indices (from 0 to 100). One way can be to normalize the highest sector/product EVAP to 100 and to assign pro-rata indices to others.

Learning Depth (Potential): The Firm as a Repository of Knowledge, Skills, and Institutional Capacity

Investment cost of starting a competitive private firm that can develop, manufacture, and sell automobiles is quite high and such an endeavour would have a relatively low probability of success. The high cost of that initiative, properly adjusted by the low probability of success, is a measure of the value of an up-and-running automotive firm.

LbD leads to the development of technical and managerial skills in the firm. A successful private automobile manufacturer, for example, encompasses the value of all the competitive skills cumulatively built over time, including the phase of investment; it is thus a repository of significant skills built over years of manufacturing operations. These skills are embedded in the firm as an intangible asset. It is manifested by diminishing unit costs as cumulative production increases.

In practice, the LbD process is based on the elasticity of unit production costs (in terms of labour hours) relative to cumulative production. Typical functional form (Fig. 12.4) for the learning curves is modelled as cN = aNb, where a represents the production costs (labour hours) of the first product, c represents the production cost of the Nth unit, and consequently N represents cumulative number production (=N). The progression rate b < 0 represents the extent of reduction in unit costs in response to the increase in cumulative production; it is calculated as b = log(L)/log(2), where the learning rate (0 < L < 1) equals 1 minus the improvement rate. The improvement rate that emanates from the learning process is defined as the reduction in the unit production cost as the cumulative production is doubled. Thus, the learning rate, the principal metric of the learning potential, is converted into the asymptotically continuous rate of improvement in production costs.
../images/463538_1_En_12_Chapter/463538_1_En_12_Fig4_HTML.png
Fig. 12.4

Learning curves for three hypothetical sectors/products

Learning depth can be defined as a combination of the extent of possible cost gains and/or the time required to reach it. These two factors, determined by the parameter b in the above equation, determine the exact shape of the curves in Fig. 12.4. The rate of accumulation of skills from LbD, b, by the representative firm may differ from sector to sector (or product to product); some sectors (or products) offer steeper progress curves. The potential for the fall in unit costs and its speed determines the ‘learning depth’ or ‘learning potential’ of the sector. For example, the learning rate is likely to be higher in aircraft manufacturing compared to bottling. In other words, not all sectors (or products) are equal in terms of the ‘learning potential’ and thus a higher potential for the accumulation of intangible assets or value that the firms may capture by entering the market and continuing to manufacture.

A sector or product with a higher ‘learning potential’ offers a higher potential for accumulation of intangible value and assets stored by the corporate sector of the country (Fig. 12.4). Industrial policies inducing firms to enter or increase production in such sector/products may yield both higher social welfare and competitiveness at the firm level.

There are many studies that have empirically estimated the learning curves for different sectors. These estimates or new ones can be used to assign an index of learning potential (ranging from 0 to 100 in this study) to different sectors/products.

Technological Depth

Technological depth of a product or sector captures its potential to give birth to the creation of new or more developed products and processes. With reference to a sector or product, technological depth is the probability of creation of new products, processes, and technical knowledge for the incumbents from that time on. Technological depth relates also to technological ‘linkages’ or externalities. For example, the imaging technology may have uses in military products, mobile phones, TV sets, or computers.

It is clear that in terms of technologies involved there is a significant gap between the electronics industry and, say, the paper industry. The technological depth offered by the electronics sector has yielded many new products during the last several decades: the digital camera, the iPod and the tablet, the cellular phone, personal computers, and so on. Clearly, the electronics industry offers a higher potential of technological learning than the paper industry. In turn, in Japan and other East Asian countries, a deliberate decision by the private firms and the public sector to concentrate on the electronics sector has positively impacted on the economy.

It is not straightforward to fix numerical values to the technological depth of a sector or product. A qualitative scoring approach can best be used to assign technological depth indices.

Linkages

Backward linkages of a firm (or sector) constitute a cluster of firms (or sectors) that will be mobilized if the firm (or sector) is to increase its production. Thus, the overall economic impact of a firm’s activities in terms of value added is multiplied by the amount of its domestic backward linkages. The manufacturing sector possesses a higher density of backward linkages than the primary or tertiary sector. However, even within the manufacturing sector, the level of linkages varies along different subsectors. As an extreme example, the shipbuilding sector has much wider backward linkages than a soda bottling plant (Table 12.2). The latter turns a liquid obtained from a well into soda, whereas the former turns a wide range of materials and equipment into a complexly designed final product.

The intensity of backward and forward linkages of a sector/product determines its total economic impact. An increase in the production of a sector with more intense linkages generates wider economic benefits across the economy. Leontief (1941) developed the concept of an input-output model which links sectors and subsectors within an economy with each other. An exogenous increase in the final demand of a product translates into economy-wide direct, indirect (primary and secondary), and induced economic impacts. These impacts can be estimated by multipliers obtained from the input-output tables.

The stronger the backward linkages of a subsector/product is, the higher is its multiplier effects. As mentioned in Chap. 5, that means its ‘social return’ (economic return to the entire society) is higher than its private return (profit to the firm or the subsector). That is because such products trigger production in many other sectors. A good example is automobile production, as mentioned in Chap. 5. Another one is the construction sector; a construction boom generates production and sales of many products such as cement, iron bars, glass, wood and furniture, cables, aluminium profiles, floor materials, and so on.

From a policy perspective this implies that it is more worthwhile for the government to support those subsectors. However, manufacturing subsectors differ in the levels of learning and technological depth—and thus in the levels of positive spillovers on the economy. Targeting two subsectors (such as automobiles and construction) with the same depth of linkages may have different economic impacts in the short run. In Fig. 12.5, the upper-right quadrant represents desirable subsectors that have maximal positive impact in both the short and long term. Some subsectors, such as aviation, can have extensive linkages and thus can contribute as a source of significant positive externalities, with only limited short-term macroeconomic effects. In the short run, the government is interested in macroeconomic performance such as growth or export earnings. In the long run however, it should focus more on developmental impact such as productive and technological capabilities.
../images/463538_1_En_12_Chapter/463538_1_En_12_Fig5_HTML.png
Fig. 12.5

Short- and long-term economic impact of sectors

Quantitatively, two short-run output multipliers can be calculated from Leontief’s input-output table. A change in final demand (direct impact) for a sector’s output multiplied by that sector’s type I (respectively type II) output multiplier will generate an estimate of direct and indirect (respectively direct + indirect + induced) impacts throughout the economy.15 Thus, a good analytical measure of the linkages of the sector is its output multiplier. It can be normalized with reference to the sector with the highest output multiplier to yield an index between 0 and 100. The long-run impact of linkages of a subsector should be incorporated through the indices measuring learning and technological potentials.

12.4 Sequencing Industrial and STI Policies

Another key question is how to sequence industrial policies and the STI policies? Should they be implemented concurrently or subsequently? Or, which one should precede the other?

The border between STI policies and industrial policies is not clear. STI policies include policies to support basic science, technology (which are ‘horizontal’ policies), and innovation (which is mostly in the form of governmental financial support or mentoring arrangements) to entrepreneurs (mostly new ones) and firms with a new idea or project. Types of general industrial policies may coincide a lot with technological or innovational policies. On the other hand, (horizontally) targeting technologies overlap with sector or product targeting. For example, supporting imaging technologies overlap with supporting electronics companies.

For the TCs at the national level (i.e. at the public and private sectors) to flourish, development of an industrial layer is a prerequisite. In other words, a certain amount of capacity building in the industrial sector will increase efficiency and effectiveness of STI policies; STI policies applied in an economy without a strong industrial layer are likely to be ineffective and cost-inefficient. This is because in the non-manufacturing sectors innovation effort is less productive compared to manufacturing sectors.

On an anecdotal basis, in many cases in the unindustrialized countries, the STI funds are wasted. The new products or processes may be developed as a result of these supports; but in many cases, they are not commercialized due to the fact that the country does not possess a strong industrial layer. Or, again due to the fact that the country lacks a strong industrial layer, the products or processes that are developed and commercialized may be just unimportant products that may not provide sufficient economies to the firm and country.

So, before a sufficiently strong industrial layer is formed in the country (or the region), the industrial policy should precede the STI policies (Fig. 12.6). In the earlier stages of industrial development, industrial policies constitute the appropriate policy response, whereas in later stages, STI policies with an emphasis on industrial research should take over, subsequently shifting to STI policies with emphasis on basic research.16
../images/463538_1_En_12_Chapter/463538_1_En_12_Fig6_HTML.png
Fig. 12.6

Sequencing industrial and STI policies. (Source: Yülek (2016))

12.5 Development-Based Public Procurement

Public procurement is potentially a very powerful industrial policy tool that can be instrumental in removing a key non-economic obstacle to different types of manufacturing firms: inadequate market access.17 Low market access is probably the most important impediment to industrialization of developing countries.

Market access is also a critical survival factor for manufacturing companies, especially for younger and smaller ones with no prior reputation in the market, little financial means and collateral even if they have unique innovative skills. Manufacturing companies face inherent impediments to growth, especially in the developing countries. These include small domestic markets, inability to access export markets, low production scale, and low TCs. In addition, they face further impediments such as limited previous references from clients, inadequate branding and reputation, and even a psychological tendency of the local administrations (and consumers) towards well-known international products independent of their quality advantage, if any, over local products. All these and other factors exacerbate the market access difficulties for local manufacturing companies in developing countries.

Public procurement policies that are designed to alleviate these difficulties are called development-based public procurement (DbPP).18 DbPP policies may be instrumental in raising the competitiveness of manufacturing companies and their technical capabilities, in assisting the development of small- and medium -sized enterprises (SMEs), and in complementing the existing STI policies. In addition, DbPP triggers secondary economic benefits accruing to the society: the observance of basic principles of public procurement of fairness, equity, transparency, competitiveness, and cost-effectiveness.19 Thus, in the wider context of the linkages between industrial policy and economic development, DbPP may play a crucial role.

In many countries, governments grant financial support to firms under their STI policies to cover the areas such as R&D and innovation. Such financial support can be considered as ‘indirect,’ whereas the DbPP policies can be classified as ‘direct’ support to businesses. This is because market opportunity is a direct driver of industrial capacity building; seeing the market opportunities, the firm will leverage resources to manufacture and deliver products triggering learning, and other processes increasing the technical capabilities. In the process, the firm may use the financial (STI) supports, which are indirect, as enablers.

The types of secondary benefits as well as the extent to which they can be derived from better public procurement policies may differ in developed and developing economies. The developed economies possess sophisticated industrial structures and TCs, and their economic growth is driven by the growth of TFP rather than factor accumulation. In these countries, public procurement could be primarily used to support innovation for maximal benefit. The procurement for innovation initiatives in the UK is an example of that.

In developing countries economic growth is driven primarily by factor accumulation, and industrial and technological capacity is constrained by various obstacles. In these countries, DbPP policies could help strengthen basic industrial and technological capabilities, thereby enhancing the competitiveness of existing industries.20

There are well-known tools of DbPP. In the military field, countertrade or offsets are used widely to build manufacturing capacity. In the civilian field, local content rules have been commonly used such as those in South Africa.21 In some countries such as the USA, India, and South Korea, ‘set asides’ from the procurement budget is utilized to provide procurement support to SMEs . Forward procurement or planned procurement is a technique to alert businesses to make preparations for future procurement plans.22 In developed economies, procurement for innovation whereby the government demands a product to be developed and manufactured as it does not exist in the market is becoming a widely debated topic.

The list of these tools can be extended. For example, ‘locality rules’ can be used as a tool of industrial as well as regional development policy.23 These rules basically specify that the procured product should be manufactured in an economically backward region that the government intends to develop. In many countries, the government extends financial incentives (such as tax holidays or exemption from employee income taxes) to private industrial investments in less developed regions. Introduction of ‘direct’ procurement incentives by locality rules of DbPP policies can complement those financial incentives in enhancing industrial investments in backward regions.