CODS 131765—12/9/2005—KALYAN—166026

Oxford Development Studies, Vol. 33, No. 3, 2005

Late Industrialization and Structural Change: Indonesia, 1975 – 2000 JOJO JACOB* ABSTRACT This paper examines economic growth and structural change in Indonesia during the period 1975– 2000 using an input – output-based structural change decomposition method. The analysis focuses on the sources and pattern of growth during three phases of economic development: the inward-oriented phase from 1975 to 1985; the outward-oriented phase from 1985 to 1995; and the recent phase of crisis and recovery from 1995 to 2000. Growth during the first phase, although impressive, was moderate in comparison with the export-led manufacturing-driven growth during the second phase. During both these phases, the Indonesian economy witnessed significant structural changes, especially within manufacturing. However, the dynamics underlying growth and structural change showed important differences. Although growth under the first two policy regimes was assisted by favourable economic circumstances, selective industrial policies may also have played a significant role. The results suggest that the long neglect of the technological foundations and human capital base of the economy could be holding back recovery and sustained growth in the present phase.

1. Introduction How do “latecomer” countries industrialize? What industries lead the process of industrialization? Are there “right” policies that can give a “kick-start” to successful industrialization? These are some of the questions that have reappeared in the aftermath of the rapid and successful “late” industrialization in East Asia’s newly industrialized countries (NICs). Industrialization in these countries, especially in Korea and Taiwan, was characterized by a sequential shift in the locus of manufacturing growth from traditional to modern industries (see, e.g. Amsden, 2001). In other words, prior to their successful entry into technology- and scale-intensive industries during the mid-1970s, they had acquired considerable manufacturing experience in traditional and light industries. The government policies and incentive structure in these countries were geared to generating an exportbased manufacturing growth, which in turn ensured learning and technological progress. *Jojo Jacob, Eindhoven Centre for Innovation Studies, Eindhoven University of Technology, 5600 MB, Eindhoven, The Netherlands. Thanks to Bart Los, Eddy Szirmai, Bart Verspagen, Ina Drejer, Abdul Erumban and two referees for their comments and suggestions, and to the Indonesian Statistical Office (BPS), Marcel Timmer, Haryo Aswicahyono, Wiwiek Arumwati, Steven Keuning and Salmet Sutomo for the provision of and advice on data. Support from the Dutch Organization for Scientific Research (NWO) is gratefully acknowledged. I alone am responsible for any remaining errors. ISSN 1360-0818 print/ISSN 1469-9966 online/05/030001-25 q 2005 International Development Centre, Oxford DOI: 10.1080/13600810500317820

CODS 131765—12/9/2005—KALYAN—166026 2

J. Jacob

Indonesia offers a slightly different example of late industrialization, which started much later than in Korea and Taiwan, from the 1970s onwards. The industrialization process began with a state-led heavy industrialization drive under an import-substituting industrialization strategy (Hill, 1996). Again unlike in Korea after 1961, the prevailing incentive structure in Indonesia did not require industrial conglomerates to fulfil any export commitments. However, the steep fall in oil prices, first in 1982 and thereafter in 1986, prompted the initiation of far-reaching economic reforms and the adoption of an export-oriented industrialization strategy. During this phase, industrialization was based mainly on resource- and labour-intensive industries. By the 1990s, science-based industries had begun to play a leading role in manufacturing exports. This was primarily the result of foreign investment from the NICs and Japan. Following the well-known flying geese pattern, labour-intensive manufacturing activities shifted to low-cost countries in the region. Indonesian manufacturing growth from the late 1960s until the economic crisis of 1997 was one of the fastest in the contemporary world. Between 1975 and 2000, manufacturing’s share in total value added and exports rose from 10.8% and 9.4% to 27% and 55.1%, respectively (see Section 2, Table 1). The economic crisis, which began in late 1997, however, reversed this trajectory of rapid growth. The annual growth rate of output in 1998 turned negative for most sectors, accompanied by widespread unemployment in manufacturing and services, worsening of the investment climate and a net outflow of capital (Hill, 1999). Against this background, the sources and patterns of growth during the inward-oriented phase from 1975 to 1985, the outward-oriented phase from 1985 to 1995, and the more recent crisis and recovery phase from 1995 to 2000 are investigated in this paper. An extended input – output (IO)-based method of structural change decomposition is employed. The IO framework is appropriate for analysing structural change due to its recognition of the sectoral interdependencies of an economy. For Indonesia, this approach has the particular advantage that the IO tables provide the most exhaustive (in terms of the coverage of the economy) and detailed (in terms of aggregation) sets of comparable data from 1975 onwards.1 Using IO tables, we decompose the sectoral value-added growth for each phase into four elements: the contributions from changes in the input coefficients (the so-called technical coefficients); the contributions from changes in the value-added coefficients; the importsubstitution effect; and the growth in the final demand components. The latter is decomposed further into the aggregate growth in each final demand component (macro demand) and the changes in their inter-sectoral allocation (reallocation effect). The results illustrate that growth during the export-promoting policy regime far outstripped an already impressive growth performance of the economy during the inwardoriented (or export-pessimistic) policy regime. The key sectors that led overall manufacturing growth and the underlying dynamics of growth also showed significant differences in the two phases.2 Although growth under the first two policy regimes was assisted by favourable economic circumstances, selective industrial policies may have played a significant role.3 The results, however, also suggest that the long neglect of the technological foundations and human capital base of the economy may be holding back recovery and sustained growth in the present phase. In Section 2, the salient features of the policy transitions in Indonesia are discussed, followed by an evaluation of the changes in the economic structure in terms of the changes

Sectoral composition of value added Sector

1975

1980

1985

1990

1995

2000

1975

1980

1985

1990

1995

2000

27.7 20.5 0.6 10.8

20.6 26.3 0.3 11.2

22.2 14.2 5.0 13.1

16.7 14.6 3.2 19.0

11.6 9.8 2.0 24.5

7.9 17.6 5.5 27.0

6.0 73.9 1.0 9.4

6.7 70.8 6.8 7.3

6.1 40.6 23.7 17.9

2.3 27.9 14.4 38.5

1.1 17.3 7.5 51.1

0.8 16.2 13.3 55.1

6.3 0.2 0.7 0.3

5.1 0.6 0.8 0.5

4.7 1.2 0.7 0.7

6.7 2.2 0.8 0.5

8.7 1.7 0.8 0.7

7.9 1.7 1.1 0.9

2.5 0.0 2.9 0.0

1.4 0.8 2.6 0.1

1.3 4.8 3.7 0.1

5.0 10.3 3.1 0.6

4.2 8.8 2.9 0.5

3.8 6.1 1.8 1.1

0.5 0.0

0.5 0.1

0.4 0.1

1.0 0.1

1.5 0.1

2.2 0.2

0.0 0.0

0.2 0.0

1.6 0.0

6.8 0.2

9.8 0.6

8.0 0.6

0.4 0.3 0.5 0.0 0.1 0.1 0.0 0.6 0.0

0.5 0.2 0.5 0.2 0.1 0.1 0.0 0.6 0.0

0.9 0.4 0.8 0.5 0.4 0.2 0.0 0.4 0.1

1.6 0.9 0.9 0.6 0.6 0.1 0.0 0.7 0.1

2.5 1.3 1.2 0.9 0.7 0.1 0.0 1.1 0.1

2.5 1.6 1.2 0.5 0.5 0.1 0.0 1.8 0.0

0.0 0.1 2.8 0.0 0.7 0.0 0.0 0.0 0.1

0.1 0.0 0.2 0.0 1.4 0.1 0.0 0.0 0.0

1.1 0.1 1.0 0.1 2.9 0.2 0.0 0.0 0.0

3.9 0.7 1.6 0.5 3.7 0.4 0.0 0.1 0.1

6.7 2.2 2.1 0.5 4.3 0.3 0.0 0.5 0.1

7.3 3.9 2.6 0.5 2.9 0.2 0.0 0.5 0.1

0.2 0.2

0.3 0.5

0.4 0.4

0.4 0.9

0.5 0.8

0.7 0.9

0.0 0.1

0.0 0.0

0.0 0.1

0.3 0.1

0.7 1.5

1.0 4.1

0.1 0.0 0.0

0.1 0.1 0.1

0.2 0.2 0.1

0.2 0.2 0.0

0.2 0.5 0.1

0.2 0.7 0.2

0.0 0.0 0.0

0.0 0.0 0.0

0.1 0.1 0.0

0.0 0.3 0.0

0.0 0.3 0.3

0.1 0.6 1.2

Late Industrialization and Structural Change

Primary Oil, gas and mining Petroleum refinery Manufacturing Resource-intensive manufacturing Food, beverages and tobacco Wood products and furniture Rubber and rubber products Non-metallic mineral products Labour-intensive manufacturing Garments and leather Other manufacturing Scale-intensive manufacturing Textiles Paper, paper products and printing Industrial chemicals Iron and steel Non-ferrous metals Shipbuilding and repairing Other transport Motor vehicles Aircraft Differentiated manufacturing Metal products Non-electrical machinery Science-based manufacturing Drugs and medicines Plastics Electrical apparatus, not classified elsewhere

Sectoral composition of exports

CODS 131765—12/9/2005—KALYAN—166026

Table 1 Pattern of structural change in Indonesia: 1975– 2000

3

J. Jacob

Sectoral composition of value added Sector Radio, TV and communication equipment Professional goods Electricity gas and water Construction Finance and insurance Other services Total:

Sectoral composition of exports

1975

1980

1985

1990

1995

2000

1975

1980

1985

1990

1995

2000

0.1

0.3

0.3

0.4

0.9

1.9

0.1

0.2

0.4

0.5

3.8

7.9

0.0 0.3 5.0 2.4 32.6 100.0

0.1 0.3 5.0 2.0 34.4 100.0

0.0 0.4 6.6 2.6 36.0 100.0

0.1 0.6 5.8 3.8 36.2 100.0

0.3 0.6 6.7 4.1 40.6 100.0

0.2 0.5 4.0 4.1 33.4 100.0

0.0 0.0 0.0 0.0 9.7 100.0

0.0 0.0 0.0 0.2 8.1 100.0

0.1 0.0 0.0 2.3 9.3 100.0

0.3 0.0 0.0 3.0 14.0 100.0

0.9 0.0 0.0 3.3 19.7 100.0

0.8 0.0 0.0 1.3 13.3 100.0

Source: Input – output tables; Statistik Industri, BPS, Jakarta, various issues.

CODS 131765—12/9/2005—KALYAN—166026

4

Table 1. Continued

CODS 131765—12/9/2005—KALYAN—166026 Late Industrialization and Structural Change

5

in sectoral shares in value added and exports, from 1975 to 2000. In Section 3, the IO decomposition techniques are discussed, and the decomposition formula derived. To contextualize the results of the decomposition analysis for the three phases, some broad expectations about the results are proposed. In Section 4, the data and sectoral classification are described. The results of the analysis are discussed and explained in Section 5. The final section explores the broad policy implications for recovery and growth in Indonesia.

2. The Indonesian Economic Performance Under the “The New Order” 2.1 The Background In a relatively short time-span of over two decades from the late 1960s until the crisis of the late 1990s, Indonesia transformed itself from a stagnant, agrarian economy into one where manufacturing exports drive rapid and sustained economic growth. Interestingly, however, this transformation of the economy took place against the backdrop of fluctuating policy regimes: regimes that were different on such important policy issues as the degree of state intervention, attitude towards foreign investment, barriers to trade, etc. Such shifts in economic policies were largely guided by the movements in international oil prices.4 When the New Order of former President Soeharto was established in 1966, the poor state of the economy prompted the relaxation of restrictions on imports and exports, liberalization of the investment policy and the adoption of orthodox monetary and fiscal policies. These policies succeeded in containing inflation, rehabilitating physical infrastructure and triggering an economic recovery. The increase in oil prices in 1973 and later in 1979 led to an expansion in state investment in industry and a return to the (pre1971) restrictive trade and foreign investment policies. The oil revenues were recycled into large-scale investment in state-owned enterprises in sectors such as iron and steel, petroleum, aluminium and fertilizers. The inward-oriented industrialization programme generated sustained growth during the period 1971 – 81. However, the fall in oil prices coupled with a slowing down of economic growth during the period 1982– 86 caused a policy response aimed at liberalizing and opening up the economy and promoting exports. The deregulation measures involved reductions in tariff and non-tariff barriers, liberalization of foreign investment regulations, financial sector reforms and efforts to reduce the monopoly power of the big businesses through (state-induced) divestiture. This phase ushered in greater independence for the private sector firms and generated substantial increases in foreign investment (Pangestu, 1991; Thee, 1991). A point to bear in mind, however, is that in both the regulated and liberalized phases of economic policy, Indonesian manufacturing experienced considerable sectoral variations in the degree of protection, monopoly power, ownership structure, and so forth (Basri & Hill, 1996; Thee, 2002). At a time when the economy was experiencing an impressive export-led industrialization drive, the country was caught in the Asian financial crisis of late 1997. Seven years on from the crisis, growth has not yet recovered to anywhere near the precrisis level. Very little domestic investment is flowing into productive sectors and, unlike other crisis-hit countries, foreign investment is still negative (Hill, 2004).

CODS 131765—12/9/2005—KALYAN—166026 6

J. Jacob

Using the IO tables for 1975, 1980, 1985, 1990, 1995 and 2000, let us examine the pattern of structural change from 1975 to 2000, with an emphasis on the effect of the policy transition during the mid-1980s and the impact of the crisis of the late 1990s. The tables for 170 sectors (with small variations in the exact number of sectors from IO table to IO table) have been aggregated to 130 sectors for comparison (see Section 4 for details on the adjustments made to the data). 2.2 The Pattern of Structural Change Table 1 presents the shares in value added and exports of 29 major sectors of the economy for the period 1975– 2000 (in constant 1983 prices). Twenty-two of these 29 sectors are in manufacturing, and are classified into five categories: resource-intensive, labour-intensive, scale-intensive, differentiated and science-based industries (classification based on OECD, 1987).5 On a broad level, the following features of structural change can be noted from the table. (1) Over the 25-year period, the combined share of the agriculture, forestry and fisheries (primary), oil, gas and mining (oil and gas) and petroleum refinery declined from nearly 50% to a quarter of the total value added. (2) In exports, the oil and gas sector alone accounted for about three-quarters of the total exports during the 1970s; a feature that, many argue, offered Indonesian policy-makers little incentive to adopt an export-oriented industrialization strategy. Its share began to fall dramatically, however, from the early 1980s, dropping to about 16% of the total by the year 2000.6 (3) With the decline of the oil and gas sector, the contribution of non-oil manufacturing to total value added and exports began to rise, especially from the mid-1980s.7 (4) The share of services in total value added remained stable, with the notable exception of the finance and insurance sector; its share increased rapidly following the banking reforms of the late 1980s. Manufacturing’s emergence as the mainstay of overall economic growth has been accompanied by important structural changes within manufacturing. These changes are discussed in the following subsections. 2.2.1 Changes in the structure of manufacturing value added. While manufacturing’s share in value added was hovering around 10% during the period 1975 –85, it had increased to 27% by the year 2000 (Table 1). The resource-intensive industries have traditionally been the leading contributors to manufacturing value added. Although their contribution to the total economy value added increased marginally over time, their share in the total manufacturing value added declined: from over 60% in the 1970s to about 40% in 2000. In this category, food, beverages and tobacco (food) has always accounted for most of the value added. Its share in the category fell substantially during the late 1980s, with the significant increases in the share of wood products and furniture (wood products). The industrial policy during the New Order regime had placed emphasis on the development of scale-intensive industries, utilizing the revenues from oil and gas. However, between 1975 and 1985, the contribution of these industries to the total value added of the economy increased only marginally (from 2.1 to 3.6%). It registered a faster

CODS 131765—12/9/2005—KALYAN—166026 Late Industrialization and Structural Change

7

increase after the liberalization of the economy (5.6% in 1990 and 8.2% in 2000). In this category, textiles, paper and printing (paper) and iron and steel were responsible for most of the early growth; motor vehicles made important contributions during the 1990s. During the latter period, labour-intensive and science-based manufacturing categories also became important contributors to total value added. In these categories, the main contribution to value added stemmed from garments and leather, and radio, TV and communication equipment (consumer electricals and electronics), respectively. 2.2.2 Changes in the structure of manufacturing exports. The rapid increase in the share of manufacturing in the total value added of the economy is dwarfed by its performance in exports, where it registered a more than fivefold increase in its share: from 9.4% in 1975 to 55.1% in 2000. The manufacturing export boom began in the 1980s, after facing a decline during 1975 –80. The early surge in manufacturing exports in the 1980s stemmed mainly from the resource- and labour-intensive industries, such as wood products and garments and leather, respectively, and, to a lesser extent, from a scale-intensive industry, such as textiles. The increase in the export share of wood products can be attributed to the ban on log exports in 1981 and the consequent increase in the exports from the plywood industry. During the 1990s, the export share of these industries remained important, though showing some decline. Consumer electricals and electronics and non-electrical machinery (which fall into the science-based and differentiated categories, respectively) showed rapid increases in their shares, especially during the latter half of the 1990s. A major reason for the surge in manufacturing exports, from the late 1980s through the 1990s, was the export-oriented investment from the four Asian NICs—South Korea, Taiwan, Hong Kong and Singapore—and Japan (Pangestu, 2002). Japan had been the single largest foreign investor during the inward-oriented phase of industrialization, with most of the investment directed to the textile and garment industries. By the 1990s, threequarters of the foreign direct investment (FDI) approvals were from the four NICs, of which Korea was the most important in terms of the number of projects (Hill, 1991). Part of this resulted from a relocation of industries from the NICs in the much acclaimed flying geese pattern, and the strategy of the international buyers to disperse production locations (Thoburn, 2001). In textiles and garments, unfulfilled market quotas under the Multi Fibre Agreement (MFA) were a major reason for foreign investment and export growth. 2.2.3 Changes in the structure of imports. Manufacturing’s share in imports declined from 70.5% in 1975 to 58.8% in 1980—the period of intense import-substituting regime (detailed results are not reported, in order to save space). Then, in tune with the shift away from import substitution to export promotion, the manufacturing import share showed a steady increase until 1990, but declined substantially during the crisis and recovery phase of 1995– 2000. In the above discussion, it was intended to provide a snapshot of the Indonesian economy’s growth trajectory over the 25-year period considered. The fast growth of the economy until the crisis was characterized by shifts in the engines of economic growth from primary and oil and gas during the 1970s to manufacturing and services from the mid-1980s onwards. While the resource- and labour-intensive manufacturing sectors dominated the manufacturing export boom of the late 1980s, science-based industries came to play a leading role during the 1990s.

CODS 131765—12/9/2005—KALYAN—166026 8

J. Jacob

Taking this overview as the context, let us examine the pattern and sources of growth in value added during the three phases mentioned earlier—the phase of inward orientation (1975 – 85), the phase of outward orientation (1985 – 1995) and the phase of crisis and recovery (1995 – 2000). 3. The Structural Change Decomposition Method and the Hypotheses Owing to its recognition of the interdependent structure of an economy, the IO framework has been an attractive tool for explaining the variations in key economic variables. These variables include output, value added, energy use, labour requirements, volume of imports and output of services (Dietzenbacher et al., 2000). In this framework, structural change and growth have usually been analysed using output as the relevant variable.8 In this paper, however, value added instead of output is employed as the variable, the growth of which will be analysed. This is because value added is a more policy-relevant variable than output at the sectoral level. This is more so in the context of a transitional economy such as Indonesia, which is undergoing changes in industrial organization and production relations. Furthermore, technological change can be assessed not only in terms of changes in technical coefficients, but also in terms of the changes in the ratio of primary inputs (land, labour and capital) to intermediate inputs (Forssell, 1989). In decomposing the change in the left-hand side variable under consideration (value added in our case), the contributing factors need to be weighted using appropriate weights. One approach is to use as weights for each contributing factor either the initial year values or the final year values of rest of the contributing factors. This approach, however, has the disadvantage of generating interaction terms in the decomposition equation, which are hard to interpret. This problem can be eliminated by using initial year values as the weights for the change in some factors and final year values as the weights for the change in the other factors. The problem here is that there can be n! decompositions when there are n factors identified as contributing to a change in the left-hand side variable. Dietzenbacher & Los (1998) suggested an ad hoc solution that involves the use of two polar equations, the unweighted average of which yields result close to the average of all the possible n! decompositions. In the first polar equation, the weights are formed by the remaining factors in such a way that those that are to the left (of the factor under consideration) take their initial year values, and those to the right their final year values. In the second equation, these weights are reversed, i.e. the initial year values are replaced by the final year values and vice versa, hence the name polar. We follow this method to derive the decomposition equation for value-added change. The final aspect to consider is whether the decomposition of the change in a variable should be undertaken in an additive or in a multiplicative framework. While the component factors sum to the value of the left-hand side term in the additive case, it is the product of the component factors that equal the left-hand side term in the multiplicative framework. The choice between the two methods is fairly arbitrary; but given the simplicity of the additive framework in deriving the polar equations, it was employed in this study. 3.1 The Decomposition Formula for Value-added Change We start by defining the well-known equation for output in the IO framework, using matrix notations (in this paper, unless mentioned otherwise: a capital letter in bold indicates

CODS 131765—12/9/2005—KALYAN—166026 Late Industrialization and Structural Change

9

a matrix; a hat over a bold capital letter indicates a diagonal matrix; a capital letter in italics indicates a vector; a small letter in italics stands for a scalar): X ¼ ðI 2 AÞ21 Y;

ð1Þ

in which, X is the (n £ 1) vector of output, I the (n £ n) identity matrix, A the (n £ n) matrix of input coefficients and Y the (n £ 1) vector of final demand. Equation (1) can be rewritten to replace the vector of output by the vector of value added (Dietzenbacher et al., 2000): ^ 2 AÞ21 Y; V ¼ KðI

ð2Þ

where, V is the (n £ 1) vector of value added and K^ the (n £ n) diagonal matrix with the ratio of value added to output on its diagonal. To capture the contribution of import substitution in production, we introduce the (n £ 1) vector of domestic supply ratio U, which is the ratio of domestic production less exports to total supply: U¼

ðX 2 EÞ ; ðF þ TÞ

ð3Þ

in which, X, E, F and T are the (n £ 1) vectors of output, export, total final demand (domestic plus imported, but excluding export) and total intermediate input demand, respectively. The value added equation (2) now takes the following form: ^ 2U ^ AÞ _ 21 ðUF ^ þ EÞ; V ¼ K½ðI

ð4Þ

ˆ is the (n £ n) diagonal matrix, whose diagonal is formed by the elements of the where U _ the (n £ n) matrix of technical coefficients (derived from the total vector U, and A intermediate input matrix, as opposed to the domestic intermediate input matrix from which the coefficient matrix A is derived in equation (1)). Greater insights into the sources of structural change, especially in a transitional economy such as Indonesia, can be gained by splitting the vectors of final demand F (which is the sum of vectors of private consumption, government consumption, investment demand and inventory changes) and exports E in the following way: F ¼ Pp þ Gg þ Cc þ Nn

and

E ¼ Qq;

where, P, G, C, N and Q are the (n £ 1) vectors of the shares of private consumption, government consumption, investment demand, inventory changes and exports. These are derived by dividing each cell of their respective final demand vectors by their column sums, i.e. by the scalars p, g, c, n and q. The latter may be called the macro final demand, in their respective final demand categories. By splitting up the final demand vectors in the above fashion, we are able to decompose the impact of the change in a final demand category, say consumption demand, on the change in value added in a sector (e.g. electronics), into the impact of reallocation of consumption demand to that sector (e.g. from the primary sector) and the impact of economy-wide or macro increase in consumption demand.

CODS 131765—12/9/2005—KALYAN—166026 10

J. Jacob

Equation (4) can now be rewritten as: ^ ^ ^ AÞ _ 21 ðUðPp þ Gg þ Cc þ NnÞ þ QqÞ: V ¼ K½ðI 2U

ð5Þ

Based on equation (5), we derive the two polar decomposition equations (as noted at the beginning of this section) for a change in value added between two periods, and take their arithmetic average (the complete derivation of the final decomposition equation can be obtained from the author). The decomposition formula, which decomposes a change in value added (DV) into 13 components, may be written in a simplified form as follows: DV ¼ DV k þ DV u þ DV a þ DV P þ DV p þ DV G þ DV g ð6Þ þ DV C þ DV c þ DV N þ DV n þ DV Q þ DV q ; where the subscripts stand for the corresponding factors in equation (5). The nature of each factor’s contribution is explained hereafter. DVk (value-added coefficient) is the change in the ratio of value added to output. A negative value for this coefficient can imply increased efficiency in the use of primary inputs (land, labour and capital). It can also indicate outsourcing and input substitution. Competitive market pressures are a major source for declines in the value-added coefficient. This is because competition can lead to improved efficiency, a widely accepted view, as well as, as suggested in the industrial organization literature, to a greater resort to outsourcing (a decline in vertical integration).9 An offsetting influence could be a reduced reliance on external supplies, especially from abroad, in industries that began with limited capabilities and developed at a later stage backward and/or forward integration with related production activities.10 DVu (import-substitution effect) captures the contribution of the domestically produced intermediate inputs and final demand goods vis-a`-vis the imports to the change in value added; DVa (technological change) shows the contribution of the technical coefficients. DVP, DVG, DVC, DVN and DVQ (reallocation effect) capture the change in value added resulting from the changes in the inter-industry allocation of private consumption, government consumption, investment, inventory changes and exports, respectively. This can also be interpreted as reflecting the changes in the preferences of final demand consumers. DVp, DVg, DVc, DVn and DVq (macro final demand effect) show the effect of an economy-wide change in the respective final demand component on sectoral value-added change. 3.2 Some Broad Expectations about the Results Based on our a priori understanding of Indonesian economic growth, some of the results of the decomposition analysis could be anticipated. For example, it is reasonable to expect that the import-substitution effect and exports were the mainstays of growth in the inward- and outward-oriented phases, respectively. Similarly, technological change can be assumed to be an important source of growth in the first phase (cf. Hill, 1996). Our analysis, while examining these points, is also expected to

CODS 131765—12/9/2005—KALYAN—166026 Late Industrialization and Structural Change

11

offer new insights into the dynamics of growth across different phases as well as sectors. Before proceeding to discuss the results, the data used, their sources, the aggregation procedures and the construction of the linked-IO tables will be outlined. 4. The Data The 170-sector (with small yearly variations) IO tables published by the Central Statistical Agency (BPS) of Indonesia for the years 1975, 1980, 1985, 1990, 1995 and 2000 are used here. Changes in the sectoral classification, especially from 1985 onwards, led us to aggregate the tables into 130 sectors for comparison (the aggregation procedure is explained in the following paragraph). In the tables for 1995 and 2000, government consumption expenditure was moved from the final demand category to the intermediate demand category; but to maintain consistency with the earlier tables, the government consumption expenditure was moved back into the final demand category in the 1995 and 2000 tables. The 130 £ 130 tables were converted into constant 1983 prices using the following price indices: the relevant wholesale price indices (WPI) for the primary and manufacturing sectors, the implicit service price indices for the services sectors. The WPI data were taken from Indikator Ekonomi published by the BPS, and implicit service price indices from the Groningen Growth and Development Centre (GGDC) database.11 We deflated each cell in a given row by the price index of the output category corresponding to that row.12 The original tables have been aggregated into 130 sectors based on the procedure suggested by Miller & Blair (1985). It can be represented in matrix notation in the following way:  ¼ eMe0 ; M  the aggregated (n £ n) matrix ((130 £ 130) where M is the original (m £ m) matrix, M matrix in our case), e the (n £ m) summation matrix of ones and zeros and e0 the transpose of the matrix e. The results for 29 major sectors of the economy are presented by pre-multiplying the decomposition equations with a (29 £ 130) summation matrix. The decomposition of value-added growth, during the interval 1975 – 2000, was carried out for all the four 5-year intervals (1975 –80, 1980 –85, 1985 –90 and 1990 – 95). However, for convenience, the results up to 1995 are reported only for the inwardoriented phase (1975 – 85) and the outward-oriented phase (1985 – 95).13 This was done by taking the simple average of the results for 1975– 80 and 1980 –85, as well as those for 1985 –90 and 1990 –95, respectively (Tables 2 and 3).14 The results for the final phase (1995 – 2000) are reported in Table 4. In the tables, the first column shows the annual compound growth rate of value added.15 The second column reports the total percentage growth rate (period growth rate), of which the decomposition is made. The third column represents the total of all contributing factors in percentage terms (equal to 100). The remaining columns to the right show the percentage contribution of each of the 11 factors. It may be noted that government consumption demand and inventory changes have been combined into “other” macro final demand and “other” reallocation effect.16

Industries Total Primary Oil, gas and mining Petroleum refinery Manufacturing Resource-intensive manufacturing Food, beverages and tobacco Wood products and furniture Rubber and rubber products Non-metallic mineral products Labour-intensive manufacturing Garments and leather Other manufacturing Scale-intensive manufacturing Textiles Paper and printing Industrial chemicals Iron and steel Non-ferrous metals Shipbuilding and repairing

Percentage contribution to value-added growthb Total ValuCf ImpSub TechCf R.Cons M.Cons R.Invs M.Invs R.Exp M.Exp R.Othr M.Othr 28.2 233.6 226.9 53.9 214.5

8.1 2 3.2 7.5 2 27.6 5.5 5.5 16.2 2 0.2 22.4 8.1

2 3.9 2 47.9 2.3 0.7 2 0.9

46.8 159.9 3.9 3.4 45.7

1.5 3.8 2.2 1.1 21.3

100.0 2107.9

9.6 2 18.7

2 29.1

239.9

0.2

12.8

9.1

7.0

10.4

14.0

7.6 2 22.0

4.7

7.1

0.0

15.5

1.9

3.2

11.6 20.1

2 39.1 27.0

14.2 18.8 2.2 24.3 17.2 23.4 210.0 42.8 37.3 18.8 72.4 2 10.2 11.3 7.7 6.0 17.1 2 38.2 2 3.0

1.7 7.6 2 2.0 0.8 2 0.3 1.5

5.0 2.5 4.4 54.7 6.9

27.8 12.9 24.3 784.7 39.4

100.0 100.0 100.0 100.0 100.0

1.8

9.5

24.9

203.5

100.0

8.8

6.4

36.7

100.0

25.6

13.4

87.7

100.0

215.1

23.1

5.1 12.0

28.2 76.4

100.0 100.0

214.2 25.2

12.7 23.2

12.3 8.4 11.0 43.4 21.1 13.2

78.5 49.6 68.4 506.1 160.9 86.1

100.0 100.0 100.0 100.0 100.0 100.0

2 1.2

17.5 2 4.8 29.8 10.5 6.5 12.0 6.6 2 56.4 146.2 2.3 30.7 2 8.3 16.1 13.7 7.5 1.2 2 16.6

1.7 14.8 9.8 20.8 1.8 9.2 2 1.2 1.4 2 0.7 3.8

14.4

4.2

2.7

33.9

2.2

1.8

1.3

2.6

21.6

77.8

1.3

4.9

15.6

45.0

3.1

0.8

5.9

1.1

72.4 24.2

0.7 1.4

2.7 3.0

60.2 2 3.9 2 26.3 2.0 2.7 2 17.0

23.3 18.5

32.0 30.9 28.5 1.5 1.4 1.9

0.0 1.0 0.2 20.1 0.8 55.4

0.6 34.9 2 3.1 2 6.7 5.2 7.0 4.1 1.9 2 11.8 23.0 6.0 2 22.5 15.9 3.6 0.2 17.6 2.7 1.2 2 8.9 4.2 5.3 61.5 4.7 2.2 20.5 41.6 12.4 4.6 5.7 0.9

CODS 131765—12/9/2005—KALYAN—166026

Period growth (%)

J. Jacob

Annual growth (%)c

12

Table 2 Value-added growth and the contributing factors, 1975– 85a

43.2 7.9 130.9

100.0 100.0 100.0

10.9 14.5

67.6 97.1

100.0 100.0

22.0 29.8

22.4 25.6 69.9

175.2 212.1 1316.7

100.0 100.0 100.0

27.2

232.7

7.3 10.1 7.6 5.3 6.1

2 1.1 0.9 2 5.6

3.5 2 61.5 3.5 2 19.1 2.7 2 1.2

35.5 17.8 1.5

44.4 37.2

0.1 2 2.2

2.3 7.4

2 1.0 3.2

4.0 3.4

0.0 21.4 21.2

0.3 5.8 10.9

2.9 2 3.5 2 0.3

1.2 0.9 1.4

1.4 3.5 2 1.0

2.9 3.0 0.5

11.3

4.5

8.7

9.2

5.3

2 0.7

2.9

51.5 31.6

7.8 1.7

16.4 7.5

26.4 2 1.5 4.9 6.6

1.5 9.3

2.8 9.1

2.9 46.5 44.9

12.3 1.2 20.8

82.4 18.2 12.8

0.0 50.6 1.3

2 0.9 2 0.7 1.7

3.0 12.6 29.2

13.1 78.8 12.0

9.3 2 63.1 18.6

12.6 149.1 3.7

28.0 82.6

17.6 7.3

5.0 2 3.3

9.3 3.5

27.6 229.3

21.5 5.0 0.3

19.7 18.3 78.7

2 0.7 17.6 2 2.9

52.0 36.6 7.8

21.7 14.2 5.7

100.0

217.3

40.6

18.6

16.9

42.5 62.0

100.0 100.0

63.0 296.7

25.1 11.0

14.4 2 107.5 55.2 59.7

44.0 29.7 34.4

100.0 100.0 100.0

20.5 0.6 2 1.5 219.1 2 15.3 2 53.9 21.2 2.6 0.8

0.9 52.1 1.5

82.3 63.0 2323.7 213.9 25.4 23.9

0.8 7.9 7.2

Results for 1975 –85 are derived as average of the 5-yearly results for 1975 – 80 and 1980 –85. The following abbreviations are used to represent the 11 factors contributing to value-added growth: ValuCf, value added coefficient; ImpSub, import-substitution effect; TechCf, technical coefficient; R.Cons, reallocation of consumption; M Cons, macro consumption; R.Invs, reallocation of investment; M.Invs, macro investment; R.Exp, reallocation of export; M.Exp, macro export; R.other, reallocation of government consumption and inventory changes; M.other, macro government consumption and macro inventory changes. c Annual compound growth. b

CODS 131765—12/9/2005—KALYAN—166026

a

231.3 2 25.4 266.8 108.7 21.9 2 3.0

7.4 1.5 18.2

Late Industrialization and Structural Change

Other transport Motor vehicles Aircraft Differentiated manufacturing Metal products Non-electrical machinery Science-based manufacturing Drugs and medicines Plastics Electrical apparatus, not classified elsewhere Radio, TV and communication equipment Professional goods Electricity gas and water Construction Finance and insurance Other services

13

Industries Total Primary Oil, gas and mining Petroleum refinery Manufacturing Resource-intensive manufacturing Food, beverages and tobacco Wood products and furniture Rubber and rubber products Non-metallic mineral products Labour-intensive manufacturing Garments and leather Other manufacturing Scale-intensive manufacturing Textiles Paper and printing Industrial chemicals Iron and steel Non-ferrous metals

Annual Period growth growth (%) Total ValuCf ImpSub TechCf R.Cons M.Cons R.Invs M.Invs (%)c 2 4.1

55.2

3.0 2 2.9 2 84.2 2 169.1 5.7 2 2.8 17.3 2 7.5 223.2 2198.2 2 957.5 2 349.5

308.8 17.8 732.1

0.5

0.8

2 2.0

20.9

20.7

R.Exp 2 2.4

10.0

61.2 100.0

3.0 6.2 0.4

16.1 100.0 35.2 100.0 2.1 100.0

17.1

120.3 100.0

4.6

0.8

4.5

9.6

35.4

20.5

13.0

14.8

16.8

117.4 100.0

15.6

2 0.3

2.7

10.2

61.3

0.0

0.2

15.9

109.5 100.0

20.7

2 0.5

16.4

2 1.2

7.9

29.9

11.4

71.2 100.0

0.2

10.9

5.8

4.3

19.4

12.1

76.9 100.0

0.2

8.2

0.7

2.4

24.8 18.7

202.3 100.0 135.6 100.0

28.4 5.7

2 5.6 2 6.0

1.9 2 4.7

22.6 23.8 14.6 16.5 16.6

176.8 190.6 97.9 114.4 115.1

100.0 3.0 100.0 3.6 100.0 24.0 100.0 24.9 100.0 214.1

2 8.0 10.0 28.2 9.4 2 0.4

3.1 15.3 2 2.0 19.6 26.3

M.Exp

R.Othr

M.Othr

24.3

0.1

7.8

23.2 13.9 2 9.8 33.7 0.6 27.7 2 95.6 116.3 214.5 489.8 21261.3 1378.3

20.7 2.0 25.7

10.6 18.5 63.3

16.4

20.3

1.6

3.7

4.8

0.6

1.0

25.7

26.9

38.8

22.4

20.9

0.7

6.3

2 18.5

73.1

2.2

24.5

13.4

213.8

70.9

8.3

9.8

23.0

3.0

18.8 0.5

17.9 29.8

0.1 20.1

0.6 2.5

46.9 46.9

26.5 21.3

0.5 0.6

0.7 3.6

18.9 9.0 2 5.4 1.0 2.8

26.3 25.6 46.5 6.8 5.9

0.1 20.5 21.0 27.6 24.3

0.4 4.8 6.2 53.8 17.6

32.8 15.8 12.8 8.5 20.0

22.5 9.3 20.2 8.6 43.6

20.4 0.9 26.5 1.6 0.0

1.2 6.1 4.9 3.1 2.6

J. Jacob

Percentage contribution to value-added growthb

CODS 131765—12/9/2005—KALYAN—166026

14

Table 3 Value-added growth and the contributing factors, 1985– 95a

32 141

2 4.3 22.9 2 25.5

48.1 31.4 42.4

11.3 19.4

70.9 100.0 142.5 100.0

0.1 23.4

34.9 2 3.2

2 34.2 6.7

2 1.7 1.7

16.3 4.3

216.9 26.9

76.3 52.7

16.6 8.0

9.8 4.6

20.4 20.1

20.7 1.8

10.3 25.1 20.5

63.2 100.0 206.6 100.0 153.6 100.0

23.8 19.8 22.1

7.5 3.8 2 4.6

1.0 1.3 4.0

2 15.8 17.7 0.0

96.7 36.3 13.5

0.0 20.2 7.2

0.8 5.8 29.4

0.7 5.4 35.6

3.5 6.0 11.5

3.7 2.4 2.8

5.6 1.8 2.6

25.1

206.5 100.0

22.7

2 9.1

3.3

19.4

18.8

10.6

13.8

30.3

12.1

1.3

2.2

36.9 13.8

381.3 100.0 210.9 91.0 100.0 59.0

24.1 2.4

0.4 2 36.6

22.1 2 4.9

16.1 55.4

0.0 20.7

5.4 10.0

28.6 6.4

14.3 9.1

20.3 23.9

0.2 3.8

10.6 15.4

65.7 100.0 104.9 100.0

23.0 25.6

0.0 2.4

2 0.5 7.0

0.0 20.4

4.1 42.4

25.9 103.3 20.6 8.0

0.2 5.2

1.2 19.3

20.5 20.8

1.0 2.3

11.5

72.2 100.0

23.1

1.5

1.4

3.5

53.4

6.9

11.0

0.4

11.6

0.0 100.0

Results for 1985 –95 are derived as average of the 5-yearly results for 1985 – 90 and 1990 –95. See notes for Table 2. c See notes for Table 2. b

2302 347 402 111 2 196.7 163.9 16.0 18.3 257.1 132.9

0.0

13.4

51 893 8.4 5.1 25.1

170 2 39 892 33 889 077 8.7 21.8 2.9 2.7 20.9 3.3 24.3 2.8 1.7

CODS 131765—12/9/2005—KALYAN—166026

a

4189

5.2 21.3 5.7

2 97 2 187 212 480 074 352 28.9 100.0 48.7 109.6 2 87.5 162.8 100.0 13.9 210.8 2 1.9 31.9 100.0 250.0 14.1 2 10.8

0.0

Late Industrialization and Structural Change

Shipbuilding and repairing Other transport Motor vehicles Aircraft Differentiated manufacturing Metal products Non-electrical machinery Science-based manufacturing Drugs and medicines Plastics Electrical apparatus, not classified elsewhere Radio, TV and communications equipment Professional goods Electricity gas and water Construction Finance and insurance Other services

15

Industries Total Primary Oil, gas and mining Petroleum refinery Manufacturing Resourceintensive manufacturing Food, beverages and tobacco Wood products and furniture Rubber and rubber products Non-metallic mineral products Labour-intensive manufacturing Garments and leather Other manufacturing Scale-intensive manufacturing Textiles Paper and printing Industrial chemicals Iron and steel

Annual Period growth growth (%) Total ValuCf ImpSub TechCf (%)b

R.Cons M.Cons R.Invs M.Invs

1.5 7.8 2 6.0 2 26.8 14.0 92.7 24.1 194.6 3.50 18.75

100.0 100.0 100.0 100.0 100.0

2 87.8 18.7 2 3.6 11.0 2 11.4

2 98.4 24.2 2 19.3 2 23.8 2 32.8

119.9 39.5 47.4 31.4 10.4

2 0.4

2 2.0

100.0

34.4

357.4

2403.8

27.8

250.2

1.6

8.5

100.0

41.8

2 25.3

286.2

2 28.7

2 9.5

8.4

49.6

100.0

38.8

2 20.3

37.1

7.3

2 3.3

2 2.6

6.3

35.6

100.0

2 9.3

0.7

95.7

8.8

2 2.4

9.2

55.1

100.0

38.0

2 2.0

0.3

2 4.6

5.5

31.0

100.0

2 31.3

2 58.6

226.3

31.4

1.1 6.9 2.0

5.4 39.4 10.6

100.0 2 253.5 36.8 100.0 2 36.1 2 34.4 100.0 2 52.5 2 194.8

2380.7 4.5 2103.1

2 9.0

2 37.7

100.0

95.6

27.0

229.4

2 14.0 2 35.3 10.1 14.8 7.6 2 1.1 14.5 2 1.2 15.6 2 16.6

R.Exp

M.Exp

R.Othr M.Othr

2.5 1.4 2.7 0.5 2 0.9

2 85.0 6.4 2 11.0 2 5.6 2 27.7

28.6 13.9 8.5 26.9 7.8

295.0 2 33.5 72.9 42.0 157.2

2 1.8 2.3 2 1.4 4.7 2 1.3

13.5 2.2 -2.6 2 0.5 2 0.4

2 0.3

8.2

65.8

2361.0

111.5

9.9

2 47.2 2 100.5

2280.0

612.6

24.2

2 1.2

2 7.6

286.8

131.3

5.8

0.3

3.5

2 89.4

39.9

50.6

2.3

2 0.2

2 4.1

2 0.2

2 0.4

240.3

114.9

2 1.3

2 0.4

2 5.8

2 0.2

2 3.1

11.8

180.9

2 0.7

2.1

2238.2 2 37.0 15.6 2 7.1 73.8 2 26.5

2 0.8 2 0.2 2 4.7

2 3.6 2 5.4 2 29.4

55.4 59.3 77.0

910.4 95.1 320.5

16.9 4.0 41.6

2 5.7 4.6 2 1.8

2 3.9

57.1

27.3

2 32.7

2 4.4

2.6

2 5.7

1.2

J. Jacob

Percentage contribution to value-added growtha

CODS 131765—12/9/2005—KALYAN—166026

16

Table 4 Value-added growth and the contributing factors, 1995– 2000

See notes for Table 2. See notes for Table 2.

100.0

85.7

50.1

172.8

219.1

21.8

2 8.7

100.0

241.0

497.7

2 298.1

28.0

20.3 10.9 23.9

2 1.7 67.9 218.0

11.5 3.1

72.5 16.5

100.0 100.0

2.6

13.7

100.0

7.2 25.4

41.8 209.5

17.1

120.4

20.4 23.3

100.0 2709.8 5317.0 2 2037.7 21339.9 100.0 12.1 2 6.6 22.2 82.0 100.0 9.3 225.0 2 73.3 238.1

6.3

43.9

219.6

2465.1

2 3.0

4.2

6.6 2138.5

138.4

469.2

2491.4

2 31.9

23.0

2975.2 2330.2 15.1 0.8 2110.8 2 3.1

8.7 0.3 20.8

4.6

685.7 0.2 247.2

21.5 22.2

10.1 2 48.9 230.5 2 102.4

210.7 2 10.4 6.1 2 250.0

100.5 8.6

64.6

2 76.3

2 80.6

149.7 237.3

20.8

100.0 100.0

214.2 24.4

2 11.4 21.3

3.4 2 7.0

94.4 211.6 22.9 20.4

100.0

2.8

20.3

2 27.3

1.9

2 1.8 215.6

100.0 1507.1 2 698.3 100.0 316.9 25.7

510.7 2 152.4

77.3 247.7

28.2 1.2

234.8 6.0

100.0 100.0

20.8 137.5

1.2 2 61.7

2.4 197.9

22.4

211.3

100.0

117.1

54.6

2 85.9

7.0 23.8

889.4 2 1507.6 2 16.0 2 2.2 98.5 34.4

99.6 27.9 6.1

17.4 265.5

34.4 218.5

1.5 2 9.9

0.6 0.2

2 3.1

12.0

34.0

27.1

10.7

23.1 3.2

2 10.0 2 9.5

20.9 61.6

40.3 40.4

2 7.4 2 2.0

21.4 20.2

21.9

3.1

2 6.6

55.2

53.7

2 1.2

20.1

126.2 22.6

202.7 0.2

273.7 25.8

17.7 21987.8 4.0 282.7

72.0 2 7.8

21.3 24.6

1.3 0.5 2149.2 257.4

215.4 29.1

92.0 2 82.4

1.0 2 265.8

23.0 381.6

2 0.3 2.2

20.6 6.4

7.1

33.8

37.8

2115.1

52.3

26.0

1.8 229.6

17

b

211.9

CODS 131765—12/9/2005—KALYAN—166026

a

22.5

Late Industrialization and Structural Change

Non-ferrous metals Shipbuilding and repairing Other transport Motor vehicles Aircraft Differentiated manufacturing Metal products Non-electrical machinery Science-based manufacturing Drugs and medicines Plastics Electrical apparatus, not classified elsewhere Radio, TV and communications equipment Professionalgoods Electricity gas and water Construction Finance and insurance Other services

CODS 131765—12/9/2005—KALYAN—166026 18

J. Jacob

5. Results and Discussion We begin by examining the changes in the influence of each factor on growth between the first and second phases. This is followed by a brief evaluation of the growth in the final phase of crisis and recovery.

5.1 Sources of Structural Change: The Inward- (1975 –85) and Outward-oriented (1985 – 95) Phases It was noted in Section 2 that, during both the inward- and outward-oriented phases, significant structural changes took place in the economy in general and in manufacturing in particular. Tables 2 and 3 reveal that growth during the first phase, although impressive, was moderate in comparison with the export-led manufacturing-driven growth during the second phase. The decomposition results of sectoral growth for the first two phases point to interesting similarities as well as differences in the underlying sources of growth. Whereas some factors were a major source of growth throughout the period 1975– 95, others were more important in one phase than in the other phase. Still more interesting is the fact that the sources that remained important for growth in both the phases showed notable differences in the sectors they influenced the most. In the following, the contribution of each factor to growth across the first two phases is discussed in some detail. 5.1.1 The role of technical change. As noted before, technological change in our framework has two components: value-added coefficient and technical coefficient. We begin by discussing the importance of the technical coefficient to value-added growth. Technical Coefficient. During the inward-oriented phase, in line with our expectations, this factor was an important contributor to value-added growth (Table 2). Although never the biggest contributor, it was either the second, third or fourth biggest contributor in about 50% of all sectors, as well as of the manufacturing sector. This can be attributed to the rapid weeding out, in the 1970s, of traditional and labour-intensive technologies, most of which were antiquated and had been in place since the 1930s. In the outward-oriented phase, technical coefficient turned out to be a less important factor for growth (Table 3). However, the number of industries where it made a positive contribution was roughly the same as that in the previous phase. A notable difference, however, was the reversal of the direction of its influence across industries. After liberalization, it became negative in most of the scale-intensive industries, but became positive in all of the science- and resource-based industries. These results point out that traditional industries have failed to improve their technologies beyond the progress they achieved during the 1970s. As for the evidence of technological progress in the science-based industries, their impact was relatively minimal. It may be noted that these industries witnessed large-scale entry of foreign firms, especially in the 1990s, with investments directed mainly towards labour-intensive activities. For example, even in a high-technology sector like electronics, production is concentrated in fairly “low-tech” activities (cf. Hill, 1996). Technological progress, therefore, would not necessarily imply a shift to more technology-intensive activities. These results thus confirm the observations of some commentators that the post-reform

CODS 131765—12/9/2005—KALYAN—166026 Late Industrialization and Structural Change

19

growth has been hindered by the insufficient technological capability of the Indonesian firms and the poor state of their human capital (see Lall, 1998).17 Value-added coefficient. This factor is the second component of technology in our framework, and reflects the effects of changes in the efficiency in primary inputs use. (It can also reflect non-technology factors as well, which will be discussed later.) In both the inward- and outward-oriented phases, efficiency improvements (a negative value-added coefficient) were less prominent than technical change. During the inward-oriented phase, efficiency improvements were highest in the services and construction sectors. In manufacturing, only a handful of industries such as motor vehicles, garments and leather and food industries benefited to any notable extent from this factor. During the outward-oriented phase, as with technical coefficients, the influence of efficiency gains also appears to have declined. A notable exception is the garment and leather industry, where efficiency improvements continued to remain substantial. Value-added coefficient is also influenced by other sector-specific factors, which may not be related directly to technology or efficiency. Consider the case of the paper industry. During its early phase of growth, thanks to the impressive overall economic growth and the resulting increases in demand, this industry relied heavily on imported pulp, the key raw material. From the mid-1980s onwards, partly as a consequence of the ban on log exports (in 1981) and the expansion of the wood processing industry, the domestic production of pulp expanded, resulting in the integration of the pulp and paper industries (van Dijk, 2003).18 This may explain the negative influence of the value-added coefficient in this industry during the first phase, and its positive influence in the following phase. In this respect, the motor vehicles industry appears to offer a different explanation. This industry experienced a substantial decline in value-added coefficient in the first phase, but an increase in the second and third phases, the reason for which may be the following. In the early stages of the development of this industry, foreign component-makers set up plants, often as joint ventures with local suppliers, with a view to circumvent domestic protection (Okamoto & Sjo¨holm, 1999). These firms initially imported most of their parts and components before starting local production. Thus, it is the supplier-dominated character of the motor vehicles industry in the initial phase, and the later shift towards own production, that might explain the nature of the contribution of the value-added coefficient in this industry. 5.1.2 The role of import substitution. In the inward-oriented phase, the contribution of the import-substitution effect, as expected, was highly important in most sectors. During this phase, it accounted for the single highest share of value-added growth in approximately 20% of the industries. Its impact is slightly more when only sectors within manufacturing are considered. Import substitution accounted for the largest share of valued-added growth in iron and steel, industrial chemicals, non-electrical machinery, electrical apparatus not classified elsewhere (electrical machinery) and consumer electricals and electronics. A point to note is that not all sectors experienced a substantial positive importsubstitution effect, and some even experienced a negative impact. This can be related to the policy of local content stipulations prevailing at the time, which differed across industries. For example, local content restrictions were high in the machinery industry, but

CODS 131765—12/9/2005—KALYAN—166026 20

J. Jacob

such restrictions were limited in the automobile industry, mainly due to the lack of domestic suppliers (Simatupang, 1996). Results for the outward-oriented phase reveal, not surprisingly, that the effect of import substitution on value-added growth not only declined in general, but also ranked as one of the lowest contributing factors in most sectors. For example, in electrical machinery and consumer electricals and electronics, the import-substitution effect was ranked first among all contributing factors in the first phase, but in the second phase its was ranked the lowest. Two major exceptions are industrial chemicals and iron and steel, where the importance of the import-substitution effect for value-added growth slipped only slightly, from the first, to the second and third positions, respectively. In fact, this evidence lends support to the view that domestic industry and trade continued to be subject to controls until the financial crisis of the late 1990s (cf. Thee, 2002). 5.1.3 The role of export demand: macro and reallocation effects. During the inwardoriented phase, the contribution of export to value-added growth in general was of only low importance compared with that of other factors. Owing to the inward-orientated character of the economy, macro increases in demand made very little contribution. At the same time, there is some evidence of the effectiveness of the shift in economic policy from the early 1980s towards promoting (comparative advantage-based) manufactured exports. This is discernable from the fact that in sectors such as wood products, textiles and nonferrous metals, reallocation of export demand from other sectors (importantly, from a traditional foreign exchange earner such as oil and gas) constituted the main source of growth. During the outward-oriented phase, there was a rapid and near-pervasive increase in the reallocation of export demand to manufacturing sectors (from primary, oil and gas and petroleum refinery). In industries such as garments, textiles, etc. the reallocation effect was already a significant factor in the first phase, and remained so in the second phase. In some other sectors, the reallocation effect emerged as the most important component of growth only in the second phase. These sectors included science-based industries such as electrical machinery, consumer electricals and electronics and professional goods. Significantly, the first two of these industries were among the most import-substituting industries of the previous phase. The latter finding is in line with the argument that industries that experience a high import-substituting effect can achieve rapid increases in exports (cf. Poot et al., 1990). However, far from being automatic, a fortuitous turn of events assists this transition. This includes, among others, a relocation of labour-intensive production activities (mainly in science-based industries) from the East Asian region (see the discussion in Section 2). Therefore, export-driven growth in the three aforementioned science-based industries cannot be interpreted as being driven by the accumulated technological knowledge from the previous era. During the second phase, macro export demand became a significant contributor to growth in all sectors. This needs little explanation, however, given the radical shift of the Indonesian economy towards export orientation. 5.1.4 The role of consumption demand and investment demand. Consumption and investment demand have been important sources of growth in both the inward- and

CODS 131765—12/9/2005—KALYAN—166026 Late Industrialization and Structural Change

21

outward-oriented phases. Their contributions resulting from macro and reallocation effects will be discussed separately. The macro effect. During both the phases, the contribution of consumption and investment demand to value-added growth stemmed mainly from macro changes. Already during the first phase, these two components had played a key role in the remarkable growth performance of the economy. After the liberalization of the economy, their contribution to growth increased substantially. They accounted for the highest growth contribution in over half of the manufacturing industries—each ranking first in about a quarter of the total. Interpreted this way, these two factors were the most important sources of value-added growth during the second phase, leaving the reallocation effect of export demand (which also emerged as a major source of manufacturing growth) to the third spot. The reallocation effect. The reallocation effect of the consumption and investment demands had a much lower impact compared to their macro effect, as well as to the reallocation effect of export demand. Important shifts occurred between the two phases, however, with respect to the industries, which they influenced significantly. During the inward-oriented phase, in the transport sector, for example, all industries except the motor vehicle industry benefited substantially from a reallocation of investment (from other sectors of the economy). In the following phase, however, investment reallocation was directed towards the motor vehicles industry and away from other industries in the transport sector. Similarly, reallocation of consumption demand during the first phase had a strong negative impact on growth in professional goods and garments and leather. During the second phase, the same factor turned out to be the third most important component of growth in these industries. 5.1.5 The role of “other” final (government consumption) demand. Among the remaining final demand components to be discussed—government consumption demand and inventory changes—only the former made any noticeable contribution.19 Its contribution to growth, however, was limited to the inward-oriented phase and to the paper industry. This industry grew essentially by feeding the domestic market and government departments, under a wall of protection, but transformed into a major export-earner for the country in the following phase (van Dijk, 2003). 5.2 Sources of Structural Change: The Crisis and Recovery Phase (1995 – 2000) During this phase, the annual value-added growth declined substantially across most sectors of the economy (Table 4). The two major exceptions to this were the oil and mining and the petroleum refinery sectors, which experienced higher growth rates than in the preceding phase. The worst affected sectors were iron and steel, construction, the primary sector and other services, all of which experienced a negative growth in value added. The sectors that continued to register a positive growth at 5% or more were electrical machinery, consumer electricals and electronics, metal products, motor vehicles, paper, garments and leather, non-metallic minerals and rubber products. What factors have contributed to the slowing down and, in some cases, declining rates of growth during this phase? Table 4 tells us that most factors, with the key exception of

CODS 131765—12/9/2005—KALYAN—166026 22

J. Jacob

macro export demand, caused a negative impact on growth in this phase. There are exceptions, however. For example, reallocation of export demand was the main contributor to growth in a few sectors, such as electrical machinery, consumer electricals and electronics and non-electrical machinery. Put differently, these industries experienced a positive average annual rate of growth mainly by increasing their share in overall exports from Indonesia. On the other hand, another industry, non-ferrous metals, which also experienced an increase in its export share, still had to face a negative annual growth rate. It may be noted that this was an industry that, in the previous phase, grew mainly as a result of the exportoriented industrialization drive of the time, i.e. through an increase in macro export demand. The slack in overall exports, therefore, has had a much more damaging effect on this industry than the other three industries that we discussed; in these industries, macro export during 1985– 95 had yet to become a leading source of growth. The supply-side factors did little to counterbalance the adverse demand-side effects on growth. The contribution of technical coefficients, for example, either remained low, as in the previous phase, or was negative in most industries. There are, however, two notable exceptions: non-metallic mineral products and metal products. In these industries, technical coefficient emerged as the single most important contributor to growth. Compared with the previous two phases, value-added coefficient appears to have been positive in a slightly higher number of sectors. Although this could be interpreted as resulting from a decline in the efficiency of primary factors, it could also reflect a decline in outsourcing from abroad, necessitated by the foreign exchange difficulties arising from the crisis. 6. Conclusions and Implications20 In this concluding section, I shall refrain from recounting all the salient features of growth, outlined in some detail in the previous section. Instead, I shall attempt briefly to take stock of the industrialization experience of Indonesia, and draw some lessons for future growth. Before proceeding, it needs to be emphasized that, in the IO-based growth decomposition framework, a large part of the (increase or decrease in) growth is explained by the demandside factors. Furthermore, this analysis is essentially a growth accounting exercise, which explains the contributions of only the proximate—as opposed to fundamental— determinants of growth. However, I have tried to establish a causal relationship between the partial effects of growth captured by the 13 factors in the decomposition equation, and specific policies with respect to industry, trade, foreign investment and so forth. Between the first two phases, significant shifts occurred in the industries driving overall manufacturing growth as well as in the sources of growth. These changes were attributed, in many instances, to the policies adopted at the industry and macro levels. The industryspecific policies relate mainly to tariffs, competition and foreign investment, adopted largely in the first phase. In the second phase, the regulations in trade and investment regimes were relaxed considerably at the macro level, aimed at boosting manufactured exports. It needs to be emphasized, however, that the success of the New Order economic policies was also assisted by a host of external factors and economic circumstances. For example, the adoption of an inward-oriented industrialization strategy was facilitated by the rise in the price of oil in the 1970s, and the resulting “boom” in foreign exchange

CODS 131765—12/9/2005—KALYAN—166026 Late Industrialization and Structural Change

23

earnings. The ensuing state-led investment drive caused rapid replacement of traditional (some dating back to the 1930s) technologies by relatively modern ones. With the fall in oil prices and the decline in foreign exchange earnings from oil exports, by the mid-1980s the emphasis had shifted towards an export-oriented industrialization strategy. Again, a set of fortuitous events helped this new strategy succeed, at least temporarily. One of the important external factors was the relocation of labour-intensive production activities to Indonesia by the NICs and Japan to take advantage of Indonesia’s cheap labour.21 The unfulfilled quotas in the textile and garment industries attracted investment into these sectors as well. As a result, growth during this phase has in fact been more impressive than that in the previous phase. During this phase, even some of the so-called import-substituting industries of the previous phase became important earners of foreign exchange through exports. This goes on to demonstrate that a simplistic dichotomy such as import substitution versus export orientation is misplaced. But, have the fortunes of the Indonesian economy finally run out? A crisis befell the economy at a time when the manufacturing industry was experiencing impressive growth. Since then, the emphasis has been directed mainly towards achieving macroeconomic stability. From this analysis, it was found that the most important “proximate” causes for the slowdown in growth were consumption and investment demand. Despite achieving macroeconomic stability, very little domestic investment is flowing into production activities and, unlike the other crisis-affected economies, foreign investment is still negative 7 years on from the crisis (Hill, 2004). The limited technological capability of the manufacturing sector in general, and a workforce that is less skilled than her South and East Asian neighbours, especially China, make Indonesia a less attractive destination for FDI in technology-intensive activities. Furthermore, ethnic tensions affecting the Chinese minority have promoted capital flight and made investment from overseas Chinese sources less attractive. The results in this paper have shown that technological improvements (as proxied by changes in the technical and value-added coefficients) in manufacturing played very little role in growth, especially since the mid-1980s. Although the shift towards labour-intensive industrialization during the 1980s paid off in the short run, the crisis has brought to the fore the cost of neglecting the upgrading of technological foundations and the human capital base of the economy. Importantly, however, given the slow recovery of the Indonesian economy from the crisis as well as the current emphasis on “fiscal discipline”, the prospects for greater investment in human capital and infrastructure in the near future appear to be rather bleak.

Notes 1

2

3

The Indonesian statistical office (BPS) in private communication with the author confirmed the superior coverage of the IO data vis-a`-vis the national accounts data. Although this analysis is limited to quantifying theoretically relevant partial effects on growth, an attempt is made to relate some of the results to the underlying forces that drive them. This is of course a complicated task in the Indonesian context, where industries enjoyed different degrees of protection, which has implications for their technological effort, export orientation, factor intensity, vertical integration and so forth. For a compelling discussion of the role of industrial policy in Indonesian economic growth, see Rock (1999).

CODS 131765—12/9/2005—KALYAN—166026 24

J. Jacob 4

5 6

7

8 9 10 11

12

13

14

15

16 17 18

19

20 21

We draw heavily on Hill (1996, 1999), who has provided comprehensive accounts of many dimensions of the Indonesian economic growth over the last 30 years. The terms sector and industry are used interchangeably. It is interesting to note the emergence of petroleum refinery as an important source of export revenue— in 2000, the combined share of oil and gas and petroleum refinery was 29.5% of total exports. Petroleum refinery consists of all petroleum refinery products as well as liquefied natural gas. Unless mentioned otherwise, by manufacturing we refer to manufacturing excluding petroleum refinery, i.e. non-oil manufacturing. For example, see Wolff (1994). See, for a discussion, Perry (1989). This point is illustrated with respect to a few industries when the results are discussed. The GGDC database provides value-added data in current and constant prices, from which the implicit price indices for services were derived (see http://www.eco.rug.nl/ggdc/index-dseries.html#top). For sectors for which price indices were not directly available, the indices of the nearest three-digit sectors and, in some cases, two-digit sectors were used. It needs to be emphasized that the choice of the time intervals for the inward-oriented (1975–85), outward-oriented (1985– 95) and crisis and recovery (1995–2000) phases were determined by the availability of IO tables. However, it appears fair to assume that these intervals capture the key features of the regimes they purport to represent. The average of the 5-yearly results is reported with the view to even out the effect of short-term fluctuations. Note that, for ease of understanding, the results of the decomposition of the growth in value added are presented, instead of the change in value added as represented in the decomposition equation. These two components are combined, because they have made only very small contributions. For a recent discussion on the importance of building up technological capabilities, see Lall (2001). It should be remembered that pulp and paper are treated as one industry in our classification. As a result, backward integration of the paper industry to the pulp industry shows up as an increase in the vertical integration in the pulp and paper industry. Although results are reported only for the combined contributions of government consumption demand and inventory changes (to keep the size of the tables within a reasonable limit), this discussion is based on their separate results. Suggestions from one of the referees helped me improve this section substantially. Liberalization of the foreign investment regime in the late 1980s has been a timely measure in this respect.

References Abramovitz, M. (1994) Catch-up and convergence in the post-war growth boom and after, in: W. J. Baumol, R. R. Nelson & E. N. Wolff (Eds) Convergence of Productivity: Cross-national Studies and Historical Evidence, Q1 pp. 86 –125 (New York: Oxford University Press). Amsden, A. H. (2001) The Rise of “The Rest”: Challenges to the West from Late-industrializing Economies (Oxford: Oxford University Press). Basri, M. C. & Hill, H. (1996) The political economy of manufacturing protection in LDCs: an Indonesian case study, Oxford Development Studies, 24, pp. 241 –259. Chenery, H. B. (1979) Structural Change and Development Policy (Oxford: Oxford University Press). Dietzenbacher, E., Hoen, A. R. & Los, B. (2000) Labour productivity in Western Europe 1975–1985: an intercountry, inter-industry analysis, Journal of Regional Science, 40, pp. 425 –452. Dietzenbacher, E. & Los, B. (1998) Structural decomposition techniques: sense and sensitivity, Economic Systems Research, 10, pp. 307–323. Forssell, O. (1989) The input –output framework for analysing transmissing of technical progress between industries, Economic Systems Research, 1, pp. 429 –445. Hill, H. (1991) The Emperor’s new clothes can now be made in Indonesia, Bulletin of Indonesian Economic Studies, 27, pp. 89–127. Hill, H. (1996) The Indonesian Economy since 1966: Southeast Asia’s Emerging Giant (Cambridge: Cambridge University Press).

CODS 131765—12/9/2005—KALYAN—166026 Late Industrialization and Structural Change

25

Hill, H. (1999) The Indonesian Economy in Crisis: Causes, Consequences and Lessons (New York: St Martin’s Press). Hill, H. (2004) Rebuilding Indonesia, The Australian, 2 November. Lall, S. (1998) Technology policies in Indonesia, in: H. Hill & K. W. Thee (Eds) Indonesia’s Technological Challenge, pp. 136– 170 (Canberra: Australian National University). Lall, S. (2001) Competitiveness, Technology and Skills (Cheltenham: Edward Elgar). Miller, R. E & Blair, P. D. (1985) Input–Output Analysis: Foundations and Extensions (Englewood Cliffs, NJ: Prentice Hall). OECD (1987) Structural Adjustment and Economic Performance (Paris: OECD). Okamoto, Y. & Sjo¨holm, F. (1999) Protection and the Dynamics of Productivity Growth: The Case of Automotive Industries in Indonesia, Working Paper No. 324, Stockholm School of Economics, Stockholm. Pangestu, M. (1991) Foreign firms and structural change in the Indonesian manufacturing sector, in: E. D. Ramstetter (Ed.) Direct Foreign Investment in Asia’s Developing Economies and Structural Change in the Asia-Pacific Region, pp. 35 –64 (Boulder, CO: Westview Press). Pangestu, M. (2002) Foreign investment policy: evolution and characteristics, in: F. Iqbal & W. E. James (Eds) Deregulation and Development in Indonesia (London: Praeger). Perry, M. K. (1989) Vertical integration: determinants and effects, in: R. Schmalensee & R. D. Willig (Eds) Handbook of Industrial Organization (Amsterdam: North-Holland). Poot, H., Kuyvenhoven, A. & Jansen, J. C. (1990) Industrialisation and Trade in Indonesia (Yogyakarta: Gadjah Mada University Press). Rock, M. T. (1999) Reassessing the effectiveness of industrial policy in Indonesia: Can the neoliberals be wrong?, World Development, 27, pp. 691–704. Simatupang, B. (1996) Economic transformation and liberalization in Indonesia, in: A. E. F. Jilberto & A. E. Mommen (Eds) Liberalization in the Developing World: Institutional and Economic Changes in Latin America, Africa and Asia (London and New York: Routledge). Thee, K. W. (1991) The surge of Asian NIC investment to Indonesia, Bulletin of Indonesian Economic Studies, 27, pp. 55–89. Thee, K. W. (2002) Competition policy in Indonesia and the new anti-monopoly and fair competition law, Bulletin of Indonesian Economic Studies, 38, pp. 331–342. Thoburn, J. T. (2001) Becoming an exporter of manufactures: the case of Indonesia, in: O. Morrissey & M. Tribe (Eds) Economic Policy and Manufacturing Performance in Developing Countries, pp. 97 – 119 (Cheltenham: Edward Elgar). van Dijk, M. (2003) Industry Evolution in Developing Countries: The Indonesian Pulp and Paper Industry, Ecis Working Paper No. 2003-02, Eindhoven Centre for Innovation Studies, Eindhoven. Wolff, E. N. (1994) Industrial composition, inter-industry effects, and the U.S. productivity slowdown, Review of Economics and Statistics, 67, pp. 268 –277.