Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

Structural change

Clearly, the region, with the exception of Myanmar, has experienced a shift out of agriculture since 1965 (Table 3.3), but the speed of the shift and the sectors which have benefited most have varied between countries (Table 3.4). This is undoubtedly due to the initial big differences in industrial structure in South-East Asian countries, and also to different policies and experiences over the period. But the share of industry increased in all ASEAN countries (but only slightly in the Philippines), partly due to the adoption of deliberate diversification strategies focused on industrialization. By contrast, in Australia and Japan, industry's share of GDP declined, reflecting a so-called 'post-industrial' move into services.

Changes in employment structure over the course of economic development generally parallel changes in the structure of production. In the case of employment, however, the shift out of agriculture is muted because output per worker (often imprecisely referred to as labour productivity) is lower in agriculture, and this continues to be the case throughout much of the development process. Thus, in countries such as Indonesia and the Philippines, agriculture's share of employment is around twice as high as its share of output (55 and 24 per cent in lndonesia, and 52 and 27 per cent in the Philippines, in the early 1980s). It remains around twice as high in developed countries such as Australia, France, Japan and the United States, even though the share of agriculture in employment in these countries has generally fallen below 10 per cent.

TABLE 3.3 Growth of Production by Sector in South-East Asian Countries, 1965-1987 (per cent)






Country 1965-80 1980-7 1965-80 1980-7 1965-80 1980-7 1965-80 1980-7
Indonesia 4.3 3.0 11.9 2.1 1 2.0 7.8 7.3 5.6
Malaysia n.a. 3.0 n.a. 5.8 n.a. 6.3 n.a. 3.8
Philippines 4.6 1.8 8.0 -2.8 7.5 -1.1 5.2 0.0
Singapore 2.8 -3.9 11.9 4.0 13.2 3.3 9.4 6.4
Thailand 4.6 3.7 9.5 5.9 11.2 6.0 7.6 6.4
Papua New Guinea 3.2 2.2 n.a. 5.3 n.a. 1.0 n.a. 2.0

Source: World Bank (1989b).
n.a. = Not available

TABLE 3.4 Structural Change in Output by Sector along the Asia-Pacific Rim, 1965 and 1987 (per cent)






Country 1965 1987 1965-87 1965 1987 1965 1987 1965 1987
Asian NICS  
Hong Kong 2 0 -100 40 29 24 22 58 70
South Korea 38 11 -71 25 43 18 30 37 46
Singapore 3 1 67 24 38 15 29 73 62
Taiwan 18 6 -67 48 51 n.a. n.a. 47 43
South-East Asia  
Indonesia 56 26 -54 13 33 8 14 31 41
Malaysia 28 22 -21 25 38 9 n.a. 47 40
Philippines 26 24 -8 28 33 20 25 46 43
Thailand 32 16 -50 23 35 14 24 45 49
Myanmar 35 37 6 13 16 n.a. n.a. 52 47
Papua New Guinea 42 34 -19 18 26 n.a. 9 41 40
Pacific Rim (Developed countries)  
Australia 9 4 -56 39 33 26 17 52 63
Canada 6 3 -50 41 35 27 19 53 62
Japan 9 3 -67 43 41 32 29 48 57
New Zealand n.a. 8 n.a. n.a. 31 n.a. 21 n.a. 61
United States 3 2 -33 38 30 28 20 59 68

Sources: ADB (1989b): World Bank (1989b).
n.a. = Not available.

Change in employment structure over 1965-80 (Table 3.5) shows that the relative shift from agriculture was achieved through a rise in both the share of employment in industry (except in the Philippines) and in services (except in Singapore). Agricultural employment remained much higher in Thailand than in the other countries. It should be noted that Thai censuses and surveys tend to overstate this figure relative to that in other countries of the region, by taking approaches that capture a very high proportion of rural women in the unpaid family worker' net. Even so, the continuing relatively high share of agricultural employment cannot be disputed; strong growth of agricultural output in Thailand between 1970 and 1985 was not achieved through rapidly rising productivity per agricultural worker (James, Naya and Meier, 1989: Table 5.1) but through continued growth in the agricultural work-force. There is an increasingly wide gap in labour productivity and income between agriculture and other sectors; value added per worker in industry and services is about ten times that in the primary sector (Lo and Kamal Salih, 1987: 48-54). A continuation of the Thai 'success story' will require a shift of large numbers of workers out of agriculture into higher-productivity sectors.

TABLE 3.5 Labour Force by Sector in South-East Asian and Other Countries, 1965 and 1980 (per cent)





Country 1965 1980 1965 1980 1965 1980
Indonesia 71 57 9 13 21 30
Philippines 58 '52 16 16 26 33
Thailand 82 71 5 10 13 19
Malaysia 59 42 13 19 29 39
Singapore 6 2 27 38 68 61
Vietnam 79 68 6 12 15 21
Laos 81 76 5 7 15 17
Cambodia 80 n.a. 4 n.a. 16 n.a.
Myanmar 64 53 14 19 23 28
Papua New Guinea 87 76 6 10 7 14
South Korea 55 36 15 27 30 37
Japan 26 11 32 34 42 55

Source: World Bank (1987b).
n.a. = Not available.

Factors responsible for strong ASEAN growth

In view of the poor economic growth performance of many developing countries in the 1970s and 1980s, it is important to explore the causes of the relatively strong showing of the ASEAN countries. It is also useful to consider why the 'ASEAN 4' have not until now grown as rapidly as the NlCs (for a thoughtful evaluation, see James, Naya and Meier, 1989: Chapter 1; J. A. C. Mackie, 1988). Explanations frequently include the relatively stable political climate, high levels of both public and private investment, efficient use of advanced technology, either home-grown (as with Malaysia's rubber and palm-oil research) or imported, and improving human-resource endowments. Williamson (1993) has stressed the role of the relatively egalitarian income distribution in ASEAN countries. Among other things, this may have fostered the development of education, an important factor in improving the stock of human capital.

Another major influence on rates of development is undoubtedly the degree of outward orientation in trade policy. The World Bank (1987b) uses this criterion to compare the economic growth performance of countries. Three groups are identified:

TABLE 3.6 Classification of South-East Asian Countries by Trade Orientation, 1963-1973 and 1973-1985

Period Strongly Outward- oriented Moderately Outward- oriented Moderately Inward- oriented Strongly Inward- oriented
1963-73 Singapore Indonesia Philippines  
1973 85 Singapore Malaysia Indonesia  
    Thailand Philippines  

Source: As for Table 3.5.

  1. Those strongly outward-oriented where trade controls are non-existent or very low, and there is little or no use of direct controls and licensing arrangements, etc.;
  2. Those moderately outward-oriented where the overall incentive structure is biased towards production for domestic, rather than export, markets, the average rate of effective protection for home markets is relatively low, and the use of direct controls and licensing arrangements is limited; and
  3. Those moderately inward-oriented where the overall incentive structure is distinctly protectionist, a wide range of relatively high and effective protection rates are applied, there is use of direct import controls and licensing is extensive, there is clear bias against exports, and the exchange rate is overvalued.

Singapore has been strongly outward-oriented throughout the 1970s and 1980s, but the rest-the 'ASEAN 4'-have had more difficulty in moving away from heavily protected import-substitution industries (Table 3.6) Malaysia and Thailand have been in the moderately outward-oriented group throughout the same period. Indonesia moved from this group in the 1973-85 period to join the Philippines as moderately inward-oriented, that is, characterized by extensive use of non-tariff trade harriers and price controls (Naya, 1987: 55).

The relative growth of ASEAN countries is consistent with the World Bank's evaluation based on data for a much larger number of countries. Economic performance of the outward-oriented economies has been superior in almost all respects to that of the inward-oriented economies. Outward orientation encourages efficient firms and discourages inefficient ones. By creating a more competitive environment for both the private and public sectors, it also promotes higher productivity and, hence, faster economic growth (see also James, Naya and Meier, 1989: 207-25; World Bank, 1987b: Chapter 5).


In 1982, manufacturing as a share of GDP ranged from 13 per cent in Indonesia to 21 per cent in Singapore. Per capita, manufacturing value added varied enormously, from just $72 in Indonesia to over $1,550 in Singapore at 1982 current prices. All ASEAN countries, with the partial exception of the Philippines, exhibited very rapid industrial growth. Malaysia, Singapore and Thailand have achieved real annual growth of manufacturing in excess of 10 per cent since 1960. Indonesia's sluggish industrial growth of the 1960s, about 5 per cent per annum, accelerated to 15 per cent in the 1970s-the highest in ASEAN and similar to South Korea's (Ariff and Hill, 1985: 25).

Throughout ASEAN, the growth of manufacturing production exceeded that of either services or agriculture over 1965-80. The trends were not so clear in the 1980s, apparently due to the economic slow-down in the mid-1980s. Only in Indonesia and Malaysia did manufacturing output grow faster than that of services and agriculture over this period. In Singapore, it lagged well behind services, and in the Philippines, manufacturing actually experienced negative product growth over the period (for the reasons, see World Bank, 1987a).

Trends in the distribution of manufacturing value added between 1970 and 1987 have been studied across very broad industrial groups (Table 3.7). Unfortunately, these groupings are not disaggregated enough to give a clear picture about the movement into higher value-added areas of manufacturing; for example, electronics is included in machinery and transport equipment, but other activities which are much more traditional in nature are also included in this category. The 'Other' category comprises a wide range of industries: wood and related products, paper and related products, petroleum and related products, basic metals and mineral products, fabricated metal products and professional goods, and other industries.

TABLE 3.7 Structure of Manufacturing Value Added by Industry in South-East Asian Countries, 1970 and 1987 (per cent, current prices)







1970 1987 1970 1987 1970 1987 1970 1987 1970 1987
Food, beverages
and tobacco n.a. 22 26 21 39 43 12 5 43 29
Textiles and                    
clothing n.a. 13 3 6 8 8 5 4 13 1 8
Machinery and transport
equipment n.a. 8 8 22 8 8 28 52 9 13
Chemicals n.a. 9 9 15 13 10 4 12 6 7
Other n.a. 48 54 37 32 30 51 27 29 33
Total manu
facturing   100 100 100 100 100 100 100 100 100

Source: World Bank ( 1990c).
n.a. = Not available.

Even so, some important trends are in evidence, consistent with Ariff and Hill's (1985: 25) observation that 'the period after 1970 witnessed a major reorientation in the structure and specialization of ASEAN manufacturing'. Although food, beverages and tobacco represents the largest subsector of ASEAN manufacturing,4 its share has shown a clear decline in Singapore, Malaysia and Thailand, offset by a rise in the machinery and transport-equipment industry and the chemical industry. In Malaysia and Thailand, the textile and clothing industry has also increased in importance. There has been much less shift in relative importance of the broad industrial groups in the Philippines, probably reflecting her industrial stagnation over the 1980s.

Malaysia's push towards the export market in the early 1970s placed special emphasis on electronics and electrical products. As a result, the share of 'non-traditional' manufactures in total exports increased from 10.4 per cent in 1968 to 21.9 per cent by 1980 (Ariff, 1983: 2). The Philippines and Thailand followed suit in the second half of the 1970s, with the former focusing on the export of manufactured wood products and clothing, and the latter concentrating on textiles, clothing and processed food.

Ariff and Hill (1985: 26) review the changes in these words:

The ASEAN manufacturing sector also displays the impact of policies designed to increase 'industrial deepening', with sectoral emphasis that differs between countries. Singapore has emphasised the production of chemicals, steel products and machinery and equipment, although it disposed of motor vehicle assembly altogether in 1979. In the other four countries, the focus of industrial deepening has been on iron and steel and basic chemicals industries, which are primarily domestic market oriented. Another interesting feature of the ASEAN manufacturing sector has been the growing importance of resource-based industries, reflecting the region's relatively rich resource endowment. Agro-based industries, which process rubber, wood and other agricultural raw materials mainly for the export market, have long been important in the region. The processing of non-renewable resources, on the other hand, represents a relatively recent phenomenon. Indonesia and Singapore are likely to become the region's major centres for petrochemical industries, while Malaysia and Thailand also have plans to set up petrochemical industries based on local natural gas. In addition, chemical fertilizers (of which the ASEAN region will soon become a sizeable net exporter), aluminium smelting (Indonesia), copper smelting (the Philippines) and zinc smelting (Thailand) are all developing rapidly.

Indonesia has undergone impressive industrialization and, because of its size, promises to develop into a significant industrial power in due course. Hill (1990: 79) writes:

The picture is one of dramatic growth and transformation since the late 1960s when Indonesia was one of the least industrialised countries for its size.... Indonesia's industrial transformation is evident not only in rapid output and employment growth but also in a transition to more capital and skill-intensive industries, a narrowing in the earlier very large (almost 'dualistic') productivity differentials, strong productivity and wage growth, a broadening of the industrial base outside of Java, and a probable reduction in concentration levels (at least by establishment).

Hill's analysis of industrialization in Indonesia is based on the 1986 Economic Census results, which also made possible revision of industrial figures for earlier years. These revisions suggest not only a higher share of manufacturing in GDP but also a more rapid industrial growth in the 1980s than was previously estimated (Hill, 1990: 82).

Indonesian manufacturing is still dominated by the oil and gas-processing industries, which together account for over 25 per cent of value added. Output of the next largest industry, food products, is just 22-27 per cent that of oil and gas, depending on whether small industry is included. A group of industries including tobacco products, textiles, wood products, food products (almost 60 per cent of which is sugar processing and coconut-oil production) and basic metals are the next most important in that order, accounting for over 50 per cent of non-oil output (Hill, 1990: 83). This industrial structure accords very much with Indonesia's strong resource base and relatively low per capita income.

ASEAN manufacturing is characterized by great diversity in terms of firm size. With the possible exception of Singapore, all countries exhibit a pronounced dualism in industrial structure, featuring the co-existence of cottage industries and large capitalintensive plants, often within the same industry. In some countries, especially Malaysia, a substantial proportion of large firms is located in export-processing zones. The large firms make the greatest contribution to manufacturing output, but in terms of employment small firms are of considerable importance (Chee, 1986). Cottage and household industries, generally defined as employing less than 10 persons, appear to be a significant source of employment only in Indonesia. Once they are excluded from the analysis, the employment and value-added shares of small-scale industries (SIs), defined as employing 10-49 persons in manufacturing activities, are far lower than those of large-scale industries (LIB, 100 or more persons), but substantially greater than those of mediumscale industries (Mls). In the Philippines, the share of small- and medium-scale industries (SMIs) in manufacturing employment and value added is relatively minor, while in Malaysian manufacturing, on the other hand, SMls appear to be well represented. Singapore lies between the two extremes. In Indonesia, the SMI share in employment is much larger than the share in value added and, to a lesser extent, this is the case in all these countries, indicating that SMI labour productivity is low (Ariff and Hill, 1985: 2832).

Contents - Previous - Next