Policy, Reserch, and External Affairs WORKING PAPERS International Trade Intsrnational Economics Department The World Bank June 1990 WPS 458 Does Japanese Direct Foreign Investment Promote Japanese Imports from Developing Countries? Kenji Takeuchi One wvay for developing cculntries to penetrate the Japanese marketcould be torely onexpaniision of Japan'sintrafirm imports -- particularly for machinery production - from Japanese maniufacturing affiliates in these CouLitries. 'IThe Pcolicy, Researchn, aid xitema 'fd. r ( ' iplc d. , An LR. WI rk,ng IPapcs t odi,srminate Lhe fInd ingi of work un progress and Io encenurge Lhe excharigc of , dilrnr.n Il.i. d! ' . ':: otr cr l d in dec:, p,rr.ent issues These papern carry the names of the authors, reflect only Lhui, s%icss, asL; sNho.c hu Iid ; .imcd a irding] I'he find;ngs, niterpreLauons, and conclusions are the authors own Thes should not hc ai rh.tu,- n dic 'A ,r,d is, k aiaid if Di ors, is managameni, or any of isu menher countnes Policy, Research, and External Affairs International Trade WPS 458 Thispaper-aproductoft.1eInternationallTrddeDivision, International EconomicsDepanment-ispart of alargerefftortinPREtoanalyzetheprospectsfordevelopingcountry exports,particularlymanufactured exports, inmajorindustrial country markets. Copies arc available free t'rom the Worid Bank,1818 H Street NW, Washington DC 20433. Pleasc contact Jean Epps, room S8-037, extension 33710 (41 pages with tables). Japanese direct foreign investment (DFI) in from 30 percent in 1980 to 77 percent in 1986), developing countries has been export-market- precision machinery (rising trom 30 percent to oriented. Exports were the dominant sales 60 percent), and general machinery (rising from destinations for the affiliates in the primary 20-24 percent to 65-75 percent). industries. For manufacturing as a whole, the share In manufacturing, although local markets increased from 15 percent in 1980-83 to more were the dominant sales destinations of'the than 2() percent in 1986. Japanese affiliates, the share of exports increased from 26 percent in 1972 to 42 percent in 1986. Thus for many types of miiachinery produc- The only subsectors in which export's sharc tion, Japanese af'filiates in Asia seem to havc remained below 30 percent in 1986 were iron! become established as a base for exporting to the steel, transport machinery, and chemicals. Japanese market through intrafitm trade. The share of Japanese affiliates in Japan's In soenc other manul'acturing subsectors. imports of manufactures from Asia (where Japanese affiliates have directcd their salzs Japanese manufacturing DFI was most active) is el'lorts to other overseas destinations, gradually found to have been particularly high in thc reducirng the share going to the local market. electrical machinery industry (50-1 00 percent), very significant for transport machinery (rising The PRE Working Paper Series disseminates tilc lindiiigs of work under kas in the Raik's Polics, Research, and External Affairs Complex. An objeetis o of the series is to get thise finidings oLut quickly, even if presentations are less than full% polished. The findings, interpretations, and conlusiSons in thiCse papers do not necessarily represent official Rank polic> IProduced at the I'RE [)issemination Center Does Japanese Direct Foreign Investment Promote Japanese Imports from Developing Countries? by Kenji Takeuchi Table of Contents I. Introduction 1 II. Data Sources 2 III. Japa..ese DFI i- the Postwar Period 5 IV. Impact of Direct Foreign Investment on Japan's 10 Imports of Natural Resources V. Japan's Manufacturing DFI in Developing Countries 20 VI. The Role of Japanese DFI in Japan's Imports of 26 Manufactures from Developing Countries VII. Conclusions 37 References Cited 40 Japanese Direct Foreign Investment: A Promoter of Japanese Imports From Developing Economies? Kenji Takeuchi I. Introdution Japan is the second largest econsiny in the world, a fast growing market for developing countries. Japan's level of manufactured imports relative to GNP is exceptionally low compared to other industrial countries, but the share of developing economies in total manufactured imports has not been lower; moreover, Japan's manufactured imports are growing rapidly, 25X per annum during 1987-89 (US$ terms).' If Japan's ratio of manufactured imports to GNP were to rise in the future to levels of those of other industrial countries, Japan's imports of manufactures from developing economies could be two to three times what they are today.2 The low ratio of manufactured imports to GNP may .tem -- at least partly -- from market behavior. Many studies have found that the Japanese distribution system and practices constitute important market access problems, and have suggested steps that the government could take to improve the openness of the marketing system. Nevertheless, private sector behavior may also affect Economist, International Economics Department, the World Bank. The author thanks Mr. Masataka Fujita of the UN Center on Transnational Corporations for his guidance in collecting data, and Messrs. Bela Balassa, Paul Meo and Shuichi Ono for their comments. I The share of developing economies in total imports of manufactures in Japan has been in the range of 27-32% over the last decade, or about the same as, or ever. higher than, those for most other industrial countries. Japan's ratio of manufactured imports form all sources to GNP was in the range of 2.1- 2.7% in the 1980-87 period as compared with 8.5-10.3% for the industrial countries as a whole and 4.7-7.2% for the United States. See Takeuchi (1989) and Takeuchi (1990). 2 Ta:euchi (1989), Takeuchi (1990). 2 trade relationships.3 One possible avenue for developing economies to enhance their penetration of the Japanese market, therefore, could be to rely on expansion of intra-firm imports from Japanese manufacturing subsidiaries and jcint ventures locaL.i in those economies. How likely are recent increases in Jaipan's direct foreign investment (DFI) in manufacturing in developing economies, especially in Asia, to lead to a significant expansion of Japan's imports of manufactures from these economies through intra-firm trade? The purpose of thi.; paper is to review the impact of japanese DFI on Japan's imports from developing economies in an effort to provide partial answers to that question. It identifies the links -- over time -- between Japanese DFI, and the exports it has generated from developing economies, and Japan's imports from them. The next section of this paper discusses the sources of statistics used in this study. The third section reviews the broad trends in the Japanese DFI in the last three decades. The fourth section reviews the changing impact of Japan's DFI in the natural resource sectors on Japan's imports. The fifth section reviews the trends in Japanese DFI in the manufacturing sector in developing economies. The sixth section examines the market orientation of the Japanese affiliates in developing economies and their role in Japan's imports of manufactures from developing economies; the last contains some conclusions. II. Data Sources There are three official sources for statistics on Japanese DFI; (1) the Ministry of Finance (MOF), (2) the Ministry of International Trade and Industry (MITI) and (3) the Bank of Japan (BOJ). Most DFI data used in this paper use MOF statistics. That ministry publishes statistics on outward DFI flows with industry and country breakdowns on approval/notification basis. These 3 For example, see Kreinin (1988). 3 have been published periodically in special issues of the Zaisei Kinvu Tokei GeRRO, covering Japanese outward DFI since 1951. The MOF defines outward direct investment to iniclude acquisition by Japanese residents of securities issued by non-residents as well as loans made to foreign corporations for the purpose of establishing long-term relationships with them. Furthermore, outward direct investment also includes outflows aimed at establishment of branch offices and factories overseas, even if they are directly owned by the investing home corporations and do not belong to any entities incorporated overseas. Investment in equity shares exceeding 10% (25% before December 1, 1979) in foreign corporations and long term lending to -- and holding of bonds issued by -- such corporations are included in outward DFI statistics. The statistics also include equity shares below 10% (25% before December 1, 1979) in foreign corporations if the investing resident corporation has a "close relationship" with ':te foreign corporations concerned.4 MOF data on DFI outflows are given in current US dollars. The "OF statistics on DFI have three major defects.5 First, firms sometimes do not actually undertake the investment announced, or do so only partially. Second, the actual implementation of an investment is often later - - sometimes much later -- than the time of reporting. Third, the amount of investment includes certain types of loans and acquisitions of bonds which can account for a substantial portion of the total investment. Repayment of these loans or bonds,however, are not covered by the statistics, nor are withdrawals 4 For example, if a representative from the resident corporation is serving as a director on the board of the foreign firm, it is considered a "close relationship." A long-term purchase, licensing or agency contract is also considered a basis for a close relationship. 5 Komiya (1988), pp. 242-247. 4 of (or capital losses on ) equity investment. For these reasons, the MOF statiqtic tend to overstate the actual value of FDI. On the other hand, no direct investment undertaken by overseas subsidiaries is included in MOF statistics on DFI except the _unding was provided by the "parent" firms and reported to the MOF. Annual DFI flows reported in the balance of payments statistics published by the Bank of Japan (BOJ)6 do reflect actual annual outflows of DFI. However, these cover only funds used tr, acquire equity shares of foreign corporations, and do not include reinvest3d earnings. And -- a fatal flow for our purposes -- the BOJ statistics do not give industry breakdowns nor host country breakdowns. MITI conducts questionnaire surveys of Japanese enterprises' activities abroad; a detailed survey once in every three years since 1981, and - -- since the early 1970s -- a less detailed survey in each year except for the years in which detailed surveys are done. Questionnaires are sent to all Japanese companies that hav3 reported to the government their participation in the management of foreign corporations through acquisition of securi.ties of these corporations (MITI, 1986, p.1). While the recovery (or response) ratio of MITI's questionnaire surveys is rather low, roughly 50%, almost all of the large corporations (those listed in the first division of the Tokyo Stock Exchange) always respond, and those which do not are relatively small firms. MITI surveys provide information on the market destinations and intra-firm trade of Japanese DFI subsidiaries among other things. This information is used in this study for identifying the sectoral pattern of Japan's imports from Japanese 6 Bank of Japan, Balance of Payments Monthly, various issues. 5 DFI affiliates in developing countries and the trends in intra-firm imports of Japan. III. Japanese DFI in the Postwar Period Although Japan's DFI outflow was resumed in 1951, its scale remained small throughi the early 1960s; stringent government restrictions were imposed because of the weak balance of payments. In the late 1960s, as Japan's strong economic growth continued and the balance of payments improved, the government began to relax its restrictions and Japan's DFI outflow grew rapidly (Chart 1).7 However, the growth of this outflow was interrupted during the period 1974-1977, with the slower economic growth triggered by the first oil price increase of 1973-74. The fear of even more adverse turns in the balance of payments caused Japan to defer further liberalization of capital outflows. Furthermore, the abrupt slowdown in the world economy and increased uncertainty of demand growth ied private business to slow their investment overseas as well as at home. By the late 1970s, fully recovered from the recession of 1975, the Japanese econc.my was back on a more stable growth th and business confidence was recovered. Beginning in 1978, capital outflow was rapidly liberalized, and Japan's DFI outflow began to recover soon thereafter. The second oil price increase in 1979-1980 and the uncertainty it created again interrupted the DFI growth in the early 1980s. 7 Original MOF data on Japanese DFI are given in current US dollars. Chart 1 shows the same data in terms of constant 1985 yen. The current US dollar values were first converted to current yen values using yearly average exchange rates for the corresponding years, and these yen values were deflated by the BOJ producer price index for capital goods to obtain the constant 1985 yen values, 6 Since 1984, with world economic recoverv, a rising current &ccount surplus, and the sharp appreciation of the yen, Japan's DFI outflows escalated. Japan's DFI outflow exceeded US $20 billion in FY 1984 and reached YS$ 47 billion in FY 1988.8 Today, Japan is the world's largest source of direct foreign investment.9 One notable feature in the evolution o. Japanese DFI is the rise and fall in the share of primary sector. In the latter half of 1960s, the share of DFI in the primary sectcr (agriculture, forestry, fisheries, mining, petroleum and gas) increased (Table 1). For the developing host region, this sector's share remained high through the decade of 1970s, although this share for the industrial host region began to decrease in the second half of the 1970s. This pattern reflected the desire of the Japanese government and businesses for secure, long-term su?plies of primary products, especially energy and raw materials. The concern was heightened, first by the rapidly rising demand for energy and raw materials with Japan's economic boom in the 1960s, but it was sustained in the 1970s by OPEC activities as well as prognostications of resource scarcities.'° In the early 1980s, the share of the primary sector declined for all host regions as concerns over resource scarcities subsided. Surplus conditions persisted in energy and raw material markets, which in turn reflected sluggish 8 The Japanese fiscal year (FY) starts in April and ends in March of the following year; e.g., FY 1988 covers the year for April 1, 1988, to March 31, 1989. 9 According to the IMF (1989, p. 68), direct investment outflow from Japan was SDR 25.5 billion in 1988, as compared to SDR 20.0 billion for the United Kingdom and SDR 13.1 billion for the United States. 10 For the background, see Tsurumi (1976), Vernon (1983), Ozawa (1977), Kojima (1977), and Crowson (1983). 7 CHART 1 Japan s DFI FIows, 1985 Yen All R199Ons 6 - 0 3~~~16 196 191 191 17 95 17 97 9i 18 95 a~~~~~~~~~~~~~~Ta InX ' I ~~ , , , , j , , . 0S~~~95 16 99 91 17 95 17 97 91 18 95 16 3~~~~~~~~~~~ra 8 Table 1. Japanese DFI: Sector Shares, J951-1983 1951f64 1965-69 1970-74 1975-79 1980-84 1985 1986 1987 1988 ,,, .. .- -....... percent .-..-.-.--...----- ALL Countries Primry Sector /a 12.9 29.6 27.8 20.4 12.6 5.3 3.3 2.0 2.? Manufacturing 37.0 21.2 34.3 35.3 28.2 19.3 17.1 23.5 29.4 Real sector total /b 49.9 50.8 62.1 55.7 40.8 24.6 20.4 25.4 32.1 All Others /c 50.1 49.2 37.9 44.3 59.2 75.4 79.6 74.6 67.9 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Develo,ing Countries Primary Sector /a 15.5 36.3 26.3 2°.2 19.2 8.3 5.0 3.2 4A1 Manufacturing 37.5 30.8 48.1 41.1 27.0 18.' 14.6 18.4 21.9 Real sector total /b 53.0 67.1 74.4 70.3 46.2 26.8 19.6 21.6 26.0 All Others /c 47.0 32.9 25.6 29.7 53.8 73.2 80.4 78.4 74.0 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 ;00.0 Industrial Countries Primary Sector /a 6.3 23.5 29.5 8.7 6.1 3.8 2.4 1.4 2.2 Manufacturing 35.4 11.9 19.0 27.7 29.4 19.6 18.3 25.7 32.2 Real sector total /b 41.7 35.4 48.5 36.4 35.5 23.4 20.7 27.1 34.4 All Others /c 58.3 64.6 51.5 63.6 64.5 76.6 79.3 72.9 65.6 Total 100.0 100.0 100.0 100.C 100.0 100.0 100.0 100.0 100.0 /a Primary sector includes egriculture, forestry, fisheries and mining/gas/oil. /b Real sector i7wLudes primary sector and manufacturing. /c All others include construction, commerce, finance, Insurance, real estate, transport, tourism, etc. Source: Ministry of Finance, Zaisei Kinyu Tookei Geppo, various issues. 9 world eaoromic growth and the sustained efforts of users to save energy and raw materials. World energy and raw material capacities had also been increased substantially through increased investment. The impact of Japanese DFI in energy and raw materials on Japan's imports of these commodities will be reviewed in the next section. Japan's direct investment in the primary sector of developing countries declined in the 1980s, not only as a share in total DFI flows to developing countries but also in absolute terms. Measured in constant 1985 yen, Japan's DFI for primary sector in developing coutries declined from 212 billion yen per year in the 1973-1981 period to 100 billion yen per year in the 1982- 1984 period and to 67 billion yen per year in the 1985-1988 period. Another notable trend is that the combined share of sectors other than primary and manufacturing sectors (w'at we call "all others" in Table 1) incrcased markedly in the 1980s, for both industrial and developing host regions. The iicrease of "all others" reflects particularly increased DFI in services; finance, real estate, transport and tourism. The increase in the share of "all others" for the developing region in the l9POs largely reflects the growth of DFI aimed at tax havens (such as the Caribbean countries) and flag-of-convenience shipping." The increase in the industrial countries reflects expanded Japanese financial activities and real estate purchases, especially in Europe and the United States. Japanese DFI in the manufacturing sector has shown .4>-.rgent trends between the industrial and developing host countries. The share of manufacturing 1 For example, Panama, the Bahamas, the Cayman Islands and Liberia together accounted for 44.2% of all Japanese DFI in all sectors in developing countries in 1988. 10 in Japan's DFI flows to develeping countries increased in the 1970s, but declined in the 1980s (Table 1). In industrial countries, the share of manufacturing declined from the mid-l1,,Os to mid-1970s, but has increased since the mid 1970s. This increase reflects, among other things, the concern of Japanese export- orientqd manufaccuring firms with the growth of protectionism in Europe and the United States and attempts by the Japar;ese producers to protect their market shares in these major markets The impact of Japanese manufacturing DFI on Japan's imports of manufactures will be analyzed in detail in the fifth section of this paper. Finallv, it should bo noted that the share of DFI directed towards developing countries has tended to decline since the mid-1970s. This trend has been particularly notable in manufacturing, agriculture/forestry, construXction, and finance (Table 2). IV. Impact of Direct Foreign Investment on Japan's Imports of Natural Resources Because of its poor endowment of natural resources, Japan has long been dependent on overseas sources for supply of raw materials, energy. and food. Trade in fuel and non-fuel minerals has traditional>. been controlled by oligopolies of multinational enterprises. In the 1960)s, in order to increase Japan's relative independence from foreign mnltinationals, the Japanese government encouraged the development of independent Japanese firms engaged in production, primary processing and distribution of these commodities. Thus, in constant yen terms, Japanese DFI in fuels and non-fuel minerals increased more than ten times between the mid-1960s and early 1970s. After peaking around 1972-74, however, the volume of DFI in these sectors stagnated for the rest of 1970s. It peaked again in the early 1980s just after 11 the second major oil price increass, but decreased substantially in the subsequent years. In contrast, Japan's food security policy has been to encourage domestic production while discouraging imports and Japan's DFI in agriculture, forestry and fisheries has been relatively tiny in comparison with that in minerals. Chart 2 shows the 3-year moving average12 of Japanese DFI in minerals, the primary sector as a whole, manufacturing, and all "real" sectors (the primary sector plus manufacturing). What has been the link between this past mineral-oriented DFI and Japanese imports? This section examines the link by reviewing major mineral groups. In the case of crude oil, until the early 1960s, the major international petroleum companies dominated the supply of crude oil to the Japanese market. In 1962, a Petroleum Industry Law assigned MITI a supervisory role over the development of the indigenous petroleum industry -- refinery licensing, financial arrangements for the industry, and approval of production and crude oil acquisition plans.13 Although a significant part of Japan's .efinery industry was Japanese-owned by the early 1970s, almost 80% of Japan's crude oil imports still came from non-Japanese multinational oil companies." In 1967, the Petroleum Development Corporation was created to subsidize overseas petroleum exploration activities, and in 1972 the Petroleum Development Technology Center was established. In 1978, these organizations were incorporated into the Japan National Oil Corporation (JNOC) which had the 12 The three-year moving average is shown to smooth out the lumpiness of yearly data as DFI flows in minerals/energy sector tend to be especially lumpy. '3 Vernon (1983), p. 94. 14 Ibid, p. 95. 12 Table 2. Japanese DFI: Developing Economies' Share in Total, by Sector, 1951-1988 1951-64 1965-69 1970-74 1975.79 1980-84 1985 1986 1987 1988 .........------ percent ------------.---------------------- Manufacturing 72.3 71.3 73.4 66.3 47.1 33.6 28.4 23.5 20.5 Agrlcul/forestry 85.7 86.8 66.2 53.5 42.9 41.7 60.0 22.7 11.7 Fisheries/marine 71.4 55.6 64.7 55.2 68.4 42.9 63.5 54.5 73.4 Mining/gas/oll 86.4 58.1 48.0 86.3 77.0 55.2 49.5 53.0 43.1 Construction 66.7 91.3 70.2 59.9 55.5 57.4 10.8 21.8 34.6 Coarce 11.8 6.0 19.1 15.3 18.0 18.3 18.4 16.6 17.0 Financial/insurance 32.5 60.7 33.4 16.3 18.6 30.0 38.8 28.4 39.4 Other /a 84.9 33.8 46.1 59.7 70.9 43.5 33.0 37.3 24.4 All Secters 71.1 49.0 52.3 57.0 49.3 34.9 33.2 30.0 27.5 a/ Alt other industries not included elsewhere, such as real estate, transport, and tourism. Source: Ministry of Finance, Zaisei Kinyu Tookei Geppo, various issues 13 additional task of stockpiling petroleum. These actions had an effect. In 1968, there were only two Japanese overseas projects producing and exporting oil to Japan (the Arabia Sekiyu project in Saudi Arabia and the North Sumatra Petroleum Development Cooperation project in Indonesia). By 1974, the number of Japanese firms engaged in oil exploration overseas increased to 47.15 The financing provided by the JNOC to Japanese firms between FY 1967 and FY 1987 (inclusive) amounted to 1,268 billion yen; about US$ 10 billion. By 1988, there were 26 overseas companies with Japanese equity participation producing crude oil in 15 countries; they together exported 21.3 million tons of crude oil to Japan in 1988 (Table 3). If an objective of the Japanese petroleum policy was to increase the share of crude oil imports coming from Japanese overseas firms, however, the policy has not been a phenomenal success. Since 1968, the share fluctuated around lOX, never exceeding 13% (see Table 4). The Japanese government, through the Petroleum Development Corporatirn, has also subsidized overseas exploration and development of natural &as supply by Japanese firms since 1972. To ship natural gas in liquefied form requires a high degree of compression and extremely low temperature; a highly capital-intensive technology. Transoceanic transportation of liquefied natural gas (LNG) requires a specialized fleet. Thus, almost all LNG projects are vertically integrated from liquefaction to distribution. Japanese participation in overseas LNG projects has expanded and Japan's imports of LNG have grown steadily -- from less than 1 million tons in 1971 to over 31 million tons in 1988. So far, all of Japan's imeorts of LNG have 15 Vernon (1983), p. 96. Japan's DFI., 1985 Yen 3-year flobmg Average 1.4 - 1.2- 0. 9 .4 0. -t~~~~~~~~~~~~~~~~~~ta o4 0.7 &, 0 0.6 0.5s 0.4 0.3 0.2 1966 196 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 Tear g Real + lIanulacturing 0 Nia/qas/joll Primary 15 come from Japanese DFI-financed projects located in five countries; Brunei, Indonesia, Malaysia, United Arab Emirates and the United States. Apart from LNG, Japan also imports liquefied petroleum gas (LPG), which does not require so much initial capital investment. Development of LPG production and exports has not required Japanese DFI financing. Although there may have been some cases where Japanese DFI was committed for LPG production, data are not readily available. Even if all of Japan's LPG imports are assumed to have been unrelated to Japanese DFI, the share of DFI-based impoLts in Japan's total liquefied gas (LPG plus LNG) imports increased from about 25% in the 1971-73 period to over 60% by early 1980s, and to 70% by 1984, where it has remained since (Table 4). Thus, for liquefied gas, the Japanese policy objective has been achieved. In non-fuel minerals, since early 1960s the government's objective was to enhance the security of long-term supply of these commodities from overseas, as in fuels. The policy was to subsidize the development of new overseas sources to give Japanese users of these raw materials some degree of supply security. In 1963, MITI established the Metal Mining Agency of Japan (MMAJ). Initially it concentrated on subsidizing domestic exploration but by 1968 the agency was promoting exploration and development overseas." By the 1960s, most attractive known ore deposits were already controlled by non-Japanese multinational mineral companies. Therefore, in addition to providing financing for exploration and development of new deposits, the Export-Import Bank of Japan offered subsidized financing to Japanese firms for engaging in long-term purchase contracts with foreign owners who developed new mines. This was called the 16 Crowson (1983). 16 Table 3. Japan's DFI-Based Imports /a/ of Selected Natural Resource Commodities Aluminum Iron Coking Liquefied Crude Year Copper /b Ingots Ore Coal Gas /c Oil --- ( 1,000 t ) --- ----- --- ( mU.l1ion tons )----------- 1960 28 NA NA NA NA NA 1961 34 NA NA NA NA NA 1962 39 NA NA NA NA NA 1963 56 NA NA NA NA NA 1964 52 NA NA NA NA NA 1965 63 NA NA NA NA NA 1966 74 NA NA NA NA NA 19b7 96 NA NA NA NA NA 1968 139 NA NA NA NA 15.5 1969 125 NA NA NA NA 15.5 1970 147 NA NA NA NA 17.3 1971 252 NA NA NA 0.8 16.8 1972 281 NA NA NA 1.0 18.0 1973 425 NA NA NA 1.9 21.2 1974 452 56 NA NA 3.4 23.7 1975 397 56 57.2 18.9 4.6 20.1 1976 468 87 60.3 19.5 5.8 20.7 1977 426 120 57.8 19.1 7.3 20.4 1978 391 127 57.6 17.1 11.2 25.6 1979 302 182 61.2 21.3 13.8 21.4 1980 405 260 69.9 20.8 16.8 19.2 1981 340 290 62.2 23.1 17.0 17.5 1982 325 319 57.2 21.0 17.5 15.9 1983 321 490 55.7 23.3 18.9 20.3 1984 371 550 43.0 32.0 25.9 20.1 1985 299 645 38.8 32.3 27.8 18.2 1986 334 563 37.6 28.0 28.4 20.6 1987 332 555 48.8 30.4 29.1 16.6 1988 404 560 47.2 30.2 31.0 21.6 a/ All imports coming from projects in which Japanese companies have equity participation and/or to which they have made sizable medium- to long-term loans are considered DFI-based imports. This is in line with the Ministry of Finance definition of DFI. b/ Since most of Japanese DFI in copper has been in production of copper ore or concentrates, DFI-based iniports of copper are in concentrates. c/ All Japanese imports of liquefied natural gas have been DFI-based. However, Liquefied gas imports in general include both LNG and lique- -fied petroleum gas. DFI-based imports include only LNG; all LPG imports are considered non-DFI-based for the purpose of this table. 17 develop-for-import formula."7 All of these loans are included in Japan's foreign direct investment data.' The develop-for-import financing -- with or without Japanese equity participation -- has been a significant factor in developing the world's new production ',apacity for minerals, especially for copper, iron ore, coking coal, and, to a much lesser extent, for nickel, bauxite, lead and zinc. Since the mid- 1970s, it has also been a key factor in linking the Japanese aluminum industry with overseas aluminum smelting projects. How successful were these efforts in increasing Japan's DFI-based imports of raw materials? Some pertinent data for aluminum, copper, coking coal and iron ore are presented in Tables 3 and 4 (along with similar data for oil and gas) and discussed below. Japan's DFI-based imports of copper concentrates increased steadily from 28,003 tons (copper content) in 1960 to over 400,000 tons in early to mid- 1970s, but they have since remained at or below 400,000 tons (Table 3). Copper has also been imported in the form of blister and refined copper as well. Japan's overall copper imports peaked at 1.1 million tons in 1973, hut varied thereafter until a new peak of 1.35 million tons was reached in 1984. In the meantime, Japan's imports of copper concentrates peaked at 458,000 tone in 1976. The share of DFI-based copper concentrate imports in Japan's overall copper imports increased from 15Z in the early 1960s to nearly 50% in 1976, but subsequently relapsed to 25-30% in the 1980s. The reasons for the recent decline in the DFI-linked share are associated with the rising energy prices and the 17 See Ozawa (1977), pp. 63-65; also Kolenda (1985). Komiya (1988). 18 Table 4. Share of DFI-Based Imports /a/ in Japan's Total Imports of Selected Natural Resource Commodities Aluminum Iron Coking Liquefied Crude Year Copper /b Ingots Ore Coal Gas /c Oil ----------------------------- percent -------------- 1960 14.9 NA NA NA NA NA 1961 15.1 NA NA NA NA NA 1962 20.4 NA NA NA NA NA 1963 24.9 NA NA NA NA NA 1964 18.0 NA NA NA NA NA 1965 22.4 NA NA NA NA NA 1966 23.0 NA NA NA NA NA 1967 18.4 NA NA NA NA NA 1968 24.5 NA NA NA NA 12.2 1969 19.3 NA NA NA NA 10.3 1970 20.5 NA NA NA NA 9.8 1971 32.8 NA NA NA 24.0 8.7 1972 32.0 NA NA NA 18.2 8.5 1973 37.4 NA NA NA 27.7 8.5 1974 44.4 16.0 NA NA 37.6 10.0 1975 46.4 15.6 47.3 31.0 44.5 8.9 1976 47.7 20.4 47.4 32.9 48.1 8.7 1977 44.3 25.5 49.0 33.9 51.0 8.5 1978 38.3 18.7 52.5 33.7 58.1 11.0 1979 26.1 29.8 51.1 37.6 59.4 9.0 1980 37.0 32.9 56.1 32.1 63.5 8.9 1981 28.7 27.3 53.2 35.1 62.1 8.8 1982 24.6 23.7 52.5 33.8 59.8 8.9 1983 29.3 35.5 53.4 37.2 63.6 11.1 1984 27.4 42.8 36.8 46.4 69.5 10.9 1985 24.4 47.7 33.9 45.2 70.7 10.7 1986 29.5 47.4 36.5 40.8 70.5 12.8 1987 28.6 29.1 44.5 43.6 69.7 10.3 1988 30.1 27.3 40.7 40.8 70.9 12.5 a/ All imports coming from projects in which Japanese companies have equity participation and/or to which they have made sizable medium- to long-term loans are considered DFI-based imports. This is in line with the Ministry of Finance definition of DFI, b/ Since most of Japanese DF1 in copper has been in production of copper ore or concentrates, DFI-based imports af copper are in concentrates, but total copper imports refer to imports of refined and blister copper as well as copper in concentrates. c/ All Japanese imports of liquefied natural gas have been DFI-based. However, Liquefied gas imports in general include both LNG and lique- -fied petroleum gas. DFI-based imports include only LNG and all LPG imports are considered non DFI-based for the purpose of this table. Source: Metal Mining Agency of Japan; Marubeni Corporation; Nippon Steel Corporation; Japan NatioialeOtltfbtpoiation. 19 increased exchange value of the yen, which made it more economical to produce blister or refined copper abroad.g'9 The case of aluminum ingots, which already incorporate energy costs, is similar. Japan's DFI-based imports rose from some 50-60,000 tons in the mid- 1970s to over 600,000 tons in the mid-1980s. They remained at around 550,000 tons in subsequent years. The share of DFI-based imports of aluminum, 15-16% in the mid 1970s, rose to over 40% in the mid-1980s, but subsequently declined to below 30%, as Japan's total ir.ports of aluminum incraeased markedly (from 1.2 million tons in 1986 to over 2.0 million tons in 1988 and 1989). In iron ore and coking coal, which are imported for Japan's steel industry, data on DFI-based imports are available only for the period since 1975. While DFI-based imports of iron ore have been relatively stable at around 40-60 million tons throughout the period, DFI-based coking coal imports have risen from below 20 million tons int he second half of 1970s to over 30 millicn tons in the period since 1984. The share of DFI-based imports in Japan's imports of iron ore from all sources rose from 47% in 1975 to 56% in 1980 but declined to the 35-44% range by the 1984-88 period. Coking cual was similar; the DFI-based share of imports increased fro. 31% in 1975 to 45-46% in 1984-85 but has decreased to 41-44% in the 1986-88 period. To summarize, Japan's DFI in the minerals and energy sector was active from mid 1960s to early 1980s but subsequently dropped both relatively and in absolute terms. DFI in this sector was support'-d by the Japanese government, whose stated objective was to increase the share of imports from 19 For background information on the pattern of world trade in the three forms of copper in the period since 1960, and the role of Japan as an important importer, see Takeuchi, et al (1987), Chapter VII and Annex C. 20 Japanese DFI-based sources in Japan's imports from all sources. This policy initially succeeded more or less in natural gas, aluminum, copper, iron ore and coking coal but was not successful in petroleum. Finally, with the exception of LNG, where vertical integration is almost mandatory, the share of DFI-based imports of all these mineral commodities has declined since mid-1980s as relative economic costs changed. Far from establishing import monopolies for Japanese firms, the evidence from Japan's raw material oriented DFI seems to indicate that although a stated policy and subsidies o. the government led to a successful increase in Japanese firms' import shares, their shares tended to decline when investments in these sectors became less attractive due to perceived worsening of the long term market outlook for the commodities in question. Will the recent apparent boom in Japan's manufacturing DFI, particularly to Asia, lead to greatly increased imports of manufactures? The next section reviews the role of Japanese DFI in the growth of manufactured imports from developing countries over the last two decades. V. Japan's Manufacturing DFI in Developing Countries Japan's DFI in manufacturing in all host country groups combined has been growing steadily since the mid 1960s; the growth accelerated since the mid 1980s. It not only increased in terms of current US dollars, it has increased in constant US dollar and yen terms as well (Table 5). However, there have been contrasting trends in Japan's manufacturing DFI flows between those destined to industrial and developing countries. The flows to industrial countries have steadily increased throughout the entire 21 Table 5. Japan DFI in Manrufacturing, by Host Region, 1965-1988 Period Averages (Per Annum) Middle Developing North Industrial World Period Asia LAC East Africa Total America Europe Oceania Total Total *-*-**-*-----*-. . .-----*--- Miltions ofCurrentUSDoLlars PerAnnum- ---- 1965-69 33 21 0 2 57 14 2 8 23 80 1970-74 268 208 20 6 503 107 39 35 181 684 1975-79 456 259 174 7 896 271 94 90 455 1,351 1980-84 642 331 56 25 1,054 891 216 75 1,182 2,236 1985-88 1,328 300 4 4 1,636 4,365 773 174 5,312 6,948 .*--------------------------- Millions of 1985 US DoIlars Per Annum ------------------------- 1965-69 96 64 1 7 168 41 5 23 69 238 1970-74 644 49e 46 15 1,202 261 94 83 437 1,639 1975-79 723 416 273 t1 1,424 423 145 140 708 2,132 1980-84 715 359 63 28 1,165 965 236 86 1,287 2,452 1985-88 1,273 291 4 4 1,572 4,173 740 167 5,080 6,652 -------------------------------- BilLions of 1985 Yen Per Annum --.----------------- . 1965-69 20 13 0 1 34 8 1 5 14 49 1970-74 117 87 8 3 214 47 17 16 80 294 1975-79 125 74 47 2 248 74 25 25 124 372 1980-84 152 79 13 6 250 211 51 18 280 529 1985-88 205 52 1 1 259 661 120 26 807 1,065 Source: Ministry of Finance, Zaisei Kinyu Tokei Geppo, various issues; Sank of Japan, Economic Statistics Annual, various issues; United States Council of Economic Advisors, Economic Report of the President, 19n0 22 period, both in terms of value and number of cases (Table 5).20 Conversely, the flows to developing countries have been relatively stagnant, at least, since the mid 1970s. As a result of these divergent trends, in the most recent period (1985-88), North America alone accounted for two-thirds of all Japanese DFI in manufacturing (Table 6). Although Japan's manufacturing DFI flows to developing countries as a group have stagnated since the mid-1970s, those going to Asia have been increasing steadily, in constant yen and $ terms and number of investment cases as well (Table 7). Thus, Asia's share in the developing economies group rose from 50-55% in the 1965-1979 period to over 80% by the 1985-88 period. The most important host economies in Asia in the recent years have been Thailand, Indonesia, Korea, Singapore and Taiwan21 (see Table 7). The decline in Japan's manufacturing DFI to other developing regions, especially Latin America and the Caribbean and the Middle East, has been dramatic. Over the last two decades the sectoral composition of Japanese manufacturing DFI flows to developing economies has also changed considerably. In the early 1970s, che;nicals, textiles and metals were the most important subsectors (Table 8). In subsequent years, while the share of textiles decreased rapidly the shares of chemicals and metals increased further in the second half of the 1970s. In the early 1980s, metals remained the most important 20 The number of investment cases directed towards industrial countries increased from 97 per annum in the 1970-74 period to 356 in the 1980-84 period and to 599 in the 1985-88 period. 21 Data for Japan's DFI to Taiwan are not available separately for manufacturing for the 1987-88 period. However, based on the fact that total Japanese DFI (including manufacturing as well as others) in Taiwan is reported to have increased frouw $405 million in the 1985-86 two-year period to $739 million in the 1987-88 two-year period, Japan's DFI for manufacturing in Taiwan is very likely to have increased substantially between the two periods. 23 Tabl 6. GeographIcal Distribution of Japans Manufacturing DfI, 1951-1988 Share in World Total, by Host Region Developin9 Countries Industrial Countries ........................................ ................................................................................. Middle Developing North Industrial Period Asia LAC East Africa Total America Europe Oceania Total ,,,,,,,,,.........--- Percent ...........-e''''''''''' 1951-64 21.0 49.1 0.0 2.7 72.9 24.7 2.1 0.3 27.1 1965-69 40.6 26.7 0.5 3.0 70.8 17.0 2.0 10.2 29.2 1970-74 39.2 30.4 3.0 0.9 73.5 15.7 5.7 5.1 26.5 1975-79 33.8 19.2 12.9 0.5 66.3 20.1 7.0 6.6 33.7 1980-84 28.7 14.8 2.5 1.1 47.1 39.8 9.7 3.4 52.9 1985-88 19.1 4.3 0.1 0.1 23.5 62.8 11.1 2.5 76.5 Source: Ministry of Finance, Zaisei Kinyu Tokei Geppo, various i.sues. 24 TabLe 7. Japanese Manufacturing DFI in Asia, by Major Host Country, 1951-88 Hong Indo Phiip- Asia Period Kong Singapore Korea Taiwan China Thailand Malaysia -nesia -pines Total (inclisive) ----------------------------- Millions of Current US Dollars ----------------------- 1973-76 64 146 292 111 0 71 154 550 78 1,496 1977-80 85 467 295 134 1 120 251 843 143 2,353 1981-82 30 323 59 96 8 99 77 476 55 1,230 1983-84 19 342 69 130 22 118 227 268 20 1,258 1985-86 66 198 178 385 46 112 97 93 57 1,265 1987-88 193 441 501 NA 273 836 NA 593 NA 4,049 Source: Ministry of Finance, Zaisei Kinyu Tokei Seppo, various issues. 25 Table 8. Subsector Shares in Japan's Manufacturing DFI in Developing Countries, 1965-1988 Period /a 1965-68 1969-72 1973-76 1977-80 1981-82 1983-84 1985-86 1987-88 Subsector ---------.-------------- ------- Percent -------------------------------- Food Processing 5.9 4.0 4.0 2.7 2.4 3.5 4.0 5.5 Textiles/Garment 23.4 35.5 19.8 7.1 5.1 9.2 2.3 4.2 Wood/pulp/paper 1.6 4.5 6.8 1.9 1.0 1.5 0.7 4.1 Chemicats 6.4 9.8 23.2 32.7 17.6 18.7 4.7 10.3 Metals, ferr/nronfer 17.6 11.9 14.2 29.1 40.8 28.8 17.3 16.4 General Machinery 8.0 6.3 6.3 5.9 7.0 8.6 9.7 8.2 Electrical Machinery 10.6 12.4 9.7 8.1 8.0 7.5 18.1 31.9 Transport Machinery 21.3 6.7 7.3 6.1 11.9 16.8 31.8 9.3 Manufacturing Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 a/ Periods are inclusive of fiscal years indicated. Source: Ministry of Finance, Zaisei Kinyu Tokei Geppo, various issues. 26 subsector but the share of chemicals declined. In the meantime, machinery subsectors2 steadily increased their importance over the two decades. Notably, transport machinery's share rose from 6-7% in the 1970s to over 30% by 1985-86. In particular, electrical machinery's share, which remained below 10% up to l1(83- 84, increased to 18X in 1985-86. 'ihe share of general machinery similarly edged up gradually in the 1980s. VI. The Role of Japanese DFI in Japan's Imports of Manufactures from Developing Countries The desirability for developing countries to have foreign direct investment in export industries as an effective means to enter world markets was questioned by Nayyer (1978), who conlcuded that the contribution that foreign direct investment b-y multinationals had made toward the expansion of the manufactured exports of host developing countries had been relatively modest, that the share of foreign-owned subsidiaries in manufactured exports of developing countries had tended to decline over time and that, considering the "costs" involved, the inducement of foreign investment in manufacturing export industries of these countries might not be the best policy for them unless more convincing evidence was found. A recent study by Blomstrom, Kravis and Lipsey (1988) re-examined the issue and found that during the 1957-1985 period the shares of United States, Japanese and Swedish owned firms in developing countries' manufactured exports had tended to increase. Unfortunately, this 2 The MITI source materials subdivide the machinery sector into four subsectors, i.e., (a) electrical machinery industry (which includes electronics equipment/apparatus), (b) transport machinery industry (which includes automobiles, bicycles, ships, aircraft, among others), (c) precision machinery industry (which includes notably cameras. other optical equipment, watches and clocks, among others) and (d) general machinery (all others not included in the first three categories mentioned above). 27 (hence, Nayyer's) approach does not show whether foreign-owned subsdidiaries are more export-oriented than local (non foreign-owned) firms. One case study of Thailand, Siburuang (1986), however, found that "TNCs and FDI (had) played an important role in generating manufactured exports from Thailand throughout the decade (of the 1970s)." (p. 250). How important has the export market (Japan and elsewhere) been for Japanese affiliates in developing countries? To examine this question, relevant data available in the various issues of MITI's surveys have been extracted and these are summarized in Table 9, which show the share of exports in total sales of Japanese DFI affiliates located in developing countries by sectors for 1972, 1980, 1983 and 1986. First, the popular impression that Japanese affiliates engaged in primary commodity production concentrate on exports, rather than local sales, is confirmed by the data. Roughly 90X of sales in the agriculture/forestry/fisheries sector and the mining/oil/gas sector has been destined to export markets, although the export share for the mining/oil/gas sector may have begun to decline in the recent few years. Second, for the manufacturing sector, local sales rather than exports have been the more important outlets for the Japanese affiliates in developing countries. The importance of exports in the total sales of Japanese manufacturing DFI affiliates in developing countries has varied from subsector to subsector and over time (see Table 9). On the one hand, export orientation has been verv strong in the natural-resource-based processing subsectors, i.e., food processing, wood processing (plywood, pulp/paper), and nonferrous metals 28 Table 9. Share of Exports in Total Sales of Japanese DFI Affiliates in Developing Countries, 1972, 1980, 1983 and 1986 Sector/Subsector 1972 1980 1983 1986 - percent --------------- Agriculture/Forest/Fish 91.8 91.4 87.8 92.5 Mining/Oil/Gas 96.0 90.6 93.2 67.6 Manufacturing 25.5 33.0 35.1 42.3 Food Processing (51.8) /a 63.1 41.0 76.7 Textile/Apparel 31.8 33.8 34.7 51.6 Wood Processing 91.7 62.8 65.3 57.1 Chemical 15.1 16.0 16.4 27.5 Iron & Steel (16.1) /a 16.0 16.4 16.5 Nonferrous Metals (25.0) /a 45.8 38.8 63.8 General Machinery 7.5 21.9 25.1 43.8 Electrical Machinery 34.0 37.5 57.0 53.1 Transport Machinery 8.9 7.7 10.6 22.9 Precision Machinery 49.2 41.6 80.2 51.6 a/ Data for all countries for 1974. Source: MITI, Kaigai Jigyo Katsudo Kihon Chosa, 1st (1983), 2nd (1986), 3rd (1989); MITI, Wagakuni Kigyo no Kaigai Jigyo Katsudo, 3rd (1974). 29 (notably, aluminum smelting).= The export share of total sales has also been high for textiles, electrical machinery and precision machinery (e.g. cameras and watches). Labor-intensiveness is a common characteristic of these last two industry groups. On the other hand, the export share has been relatively low for three industries, i.e., chemical, iron/steel and transport machinery. DFI in these industries have largely been made in response to import substitution incentives provided by host countries. Third, the export share in the sales of Japanese affilia-tes has increased for most subsectors over time. The export share has clearly increased over time in textiles, electrical machinery, precision machinery and general machinery. Furthermore, the export share for transport machinery increased visibly in the 1980s. Thus, the export share has indeed increased for most industries, and by 1986, the only industries for which the export share still remained below 30% were chemical, iron/steel and transport machinery. How important have exports to Japan been in the total sales of Japanese affiliates ili developing countries? Table 10 shows the share of exports to Japan in the sales of Japanese affiliates located in all developing countries and the share of exports to Japan in the total exports of these affiliates, in 1930, 1983 and 1986. For the manufacturing sector as a whole, about 40% of total sales were exported, and 40Z of the exports were destined to Japan; thus, about 15% of total sales of these affiliates were shipped to Japan. The industries which recorded relatively high shares of exports to Japan in their total sales were industries which recorded relatively high shares 23 Because Japanese foreign direct investment statistics include nonferrous metals in the manufacturing sector, smelting and refining of nonferrous metals are included in manufacturing for the purpose of this section. 30 Table 10. Shares of Sales to Japan and All Exports in Total Sates of Japan.s. Affiliates Located in Developing Countries, by manufacturing Subsector Share of Exports to Share of Exports in Share of Exports to Japan Japan in Total Sales Total Sales in Total Exports of Affiliates of Affitiates of Affitiates 1980 1983 1986 1980 1983 1986 1980 1983 1986 -- percent ....... ---.---.percent -------- ----.--.percent -----.-- Manufacturing Total 13.0 14.6 16.4 33.0 35.1 42.3 39.4 41.6 38.8 Food Processing 26.9 15.1 26.5 63.1 41.0 76.7 42.6 36.8 34.6 Textiles 3.3 4.6 9.7 33.8 34.7 51.6 9.8 13.3 18.8 Wood/pulp/paper 36.0 43.3 28.8 62.8 65.3 57.1 57.3 66.3 50.4 Chemicals 8.8 9.5 8.1 16.0 16.4 27.5 55.0 57.9 29.5 Iron/steel 7.4 5.7 3.6 16.0 16.4 16.5 46.3 34.8 21.8 Nonferrous metals 29.8 16.0 28.8 45.8 38.8 63.8 65.1 41.2 45.1 General machinery 3.3 14.1 24.5 21.9 25.1 43.8 15.1 56.2 55.9 Electrical machinery 12.4 19.1 20.4 37.5 57.0 53.1 33.1 33.5 38.4 Transport machinery 1.4 4.1 3.9 7.7 10.6 22.9 18.2 38.7 17.0 Precision machinery 8.6 29.0 21.7 41.6 80.2 51.6 20.7 36.2 42.1 Source: MITI, TARS 31 of exports in total sales. When the share of exports to Japan in the total sales was regressed on the share of exports to all destinations in the total sales, a correlation coefficient of 0.7 (uncorrected) was found (data for 1980, 1983 and 1986 for 10 industries were pooled for 30 observations). However, when the share of exports to Japan was regressed on the share of exports to all destinations in the total sales, the resulting correlation coefficient was low (0.06) and highly insignificant. This implies that at least in the 1980-1986 period, export sales to Japan were not a common driving force for the export activities of these affiliates, although it may have been such a driving force for some specific industries and specific firms. Japanese DFI affiliates in Asia have been more export-oriented than those in the LAC region. Between 1972 and 1986, the share of exports in total sales of Japanese manufacturing DFI affiliates in Asia was consistently higher than that of those located in the LAC region (Table 11). Also, the share of exports to Japan in the affiliates' total exports was consistently higher for those in Asia than for those in the LAC region. Within Asia, Urata (1989) has found a contrasting pattern in the importance of Japan as a destination of the affiliates' exports between ASEAN countries and the rest of Asia. For the affiliates in ASEAN countries, the share of exports to Japan in their total sales is especially high in the wood processing and non-ferrous metals subsectots. In contrast, for the affiliates in non-ASEAN countries (where Korea, Hong Kong and Taiwan are dominating), the share going to the Japanese market is relatively high in general machinery, 32 Table 11. Share of Total Exports and Exports to Japan in Total Sales of Japanese manufacturing DFI Affiliates and Share of Exports to Japan in Their Total Exports: Asia Vs. LAC Asia LAC Share in Total Sales Share of Share in Total Sales Share of of DFI Affiliates Exports to of DFI Affiliates Exports to ................................. Japan in . . . . Japan in Exports Exports to All Exports of Exports Exports to All Exports co. to Japan Others Exports Affiliates to Japan Others Exports Affiliates (X) (X) (X) (X) (X) (X) (X) (X) 1972 5.8 31.2 37.0 15.7 1.0 8.2 9.2 10.9 1974 26.3 20.5 46.8 56.2 7.3 9.9 17.2 42.4 1975 23.8 18.8 42.6 55.9 4.5 9.8 14.3 31.5 1977 10.0 23.3 33.3 30.0 1.1 7.0 8.1 13.6 1978 8.9 23.7 32.6 27.3 1.0 7.3 8.3 12.0 1980 9.8 26.4 36.2 27.1 9.4 8.0 17.4 54.0 1983 10.8 22.3 33.1 32.6 12.2 15.8 28.0 43.6 1986 15.8 29.5 45.3 34.9 4.1 15.4 19.5 21.0 1987 16.7 24.3 41.0 40.7 NA NA NA NA Source: MITI, Wagakuni Kigyo no Kaigai Jigyo Katsudo, various issues; MITI, Kaigai Jigyo 'atsudo Kihon Chosa, 1st, 2nd and 3rd. 33 electrical machinery, precision machinery as well as in food processing and non- ferrous metals. Urata argues that the above contrast is based on the fact that non-ASEAN Asian countries have a comparative advantage in the production of capital-intensive and human-capital-intensive products while ASEAN countries have a comparative advantage in the production of natural resource based and labor- intensive products (Urata, 1989, p. 29). How important are imports from Japanese affiliates in Japan's imports of the same products from all sources? Because of the apparent importance of this question for Japanese manufacturing DFI in Asia, relevant data for Asia have been collated in Table 12. For various subsectors, imports from Japanese affiliates in Asia, total imports from Asia and the share of the former in the latter are shown. Data presented in Table 12, however, have some major problems which must be acknowledged outright. First, because of very large discrepancies between the data for Japanese imports as reported by the importer (Japan) and those reported by the exporters, Japan's imports by commodity are shown both ways. Since the data for Japan's imports from Japanese DFI affiliates are based on the reporting from these affiliates located in these Asian countries (taken from the MITI surveys), the Japanese imports (namely, exports to Japan) as reported by the exporting countries are just as relevant as the imports reported by Japan for the purposes at hand. Second, it should be recalled that the response ratio of the MITI questionnaire surveys (on which the data on imports from affiliates are based) has been rather low. Therefore, the amounts shown for imports from affiliates tend to understate the true values. Third, while imports from affiliates are reported on a fiscal year basis, both Japan's imports as reported by Japan and those as reported by the exporters are based on calendar year data. 34 Table 12. Share of Imports from Affiliates in Japan's Imports from Asia by Manufacturing Sub Sector, 1980, 1983 A 1986 Imports from Asia Imports from ............ .......Share of Imports Fiscal Affiliates As reported As reported by from Affiliates Sector/subsector Year (A) by Japan(8) Exporters(C) A/B A/C - - ---hMillion US S - ---- percent - Manufacturing Total 1980 1,081 6,376 7,244 16.9 14.9 (SITC 5+6+.78+9) 1983 1,061 7,073 7,897 15.0 13.4 1986 3,003 12,966 13,746 23.2 21.8 Food Processfne 1980 106 1,573 1,371 6.8 7.8 (ISIC 311+312+313) 1983 40 1,693 1,548 2.4 2.6 1986 80 2,545 2,328 3.1 3.4 Wood Processing 1980 33 248 232 13.4 14.4 (inct. pulp/paper) 1983 29 204 246 14.0 11.6 (SITC 25+63+64) 1986 30 482 550 6.3 5.5 Chemical 1980 85 795 1,228 10.7 6.9 (SITC 5) 1983 103 845 735 12.2 14.0 1986 83 1,366 1,307 6.0 6.3 Iron & Steel 1980 60 458 424 13.2 14.2 (SITC 67) 1983 53 672 617 7.8 8.5 1986 38 896 878 4.3 4.3 Nonferrous Metals 1980 8 632 547 1.2 1.4 (SITC 68) 1983 2 671 641 0.3 0.3 1986 204 639 563 32.0 36.3 Textile/Apparel 1980 77 2,188 2,152 3.5 3.6 (SITC 65+84) 1983 88 2,310 2,329 3.8 3.8 1986 224 4,242 4,177 5.3 5.4 General Machinery 1980 32 123 177 25.9 18.0 (SITC 71) 1983 47 170 274 27.9 17.3 1986 404 526 608 76.8 66.4 Electrical/Electronics 1980 524 673 652 78.0 80.4 Machinery 1983 473 743 807 63.7 58.6 (SITC 72) 1986 1,499 1,180 1,537 127.0 97.5 Transport Machinery 1980 21 69 67 29.8 30.7 (SITC 73) 1983 81 78 183 104.7 44.3 1986 139 91 182 153.0 76.7 Precision Equipment 1980 45 168 161 27.1 28.3 (SITC 861+864) 1983 48 171 167 28.2 29.0 1986 179 300 296 59.6 60.3 Notes: (a) Inports from affiliates are reported in yen on fiscaL year basis. Yen values have been converted to USS values using calendar year average exchange rates. (b) The FY equivalent of Japan's imports which are reported on calendar year basis has been obtained by adding one fourth of the following CY's value to three fourths of the current CY's value. Source: MITI Surveys and UN Trade Statistics 35 These calendar year data were crudely adjusted to derive the fiscal year equivalents.24 Because of these data problems, we must be careful in drawing too precise conclusions from the data presented in Table 12. Nevertheless, some broad conclusions can be safely drawn. First, for the manufacturing sector as a whole, the share of imports from affiliates ranged around 15-20% and the share seems to have increased between 1983 and 1986. Second, in the four machinery subsectors, the share of imports from affiliates has been highly significant; over 60X by 1986. In the electrical machinery subsector, it has been the dominating force. Tran Van Tho argues that the share of imports from affiliates is particularly high in the various machinery subsector because these industries "are characterized as industries of multiproduction stage which have different factor intensities" and that "this characteristic tends to induce, or facilitate, the intra-firm division of labor.. ." (Tran Van Tho, 1987, p.36) Third, in other subsectors, imports from affiliates have been a relatively minor factor, perhaps with the exception of nonferrous metals subsector in 1986. It is conspicuous that the share has been particularly low in the case of textiles/apparels, only 4-5X. It has been popularly supposed that predominant portions of the products exported by Japanese affiliates to Japan represent their sales to their parent companies in Japan. This popular supposition is confirmed by the data presented in Table 13, which shows the share of affiliates' exports to parent companies in the exports of the affiliates in Asia to Jap.'n in 1974, 1980, 1983 and 1986, by industry. For manufacturing as a whole, the share remained close to 90% in 1974 and 1980, but declined somewhat in 1983 and 1986, to around 75%. > See notes to Table 12 for the method of conversion used. 36 Table 13. Share of Exports to "Parents" in Total Exports to Japan from Affiliates in Asia 1974 1980 '983 1986 - percent ------------- Manufacturing Total 87.8 89.2 74.4 76.5 Food Processing 88.0 80.4 64.1 87.0 Textiles /a 84.7 86.2 74.9 57.7 Apparels /a 78.4 NA NA NA Wood Processing /b 73.6 100.0 89.5 27.7 Pulp/Paper /b 83.8 NA NA NA Chemical 82.0 79.4 49.8 83.9 Iron/Steel 100.0 85.4 97.9 100.0 Nonferrous Metals 99.5 93.3 77.1 99.2 General Machinery 98.9 100.0 88.8 94.7 Electrical Machinery 85.6 96.0 76.3 73.0 Transport Machinery 99.2 51.2 67.3 46.0 Precision Mach.nery 98.6 83.9 91.5 86.1 Other Misc. NA /c 76.8 67.5 88.5 a/ Textiles include secondary manufactures such as apparels, except for 1974. For 1974, data for apparels and other secondary manufactures are shown separately. b/ Wood processing includes mechanically processed products such as plywood, veneers, particleboard, wooden furniture as well as pulp and paper, except for 1974. For 1974, pulp and paper are shown separately. c/ Industry classification for 1974 is generally more detailed than that for 1980 and after. Especially data for "other misc," for 1974 are not comparable for data for the same category for other years. Source: MITI, Kaigai Jigyo Katsudo Kihon Chosa, 1st (1983), 2nd (1986) and 3rd (1989); MITI, Wagakuni Kigyo no Kaigai Jigyo Katsudo, 5th (1976). 37 Among the subsectors, textiles, wood processing, electrical machinery and transport machinery followed this pattern of declining share. VII. Conclusions Japanese outward DFI flows, measured in constant yen, increased in the 1965-1973 period but stagnated in the 1974-1980 period. They increased rapidly in the 1980s, especially after 1983. However, the most important thrust of the rapid increase in the 1980s was a flood of DFI flows in the service sectors, especially financial and real estate sectors. Japanese outward DFI in the primary sector, part; ularly mining, gas and oil, was boosted steadily in the decade from mid 1960s to mid 1970s, mainly because of the concern for the long-term security of raw materials supply on the part of both the government and the private sector. This concern lingered on through the early 1980s with respect to energy (oil, gas, coal, uranium, etc.) and aluminum smelting, which is highly energy intensive. This led to a burst of bunching of overseas investments in the early 1980s. A major objective of the Japanese government in promoting direct investment by Japanese firms in overseas production of fuel and nonfuel minerals was to increase the share of Japanese DFI-based supply in Japan's imports of these commodities. This objective was successfully achieved in several key commodities such as copper, aluminum, iron ore, coking coal and natural gas but, in the case of the most important commodity, crude oil, the achievement was only modest, at best. Even in the commodities where the share of DFI-based imports became large, it peaked in the early to mid 1980s as the concern of the long- term supply security subsided in the face of prolonged surplus market conditions. 38 Japanese outward DFI in manufacturing, measured in constant yen, began to grow steadily in the late 1960s. A peak around 1972-1974 was followed by a brief stagnation in 1975-1976, but steady growth was resumed in 1977 and the pace accelerated in the latter half of the 1980s, in contrast to the stagnation for DFI in the primary sector. However, this steady growth in manufacturing DFI since the mid 1970s hides some divergent geographical trends. Notably, the overall steady growth reflected a sharp increase in manufacturing DFI flows to industrial countries (particularly Europe and the United States) which were likely mainly motivated to protect market shares in the face of growing new protectionism there. Japanese manufacturing DFI flows to developing countries have shown only a modest growth since early 1970s. The significant growth in the flows to Asia has been offset by a decline in the flows to developing countries in other regions. The sub-sectoral composition of Japanese manufacturing DFI in developing countries has changed considerably over the last two decades or so. In the 1965-1972 period, the mast important subsectors were textiles and clothing, metals (ferrous and nonferrous) and transport machinery. In the 1972- 1980 period, the importance of textiles and transport machinery subsided, and the importance of chemicals and metals increased. In the 1981-1984 period, metals and chemicals continued to be the most important subsectors but transport machinery's importance was revived. Since 1985, electrical machinery became the leading subsector. Japanese DFI in developing countries has been significantly export- market oriented. Exports were the dominant sales destination for Japanese affiliates in the primary industries, agriculture/forestry/fisheries and mining/oil/gas. In manufacturing, the share of exports increased steadily -- 39 from 26X in 1972 to 42X in 1986. In the majority of manufacturing subsectors, the share of exports in the total sales of Japanese manufacturing affiliates increased significantly between 1972 and 1986. The only subsectors in which the export share did not exceed 30X in 1986 were iron/steel, transport machinery and chemicals. The role of Japanese affiliates in Japan's imports of manufactures from developing countries was examined, subsector by subsector, for Asia (where Japanese manufacturing DFI was most active among developing countries), based on the data available for 1980, 1983, and 1986, This share was very high in the electrical machinery industry (50-100), significant for transpo.t machinery industry (increasing from 302 to 772), and for the precision machinery subsector (increasing from 301 to 602) . For general machinery, the share increased from around 20-242 to the 65-75X range. In other subsectors, th.s share is found to have been relatively minor, usually less than 20% (except for non-ferrous metals in 1986). For the manufactures as a whole, the share increased from around 15% in 1980-1983 to over 20X in 1986. In sum, for many types of machinery production, Japanese affiliates in Asia seem to have become established as a base to export to the Japanese market. In some other manufacturing subsector, Japanese affiliates have directed their sales efforts to other overseas destinations gradually reducing the share going to the local markets. 40 References Cited Blomstrom, Magnus, Irving B. Kravis and Robert E. Lipsey (1988), "Multinational Firms and Manufactured Exports from Developing Countries," National Bureau of Economic Research Working Paper No. 2493, January 1988. Bank of Japan, Balance of Payments Monthly, various issues. Bank of Japan (1990), Economic Statistics Annual 1989, Tokyo, March 1990. Crowson, Phillip, "Non-Fuel Mineral Procurement Policies," in Nobutoshi Akao (ed.), Japan's Economic Security, London, Cower, 1983, pp. 145-167. Goto, Akira (1988), "Japan: A Sunset Industry," in Merton J. Peck (ed.), The World Aluminum Indus;ry in a Chancing Energy Era, The Johns Hopkins University Press, 1988, pp. 90-120. Hyun, Jung Taik, and Katharine Whitmore (1989), "Japanese Direct Foreign Investment: Patterns and Implications for Developing Countries," the World Bank Industry and Energy Department Working Paper Industry Series Paper No. 1, February 1989. International Monetary Fund (1989), International Financial Statistics. Yearbook 18, Washington, D.C. International Monetary Fund (1989), Balance of Payments Statistics, vol. 40, Yearbook, Part 2, 1989. Japan, MITI (1986), Dai-ni-kai Kaigai Jigvo Katsudo Kihon Choosa. Kaigai Tooshi Tokei Soran (Second Basic Survey on Overseas Investment: Comprehensive Statistics), Keibun Shuppan, 1986 (in Japanese). Japan, MITI (1989), Dai-San-kai Kaigai Jigyo Katsudo Kihon Choose-Kaizai Toshi Tokei Sogan (The Third Basic Survey on Overseas Investment: Comprehensive Statistics, Keibun Shuppan, 1989. Japan, MITI (July 1989), "Dai 18-kai Wagakuni Kigyo no Kaigai Jigyo Katsudo Dookoo Choosa" in MITI, Industrial Policy Department (Ed.) Kyozon-teki Kyoso-eno Michi (Globalization Re2ort), Tsusho Sangyo Chosa-kai, July 1989. Japan, Ministry of Finance, Zaisei Kinvu Tookei Geppo, no. 305 (September 1977); no. 356 (December 1981); no. 428 (December 1987); r.o. 452 (December 1989). Japan, Ministry of Finance, Showa 63-nendo ni okeru Taigai oyoi Tainai Chokusetsu Tooshi Todokede Jittseki (Reported Outgoing and Incoming Direct Foreign Investment for FY 1988), Press Release, May 30, 1989. Japan National Oil Corporation (JNOC), Sekivu Koodan no Gaivo (Japan National Oil Corporation: A Resume), JNOC, July 1988. 41 Kolenda, T.E. (1985), 'Japan's Develop-for Import Policy," Resources Policy, vol 11, no. 4, December 1985, pp. 257-266. Komiya, Ryutaro (1988), "Japan's Foreign Direct Investment: Facts and Theoretical Considerations," in Silvio Borner (ed.), International Finance and Trade in a Polycentric World, Macmillan, 1988, pp. 241-289. Kreinin, Modechai E. (1988), "How Closed is Japan's Market? Additional Evidence," The World Economy, vol 11, no. 4 (December 1988), pp. 529-542. Lawrence, Robert Z. (1987), "Imports in Japan" Closed Markets or Minds?" Brookings Papers on Economic Activity, no. 2, 1987, pp. 517-554. Nayyer, Deepak (1978), "Transnational Corporations and Manufactured Exports from Poor Countries," Economic Journal, vol. 88, March 1978, pp. 59-84. Ozawa, Terutomo (1977). "Japan's Resource Dependency and Overseas Investment," Journal of World Trade Law, vol. II, no. 1, January-February 1977, pp. 52- 73. Organization for Economic Cooperation and Development (1987), Recent Trends in International Investment. Sibunruang, Atchaka (1986), Foreign Investment and Manufactured ExRorts in Thailand, Chulalongkorn University Social Research Institute, January 1986. Takeuchi, Kenji (1989), "Does Japan Import Less Than it Should? A Review of the Econometric Literature," the Asian Economic Journal, vol. III, no. 2, September. Takeuchi, Kenji (1990), "Problems in Expanding Japan's Imports of Manufactures from Developing Economies: A Survey," the Asian Economic Journal (forthcoming). Takeuchi, Kenji, John E. Strongman, Shunichi Maeda and C. Suan Tan (1987), The World CoRper Industry. Its Changing Structure and Future Prospects, World Bank Staff Commodity Working Papers No. 15. Tran Van Tho (1987), "New Trends in Japan's Foreign Direct Investment and Implication.s for the Division of Labor in the Asian Pacific Region; A Preliminary Examination," Japan Economic Research Center, JERC Discussion, Paper No. 1, June 1987. Urata, Shujiro (1989), "Recent Economic Developments in the Pacific Region and Changing Role of Japan in the Regional Interdependence," presented at FAIR Conference, Fukuoka, Japan, August 1989. Vernon, Raymond (1983), Two Hungry Giants, Harvard University Press, 1983. PRE Working Pager Series Co ntact ilia Author bia for gaper WPS432 Policy Choices in the Newly Bela Balassa May 1990 N. Campbell Industrializing Countries 33769 WPS433 India: Protection Structure and Francois Ettori Competitiveness of Industry WPS434 Tax Sensitivity of Foreign Direct Anwar Shah June 1990 A. Bhalla Investment: An Empirical Joel Slemrod 37699 Assessment WPS435 Rational Expectations and Boum-Jong Choe June 1990 S. Lipscomb Commodity Price Forecasts 33718 WPS436 Commodity Price Forecasts and Boum-Jong Choe June 1990 S. Lipscomb Futures Prices 33718 WPS437 Institutional Development Work in Cheryl W. Gray June 1990 L. Lockyer the Bank: A Review of 84 Bank Lynn S. Khadiagala 36969 Projects Richard J. Moore WPS438 How Redistribution Hurts Milan Vodopivec June 1990 J. Lutz Productivity in a Socialist Economy 36970 (Yugoslavia) WPS439 Indicative Planning in Developing Bela Balassa May 1990 N. Campbell Countries 33769 WPS440 Financial Sector Policy in Thailand: William Easterly June 1990 R. Luz A Macroeconomic Perspective Patrick Honohan 34303 WPS441 Inefficient Private Renegotiation Kenneth Kletzer of Sovereign Debt WPS442 Indian Women, Health, and Meera Chatterjee Productivitv WPS443 The Inflation-Stabilization Cycles Miguel A. Kiguel in Argentina and Brazil Nissan Liviatan WPS444 The Political Economy of Inflation Stephan Haggard June 1990 A. Oropesa and Stabilization in Middle-Income Robert Kaufman 39176 Countries WPS445 Pricing, Cost Recovery, and Rachel E. Kranton June 1990 W. Wright Production Efficiency in Transport 33744 A Critique WPS446 A Model of Crop and Livestock Gerald T. O'Mara Markets in Mexico Merlinda Ingco PRE Working Paper Series Co ntact I& ~Author Date for paper WPS447 Analyzing the Effects of U.S. Gei,71d T. O'Mara Agricultural Policy on Mexican Agricultural Markets using the MEXAGMKTS Model WPS448 A Model of U.S. Corn, Sorghum, Richard E. Just and Soybean Markets and the Role of Government Programs (USAGMKTS) WPS449 Analysis of the Effects of U.S. Richard E. Just Macroeconomic Policy on U.S. Agricufturre Using the USAGMKTS Model WPS450 Portfolio Effects of Debt-Equity Daniel Oks June 1990 S. King-Watson Swaps and Debt Exchanges 31047 with Some Applications to Latin America WPS451 Productivity, Imperfect Competition Ann E. Harrison and Trade Liberalization in the C6te dilvoire WPS452 Modeling Investment Behavior in Nemat Shafik June 1990 J. Israel Developing Countries: An 31285 Application to Egypt WPS453 Do Steel Prices Move Together? Ying Qian June 1990 S. Lipscomb A Cointegration Test 33718 WPS454 Asset and Liability Management Toshiya Masuoka June 1990 S. Bertelsmeier in the Developing Countries: Modern 33767 Financial Techniques -- A Primer WPS455 A Formal Estimation of the Effect Junichi Goto June 1990 M. T. Sanchez of the MFA on Clothing Exports 33731 from LDCs WPS456 Improving the Supply and Use of S. D. Foster June 1990 Z. Vania Essential Drugs in Sub-Saharan Africa 33664 WPS457 Financing Health Services in Africa: Germano Mwabu June 1990 Z. Vania An Assessment of Alternative 33664 Approaches WPS458 Does Japanese Direct Foreign Kenji Takeuchi June 1990 J. Epps Investment Promote Japanese Imports 33710 from Developing Countries?