PRI-81492 Market Outlook for Major Primary Commodities (In Two Volumes) Volume I: Summary, Energy, and Metals and Minerals October 1992 International Trade Division International Economics Department FOR OFFICIAL USE ONLY Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  FOR OFFICIAL USE ONLY Market Outlook For Major Primary Commodities Report 814/92 Volume I: Summary, Energy, and Metals and Minerals This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. Contributors to this volume were: Boum-Jong Choe (coal, copper, lead and zinc); CPM Group (gold, silver); Mudassar Imran (petroleum); Ying Qian (iron ore, nickel, steel), Kenji Takeuchi (aluminum and bauxite, tin). Contents Preface ........ v Notes and Definitions ....... vi Commodity Description .............................................. vii Summary ....................................................... viii Highlights of Forecasts to 2005 ......................................... viii Commodity Price Forecasts ........................................... ix Commodity Market Outlook-Agriculture.................................... ix Commodity Market Outlook-Metals and Minerals .......................... . . . xiii Commodity Market Outlook-Energy ..................................... xv Energy Petroleum ........................................................ 1 Summ ary . ...................................................... 1 Demand Outlook................................................... 6 Supply Outlook.................................................. 13 Petroleum Price Outlook ............................................ 22 Uncertainties in the Petroleum Market Outlook ................................ 26 Annex I: Investment Needs of the International Oil Industry ..................... 31 Annex II: World Refining Outlook . ..................................... 33 Coal . .......................................................... 42 Summary ........................................................ 42 Demand Outlook................................................. 43 Supply Outlook.................................................. 46 Price Outlook................................................... 49 Coal and the Environment ........................................... 50 Metals and Minerals Copper Summary ........................................................ 56 Demand Outlook................................................. 57 Supply Outlook.................................................. 60 Price Outlook................................................... 66 Tm ........................................................... 80 Summary........................................................ 80 Demand Outlook................................................. 81 Supply Outlook.................................................. 83 Price Outlook ................................................... 85 i Nickel ......................................................... 88 Summary........................................................ 88 Demand Outlook. ................................................. 89 Supply Prospects ................................................. 91 Price Prospects .................................................. 96 Aluminum and Bauxite............................................... 99 Summary ..................................................... 99 Demand Outlook................................................... 100 Supply Outlook.................................................. 103 Trade Outlook .................................................. 107 Price Outlook ................................................... 110 Ion Ore ........................................................ 124 Summary ..................................................... 124 Demand Outlook ................................................. 125 Supply Outlook.................................................. 126 Trade Outlook .................................................. 129 Price Prospects .................................................. 132 Steel .......................................................... 140 Summary ..................................................... 140 Demand Outlook................................................. 141 Supply Outlook.................................................. 143 Trade Prospects ................................................. 149 Price Outlook ................................................... 149 Lead and Zinc .................................................... 155 Summary ..................................................... 155 Demand Outlook................................................. 156 Supply Outlook.................................................. 157 Price Outlook. ................................................... 159 Gold ..... ..................................................... 163 Summary..................................................... 163 Price Outlook. .................................................. 164 Supply Outlook. ................................................. 165 Demand Outlook ..................................................... 167 Silver .... ..................................................... 174 SummaryPieto..................................................... 174 Price Outlook. ................................................... 175 Demand Outlook.................................................. 175 Supply Outlook.I..................................................177 iv Preface This report is a compilation of studies which review the market prospects for major primary commodities exported by developing countries. The forecasts are mainly used in forecasting the balance of payments of countries that borrow from the World Bank and in appraising investment projects that include these commodities as inputs or outputs. Because of the multiple purposes they are intended to serve, the price forecasts are presented in current (nominal) as well as 1990 constant dollar (real) terms.! For the period 1992-95, the price forecasts are in terms of prices expected for the individual years. For 2000 and 2005, the price forecasts are forecasts of the average price levels expected during that period. The forecasts are conditional on the various macroeconomic and commodity-specific assumptions used-all of which are subject to uncertainty. The macroeconomic assumptions forming the basis for the forecasts are available in the World Bank's report Global Economic Prospects and the Devsloing Countries. 1992. Because of the uncertainty which is inherent in commodity price forecasts, the International Trade Division periodically prepares probability distributions of its price forecasts. These are available upon request. The primary commodity market outlooks are discussed by commodity. For most commodities or group of commodities, there is a standardized set of tables giving historical and forecast values for production, consumption, exports, and imports; these tables give details in terms of major economic regions as well as for countries which are major participants in these markets. For most commodities, the forecasts have been based on simulation runs of global commodity models maintained within the International Trade Division of the World Bank's International Economics Department. Details of these models can be obtained directly from the Division. The assistance given to the Division in preparing this report is gratefully acknowledged.. People in both public and private organizations have been most forthcoming in providing data and in discussing the outlook for the various commodity markets. Their cooperation has added greatly to the usefulness of the report. ' Commodity prices have been deflated by the World Bank's Manufacturing Unit Value (MUV) index and the G-7 CPI. The MUV index is the c.i.f. index of US dollar prices of industrial countries' manufactured exports (SITC 5-8) to the developing countries and may be regarded as a useful deflator to measure changes in the net barter terms of trade of developing countries highly dependent on exports of primary commodities. The G-7 CPI may be a useful deflator to use in circumstances where the major countries' inflation rate is believed to be an appropriate measure of changes in the overall price or cost level. v  Notes and Deflaloas * Commodity market-price descriptions are shown on the next page. * Dollars are United States dollars unless otherwise specified. * All tons refer to metric tons (1,000 kilograms) unless otherwise noted. Abbreviations and Symbol tons = metric tons lb -pounds cum = cubic meters bbl =barrels kg = kilograms abid - million barrels per day ha = hectares N.A. - not available MTOE = million tons of oil equivalent ../-I = no data Country Clssiflcations High-Income Countries OECD Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Iceland, Ireland, Italy, Japan, Luxembourg, Netherlands, New Zealand, Norway, Spain, Sweden, Switzerland, United States, United Kingdom. Non-OECD Andorra, Aruba, Bahamas, Bermuda, Brune, Channel Islands, Cyprus, Facroe Islands, French Polynesia, Greenland, Hong Kong, Israel, Kuwait, Mayotte, OAE , Qatar, Singapore, United Arab Emirates, Virgin Islands (US). IAw- and Middle-Income Countries (LMICs) Africa Angola, Benin, Botswana, BurkinaFaso, Burundi, Cameroon, CapeVerde, Central African Rep., Chad, Comoros, Congo, Rep., C6te d'Ivoire, Djibouti, Equatorial Guinea, Ethiopia, Gabon, the Gambia, Ghana, Guinea, Guinea- Bissau, Kenya, Lesotho, Liberia, Madagascar, Malawi, Mali, Mauritania, Mauritius, Mozambique, Namibia, Niger, Nigeria, Reunion, Rwanda, Slo Tomd and Principe, Senegal, Seychelles, Sierra Leone, Somalia, South Africa, Sudan, Swaziland, Tanzania, Togo, Uganda, Zaire, Zambia, Zimbabwe. America Antigua and Barbuda, Argentina, Barbados, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, Cuba, Dominica, Dominican Rep., Ecuador, BI Salvador, French Guiana, Grenada, Guadeloupe, Guatemala, Guyana, Haiti, Honduras, Jamaica, Martinique, Mexico, Netherlands Antilles, Nicaragua, Panama, Paraguay, Peru, Puerto Rico, St. Kitts and Nevis, St. Vincent, St. Lucia, Suriname, Trinidad and Tobago, Uruguay, Venezuela. Asia and Facific American Samoa, Bangladesh, Bhutan, Cambodia, Fiji, Guam, Indonesia, Kiribati, Korea, Dem. Rep., Korea, Rep., Lao PDR, Macao, Malaysia, Maldives, Mongolia, Myanmar, Nepal, New Caledonia, Pacific Islands, Trust Territory, Pakistan, Papua Now Guinea, Philippines, Solomon Islands, Sri Lanka, Thailand, Tonga, Vanuatu, Viet Nam, Western Samoa. Europe Albania, Bulgaria, Czechoslovakia, Former Soviet Union (FSU)w, Gibraltar, Greece, Hungary, Isle of Man, Mak, Poland, Portugal, Romania, Turkey, Yugoslavia. Middle East and North Africa Afghanistan, Algeria, Bahrain, Egypt, Arab Rep., Iran, Islamic Rep., Iraq, Jordan, Lebanon, Libya, Morocco, Oman, Saudi Arabia, Syrian Arab Rep., Tunisia, Yemen, Rep. a/ Other Asian economies-Taiwan, China. b/ Data are not available for individual republics. Vi Commodity Description Energy Petroleum, average OPEC price: OPEC government sales weighted by export volumes through 1981; beginning 1982 OPEC spot prices weighted by OPEC export volumes. Thermal Coal, (12,000 BTU/lb, lea than 1.0% sulfur, 12% ash), f.o.b. Piers, Hampton Roads, Norfolk. Food Coffee (ICO), indicator price, other mild Arabicas, average Now York and Bremen/Hamburg markets, ex-dock for prompt shipment. Cocoa (ICCO), daily average price, New York and London, nearest three future trading months. Tea (London Auction), average price received for all teas. Sugar (World), ISA i f.o.b. and stowed at greater Caribbean ports. Beef (Australian/New Zealand), cow forequarters, frozen boneless, 85% chemical lean, c.i.f. US port (East Coast), ex-dock. Bananas (Central and South American), first-class quality tropical pack, importer's price to jobber or processor, f.o.r. U.S. ports. Oranges (Mediterranean Exporters), navel, EEC indicative iport price, ca.f. Paris. Cereals Rice (Thai), white, milled, 5% broken, government standard, Board of Trade posted export price, f.o.b. Bangkok. Wheat (Cadian), No. 1 Western Red Spring (CWRS) 13.5%, basis in store Thunder Bay, domestic through March 198 ; subsequently St. Lawrence, export. Maize (US), No. 2, yellow, f.o.b. Gulf ports. Grain Sorghum (US), No. 2, Milo yellow, f.o.b. Gulf ports. Fats and Oils Palm Oil (Malaysian), 5% bulk, c.i.f. N.W. Europ. Coconut Oil (Philmppias/Indonesian),bulk, c.i.f. Rotterdam. Groundnut Oil (Nigerian/West African), bulk c.i.f. UK, through January 1977; subsequently (any origin), c.i.f. Rotterdam. Soybean Oil (Dutch), crude, f.o.b. ex-mill. Soybeans TS), c.i.f. Rotterdam. p.pines/Indonesian),bulk, c.i.f. N.W. Europe. p elsar I" (NierianM), c.i.f. UK. Groundnut Meal lndian), 48%, c.i.f. Rotterdam through 1981; thereafter Argentine, 48/50%. Soybean Meal (US), 44% extraction, c.i.f. Rotterdam. Non-Food Cotton (Outlook A" Index), Middling (1-3/32, c.i.f. Europe. Jute (Bangladesh), white D, f.o.b. ChittagongClna. Rubber (lSS No. 1), in bales, spot New York. Tobacco (Indian), flue-cured, average export unit value. Timber Log (Southeast Asian), Philippines, Lauan for plywood and vencer, length over 6.0 M, diameter over 60 CM, average wholesale price in Iapan through 1976, be gi 1977, Malysian, Meranti, Sabah SQ Best Quality, sale price charged by importers, Japan. Logs (West African), S hi quality, loyal and marchand, f.o.b. Cameroon. Sawnwood (Malaysian), Dark Red Morantsi, select and better quality, standard density, c.i.f. French ports. Metals and Minerals Copper (IME), cash wirebars through November 1981; from December 1981 through June 1986, high grade cathodes, settlement price; subeuetdi, frade- A. Tin (Malaysian , traits quality, ex-smelter, Penang, official settlement price. Nickel (Canadian), electrolytic cathodes, Ni 99.9 % shipping point through 1979; subsequently cathodes, minimum 99.8% purity, official mornig session weekly average bid/asked price. Aluminum, indicative price of U.S. unalloyed primary ingot in the European Market through 1978; subsequently IME standard grade, minimum 99.5% purity, cash price. Lead L , Ettlement price, refied lead, purity 99.97%. Zinc , settlement prico,good ordinary brand through August 1984; thereafter High Grade brand. Iron Or (Brazilian), CVRD Itabira sinter fed produced from 64.2prity ore, metal content weight, contract to Germany, Federal Repuic, f.o.b. reference price through 1974; from 1975 throug 1985, standard sinter food from 64% purity ore; starting 1986 Southern System (Itabira and other southern mines) 64%; during 1988 and 1989, 64.2%; beginning 1990 64.3% purity ores. Bauxite, crude and dried, US import reference price based on imports from Jamaica through 1974; from 1975 US import price, c.i.f. US port. Gold (UK), 99.5% fine, London afternoon fixing, average of daily rates. Silver (Handy & Harman), 99.9% grade refined, New York. Fertilizers Phosphate Rock (Moroccan), 72% BPL, PAS Casablanca through 1980; from 1981, 70% TPL contract. Urea y o in), bagged, f.o.b. N.W. Europe. TSP (Triple uporphshate), bulk, f.o.b. US Gulf. DAP (Diammonium osphatc), bulk, f.o.b. US Gulf. Potassium Chloride (Muriate of Potash), bulk, f.o.b. Vancouver. World Bank, International Economics Department. November 5, 1992 vii Summary HIoLmIrrs oF FORECASTS To 2005 Prices * Non-fiel commodity prices are expected to fall even further in real terms from the 1991 record low. No significant increase is foreseen until the latter half of this decade. * Real price declines over the next few years are expected in all commodity groups with the exception of beverages (the price index for this group has declined an extraordinary 68% in real terms since 1986) timber (which is expected to continue increasing in real terms), and vegetable oils. * Over the longer term, it is mainly the expected recovery in beverage prices which supports the upturn in non-fiel prices. * Petroleum prices are expected to average near present levels in current dollar terms for several years before increasing due to tightening supplies. Production, Consumption, and Trade * Recent developments in Eastern Europe, and particularly in the FSU, have increased the uncertainty in the outlook for most commodities. Because of their reductions in consumption, reverberations will be felt for many years in the form of reductions in imports (e.g., cocoa, coffee, tea, sugar, grains) or increases in exports (e.g., metals). * Production growth in the beverages should slow rapidly as a result of reduced new plantings and replantings and reduced inputs, setting the stage for a recovery in prices. * The main growth markets for grains imports will be the developing countries, particularly those in the Asia-Pacific region. Industrial country exporters will supply the bulk of these imports. * Most metals markets are in surplus and prices should increase over the next few years as balance is restored. Mining, smelting, and refining are expected to continue shifting towards the developing countries. * Petroleum consumption is expected to grow at 1.3% p.a. on average but with growth being increasingly concentrated in the LMICs. * Petroleum supplies will be more concentrated in the Middle East. Viii CoMmoDITY PRICE FORECASTS The historical prices and the price forecasts made in this report are brought together in constant 1990 dollars in the Summary Table 1 and in current dollars in Table 2. The various commodity price indices maintained by the World Bank are shown in Table 3 (constant dollars) and Table 4 (current dollars), with the indices based on weights for the period 1979-81. The deflator used to derive the constant dollar values in Tables 1 and 3 is the Manufacturing Unit Value (MUV) Index. This and other price indices are presented in Table 5 for the period 1948-2005. The price indices for the major commodity groups, in current and constant dollars, are graphed in the two figures below. The constant dollar, non-fuel commodity price index is expected to fall six percentage points in 1992, matching the fall in 1991, and to remain at that level over the following two years. The main reason for the decline in 1992 is the 20% decline in the beverage index. The constant dollar beverage price index has now fallen 68% since 1986. With beverage prices expected to recover somewhat in 1993, offsetting declines in the aggregate index are expected mainly to come from cereals and other foods. The turnaround in the aggregate non-fuel index after 1994 is supported by all commodity groups with the exception of vegetable fats and oils. This turnaround is conditional mainly on two assumptions: that there will be an upturn in economic activity in the industrial countries and that production growth in the perennial crops will slow fairly rapidly. A brief summary of the market outlook for the major primary commodities is provided below. For more details, readers are referred to the chapters reporting on the market outlook for each commodity. CoMmoDrrY MARKET OUTLOOK-AGRICULTURE (i) BEVERAGES Cocoa prices to increase eventually, as production growth slows. In real terms, the market outlook for the 1992-95 period is for cocoa prices to remain near present levels, given the large stocks and the continuing increases in production. After the mid-1990s, the low level of new plantings and replantings in recent years should result in a production decline in some of the major producers (Brazil, Cameroon, COte d'Ivoire, and Malaysia) and a slowdown in most other. Only Indonesia is expected to have significant growth in production over the forecast period. Long-term prices should increase somewhat in real terms with the slowing of production growth. Consumption growth will be likely slow, averaging less than 2% p.a. Recent developments in the FSU and Eastern Europe, which have been substantial cocoa importers, increase the range of uncertainty about this market. Growth in consumption is expected from some nontraditional markets such as China, Japan, and the Republic of Korea. ix Index of Non-Fuel Primary Commodity Prices (Current), 1948-2005 (1990=100) 300 250 200 150 - 50 1948 1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003 Index of Non-Fuel Primary Commodity Prices (Constant), 1948-2005 (Deflated by MUV, 1990=100) 600 500 400- 300- 200- o 1 . . . 1 . . . ¡ . . . : . . . 1 . . . . . . . , . . . . . 1 . . ! . : . . . ....1.... 1948 1953 1958 1963 1968 1973 1978 1983 198B 1993 1998 2003 Beverages Cereals Non-Food Metals & Minerals Source: World Bank, International Economics Department. x Coffee price outlook is also dependent on production growth,* demand growth expected to be only 1%. The current low international prices are below variable production costs for many producers, dampening new plantings and replantings. Production growth should slow; with output growth projected to average only 0.6% p.a. over the forecast period. Supply prospects differ significantly between countries; output in Brazil, Colombia, and COte d'Ivoire is expected to fall. Indonesia is expected to increase production during the forecast period because of its abundance of land and low-cost labor. World coffee demand is expected to average about 1% p.a. growth for the 1991-2005 period. Very slow population growth and already high consumption levels in the high-income countries mean faster consumption growth is unlikely. Japan and LMIC Asia and Europe are the main prospective growth areas. International coffee price prospects point to some recovery in 1993. But any possibility of a significant increase during the 1990s, absent some supply shock or re-introduction of an international coffee agreement, has been made less likely by the large stocks now held in the importing countries. The demand for robusta coffees is projected to grow more slowly than the demand for arabica coffees. The switch from robustas to arabicas in the OECD countries is expected to continue. This, together with the expected increase in robusta production in Brazil, Indonesia, and Viet Nam should lead to an increasing price differential between arabicas and robustas, which is bad news for African producers of robustas such as Cameroon, COte d'Ivoire, and Uganda. Tea price increase expected with Kenyan production growth slowing; recovery of FSU and Middle East demand would assist. World tea prices at London auctions were at their lowest level ever in real terms in 1991. The main cause has been continuing output increases while world import demand has been stagnant. One of the key reasons for the continued growth in output in major producing countries has been depreciations in their real exchange rates. Tea prices are recovering somewhat at the end of 1992 and this should continue into 1993 as a result of reduced output in South India, Sri Lanka, and southern Africa due to drought. In the second half of the 1990s, real tea prices should increase as production growth slows in Kenya and demand increases in the FSU and the Middle East. World tea output is projected at 2% p.a. over the forecast period-a considerably slower rate than for the 1970-90 period when it averaged 4.8% p.a. (ii) GRAINs, OILSEEDS, AND SUGAR Grain import gap in developing countries will continue to widen; to be fdled by high- income exporters. The LMICs other than Eastern Europe and the FSU are expected to continue expanding their net grain imports to satisfy the increased demand generated by increasing incomes and populations. They are projected to double their grain imports over the period to 2005. The industrial countries will be the main sources of exports-increasing exports by about 40% over the forecast period. Wheat is projected to be the fastest growing export crop. xi Forecasts for Eastern Europe and the FSU are highly uncertain but our best guess is that grain consumption will be depressed for several years, production will begin to grow more quickly than over the past two decades, and thus imports will decline-probably halving over the forecast period. Grain prices are expected to remain nearly constant in nominal dollars for the next several years with grains markets in relative surplus. However, prices are expected to rise slightly in real terms in the latter half of the 1990s given the continued fast growth projected for the developing countries. Vegetable oils/meals markets to have robust growth, but dedining real prices. Over the long term, vegetable oil prices are expected to be constant to downward trending in real terms, despite vigorous growth in demand. The recent fast pace of production increases in the low-cost producing countries, especially Argentina, Brazil, Indonesia, and Malaysia, is expected to continue. Because of earlier plantings, Indonesia and Malaysia are locked onto a path of rapid expansion of palm oil output throughout this decade. In an environment of falling real prices and substantial income growth, especially in the many populous countries in Asia, vegetable oil demand should grow strongly. Demand for oilseed meals should also grow rapidly in developing countries, especially in the Asia-Pacific region, as their demand for meats rises. Sugar consumption growth expected to return to 2% p.a. after China and FSU shocks; long-term realprice about (13/1b. Since 1988-89, political developments in China, Eastern Europe, and the FSU have resulted in a halving of the usually steady 2-2.5% p.a. growth in sugar consumption. Prices fell from around 015/lb in 1989 to around 010/1b in mid-1992. Long-term world sugar consumption growth is expected to average 2% p.a. with strongest growth in Asia (3.1%), Africa (2.5%), and Latin America (2.1%). Sugar production is expected to decline in Cuba, the FSU, and the United States; growth in production should slow in China and India; while growth should increase in Australia, Brazil, the EC, and Thailand. The long-term average price forecast is C12-13/lb (1990 dollars). However, price volatility is expected to remain very high. (iii) AGRicuLTuRAL RAW MATERiALS Cotton market to be dominated by Asia region. Cotton consumption declined in the 1990-92 period due to the poor global economic performance, particularly in the Eastern European countries and the FSU. Over the long term, the declining trend in population growth is expected to reduce the growth rate of cotton consumption below the 2% p.a. rate of the 1964-90 period. The region with the largest population and fastest expected economic growth-LMIC Asia-Pacific-is expected to see the fastest increase in cotton consumption. A large part of this increase should be in China, although China's cotton demand growth should be at a slower pace than in recent years due to competition from domestically produced synthetic fibers. The Asia region (major producers China, India, and Pakistan) is expected to continue to be the largest cotton-producing area. LMIC Europe is likely to see production declining. xii Some recovery in prices is expected in 1993/94 as producers reduce planted area and yields fall from the record levels of 1991/92. Long term, the declining trend in real prices is expected to continue. Rubber markets' main growth to be in Asia-Pacific region, Indonesia to become leading producer. Natural rubber consumption is expected to grow fastest in the LMICs of the Asia-Pacific region. The impact of rapid income growth on rubber consumption in these countries is illustrated by the 50% increase in China's consumption from 450,000 tons to 660,000 tons during the 1986-88 period and by the 30% increase in India's consumption from 278,000 tons to 358,000 tons in the 1987-90 period. Rubber consumption in the Asia-Pacific region is expected to increase from 1.9 million tons to 3.1 million tons over the forecast period. World rubber consumption is expected to increase at 2.6% p.a. to reach 7.6 million tons in 2005. The fastest growth in the OECD countries should be in Japan with 1.7% p.a. increase. The economic reforms in Eastern Europe and the FSU should reduce demand for synthetic rubber and increase natural rubber consumption. However, the time path of this shift is highly uncertain. Of the three major producers, Indonesia is projected to become the leading producer with output growth of 3.5% p.a.-going from 1.1 million tons in 1992 to 1.9 million tons in 2005. Higher investment than previously expected in Malaysia should lead to 2.4% growth, with output increasing from 1.2 million tons in 1992 to 1.8 million tons in 2005. Because of its land constraint, growth in Thailand is expected to slow to 1.3% p.a. Fast growth is expected in the upcoming producers China, India, and Viet Nam. An upturn in real prices is expected over the 1995-2000 period, given the economic growth assumptions. But over the long run, major improvements in productivity-including shifts to lower-cost producers-are expected to cause real prices to decline. CoMmoDrry MARKET OUTLooK-METALs AND MINERALS Copper market outlook reasonably optimistic for producers. World consumption of refined copper is forecast to be relatively robust, with an average growth rate of 2% p.a. over the 1991- 2005 period. The newly industrializing countries are expected to continue to be the main growth point. Demand in the FSU and Eastern Europe is expected to remain depressed, at least in the near term, but could grow rapidly from the mid-1990s as these economies require large investments in infrastructure and equipment. In the industrial countries, the copper intensity of industrial production should decline at a slower rate than over the past two decades owing to lower energy prices and faster expansion of the capital goods sector. A tally of announced projects indicates that over the 1992-95 period world mine and refined copper capacities will each expand by as much as one million tons. Most of these increases will take place in Latin America, North America, and Asia. xiii Over the period to 1996, the market balance is expected to turn to a moderate surplus as supplies increase. Prices are likely to decline. However, the constraints of smelter capacity should persist through 1995. If the historical pattern of the investment cycle is repeated, copper prices are likely to trend upwards again in the latter part of this decade. Aluminum market should continue its recent slow growth. Aluminum consumption growth in the 1980s was a mere 1.6% p.a. Some increase in the growth rate is expected in the forecast period (to 2.2% p.a.), with the share of LMICs rising from 31% in 1991 to 36% in 2005. As aluminum prices are also expected to remain low by historical standards, smelting capacity growth should be slow. Most new smelting capacity is likely to be in Australia, Canada, the Middle East, South America, and southern Africa where bauxite and low-cost energy are in good supply. Prices of aluminum and bauxite are expected to increase in the 1992-94 period as the market balance adjusts from the surplus in 1991. After that, with declining average costs of production, aluminum prices should fall in real terms. The greatest source of uncertainty in the medium term for aluminum prices is the level of FSU exports to international markets. These will depend on the pace of domestic demand recovery, the changes in energy prices, and the renovation of antiquated smelting facilities. Steel consumption in high-income countries fell 7% in 1991 due to the global economic slowdown. Consumption in the LMICs fell 8% in the 1989-91 period-mainly because of the sharp declines in Eastern Europe and the FSU. High-income country consumption is expected to grow at only 0.7% p.a. over the period to 2005. LMIC steel consumption growth is forecast at 2.1% p.a. Steel production to continue its shifts to LMICs. World steel production is undergoing a substantial change with capacity shifting from the high-income countries to the LMICs,, In 1980-91, steelmaking capacity in the high-income countries fell 120 million tons while LMIC capacity grew by 110 million tons. A further 20 million tons decline is expected in the high-income countries by 2000, which should be more than offset by a 30 million tons increase anticipated in the LMICs. Low-cost iron ore producers will increase domination. Iron ore consumption will track steel production, increasing at about 1.3% p.a. over the 1991-2005 period. About 30 million tons of additional iron ore production capacity is expected to come on-line around the mid-1990s, mainly from upgrading existing operations in major producing countries. Australia and Brazil should strengthen their dominant positions as producers. With the increasing dominance of the large low-cost producers, iron ore prices are expected to decline at an average rate of 0.7% p.a. in real terms over the forecast period. Nickel market is one of the brighter spots for producers. Long-term growth in nickel consumption should be sustained at around 2% p.a. The outlook for nickel is generally upbeat because of its use in products that have strong growth potential. Environmental protection pressures should boost xiv nickel demand in the form of stainless steel and high nickel alloys used in the flue gas desulphurization in power plants. Nickel production capacity is expected to increase by at least 100,000 tons over the next five years, with major expansions in Australia, Brazil, China, Cuba, and Colombia. Some increase in real prices is expected in the latter half of the 1990s, with capacity expansions lagging demand increases because of current low prices. Lead and zinc markets to remain in excess supply for several years. New mines will lower production costs. Lead and zinc prices have been at record lows in real terms in the past year due to weak demand-particularly in the automobile sector-and to increased supplies. The excess supply situation was exacerbated by exports from the FSU and Eastern Europe. Prior to 1990, these countries had been net importers of these metals. Over the short term, lead and zinc prices are not expected to increase much, if at all, as the excess supply situation is expected to continue for the next two to three years. The market balance is expected to improve by the mid-1990s as demand catches up with production capacity. Although greenfield lead and zinc projects have moved to more unfriendly environments in recent years (e.g., Red Dog mine in Alaska), costs of production have tended to fall. New mines projected to come on-stream in the latter half of the 1990s are expected to have even lower costs. So prices over the long term are expected to average at near the currrent low levels in real terms. COMMODrrY MARKur OtrLOOK-ENERGY Petroleum consumption will grow most rapidly in LMICs. Petroleum consumption is expected to increase, on average, by 1.3% p.a. during the forecast period. Fastest growth should be seen in the LMICs (excluding Eastern Europe and the FSU), with their share growing from 16% in 1991 to 36% by 2005. Increased use of motor transport will be the main reason for this growth. Supplies from non-OPEC producers (excluding the FSU) will contribute about one half of the incremental global output up to the mid-1990s. But beyond that time, dependence on OPEC will increase as several producing regions reach maturity. OPEC's share to increase. OPEC's share of global output is expected to increase from 38% in 1991 to around 46% in 2005. The share of Middle East producers is expected to grow from 28% to 37%. Which could mean higher prices for a period. The market outlook is for prices to decline slightly in real terms in the near term. Although industrial activity is expected to pick up after 1994, supplies should be ample to forestall price increases for some time. However, in the latter half of the decade, we expect prices to increase in real terms as production begins to decline in several OPEC and non-OPEC producers. xv Coal prices expected to rise In long ten with Increased conswnption and higher production costs. World coal consumption is expected to increase at 1.8% p.a. over the 1991-2005 period. Thermal coal consumption growth is forecast at 2.5% p.a. Future production increases should come mostly from Australia, China, India, South Africa, and the United States. Most European countries and Japan can be expected to phase down their coal industries due to high costs, while Colombia, Indonesia, and Venezuela are likely to become more important exporters. No increase is expected in coal prices in real terms up to 1995. Beyond that time, prices are expected to rise somewhat in real terms in line with increases in petroleum prices and with expected increases in production costs, xvi Table 1: Com odity Pucc, and Pric Projcons in 1990 Ck~an Dollars af -Sh Tärm- --Ioa-Trm- 1970 1980 195 1938 1989 1990 1991 1992 1993 1994 1995 2000 2005 BNEG Peroleum $=BBL 5.2 42.4 38.9 14.3 11.2 21.2 17.0 16.5 15.7 153 15.6 18.0 17.0 Cosi S/Mf N.A. 60 68 39 43 4 41 38 38 38 39 44 43 Cofeo /KG 457 478 468 318 252 197 183 130 142 164 195 226 216 Coco e/KG 269 34 329 166 131 127 117 106 108 111 113 124 136 Te. w/KG 437 310 289 188 213 203 181 193 188 182 190 198 204 Sutga /Mr 323 878 130 236 298 277 194 188 185 195 220 255 280 Beef 3/KO 520 384 314 264 271 256 261 229 228 227 228 274 253 Ba$aWas $lMr 659 527 551 502 578 541 548 471 443 439 434 422 414 Or$Wges $lMT 670 543 581 476 471 531 510 488 473 471 470 453 436 Rico $/Mr 574 603 315 316 338 287 308 272 261 258 260 245 237 Whet $/M 250 265 253 188 213 156 140 161 147 146 148 159 129 Maiz W/Mr 233 174 164 112 118 109 105 99 94 95 95 101 82 Grain ohu 8/Mw 207 179 150 103 112 104 103 97 92 91 91 98 79 Palm oil $/MW 1,037 811 730 459 370 290 332 371 380 353 343 303 266 Coco~* 0il 3/MT 1,584 936 860 593 546 337 424 554 498 509 535 564 457 GroundauI t 0i $/ 1,510 1,194 1,319 619 319 964 875 568 584 383 617 554 421 SoybemaOil W/M 1,224 829 834 486 456 447 444 404 416 443 433 409 374 8ybems $/MT 466 412 327 318 291 247 235 221 224 234 234 219 233 C4pra $/Mr 897 629 563 417 368 231 280 357 326 335 350 396 322 Palm YCernes $/mT 670 480 424 280 265 185 215 216 214 219 234 243 227 Gro~uda tMeul $/mT 407 334 208 220 211 185 147 147 149 163 172 157 175 SeybeaaMeal $3M 411 364 229 281 260 209 195 195 199 204 209 185 210 IN-Pom Cmoa 0/KG 252 284 192 147 177 182 164 122 117 136 140 150 145 lute $/MT 1,092 428 850 388 394 408 370 300 308 311 308 321 313 Rn~ 0/KG 185 226 135 135 118 102 99 96 96 99 103 114 112 Tobacco / 4 ,938 3,196 2,843 2.037 1,998 1,964 2,182 1,849 1,787 1,742 1,733 1,712 1,690 I ER Logg(Meraai) $/CM 148 271 199 245 237 210 217 236 238 241 243 258 276 Log (S$elli) 8/CM 171 350 253 285 289 344 309 316 322 324 327 346 361 Sawawood 1/CM 370 507 403 322 446 524 462 483 484 489 493 517 543 MEALS & MNRL CM $MT 5,634 3,032 2,066 2,730 3,009 2,6=2 2,291 2,173 2,153 1,865 1,704 1,974 1,840 Tia /KOG 1,432 2,284 1,682 740 902 609 536 577 597 640 675 620 610 ickel $/MT 11,348 9,058 7,142 14,457 14,061 ,864 7,987 6,759 6,976 7,195 7,527 8,377 8,566 Ahuninuf $/Mr 2,153 2,466 1,517 2,676 2,062 1,639 1,275 1,183 1,312 1,466 1,601 1,465 1,435 Lead $/M 1,212 [,259 570 688 711 811 546 530 543 551 597 656 539 Zia $/MT 1,176 1,057 1,141 1,303 1,753 1,513 1,094 1,220 1,131 1,110 1,107 1,173 1,034 Iron Om. $/MW 39.2 39.0 38.7 24.7 28.0 30.8 32.6 29.7 27.4 26.4 27.8 27.9 27.1 BauxtC 8/MT 47.8 44.5 52.0 31.8 36.3 34.4 33.1 30.0 29.9 30.2 31.1 30.0 30.0 Od $/rOZ 143 845 463 459 40 384 355 324 317 335 340 360 360 Silver «/rOz 706 2,867 895 686 581 482 396 375 353 380 389 450 450 ERTILMZERS rwospateRock $/Mw 44 65 49 38 43 41 42 40 40 41 42 42 41 Urea 8/Mr 193 309 199 163 140 157 168 134 141 148 153 169 159 TSP 8/MT 169 251 177 166 152 132 130 114 110 115 120 125 123 DAP $/MW 215 309 246 206 183 171 169 136 127 135 140 148 145 Potasiuum lorb/ /M 126 161 122 92 104 98 107 105 104 108 108 107 105 W.A. - NMt Available. al Coipeted fra uamdeddta and de~kv MUV (1990-100). b/ Pola"ma, Qlorde, also knowe as Muri~te Pa . Soure: Word Bank, htseaionam m-mma Darwtme. November 9, 1992 Xvå Table 2: Co~~dy Pricms and Prico Profctiom in Cu~ Dollars a --4ot.Tc.-Lmng-Term- 1970 1980 1985 1913 1989 1990 1991 1992 1993 1994 1995 2000 2005 ENERG Potetum S/BBL 1.3 30.5 26.7 13.6 16.3 21.2 17.3 17.6 17.3 17.4 18.0 24.7 26.8 Coml 3/MT N.A. 43 47 37 41 42 42 40 42 43 45 60 68 Coffe 0/KG 115 344 321 303 239 197 187 139 157 185 225 :310 340 Cocoa C/KO 68 260 225 159 124 127 120 113 119 125 131 170 215 Tea /KG 110 223 198 179 202 203 184 206 208 205 220 272 322 Sugr S/MT 81 632 90 225 282 277 198 200 205 220 254 350 441 Beef */KG 130 276 215 252 257 256 266 244 252 256 264 376 399 Ba~aa- 3/MT 165 379 378 478 547 541 560 502 490 494 502 579 652 Orange$ S/MT 168 391 398 453 445 531 521 520 523 530 543 622 687 Rico 3/MT 144 434 216 301 320 287 314 290 288 290 300 336 374 $Weat 3/MT 63 191 173 180 201 156 143 172 162 164 171 218 204 Maize 3/MT 58 125 112 107 112 109 107 105 104 107 110 139 130 GraiaSorghum 3/MT 52 129 103 99 106 104 105 103 102 103 105 134 125 FATS a o0lu Palm Oil 3/MT 260 584 501 437 350 290 339 395 420 397 396 416 420 Coconut Oil 3/MT 397 674 590 565 517 337 433 590 550 573 618 774 721 Grount Oil 3/MT 379 859 905 590 775 964 894 605 645 656 713 760 664 SoybeaOil 3/MT 307 597 572 463 432 447 454 430 460 499 500 562 590 Soybeans $/MT 117 296 224 304 275 247 240 235 248 263 270 300 368 Copr $ffMT 225 453 386 398 348 231 286 380 360 377 404 544 508 PaLEKcRnels 3/MT 168 345 291 267 251 185 220 230 237 246 270 334 358 GroundnutMeal /MT 102 240 143 210 200 185 150 157 165 184 199 216 276 ScybeanMed $/MT 103 262 157 268 246 209 199 208 220 230 241 254 331 NON-POOD Cotton /KG 63 205 132 140 167 182 168 130 129 153 162 206 229 Jueu $/MT 274 308 583 370 373 408 378 320 340 350 356 441 493 Rubbe 0/KG 46 162 92 129 112 102 101 102 106 112 119 156 177 Tobaco /MT 988 2,300 1,950 1,941 1,891 1.964 2,228 1,970 1,975 1,961 2,003 2,350 2,663 Logs PMeranti) $/CM 37 195 136 233 225 210 2 251 263 271 281 354 435 Logs (Sapelli) 3/CM 43 252 174 271 274 344 316 337 356 365 378 475 580 Sawwood 3/CM 93 365 276 307 422 524 472 514 535 550 570 710 855 METALS & MINRALJ Cpper 3/MT 1,413 2,182 1,417 2,602 2,848 2,662 2,339 2,315 2,380 2,100 1,97o 2,710 2,900 Tia 0/KG 359 1,644 1,154 705 853 609 548 615 660 720 780 851 961 Nickel $/MT 2,846 6,519 4,899 13,778 13,308 8,864 8,156 7,200 7,600 8,100 8,700 11,500 13,500 Aluminum $/MT 540 1,775 1,041 2,551 1,951 1,639 1,302 1,260 1,450 1,650 1,850 2,011 2,261 Lead 3/Mr 304 906 391 656 673 811 558 565 600 620 690 900 850 Zinc 3/MT 295 761 783 1,242 1,659 1,513 1,117 1,300 1,250 1,250 1,280 1,610 1,630 Iron Oro 3/MT 9.8 28.1 26.6 23.5 26.5 30.8 33.3 31.6 30.3 29.7 32.1 38.3 42.7 Bauxisa 3/MT 12 32 35.7 30.3 34.4 34.4 33J 32 33 34 36 41.2 47.3 Gold 3/rOZ 36 608 318 437 381 384 362 345 350 377 393 494 567 Silver /TOZ 177 2,064 614 654 550 482 404 400 390 428 450 618 709 FERTIZERS Phspha e Rock 3/MT 11 47 34 36 41 41 43 43 44 46 48 58 65 Urea 3/MT 48 136 155 132 157 172 143 156 167 177 232 251 TSP 3/MT 43 180 121 158 144 132 133 121 122 129 139 172 194 DAP $lMT 54 222 169 197 173 171 173 145 140 152 162 203 229 Potasium lor b/ $/MT 32 116 84 88 99 98 109 112 115 122 125 147 165 N.A. - Not Av~ilable. al Data have been unnt~ot. b/ Po~as. Chlorde, also kown an MIat of Potush. Source Word Bank, bterna~ional Economics Dqartmect. November 9, 1992 xviii Table 3: Weighted Index of Commodity Prices Contain1 Dollars) (1990= 100) Petrolaum 33 Commodities Timber Metals (Bxcluding Total Fn PA Noe-Food & (Weigts Bner) Total Beverages Cereal Fate & Oil Other Minerals %Share) a/ (100.0) (67.7) (53.2) (22.3) (9.4) (9.3) (12.3) (14.4) (5.2) (27.1) 1948 55.9 176.2 228.7 226.8 203.4 280.0 396.7 137.6 235.0 50.7 111.9 1949 48.1 175.7 223.8 226.1 234.8 295.5 309.7 136.1 216.2 55.7 118.1 1950 48.8 232.3 308.4 286.4 343.7 320.6 370.2 165.4 380.0 66.5 136.1 1951 42.4 242.9 317.8 276.3 331.3 301.3 383.5 151.5 453.2 84.9 147.1 1952 40. 210.5 258.3 245.5 301.5 285.1 308.1 133.1 300.3 60.7 159.5 1953 44.1 2023 251.2 254.8 316.4 293.9 317.0 137.1 239.2 57.1 148.6 1954 47.6 224.0 285.5 299.0 458.6 279.1 300.8 141.6 241.5 80.1 148.1 1955 46.7 217.3 262.3 253.6 357.6 242.1 274.5 140.5 290.6 63.0 172.1 1956 45.1 211.5 254.1 253.5 362.0 236.4 271.6 140.3 256.1 58.6 170.1 1957 44.1 195.7 242.1 241.8 324.1 220. 267.8 154.6 243.3 55.3 145.1 1958 41.1 179.3 220.7 220.9 290.3 219.6 247.8 135.4 220.3 51.8 134.5 1959 37.0 181.3 220.8 210.2 258.9 211.8 268.0 129.6 255.5 61.9 138.7 1960 34.0 177.9 215.2 200.7 246.1 197.6 249.5 130.7 262.6 66.6 137.0 1961 33.5 166.4 199.9 193.7 229.2 208.9 248.2 120.9 220.3 67.7 129.3 1962 30.6 163.4 196.4 190.8 214.1 226.2 238.3 122.8 214.6 72.8 125.5 1963 31.2 170.9 207.7 209.1 215.7 230.6 256.2 166.7 203.1 72.3 128.2 1964 28.5 181.4 210.9 213.3 243.6 225.9 259.6 151.3 203.3 60.9 155.7 1965 28.2 184.4 201.7 203.1 231.7 219.3 283.3 124.3 197.4 69.7 178.6 1966 27.4 184.8 197.7 198.9 218.9 239.5 265.8 121.7 193.7 70.9 186.4 1967 27.0 172.1 193.3 196.7 210.7 249.2 251.7 125.0 182.0 74.9 156.0 1968 27.3 175.3 193.9 195.0 210.3 244.8 242.9 126.8 190.3 77.5 163.7 1969 25.9 177.9 191.9 191.4 207.2 230.5 228.2 134.3 193.7 71.0 176.1 1970 24.4 172.9 189.7 195.6 226.6 198.4 253.9 132.7 170.5 70.6 165.5 1971 30.2 153.4 172.4 175.5 187.2 181.8 240.9 127.5 162.2 68.3 138.6 1972 31.0 148.0 170.3 175.6 189.0 175.7 218.1 140.7 153.1 62.0 127.6 1973 38.0 194.1 229.2 235.4 207.5 310.0 389.3 1463 209.1 93.5 154.6 1974 129.4 212.3 249.2 266.8 196.2 368.8 361.4 236.2 191.7 91.9 173.9 1975 113.3 158.9 184.7 195.3 167.5 2593 218.3 176.5 150.0 62.4 134.6 1976 119.9 179.3 214.4 223.1 316.2 209.1 230.5 129.4 186.2 82.6 139.0 1977 119.4 197.0 247.8 271.0 484.8 174.8 260.1 106.5 172.2 87.4 132.5 1978 104.6 163.2 199.6 210.9 312.3 182.3 236.4 108.1 162.9 80.0 117.9 1979 133.1 167.9 195.2 202.5 285.6 168.0 238.9 116.9 171.4 123.2 130.2 1980 199.0 168.2 193.0 197.4 233.6 182.3 201.7 165.4 178.7 129.1 133.8 1981 222.9 146.6 167.9 172.0 193.6 194.4 193.7 125.8 154.7 102.4 119.2 1982 204.6 134.0 150.6 154.3 200.9 143.8 160.2 108.4 138.4 104.2 111.8 1983 189.8 144.2 165.3 167.3 207.8 159.1 193.3 116.6 159.0 99.3 117.1 1984 189.7 148.8 174.8 182.0 245.5 155.3 232.3 105.9 151.3 116.8 110.6 1985 182.7 131.3 151.0 158.7 224.4 134.7 161.4 101.8 126.1 94.4 105.1 1986 78.4 112.1 132.1 142.8 232.8 100.0 105.6 91.1 97.2 89.0 82.4 1987 90.9 102.0 109.6 109.6 139.2 85.9 111.9 90.8 109.9 118.3 85.4 1988 67.0 114.0 116.8 120.8 144.9 108.1 136.9 94.8 103.9 116.4 108.7 1989 80.9 112.8 112.2 114.5 120.4 116.3 124.3 102.3 104.6 112.9 113.8 1990 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 1991 79.6 93.9 95.5 95.1 92.2 100.0 100.8 92. 97.0 103.2 89.2 1992 77.6 87.2 85.8 86.8 73.8 94.4 104.0 87.7 82.4 112.0 84.4 1993 73.5 86.6 84.9 86.3 77.5 89.5 104.1 84.9 80.6 113.1 83.8 1994 72.6 87.5 88.1 88.9 85.0 89.2 104.5 85.2 85.6 114.4 80.9 1995 73.2 90.1 92.2 93.5 96.6 89.7 105.8 86.5 87.7 115.6 81.3 2000 84.5 94.2 96.4 97.4 109.1 91.2 98.0 88.2 93.4 122.6 84.6 2005 79.9 92.2 94.1 94.9 107.6 79.6 96.9 89.1 91.6 131.2 80.9 s/ Weighted by 1979-1981 Developing Coutries Expot Values. Source: World Bank, International Economics Department. November 9, 1992 XiX Table 4: Weightd Inde of Commodity Prom (Current US Dollars) (1990-100) lP.troleum 33 Comnmodides. A3am u an . . Timber Metuls (Exluding TOal Food N~OM&a (Wighgy Total Bvrag Creal & Oil oMnral %S~ar.) al (100.0) (67.7) (53.2) (22.3) (9.4) (9.3) (12.3) (14.4) (5.2) (27.1) 1948 10.8 34.0 44.1 43.8 39.3 54.1 76.6 26.6 45.4 9.8 21.6 1949 8.9 32.6 41.5 41.9 43.5 54.8 57.4 25.2 40.1 10.3 21.9 1950 8.0 38.0 50.5 46.9 56.2 52.5 60.6 27.1 62.2 10.9 22.3 1951 8.0 45.7 59.8 52.0 62.3 56.7 72.1 28.5 85.3 16.0 27.7 1952 8.0 41.5 50.9 48.4 59.5 56.2 60.8 26.2 59.2 12.0 31.5 1953 8.5 38.8 48.1 48.8 60.6 56.3 60.7 26.3 45.8 10.9 28.5 1954 8.9 42.0 53.5 56.0 86.0 52.3 56.4 26.5 45.3 15.0 27.8 1955 8.9 41.5 50.1 48.4 68.3 46.2 52.4 26.8 55.5 12.0 32.9 1956 8.9 41.9 50.3 50.2 71.6 46.8 53.7 27.8 50.7 11.6 33.7 1957 8.9 39.6 48.9 48.9 65.5 44.6 54.1 31.2 49.2 11.2 29.3 1958 8.5 36.9 45.4 45.4 59.7 45.2 50.9 27.8 45.3 10.6 27.7 1959 7.5 36.8 44.8 42.6 52.5 42.9 54.4 26.3 51.8 12.5 28.1 1960 7.0 36.8 44.5 41.5 50.9 40.9 51.6 27.1 54.4 13.8 28.3 1961 7.0 35.0 42.1 40.8 48.2 44.0 52.2 25.4 46.4 14.3 27.2 1962 6.6 35.1 42.2 41.0 46.0 48.6 51.2 26.4 46.1 15.6 27.0 1963 6.6 36.0 43.7 44.0 45.4 48.5 53.9 35.1 42.7 15.3 27.0 1964 6.1 38.8 45.1 45.6 52.1 48.3 55.6 32.4 43.5 13.0 33.3 1965 6.1 39.9 43.6 43.9 50.1 47.4 61.2 26.9 42.6 15.1 38.6 1966 6.1 41.2 44.1 44.4 48.8 53.4 59.3 27.1 43.2 15.8 41.6 1967 6.1 38.9 43.7 44.4 47.6 56.3 56.8 28.2 41.1 16.9 35.2 1968 6.1 39.2 43.4 43.6 47.1 54.8 54.4 28.4 42.6 17.3 36.6 1969 6.1 41.9 45.2 45.1 48.8 54.3 53.8 31.7 45.6 16.7 41.5 1970 6.1 43.3 47.5 49.0 56.7 49.7 63.6 33.2 42.7 17.7 41.4 1971 8.0 40.6 45.6 46.4 49.5 48.1 63.7 33.7 42.9 18.1 36.6 1972 8.9 42.7 49.1 50.6 54.4 50.6 62.8 40.5 44.1 17.9 36.8 1973 12.7 64.7 76.5 78.5 69.2 103.4 129.9 48.8 69.7 31.2 51.6 1974 52.6 86.3 101.2 108.4 79.7 149.8 146.9 96.0 77.9 37.4 70.7 1975 51.2 71.8 83.4 88.2 75.7 117.2 98.6 79.7 67.8 28.2 60.8 1976 54.9 82.1 98.2 102.2 144.9 95.8 105.6 59.3 85.3 37.8 63.7 1977 60.1 99.2 124.8 136.4 244.1 88.0 131.0 53.6 86.7 44.0 66.7 1978 60.6 94.5 115.6 122.1 180.8 105.5 136.9 62.6 94.3 46.3 68.3 1979 87.3 110.1 128.0 132.8 187.3 110.2 156.7 76.7 112.4 80.8 85.4 1980 143.2 121.1 138.9 142.0 168.1 131.6 145.1 119.0 128.6 92.9 96.2 1981 161.0 105.9 121.3 124.3 139.8 140.4 140.0 90.9 111.8 74.0 86.1 1982 145.5 95.3 107.1 109.7 142.9 102.3 113.9 77.1 98.4 74.1 79.5 1983 131.9 100.2 114.9 116.3 144.5 110.6 134.4 81.0 110.5 69.0 81.4 1984 129.1 101.2 118.9 123.8 167.0 105.6 158.0 72.1 102.9 79.5 75.3 1985 125.4 90.1 103.6 108.8 154.0 92.4 110.7 69.8 86.5 64.7 72.1 1986 63.4 90.6 106.8 115.5 188.2 80.8 85.4 73.6 78.6 71.9 66.6 1987 80.8 90.6 97.4 97.3 123.6 76.3 99.4 80.7 97.6 105.2 75.8 1988 63.8 108.7 111.4 115.1 138.1 103.0 130.5 90.3 99.1 110.9 103.7 1989 76.5 106.7 106.1 108.3 113.9 110.0 117.6 96.8 99.0 106.8 107.7 1990 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 1991 81.2 95.9 97.5 97.1 94.1 102.1 102.9 94.4 99.1 105.4 91.0 1992 82.6 92.9 91.4 92.5 78.6 100.6 110.8 93.4 87.7 119.3 89.9 1993 81.2 95.6 93.9 95.3 85.7 98.9 115.0 93.8 89.1 125.0 92.6 1994 81.7 98.5 99.2 100.1 95.8 100.4 117.6 95.9 96.4 128.8 91.1 1995 84.5 104.1 106.5 108.0 111.6 103.6 122.2 99.9 101.3 133.6 94.0 2000 116.0 129.4 132.4 133.7 149.8 125.2 134.5 121.1 128.2 168.3 116.1 2005 125.8 145.3 148.2 149.4 169.5 125.5 152.7 140.4 144.3 206.7 127.5 al W ~ighted by 1979-1981 Dveloping Couti Export Valm. Sourc: Word Bank, Internadonul Economica Departnt. Novmbr9, 1992 xx Table 5: Inflation Indice: 1948-2005 a/ --G5 MUV ndex b/- - US GNP Ddlator- - G-5 GNP D~aor c/- -7 CPI d/- Year 1990=100 % Oange 1990=100 % ~aage 1990=100 % Cange 1990-100 % Qamge 1948 19.29 17.89 7.01 1949 18.54 -3.90 17.80 0.52 13.25 1950 16.33 -11.90 18.15 2.01 12.40 -6.46 1951 18.83 15.30 19.03 4.82 13.74 10.84 1952 19.73 4.80 19.32 1.51 14.33 4.29 1953 19.18 -2.80 19.63 1.61 14.38 14.45 0.87 1954 18.16 -2.20 19.94 1.60 14.63 1.75 14.67 1.48 1955 19.11 1.90 20.58 3.24 14.94 2.16 14.72 0.38 1956 19.80 3.60 21.28 3.36 15.57 4.22 15.04 2.13 1957 20.22 2.10 22.04 3.60 16.08 3.24 15.37 2.24 1958 20.56 1.70 22.49 2.05 16.27 1.18 15.66 1.86 1959 20.27 -1.40 23.07 2.57 16.27 0.00 15.63 -0.21 1960 20.70 2.10 23.46 1.68 16.65 2.33 15.93 1.91 1961 21.05 1.70 23.68 0.96 17.21 3.41 16.30 2.36 1962 21.47 2.00 24.21 2.22 17.84 3.66 16.76 2.84 1963 21.06 -1.90 24.56 1.44 18.35 2.83 17.30 3.21 1964 21.43 1.75 24.96 1.64 18.85 2.75 17.75 2.60 1965 21.59 0.74 25.60 2.57 19.42 3.01 18.32 3.18 1966 22.35 3.50 26.50 3.50 20.05 3.25 18.94 3.39 1967 22.61 1.16 27.25 2.832 20.68 3.14 19.43 2.58 1968 22.39 -0.96 28.61 4.98 21.44 3.66 19.97 2.78 1969 23.60 5.40 30.16 5.42 22.07 2.94 20.84 4.38 1970 25.08 6.27 31.86 5.65 23.55 6.71 22.11 6.10 1971 26.44 5.42 33.64 5.59 25.44 8.03 23.60 6.71 1972 28.81 8.97 35.24 4.77 28.05 10.26 26.04 10.38 1973 33.37 15.84 37.55 6.55 31.82 13.44 29.76 14.27 1974 40.66 21.84 40.91 8.93 34.73 9.11 32.96 10.74 1975 45.21 11.18 44.96 9.91 39.28 13.11 37.03 12.36 1976 45.83 1.38 47.81 6.34 40.39 2.83 38.26 3.30 1977 50.34 9.85 51.00 6.67 44.03 9.02 42.29 10.54 1978 57.94 15.08 54.75 7.34 51.23 16.35 51.82 22.54 1979 65.62 13.26 59.56 8.79 57.10 11.46 54.84 5.82 1980 71.97 9.68 64.98 9.10 63.06 10.42 61.32 11.82 1981 72.26 0.41 71.23 9.62 62.65 -0.65 61.82 0.82 1982 71.15 -1.53 75.81 6.43 62.20 -0.72 61.04 -1.27 1983 69.53 -2.28 78.73 3.86 62.49 0.48 62.01 1.58 1984 68.05 -2.14 81.67 3.73 61.74 -1.20 61.84 -0.26 1985 68.60 0.81 84.11 2.98 63.11 2.21 63.10. 2.02 1986 80.89 17.91 86.37 2.49 76.53 21.27 76.14 20.67 1987 88.84 9.84 89.20 3.28 8632 12.78 86.01 12.97 1988 95.31 7.28 92.28 3.45 92.11 6.71 92.60 7.65 1989 94.65 -0.70 96.08 4.12 91.25 -0.94 92.18 -0.45 1990 100.00 5.66 100.00 4.08 100.00 9.59 100.00 8.49 1991 102.11 2.11 103.63 3.63 104.01 4.01 105.11 5.11 1992 106.53 4.33 111.65 6.22 1993 110.53 3.75 115.65 3.58 1994 112.58 1.86 118.37 2.35 1995 115.59 2.67 121.78 2.88 1996 119.63 3.50 126.40 3.80 1997 123.82 3.50 131.21 3.80 1998 128.15 3.50 136.19 3.80 1999 132.64 3.50 141.37 3.80 2000 137.28 3.50 146.74 3.80 2005 157.60 2.80 17535 3.65 a/ Fgurem for 1992-2005 a b/ Unit valu indacx iaUS d ms ofacæ- expored from the h -5 countries (Prae Germany, J1a, UK, ad US) weightd prøporially to awe countria' ~xot to b.davC1pC i = !. e/ Ag~ inde= ofGNP df a d~ rt~ for to h -5 cow~ mi« SDR-buaed moving d/ iua0 coum 0 nd~ iM US dollartms f~t the G- cmFrnce Grmany. aly pn, UK and US) weight by the ommrics' 1 avrage in crren USdollars. Sources 0-5 MUN ndex, -5 GNP De~ator, ad 0-7 CPI - Word Bank. US GNP De8ator - Inm&rauionalMontary Fund. World Bank, International o ics Dep~atment, Intenational Trede Diviio. Octobr 26, 1992 Mi Petroleum Swnmary * Petroleum consumption is expected to grow at an average 1.3% p.a. throughout the 1992-2005 period. * Fastest consumption growth should be seen in the LMICs (excluding Eastern Europe and the FSU), with their share of global consumption growing from 16% in 1991 to 36% by 2005. Increased use of motor transport will be the main reason for this growth. * Supplies from non-OPEC producers (excluding the FSU) will contribute about one half of incremental global output up to the mid-1990s. But beyond that time, the world will depend increasingly on OPEC as many producing regions reach maturity. * OPEC's share of global output is expected to increase from 38% in 1991 to around 46% in 2005. The share of Middle East producers is expected to grow from 28% to 37%. * The market outlook is for prices to decline slightly in real terms in the near term. In the latter half of the decade, we expect prices to increase in real terms as production begins to decline in several OPEC and non-OPEC producers. Projected Petroleum Production, by Reagio Phjected Petroleum Coasanption, by Region 3500 3500 3000 3000 2000 2000 0 ISM 1970 180 1990 los 2000 2001 1970 10 190 1995 2000 2005 |aws...mE Paw mum wD EmapeJMWW tiune UIC M. LUE smu Wsrep Petroleum Prices, a/ 1960-2005 C$/barrel) 40 Constant 1990 $ b/ A J % 30 % 101 20 /1 I /1 I 10 - Current $ I 0 I 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 a/ OPEC weighted average price b/ Deflated by G-5 MW Index Global petroleum consumption increased at a rate of 2.5% p.a. over the period 1985-89, reversing the declining trend of the early 1980s. However, since then, consumption growth has slowed significantly (to 0.2% p.a.), with modest increases in the industrial countries and developing countries partially offset by the decline in the FSU. In the FSU and Eastern Europe, petroleum consumption declined from 540 mtoe in 1989 to 503 mtoe in 1991. In the OECD and developing countries, sluggish growth in industrial production and incomes contributed to the slowdown in oil consumption growth. The oil price shock in late 1990 and the subsequent increased uncertainty about oil supplies may also have had an impact. Under our long-term baseline economic growth assumption of 2.8% p.a. for industrial countries and 5% for developing countries, petroleum consumption is expected to grow throughout the forecast period (to 2005) at an average rate of 1.3% p.a. (Table 1). Three fourths of the incremental oil consumption expected is anticipated to occur in developing countries. As a result, their share in global oil consumption should increase to 36% by 2005 (up from 16% in 1991). Increased motorization will be one of the main factors behind this strong performance. The transportation sector is likely to remain tied to liquid fuels over the forecast period, while the power generation and industrial and residential/commercial sectors are expected to be the main areas of substitution for petroleum products. However, the continuing introduction of more energy- efficient transport equipment should reduce the rate of growth of demand for transport fuels. In addition, higher gasoline prices resulting from compliance by refineries with environmental standards in the OECD and the continuation of the recent trends in developing countries to align domestic oil prices to reflect prices in the international market, are likely to constrain the increase in oil demand. 2 Table 1: Global LAng-Run Petroleum Market Outlook to 2005 1991 2005 Incremental Change Regions MTOE % MTOE % MTOE % % p.a. Oil Consumption Industrial 1,740 56 1,884 50 144 23 0.6 Eastern Europe & FSU 503 28 517 14 14 2 0.2 Other Countries 879 16 1,343 36 464 75 3.1 Total 3,122 100 3,744 100 622 100 1.3 Oil Production Industrial Countries 779 26 708 19 71 -11 -0.7 Eastern Europe & FSU 541 17 550 15 9 1 0.1 Other Countries 1,802 58 2,486 66 684 110 2.3 Memo: Non-OPEC Developing Countries 625 20 773 21 148 24 1.5 Non-OPEC (excl. FSU) 1,404 45 1,481 40 77 12 0.4 Total Non-OPEC 1,945 62 2,031 54 86 14 0.3 OPEC 1,177 38 1,713 46 536 86 2.7 Middle East 864 28 1,379 37 515 83 3.4 Prices ($/bbl) Current 17.3 26.8 9.5 3.2 Constant ($1990) 17.0 17.0 0.0 0.0 Source: World Bank, International Economics Department. We expect the declining trend in FSU oil consumption to continue up to the mid- to late 1990s. This decline will be forced by several factors including: (i) the expected decline in FSU oil output; (ii) the shift to market pricing; and (iii) improvements in the efficiency of oil/energy use. However, over the longer run, we expect oil consumption to increase under the influence of rising per capita incomes and transport demand. Since 1980 non-OPEC crude oil output has increased by around 1.1% p.a. to 1,945 mtoe in 1991. Declines in the United States and the FSU (in the late 1980s) were more than offset by increases in the North Sea region and in developing countries. Based on estimates of recoverable oil reserves (and reserves-to-production ratios) and expected investments, the outlook for supplies from non-OPEC sources (excluding the FSU) is for an increase of around 0.4% p.a. to 1,481 mtoe in 2005 (up from 1,404 mtoe in 1991). Crude oil production in the FSU is expected to decline over the medium term, but 'due to the impact of foreign investment and the supply response to increases in domestic prices, is expected to recover to about 550 mtoe by the year 2005. In aggregate, therefore, non-OPEC production is expected to supply roughly 14% of the increment in the global oil demand up to 2005. Up to the mid-1990s, the 3 increase in output in industrial countries and in the developing countries should more than offset the decline in the FSU. While increased foreign participation and investment should increase supplies from non-OPEC developing countries, at the same time liberal fiscal terms for the development of new fields and the application of improved technologies are likely to support increases in the OECD (mainly in the North Sea region). However, given the high rate of depletion of OECD oil reserves in the 1980s, production from the industrial countries is projected to decline after 1995. Non-OPEC (excluding the FSU) producers will contribute about one half of the incremental global output up to 1995. However, beyond the mid-1990s, the world will depend increasingly on OPEC for additional supplies as many producing regions will reach maturity and begin to deplete. Since a few OPEC members will also reach production plateaus by the mid-1990s and several others will enter the depletion phase towards the end of the century, a smaller group of producers including Saudi Arabia, Iran, the United Arab Emirates (UAE), Iraq, and Kuwait will attain a more dominant position in the world oil market. However, we do not believe that OPEC will be able to exert effective control over the market. It is anticipated that by 2005, production in OPEC countries will increase to 1,713 mtoe with OPEC's share in total petroleum production rising to around 46% (up from 38% in 1991). Middle East OPEC production is expected to increase at a trend rate of roughly 3.4% p.a. over the forecast period and its market share to increase to over 37% in 2005 (up from 28% in 1991). The market outlook for crude oil prices is presently for an increase to $18.40/bbl in the second half of 1992 compared with a $16.70/bbl first half 1992 average. Market expectations that crude oil supplies will constrain increases in oil consumption and the view that the UN embargo on Iraqi crude oil exports will continue underlie the expected price increase. However, in 1993 prices are expected to decline to an average of $17.30/bbl with any increase in global demand being more than offset by increases in supplies from OPEC and non-OPEC sources. Over the period 1993-95, we expect (nominal) prices to increase by 2% p.a. However, in real terms, they are expected to remain unchanged. The impact on prices of economic recovery and growth, particularly in the United States, and further declines in supplies from the FSU is offset by increases in supplies from OPEC sources. Based on our long-run macroeconomic assumptions, we expect global oil demand to increase at a rate of 1.5% p.a. (by 259 mtoe) over the period 1995-2000. Only modest increases are expected in non-OPEC supplies during this period which will put pressures on oil-producing capacity. Based on the expected increase in OPEC output capacity to 1,715 mtoe, an increase in OPEC output to 1,480 mtoe to meet demand translates into an OPEC capacity utilization rate of over 87%. Therefore, with the tightening market balance, prices should increase in real terms from $15.60/bbl to $18/bbl during this period. To test the robustness of these results, we also carried out a model simulation which assumes flat prices (in real term) over the forecast period. We have found that if oil prices are held constant at $16.50/bbl (the average price expected in 1992 in constant 1990 US dollars) over the period 1992-2000, the projected demand and supply responses exert strong pressures on OPEC's oil production capacity. We estimate that in this situation, global oil demand would be roughly 100 mtoe (2 mb/d) higher and non-OPEC supply would be roughly 50 mtoe (1 mb/d) lower than assumed in the base case by the year 2000. This would lead to an increase in OPEC's capacity utilization rates to roughly 95% 4 thereby exerting strong upward pressure on prices. Alternatively, at a utilization rate of around 81%, it is estimated that OPEC's output capacity would need to expand to roughly 1,850 mtoe (37 nb/d) by 2000 to meet oil demand growth generated by flat oil prices. These estimates are outside the range of industry estimates (or at the very high end) for OPEC's output capacity. Beyond the year 2000, we expect a downturn in prices. OPEC output capacity is expected to reach 2,000 mtoe by the year 2005 mainly in response to large investment incentives provided by the higher prices in the late 1990s. FSU output is also expected to increase as a result of increased foreign investments and application of improved technologies. At the same time, oil demand growth in the OECD should decline and developing countries' consumption growth should slowsdue to improvements in the efficiency of oil use and increased use of natural gas-primarily for environmental reasons. The long-run base case price forecasts are based upon the following key assumptions: (i) GDP growth in the OECD and developing countries will be 2.8% p.a. and 5% p.a., respectively, through the year 2000; (ii) OPEC output capacity will increase to about 1,700 mtoe (34 mb/d) by 2000 and to over 2,000 mtoe (40 mb/d) by 2005; (iii) the income elasticity of oil demand is close to unity; (iv) non- OPEC oil output (excluding the FSU) will increase to 1,475 mtoe (29.5 mb/d) in 2000 and remain around this level; and (v) FSU output will continue to decline up to the mid-1990s but increase steadily thereafter to reach 550 mtoe (11 mb/d) in 2005 (see Table 1). Different assumptions for any of these variables would give different forecasts. Under an assumption of higher OECD economic growth (from the 2.8% p.a. assumed in the base case up to 3.2% p.a.), the market-clearing price of oil (in real terms) trends above the baseline by about 2.5% in the medium term and about 5% in the long term. However, a corresponding decline in the economic growth assumption (down to about 2.5% p.a.) has a greater impact on prices. In this situation, prices trend below the projected baseline by about 3% in the medium term (1995) and about 8% in the long term (2005). We estimate the impact on world oil prices of alternative assumptions about the income elasticity of oil demand to be as follows: under an assumed higher income elasticity of demand of 1.4, prices trend about 3% higher in the medium term and 15% higher in the long term. Alternatively, a reduced income elasticity of demand to 0.6 results in price trending 3% and 14% lower in the medium and long term, respectively. Adoption of environmental policies such as use of carbon taxes to achieve CO2 reduction is assumed in the baseline. If more aggressive OECD policies are implemented to reduce dependence on oil imports than we have assumed, this would reduce crude oil prices even further. We estimate that reduction in oil demand (of 1-2 mb/d in the medium term and 3-4 mb/d in the long term) due to environmental policies would lower oil baseline forecast prices by 4-6% in the medium term and 8-10% in the long run, ceteris paribus. Output forecasts for non-OPEC suppliers to the year 2000 have been made within a range of 1,200 to 1,700 mtoe. An assessment of the impact of different assumptions about non-OPEC output on the long-term price shows that an increase in non-OPEC oil supplies to 1,700 mtoe (as compared to our base case assumption of 1,475 mtoe, excluding the FSU) would result in the long-run price of oil trending about 11% below the baseline. However, assuming production from non-OPEC sources at the 5 lower end of the range (1,200 mtoe) results in the long-term oil prices trending about 18% above the baseline forecast, since it would lead to sharp increases in the utilization of OPEC's oil production capacity. In this case, the impact on prices would be lessened beyond the year 2000 because of lagged demand and supply (OPEC) side responses to the higher prices. In assessing the impact of different FSU supply prospects, we see the price risks associated with unexpected changes in FSU oil supplies to be asymmetric. The medium-term impact on prices would be greater where the FSU output increases than would be the case of a decline. This is based on our view that a decline in FSU output would be offset by increases from other producers. However, in a situation of a higher-than- anticipated output, other producers would be reluctant to reduce output below prevailing levels. In the case of sharply lower FSU output, say down to 400 mtoe in 1995 and 300 mtoe in 2000 (assumed under a disruption case), prices would trend only 6-10% higher than in the base case. By comparison, prices would trend 16-21% lower in a situation of higher-than-expected increases in FSU oil supplies (assuming 625 mtoe in 1995 and 700 mtoe in 2000). However, where increases in FSU output are accompanied by supply adjustments by OPEC producers (say mothballing of output capacities), or by increases in demand, the price decline would be moderated. Given the uncertainties surrounding the key assumptions, we recommend strongly that the price forecasts be used with appropriate caution. In particular, since supplies will be increasingly concentrated in a few countries in the Middle East, the probability of supply-side shocks will remain high. Therefore, to cope with the inherent uncertainty in crude oil and petroleum product markets, decisionmakers should adopt suitable risk management strategies. Demand Outlook RECENT DEVELOPMENTS. Global petroleum consumption increased at a rate of about 2.5% p.a. over the period 1985-89, reversing the declining trend of the early 1980s (Table 2). The oil price shock in 1979, and the expectation that prices would continue to rise under the influence of OPEC production controls, resulted in declines in global oil consumption by about 2% p.a. over the period 1979-85. As a result, the share of oil in primary energy declined from 46.8% to 39.3%. The consumption increase during the late 1980s was in response to the sharp price decline in 1986, as well as the rapid growth in industrial production and income, both in industrial and developing countries. However, despite the resumption of consumption growth, global consumption is still below the annual average in 1979. The increases in the developing countries have been more than offset by declines in the OECD and FSU. In the OECD, lower oil consumption is attributed mainly to a reduction in the use of heavy fuel oil for industry and for power generation. In both these sectors, oil lost market share mainly to coal as well as to nuclear energy and natural gas in power generation. In the FSU, supply constraints and economic uncertainty resulted in an overall decline in oil use. Global oi demand rose in 1991 following dedine late in 1990. Petroleum consumption declined slightly in late 1990 in response to the sharp increase in price due to the hostilities in the Middle East. Initially, for inventory accumulation purposes, petroleum demand increased sharply in the third quarter of 1990, mainly in the industrial countries. However, in the fourth quarter, global oil offtake declined sharply. In the OECD region, the decline reflected higher end-use prices, as well as a slowdown in both inventory buildup and economic activity. In developing countries, such as India, the Philippines, and Pakistan, the slowdown in demand growth was a result of government policies that mandated cuts 6 Table 2: Recent Changes in World Petroleum Consumption, 1979-91 1979 1985 1989 1991 GrowthRates __________________________________ - % jp..)-- Regions MTOE % MTOB % MTOB % M'IE % 1979-85 1985-89 1989-91 Industrial 1,996.3 62.8 1,585 52.2 1,735 55.8 1,739.7 55.7 -3.8 2.3 1.3 E. Europe & PSU 550.7 17.3 545 19.3 540 17.4 503.0 16.1 -0.2 -0.2 -3.5 Other IMICs 631.0 19.9 690 24.5 832 26.8 879.0 28.2 1.5 4.8 2.8 Total 3,178 100.0 2,820 100 3,107 100 3,121.7 28.2 -2.0 2.5 0.02 Note: Details may not add up due to rounding. Sources: United Nations Energy Statistics; Petroleum Economics Limited.; World Bank, International Economics Department. in consumption to alleviate the economic impact of the high costs of petroleum imports. Despite that, developing countries registered an increase in 1990 which offset the decline in the industrial countries and the FSU. For 1990 as a whole, global oil demand remained largely unchanged at around 3,106 mtoe. Consumption increased by roughly 0.5% in 1991 to 3,122 mtoe as a result of the sharp decline in oil prices following the end of the Middle East hostilities. The most significant increases occurred in the developing countries. There were, however, sharp declines in the FSU and Eastern Europe (5%) in response to the political and economic uncertainties there, and production declined in the FSU, reducing the availability of petroleum. In the industrial countries consumption was unchanged in aggregate with increases in Western Europe (mainly Germany) offset by declines in the United States due to sluggish economic growth. DEMAND OUTLOOK: GLOBAL. Demand to increase through 1995 at 0.9% p.a., long- term growth rate 1.3% p.a. Based upon our assumptions about economic growth rates, over the near and medium term (1991-95) we expect global petroleum demand to increase by 0.9% p.a. (Table 3). Increases in developing countries (3.2% p.a.) and industrial countries (0.9% p.a.) will more than offset expected declines in the FSU (3% p.a.). OECD oil consumption is expected to increase by over 1% in 1992. The upturn in the industrial countries is predicated on economic recovery. The expected decline in oil consumption in the FSU is based on the response to higher oil prices and weak economic performance. Also, supply bottlenecks are likely to restrain consumption in the FSU. Under our long-term baseline economic growth assumption of 2.8% p.a. for the industrial countries and 5% p.a. for the developing countries, we expect global oil demand to increase by 1.3% p.a. between 1991 and 2005, to reach 3,744 mtoe by the year 2005. This growth rate is slower than the 1.8% p.a. growth recorded over the 1985-91 period. Improvements in the efficiency of oil use and other conservation practices, and increases in energy taxes are expected to partially offset the impact of economic growth on oil demand. Another factor expected to moderate the long-run growth in global oil demand is the increased use of alternative fuels, such as natural gas in industrial and power generation applications, especially in the late 1990s and beyond. The substitution of natural gas for heating (gas) oil in the OECD's residential and commercial markets should contribute to a slowdown in oil demand. Over the long run, natural gas use is expected to increase rapidly in the FSU where gas reserves are 7 Table 3: Short- and Medium-Term Oil Demand Forecasts by Economic Regions, 1992-95 Growth Rates 1992 1993 1995 -(% p.a.)- Regions MTOE % MTOE % MIWE % 1991-93 1993-95 1991-95 industrial 1,759 56.0 1,776 56.0 1,807.0 56.0 1.0 0.9 0.9 B. Europe & FSU 480 15.0 460 14.5 430.0 13.0 -4.2 -3.4 -3.0 Other LMICs 907 29.0 935 29.5 998.0 31.0 3.2 3.2 3.2 Total 3,145 100.0 3,172 100.0 3,235.0 100.0 0.9 1.0 0.9 Source: World Bank, International Economics Department. abundant. However, in the developing countries, growing populations and rising per capita incomes will increase the demand for oil more rapidly than in other regions of the world. Within the developing countries, oil demand growth in the Asia-Pacific region is expected to be the fastest for transportation, industry, and electricity generation needs. DEMAND OUTLOOK: INDUSTRIAL COUNTRIEs. Industrial country demnd to moderate. Oil demand in the industrial countries over the forecast period 1991-2005 is expected to increase at about 0.6% p.a. Growth will be tied mainly to petroleum's use in the transport sector, while there are likely to be declines in other major sectors (industrial and power). The introduction of more efficient transport vehicles, higher gasoline taxes, and tighter regulations should moderate the rate of growth in transport demand. Other factors expected to contribute to the slowdown in demand growth include the perceived need to reduce dependence on imported oil and efforts to reduce CO2 emissions. These two concerns have led to the development of a host of regulatory measures in the OECD to manage oil demand. However, contrasting approaches are being pursued in OECD countries. In the United States, reductions in gasoline demand growth are expected to be achieved through regulatory measures such as the Corporate Automobile Fuel Efficiency (CAFE) standards. This regulation mandates improvements in automobile fuel efficiency. In Western Europe, reductions are expected to be achieved through increases in taxes on gasoline. Gasoline taxes already make up about 70% of end-use prices in Western Europe compared with only 30% in the United States. Providing public information on the fuel efficiency of car models is a measure proposed to promote efficient use. It is argued that this information will encourage competitiveness between car manufacturing companies in improving their efficiency. Some governments are exploring the option of road pricing and greater use of congestion tolls. Despite these measures, the use of transport fuels such as gasoline is expected to increase by 1.2% p.a., with the impact of fuel efficiency gains more than offset by the increase in the average miles driven per car as well as an increase in the number of cars (Table 4). However, the long-run demand for fuel oil will decline (-1.7% p.a.) mainly because it is highly substitutable and because its use will be influenced by environmental considerations. Fuel oil is expected to lose further market share to natural gas in the industrial and power-generation sectors as governments pursue a policy of favoring use of natural gas. Environmental concerns are likely to be manifested in the form of "carbon taxes" which should increase the price of all polluting fuels and reduce their consumption. However, the relative price impact should be greater on boiler fuels such as coal which is highly polluting. The substitution effect 8 Table 4: OECD: Forecast Petroleum Product Demand, 2005 1990 2005 Increment Growth MTOE % MTOB % MTOE % p.a. Gasoline 547 31.5 650 34.5 103.0 1.2 Middle distillates 565 32.5 607 32.2 28.5 0.5 Fuel oil 208 12.0 160 8.5 -48.0 -1.7 Others 417 24.0 467 24.8 50.0 0.8 Total 1,737 100.0 1,884 100.0 147.0 0.6 Sources: Petroleum Economics Limited and World Bank (1990); World Bank, International Economics Department (forecasts). may favor fuel oil use vis-a-vis coal and therefore may moderate the decline in fuel-oil use. Switching to natural gas in the OECD residential sector is likely to moderate demand growth for home-heating oil. The slowdown (or reduction in some OECD countries) in oil demand is likely to be influenced by energy-security concerns. However, in most cases the high costs of substituting for imported oil are likely to impose limits on the reductions that will be achieved. Increased flexibility in the use of factors of production and a buildup of strategic petroleum stocks are some of the policies being pursued in the OECD to reduce exposure to oil imports. DEMAND OUTLOOK: DEVELOPING COUNTRIs. Developing country demand continues to be strong. Under our baseline long-run economic growth assumption of 5% p.a., oil demand growth in developing countries should remain robust. However, sharp regional and sectoral differences are expected. Petroleum demand is expected to increase most rapidly in the Asia-Pacific region-although demand growth should be reasonably strong in other regions such as the Middle East and Latin America (Tables 5 and 6). In most countries in the Asia-Pacific region, the income elasticity of energy demand appears to be close to unity. In several countries such as India, Indonesia, Malaysia, Pakistan, the Republic of Korea, and Thailand oil consumption increased by over 6% p.a. between 1985 and 1990. However, the forecast growth rates are much lower than recent growth rates in these countries. In the oil-importing developing countries, balance of payments and debt-servicing difficulties are likely to lead to policies such as market- based energy pricing to slow the rate of increase in demand. This will involve elimination of subsidies on oil products. Oil-producing countries such as Indonesia and Malaysia are expected to try to limit domestic use to maximize revenues from oil exports. Indonesia and Malaysia are major producers of natural gas. Greater market penetration of natural gas is at the heart of fuel diversification policies in these countries. In the Middle East, diversification of the industrial base, increases in oil-fired power generation, and a shift in the mode of transportation from animals to greater use of cars and trucks should lead to oil demand increasing at a rate of 3% p.a. over the forecast period. In Latin America, structural adjustment policies and the resulting economic recovery is expected to lead to oil demand increasing at a rate of 2.8% over the long term (up from 2.2% p.a. in the 1980s) (see Tables 5 and 6). 9 Table 5: Non-OECD Oil Demand Forecasts by Income Group to 2005 1991 a/ 1993 1995 2005 MTOE % MTOB % MTOB % MTOE % High-Income 51 3.7 53 3.8 55 3.8 73 3.9 LMICs 1,329 96.3 1,343 96.2 1,373 96.2 1,787 96.1 Africa 52 3.8 54 3.9 57 4.0 70 3.7 Americas 262 19.0 277 19.8 293 20.5 386 20.8 Asia and Pacific 325 23.6 354 25.4 382 26.7 530 28.5 Europe b/ 503 36.4 460 32.9 430 30.10 517 27.8 Others 187 13.5 198 14.2 211 14.8 284 15.3 Total 1,380 100.0 1,398 100.0 1,428 100.0 1,860 100.0 a/ World Bank estimates. b/ Including the FSU. Source: World Bank, International Economics Department. Table 6: Forecast Growth Rates of Non-OECD Oil Demand by Income Group Growth Rates (% p.a.) 1991-93 1993-95 1995-2005 1991-2005 High-Income 1.9 1.9 2.9 2.6 LMICs 0.5 1.1 2.7 2.2 Africa 1.9 2.7 2.1 2.1 America 2.8 2.8 2.8 2.8 Asia and Pacific 4.4 4.0 3.3 3.6 Europe a/ -4.5 -3.4 1.9 0.2 Others 2.9 3.2 3.0 3.0 Total 0.6 1.1 2.7 2.2 a/ Including the FSU. Source: World Bank, International Economics Department. 10 While all sectors are expected to contribute to the increase, oil demand growth should remain strongest in the transport sector. Rising per capita incomes and the growing demands for personal and commercial mobility should support increases in the use of both gasoline and diesel oil. Based on the anticipated growth in the number of vehicles per thousand people (expected to quadruple over the next 20 to 25 years), and given our economic growth assumptions, we expect the demand for transport fuels (gasoline) to increase at an average rate of 4.2% p.a. over the period to 2005 and their share to increase from 18% to 21% (Table 7). Supported by growth in the residential and commercial sectors, the long- run growth in middle distillate demand is expected to average 3.4% p.a. Factors contributing to the growth include: (i) the continued transition from biomass fuels to petroleum products; (ii) growth in the number of urban households; and (iii) rising levels of income and population. Distillate demand is also expected to be supported by its use for power generation. Fuel oil demand (for power generation) is expected to increase almost as rapidly as the demand for residential fuels (3.2% p.a.), supported by its use in electricity generation. The demand for electricity is expected to grow faster than GDP in many developing countries. The implied income elasticity is between 1.2 and 1.5. The political limits to nuclear-power generation and the expectation that natural gas demand will be constrained by supply until the end of this decade are the key factors behind these forecasts. Given that gas is not likely to be available (because of the heavy investment needed to develop and transport natural gas), gas turbines should be fueled by distillates. Additional generation capacity-at least in the 1990s-therefore, is likely to be oil-fired. Countries facing severe capital constraints and therefore with limitations on their ability to install natural gas facilities are likely to put greater reliance on oil for new generation capacity and industrial installation, further supporting the use of oil. Several developing countries, including the Republic of Korea, are starting up oil-fired plants. These trends are also noticeable in Japan. Desulphurization of fuel oil should stimulate demand for low sulphur fuel oil at the expense of coal-powered generation. DEMAND OUTLOOK: THE FSU AND EAsFERN EUROPE. Declining trend in FSU and Eastern Europe to continue to late 1990s. Oil consumption in the FSU and Eastern Europe has declined sharply-from 560 mtoe in 1989 to 530 mtoe in 1990, and to an estimated 503 mtoe in 1991. The decline in demand is attributed to reduced oil supplies in the FSU, increases in domestic energy prices, and Table 7: Developing Countries: Forecast Petroleum Product Demand to 2005 a/ 1990 2005 Increment Growth MTOE % MTOE % MTOE % p.a. Gasoline 151 18.0 280 20.8 129 4.2 Middle Distillates 302 36.0 498 37.1 196 3.4 Fuel Oil 201 24.0 320 23.8 110 3.2 Others 185 22.0 245 18.3 60 2.0 Total 839.0 100.0 1,343 100.0 504 3.2 a/ Excluding FSU. Sources: Petroleum Economics Limited, and World Bank (1990); World Bank, International Economics Department (forecasts). 11 economic contraction. These declines in production are attributed, inter alia, to the lack of replacement parts for rapidly deteriorating equipment. Roughly 12% of Russian wells have been shut-in for this reason. In Western Siberia, 20% of the wells are shut down due to the lack of spare parts. Energy prices in the FSU and Eastern Europe have been kept at very low levels in comparison with international prices. Energy prices have been increased sharply since January 1992. Despite the price increase in May to between 1,800 and 2,000 rubles/ton (roughly $2.50-3/bbl), they are still significantly below world levels. In some Russian commodity exchanges, crude oil has at times been traded as high as 10,000 rubles ($14/bbl), even prior to the most recent price increase. The complete deregulation of domestic prices expected over the next one to two years will result in a significant increase in prices, thereby lowering oil demand further. The inefficient practices of the past such as price controls and preferential trade have resulted in wasteful use of energy resources. Energy use is high both on the basis of per capita consumption and intensity of use per unit of output. In 1990, energy intensity in the FSU and Eastern Europe was roughly 32 barrels and 19 barrels, respectively, per 1980 US$1,000 of GDP, compared with 2.6 barrels in Western Europe. In the same year, per capita energy consumption in the FSU and Eastern Europe was roughly 32 barrels compared with 39 barrels in the OECD and only 4 barrels in the developing countries. Thus, there exists large scope for improvement in the efficiency of energy use. The market-oriented reforms should lead to significant improvements in energy use in areas where initial low- cost savings are possible. The scope appears largest in the energy-intensive heavy industries such as steel, nonferrous metal fabrication, chemical industries and material building. In the transport sector where bad roads and large and heavy vehicles have contributed to high energy use, efficiency can be improved by reducing the size and weight of the vehicles. Economic activity in the FSU is estimated to have fallen by about 20% between 1990-91 and 1991-92. We expect the declining trend in oil consumption to continue up to the mid- to late 1990s because of: (i) the expected decline in the FSU output; (ii) the elimination of preferential trade between the FSU and Eastern Europe and the shift to market pricing; and (iii) improvements in the efficiency of oil/energy use. However, over the long run, we expect oil consumption to increase under the influence of rising per capita income levels and transportation demands. But higher energy prices, due to market- based reforms, and efficiency improvements in the industry and transport sectors are expected to limit the growth in oil demand. Increased availability of natural gas is also likely to restrict the increase in oil demand in end-use markets. We expect oil consumption in the FSU and Eastern Europe to decline initially from 530 mtoe in 1990 to 430 mtoe in 1995, but to rise gradually in the late 1990s to reach 517 mtoe in 2005 (giving an average of 0.2% p.a. growth over the period 1991-2005) (see Table 5). However, the outlook for oil demand in the FSU and Eastern Europe is highly uncertain. Underlying uncertainties are the economic and political developments and the speed with which any reforms will be implemented. While most forecasts of FSU oil demand foresee declines up to the mid- 1990s, longer-run forecasts vary greatly. For instance, the SNS Energy group forecasts FSU oil demand to decline to as low as 290 mtoe (5.8 mb/d) by 2000.' The US Department of Energy forecasts oil demand in the FSU and Eastern Europe to fall within a range of 400-575 mtoe by 2000 and 435-750 mtoe 'Marian Radetzki et al., "Economic Reforms and Energy in the CIS," Jue 1992. 12 by 2010. Its base case forecasts for 1995, 2000 and 2010 are 425 mtoe, 490 mtoe, and 590 mtoe, respectively (compared with 515 mtoe in 1990).2 Supply Outlook NoN-OPEC OIL SUPPLY FoREcASrs. Long-run non-OPEC supplies to rise at 0.3% p.a. Since 1980, non-OPEC oil supplies have increased by around 1.1% p.a. and reached 1,945 mtoe in 1991. Declines in the United States and the FSU were more than offset by increases in the North Sea and in developing countries (Table 8). Based on the estimates of proven recoverable reserves (reserves-to-production ratios) and expected investments (to increase production capacities), the outlook for supplies from non-OPEC sources (excluding the FSU) is for an increase of around 0.4% p.a. to reach 1,482 mtoe in 2005 (up from 1,404 mtoe in 1991). With the inclusion of the FSU, non-OPEC supplies are expected to increase at an average rate of 0.3% p.a. and reach 2,032 mtoe in 2005 (from 1,945 mtoe in 1991). Non-OPEC oil production is expected to supply roughly 15% of the increase in the global demand for oil over the forecast period. The expected changes in production capacities of selected non-OPEC countries are presented in Table 9. NORTH SEA PROSPECTS. North Sea production expands slowly. North Sea production increased by about 7% in 1991 to reach 196 mtoe. Increases in Norway (14%) and Denmark (17%) more than offset declines in United Kingdom (0.3%) and Netherlands (-8.7%). In 1992, North Sea Table 8: Non-OPEC Oil Production, 1980-91 Growth Rates (mtoe) (%) 1980 1989 1991 a/ 1980-89 1988-91 Total Non-OPEC 1,732.0 1,978.0 1,945.0 1.5 -0.8 Norway 24.5 78.8 97.5 14.0 10.5 United Kingdom 80.6 92.2 92.0 1.5 0.0 United States 479.9 437.5 444.8 -1.0 0.8 FSU 612.4 615.7 544.0 0.06 -6.2 Developing Countries 397.0 597.0 622.0 4.6 2.1 Mexico 107.1 143.7 149.6 3.3 2.0 a/ Estimates. Source: United Nations, Energy Statistics; Petroleum Economics Limited; World Bank, International Economics Department. 2 US Department of Energy and Energy Information Agency, "International Energy Outlook 1992." 13 Table 9: Potential Petroleum Capacity Changes in Major Non-OPEC Countries a/ Change 1991 1995 2000 1991-2000 (--mbld) Developing Countries 10.12 11.44 12.64 22 Angola 0.50 0.60 0.60 0.10 Brazil 0.68 0.80 0.90 0.22 China 2.77 3.00 3.20 0.43 Colombia 0.46 0.55 0.55 0.09 India 0.69 0.79 0.90 0.21 Malaysia 0.65 0.75 0.85 0.20 Mexico 3.00 3.10 3.50 0.50 Oman 0.70 0.75 0.72 0.02 PNG 0.00 0.15 0.27 0.27 Syria 0.45 0.50 0.40 -0.06 Yemen (Unified) 0.21 0.45 0.75 0.54 Industrial Countries 14.87 15.20 13.90 -0.97 Canada 1.95 2.10 2.10 0.15 Norway 1.95 2.30 2.00 0.05 United Kingdom 1.97 2.60 2.00 0.03 United States 9.00 8.20 7.80 -1.20 FSU 10.30 10.00 10.50 0.10 Total 35.29 36.64 37.04 1.75 a/ Includes natural gas liquids. Source: Petroleum Economics Limited; World Bank, International Economics Department. output is expected to increase to about 215 mtoe with both the United Kingdom and Norway contributing to the increase. It is expected that North Sea production will increase up to the mid-1990s. The resource base and the incentives provided by the recent changes in the tax system are expected to support increases. In a bid to encourage exploration and development, several North Sea producing countries have introduced liberal fiscal terms for the development of new fields. As a result of the favorable tax regimes, North Sea output is expected to increase by about 45 mtoe (or 0.9 mb/d) to reach 245 mtoe in 1995. The application of improved technologies (which should reduce the rate of decline in mature fields), the development of new, small fields, and adoption of various cost-cutting methods are expected to support the increase. In the United Kingdom, the projected increase partially reflects a recovery in production from fields affected by a series of accidents in 1989 and partially the result of policy changes which 14 provide incentives to develop new (less costly) fields. However, beyond the mid-1990s, output is expected to decline from large fields including the Brent, Forties, Ninian and Piper fields, it should also fall as replacement production from newer fields becomes more costly. Supported by its reserves of about 7.6 billion barrels and with the expansion and enhanced recovery program on large existing fields (Gullfak and Oseberg fields), Norway's production should continue to increase beyond the mid-1990s. Development of smaller fields should also support increases in production. As a result, Norway's output is expected to increase to about 115 mtoe by the turn of the century. UNrrED STATEs. US prduction to continue faling as investment declines. Following a brief period of increase in late 1990 and early 1991, mainly from Alaskan fields in response to higher prices due to the Middle East crisis, crude oil production in the United States has resumed its decline. In 1992, US output is expected to decline to an average of 439 mtoe (from 445 mtoe in 1991). Declines in production from aging fields in Alaska and a slump in drilling activity are contributing to the decline. Total wells drilled in the United States are now roughly one third of the levels in 1985. A shift in oil company investment out of the United States has contributed to the decline in production. Amid the general perception that most of the oil-producing areas in the United States have been tapped and because of the high costs of the fields (due to their size and age), oil companies have reduced exploration and development (E & D) investments in the United States in favor of more profitable regions elsewhere. The US share of worldwide capital spending for upstream activity declined from about 50% in 1988 to roughly 43% in 1990. These trends are expected to continue. As a result of declining investment, US oil production is expected to decline to 374 mtoe by the year 2005. This expectation is supported by the low levels of US proven reserves, the recent poor record in finding replacement reserves, and the constraints on E & D from environmental concerns. At the end of 1991, US proven reserves were estimated at about 26.25 billion barrels-a slight decline from the previous year. At current production levels, this translates into another ten years of reserves. In 1970, US proven reserves tallied about 39 billion barrels with a reserves-to-production ratio of 12:1. Given that the best prospects for discoveries lie in the environmentally sensitive areas such as the coastal plains of the arctic national wildlife refuge in Alaska (ANWRA) and in the outer continental shelf (Gulf of Mexico) environmental concerns will severely limit US prospects. In the ANWRA, reserves are assessed at about 3.57 billion barrels with a 5% chance of finding up to 9.24 billion barrels. FSU AND EASTERN EUROPE. Significant capital ifecdion needed in FSU and Eastern Europe. From its peak of 628 mtoe in 1987, crude oil production in the FSU declined to 596 mtoe in 1990. Output in 1991 is estimated at 541 mtoe. The decline in output in the Russian Republic, mainly in the Western Siberian province of Tyumen, accounted for about 80% of the decline in FSU oil production in 1991. Production in Western Siberia declined by about 45 mtoe. Current FSU output (early 1992) is estimated to be below 500 mb/d. The Russian Republic accounts for about 90% of total FSU oil production (Table 10). Several factors have contributed to the recent sharp decline in FSU oil output: (i) ethnic disturbances; (ii) transportation problems; (iii) drastic cutbacks in investments; and (iv) oil well depletion caused by excessive injection of pressurized water. Capital constraints not only have restricted the replacement of aged and obsolete technology, but have also caused a severe shortage of spare parts and steel pipes needed for the repair and replacement of defective equipment. 15 Table 10: FSU Oil Production by Republics (1991) Production (MTOE) % Azerbaijan 11.0 2.1 Kazakhstan 32.0 6.1 Russia 455.0 88.6 Turkmenistan 8.0 1.5 Ukraine 5.0 0.8 Other 5.0 0.8 Total 516.0 100.0 Source: Petroleum Economics Limited; US Department of Energy; World Bank, International Economics Department. Without significant injection of capital, importantly foreign capital which is accompanied by western technology, oil production should continue to decline in the FSU. In 1988, the oil sector accounted for about 20% of the total investment expenditure. The share has declined sharply since then as scarce capital was channelled into other activities, such as consumer goods. Capital needs are made more acute by the fact that new prospects lie in remote and difficult to explore areas. In an effort to prevent further declines in energy production, the FSU republics are seeking joint ventures with international firms in oil/gas exploration and development activity. Bringing these to fruition will greatly improve the long-term prospects for oil supplies. While there is keen interest among leading energy companies in joint venture activities, progress appears to be slow and prospects remain very uncertain. Factors such as the following pose risks to such investments: (i) the political and economic instability in the Republics; (ii) unclear legal framework and ownership structure; (iii) export tariff on oil in Russia; and (iv) the underdeveloped commercial infrastructure. The deals being finalized are far fewer than generally perceived. Amid the rapid structural changes in the FSU's energy industry, international companies are confronted with lengthy negotiations. It is not always clear who the responsible authorities are and who owns the resources. Ill-defined property rights are causing costly delays in finalizing deals. The issue of ruble convertability and the uncertainty over its future value are further undermining international companies' efforts to invest. At the same time, foreign companies are no longer content to be contractors only They are likely to advance cautiously until the situation improves. Therefore, it will take some time before the desired investments needed to revitalize the FSU oil sector and the resulting output response are realized. We, therefore, do not expect the declining trends in oil production to reverse in the short to medium term. However, it is generally perceived that the long-term prospects are favorable. Over the long run, if foreign participation grows and investments bear fruit, FSU oil production could increase to about 550 mtoe by the year 2005. DEVELOPING COUNTRIMs. New fields promise future production increases. While the bulk (about 60%) of oil discovered in recent years has been in traditional areas in the Middle East, Latin America (particularly, Venezuela) and the North Sea, wildcat drilling has uncovered some new fields in other areas. The latter has contributed to a significant increase in global oil reserves (Table 11). 16 Table 11: Global Oil Reserves, by Geographic Location, 1970-91 Region 1970 1980 1985 1991 (billion barrels) Asia & Pacific 14.41 19.63 18.85 44.07 Western Europe 3.71 23.08 26.41 14.50 Middle East 344.57 362.07 398.02 661.57 Africa 74.76 55.15 56.73 60.49 Western Hemisphere 73.95 102.39 118.74 :151.60 Eastern Europe, FSU, and China 100.00 86.30 81.37 58.77 Total 611.40 648.50 700.14 991.01 Source: Oil and Gas Journal (various issues). In LATIN AMERICA, Brazil and Colombia have been the most active areas of E & D. In Brazil, discoveries in the Campos basin and the recent discoveries at the mouth of the Amazon River (deep water fields) have improved the outlook for Brazilian oil supplies considerably. While estimates are still highly speculative, reserves from these fields are reported to be as high as 10 billion barrels. However, the difficulty of developing deep water fields is likely to restrict output from these fields. In Colombia, the recent discovery of the Cusiana field is expected to add between 2 and 5 billion barrels to its present reserves of about 2 billion barrels. Given its large resource base (about 51 billion barrels) and high reserves-to-production ratio (50:1), the prospects for increased output in Mexico are favorable. However, the development of these reserves has suffered from reduced government investments as well as from restrictions on foreign investment. More accommodating government policies towards domestic and foreign investment could increase Mexico's output to about 173 mtoe by the turn of the century. In the MIDDLE EAsr, the Yemen, Syria, and Oman provide the most favorable outlook for oil supplies outside OPEC. Since oil was discovered in 1984, significant additions to Yemen's reserves have been made. Unified Yemen's reserves are now estimated at 4 billion barrels. These have been built up by increased participation of foreign oil companies. Based on the prospects for its several fields including Alif in the Mareb-Janof concessions and the Shebwa fields, it is expected that annual production could increase to over 37 mtoe by the turn of the century. In Syria, the discovery of large fields at Al-Thayyem in the mid-1980s improved the prospects for increased supplies. Prospects have also been supported by the recent reversal of the decline in the Omar field. As a result, Syria's production reached 23.5 mtoe in 1991 (up from 19.5 mtoe in 1990). Extensive wildcat drilling has added about 1.5 billion barrels of reserves which are currently about 1.7 billion barrels. By the year 2000, Syria's crude oil output is expected to increase (mainly by additions from small fields) to 30 mtoe. In the ASIA-PACIFIc region, significant production increases are expected in India and Malaysia. In recent years, foreign oil company participation has increased significantly in these two countries. Recent discoveries of three fields in the Northern Malay Basin (by Exxon) and improved 17 estimates of probable reserves in the frontier areas have resulted in the upward adjustment of reserves to about 3.6 billion barrels (from 3 billion barrels at the end of 1991). As a result, Malaysia's production is forecast to increase to about 37 mtoe (about a 19% increase from the 1991 level). Indian oil production has more than doubled since 1981 to 32.4 mtoe in 1991. About two thirds of production comes from a single field-Bombay High off the West Coast. Although production has stabilized in this field (currently sustained by water injection), with prospects for developing small to marginal fields in Bombay and other regions (namely, Heera, Ratna, and Panra), Indian production is likely to increase to around 40 mtoe by the turn of the century. India has recently signed several new exploration and production contracts. If current and new exploration and production activity pays dividends, Indian production could exceed 40 mtoe annually by the year 2000. Renewed talks of soft Japanese loans to China have improved the oil supply outlook from China. The recent decline in oil output can be attributed to the decline in industry investment (25% in 1989) and the suspension of loans by the United States and Japan, as well as the poor outcome of offshore exploration activity. However, the recent discoveries of the Weizhou Field in the South Sea, the Tarim on-shore basin in Xinjiang province, and another in the Liaodong Bay, north east of Bohai-together with the trend towards more liberal policies which allow increased foreign oil company participation in exploration and development-are likely to support increases in production. The promising potential offered by several fields, including the huge Xinjiang Field, should provide increases in production beyond the turn of the century. It is expected that China's annual oil production could reach 155 mtoe by the year 2000 and 165 mtoe by 2005. However, given the huge costs of developing infrastructure, these increases are contingent on active foreign participation in oil E & D. There are many other developing countries that will provide additional sources of oil supplies. Supported by its estimated proven reserves of 4.2 billion barrels, Oman plans to develop a small reservoir in the future which is likely to result in marginal increases in output. Egypt has increased its proven reserves from 430 mtoe in 1981 to 616 mtoe, and modest increases in output are possible at least up to the mid-1990s. In a bid to raise capital for E & D, a move towards privatization and removal of restrictions on foreign investments in upstream operations have improved the supply prospects in Argentina and Peru. Foreign participation has recently been allowed by the Argentinean government in four key oil-producing areas. The Peruvian government's decision in late August 1991 to allow local private companies and foreigners to invest in both upstream and downstream activity has reduced Petroperu's monopoly position considerably. As a result of expected increases in investments, Argentina's output should increase by about 50% and Peru's by 7% by the year 2005. Proven reserves in Argentina and Peru are estimated at 1.57 billion barrels and 0.38 billion barrels, respectively. In addition to the supply prospects referred to in the previous paragraphs, small increases are expected in many other countries such as Papua New Guinea, Viet Nam, Pakistan, and Thailand. However, these will necessarily involve significant foreign participation. The supply forecasts for non-OECD economic regions are presented in Table 12. The outlook for supplies from non-OECD regions is for an increase from 1,802 mtoe in 1991 to 2,478 mtoe in 2005. Excluding the FSU and OPEC, supplies are expected to increase at an average rate of 1.4% p.a. and reach 765 mtoe by 2005. OPEC OiL PRODUCTION FoREcAsrs. OPEC's world mar*et share to increase. OPEC crude oil production has increased steadily since 1986. Between 1986 and 1989, OPEC oil production 18 Table 12: Non-OECD Petroleum Output Forecasts, by Income Group, to 2005 Growth 1991 a/ 2005 Increment 1991-2005 Countries/Regions MTOE % MTOE % MTOE % p.a High-Income 147 6.2 345 11.2 198 6.3 LMICs 2,196 93.8 2,733 88.8 537 1.6 Africa 149 6.4 146 4.7 -3 -0.01 America 380 16.2 463 15.0 83 1.4 Asia-Pacific 286 12.2 347 11.3 61 1.4 Europe 541 23.1 550 19.5 9 0.2 Middle East & North Africa 841 35.9 1,177 38.3 336 2.4 Total 2,343 100.0 3,028 100.0 685 1.9 Memo: Excl. FSU 1,802 77.0 2,478 80.5 676 2.3 Excl. OPEC 1,166 50.0 1,315 44.3 149 0.9 Excl. FSU & OPEC 625 27.0 765 25.0 140 1.4 a! World Bank estimates. Source: World Bank, International Economics Department. increased by 180 mtoe (3.6 mb/d). Despite the Iraqi invasion of Kuwait in August 1990, and the subsequent embargo on Iraqi crude oil exports which affected supplies temporarily, OPEC production increased by 46 mtoe (0.92 mb/d) in 1990. The immediate impact of the embargo on IraqilKuwaiti oil was a reduction by about 195 mtoe in world oil exports. However, because of the increased availabilities of replacement supplies, mainly from other OPEC producers, the shortfall in crude oil supplies was more than offset in volume terms. By December 1991, OPEC output reached 1,161 mtoe (23.22 mb/d), surpassing pre-invasion OPEC production of 1,153 mtoe (23.06 mb/d) in July 1990. Increases in supplies came mainly from Saudi Arabia (2.68 mb/d) and the UAE (0.43 mb/d). Other producers that contributed to the increases include Indonesia, Iran, Libya and Venezuela. OPEC's crude oil production increased further in 1992 to average 1,193 mtoe; in the first half of the year the increases largely reflected the recovery of output in Kuwait. At the end of June 1992, Kuwait's production capacity reached 50 mtoe. OPEC production is expected to increase over the forecast period in line with the expected increase in their output capacities (Table 13). In the short and medium term (1992-95), the increases in OPEC output are expected to be mainly from Iraq and Kuwait as they rebuild capacity. However, short- term exports from Iraq will be contingent on the removal of the UN sanctions on its exports. Based on oil field rehabilitation plans, Iraqi output capacity is expected to reach roughly 3 mb/d by the mid-1990s, up from 0.4 mb/d in 1991. Kuwaiti output capacity is expected to increase to 2.5 mb/d by that time. The Middle East hostilities reduced Kuwait's output capacity to near zero. Prior to the Iraqi invasion of Kuwait, crude oil production capacities of Iraq and Kuwait were 3.5 mb/d and 2.4 mb/d, respectively. Kuwait is negotiating contract terms with several international companies for technical assistance in bringing fields back on-stream. In an effort to attract foreign investment, Iraq has begun talks with several international oil companies on oil exploration contracts. 19 Table 13: OPEC: Sustainable Capacity and Output in 1991, and Forecast Capacity to 2000 Output Output Change in Capacity Capacity Capacity 1991 2000 1991-2000 (mb/d) Algeria 0.8 0.9 +0.1 Ecuador 0.3 0.3 0.0 Gabon 0.3 0.3 0.0 Indonesia 1.5 1.3 -0.2 Iran 3.4 4.0 0.6 Libya 1.6 1.8 0.2 Nigeria 1.9 2.3 0.4 Qatar 0.4 0.4 0.0 Saudi Arabia a/ 8.9 10.5 1.6 United Arab Emirates 2.5 3.3 0.8 Venezuela 2.5 3.4 0.9 Iraq 0.4 3.0 2.6 Kuwait 0.3 2.9 2.6 Total 24.4 34.0 9.6 a/ Includes 50% share of neutral zone. Sources: Cambridge Energy Associates; Middle East Economic Survey; Petroleum Economics Ltd. The recent Middle East conflict has had a significant impact on OPEC's long-term capacity expansion programs. While it has delayed expansion in Iraq and Kuwait, the development plans of other OPEC members have accelerated. Iraqi and Kuwaiti pre-war plans to expand output capacity to 4.5 mb/d and 3.5 mb/d by 1995, respectively, are now likely to be realized well after the turn of the century. Saudi Arabia's pre-war target of 10 mb/d by mid- to late 1990s is now likely to be achieved before 1995. Saudi plans to increase output capacity involve the development of new fields and the expansion and upgrading of old fields. Plans also call for the installation of sea water injection capacity and wet crude handling facilities. In the north, plans include development of the Zuluf offshore field and the Marzen field. In the south, plans include the expansion of the offshore Safania field and the Uthmeniya field which is in the Howaiyah area of the Ghawer field. There are also plans for the development of the three new fields believed to be part of a giant field. By the year 2000, Saudi Arabia's capacity is expected to reach 10.5 mb/d (its level of the late 1970s). This capacity expansion program is likely to put Saudi Arabia in a more dominant position within OPEC. Iran has plans to develop new oil and gas reserves, as well as reconstructing its oil production and export facilities damaged during the eight-year war with Iraq. These plans envisage increases in sustainable output capacity to as high as 5 mb/d by mid-1993. However, this is an optimistic target, one which is not likely to be achieved until after the turn of the century despite foreign involvement. 20 Venezuela expects to increase supply by roughly 0.2-0.3 mb/d from its more than 50 marginal fields over the next three years or so. Indonesia's plans to explore and develop the Timor Island region and the Belida oil field in the South China sea have marginally improved its prospects for oil supplies. Plans also call for increasing investment in enhanced recovery projects. OPEC should supply increasing quantities of crude over the long term. However, since a few OPEC members will reach a production plateau by the early 1990s, and several others will enter the depletion phase towards the end of the century, a small group of producers including Saudi Arabia, Iran, the UAE, Iraq, Kuwait, and possibly Venezuela should attain a more dominant position in the world oil market, largely due to the enormous low-cost oil reserves in these countries (Table 14). Of the proven recoverable world oil reserves, only 236 billion barrels (22%) are located in the non-OPEC countries. The reserves-to-production ratio of the non-OPEC suppliers is around 16 compared with 89 for OPEC (see Table 14). Therefore, by the late 1990s, the world is expected to be increasingly dependent on OPEC for additional supplies. It is anticipated that by 2005 production in OPEC countries should increase to 1,713 mtoe with OPEC's share in total petroleum production rising to around 46% (up from 38% in 1991). Middle East OPEC production is expected to increase at a trend rate of roughly 3.4% p.a. over the forecast period and its market share should increase to over 36% by 2005 (Table 15). Table 14: Global Oil Reserves, 1991 Reserves-to-Production a/ Regions/Countries billion barrels % Years Industrial 48.1 4.8 10 Eastern Europe & FSU 58.8 5.9 15 Other Countries 884.1 89.3 70 Total 991.0 100.0 44 Memo: Non-OPEC 221.6 22.4 16 OPEC 769.4 77.6 89 Saudi Arabia 257.8 26.0 86 Iraq b/ 100.0 10.1 1,000 Iran 92.9 9.4 77 Kuwait b/ 94.0 9.5 1,567 United Arab Emirates 96.3 9.7 119 Venezuela 59.1 6.0 67 Other OPEC 69.3 6.9 28 a/ Based on estimates of reserves at the end of 1991 and average production in 1991. b/ Based on production estimates for the first half of 1990 the reserves-to-production ratio for Iraq and Kuwait is roughly 90 years and 140 years, respectively. Note: Estimates of reserves range considerably from one source to another. Source: Oil and Gas Journal. 21 Table 15: Petroleum Production Forecasts, to 2005 by Economic Regions 1991 1995 2000 2005 (%rowthr Regions MTOE % MTO % MTOB % MTOE % 1991-2005 Industrial 779 25.9 794 24.0 735 21.0 708 18.9 -0.7 Eastern Europe & FSU 541 17.3 500 15.4 530 15.5 550 14.9 0.1 Other Countries 1,802 57.7 1,941 60.0 2,219 63.5 2,486 66.4 2.3 Total 3,122 100.0 3,235 100.0 3,484 100.0 3,744 100.0 1.3 Memo: Non-OPEC 1,945 62.3 1,981 61.2 2,004 53.5 2,031 54.2 0.3 OPEC 1,177 37.7 1,254 38.8 1,480 42.4 1,713 45.8 2.7 Mid-East OPEC 864.1 27.7 976 30.1 1,152 33.0 1,379 36.8 3.4 Source: World Bank, International Economics Departiment. Petroleum Price Outlook We rely on three sources of information for our oil price forecasts. For the short run .(1992-93), we use futures price quotations. For the medium term (1994-95), we use a combination of futures prices and, where available, over-the-counter quotations for long-dated underlying financial derivatives such as oil swaps. These quotations are acquired from industry sources engaged in swap trading activity. For the long-term price projections, we rely on our oil market model-based analysis. The price projections are presented in Table 16. OIL MARKET MODEL. The oil market model includes the behavior of three key sets of market factors, namely, oil/energy demand, non-OPEC oil/energy supplies, and OPEC output capacity and capacity utilization rates. The model is used to study the interaction of various assumptions about these key market factors and to determine the market-clearing price of oil under various scenarios. Oil demand is divided into residual and nonresidual fuel oil and is influenced by economic growth and the opportunity for substitution between residual fuel oil and other energy sources. The magnitude of the price change depends upon the degree (elasticities) to which demand and supply are responsive to (past and current) prices and OPEC's capacity utilization rate. The demand side consists of the following subregions: United States, Japan, OECD Europe, other OECD, OPEC, high-growth newly-industrializing countries, low-growth, newly- industrializing countries, low-growth oil-importing developing countries, and other developing countries. Non-OPEC supply is the sum of the exports of the FSU and supplies from the United States, Canada, United Kingdom and Norway, other OECD, and the developing countries. Net exports from the FSU are exogenous. OPEC is assumed to be the residual supplier to the market. Its production capacity is specified exogenously. For expected changes in the world demand for oil (in response to changes in income and the lagged effects of past prices), the model solves interactively for the market-clearing price at which the demand for OPEC oil equals a specified amount, given expected changes in non-OPEC supplies. For a given year, therefore, price is determined simultaneously by the interaction of OPEC's supply curve and the demand for OPEC oil. 22 Table 16: Petroleum Prices a/ 1989-91 (Actual) and 1992-2005 (Projected) Current 1990 Constant $ b/ /bbl)- 1989 16.3 17.2 1990 21.2 21.2 1991 17.3 17.0 1992 17.6 16.5 1993 17.3 15.7 1994 17.4 15.5 1995 18.0 15.6 2000 24.7 18.0 2005 26.8 17.0 a/ Refers to the weighted average f.o.b. price of petroleum exports from OPEC countries. b/ Deflated by Manufacturing Unit Value (MUV) Index. Sources: Platt's Oil Price Renort and World Bank, International Economics Department (actual); and World Bank, International Economics Department (projected). RATIONALE FOR USING FUTURES PRICEs. The use of futures prices for short- and medium-term forecasts is based on the following argument. In recent years, the volume of oil traded in the futures markets and the number of active participants in this market have increased rapidly. Spot and futures tradings are now used by various market participants including national oil companies, refiners, producers and speculators. Many oil producers are now adopting pricing strategies that link term contract supplies to spot and futures prices. In fact, this practice has become so dominant that many of the contracts entered into by producers have price clauses that are in some way linked to this market-based pricing. OPEC exports based on spot pricing increased to 44% (in late 1991) up from about 26% in 1985. Due to the low transaction costs involved, information about perceived changes in market factors (demand, production, supply, OPEC policy) is instantaneously reflected in futures price. This has given credence to the view that the oil futures market performs a price discovery function. New exchanges have emerged around the world, and new oil products have been introduced to these markets. This has allowed arbitrage trading between contracts and between markets. The futures market is likely to continue to flourish in the presence of high price volatility in the petroleum market. To cope with the inherent uncertainty in crude oil and petroleum product prices, decisionmakers are adopting appropriate risk management strategies. Other financial instruments such as forwards, options, and oil swaps are now being used extensively by market participants to hedge risks stemming from fluctuations in the price of oil. This trend is expected to increase. However, the use of long-dated instruments has not evolved as rapidly as futures. The use of complex hedging techniques by many financial institutions and the wide variation in costs to users are some of the reasons restricting their growth. Because the markets for these financial instruments are not very liquid, the underlying price quotations cannot be thought of as being as efficient as futures markets in incorporating information. 23 RECENT PRICE DEVELOPMENTS. Our indicator price for crude oil, OPEC average spot, declined 18.2% in 1991-from $21.20/bbl in 1990 to $17.35/bbl. The main factors contributing to this decline were: (i) reduced supply uncertainty following the cessation of the hostilities in the Middle East in early 1991; (ii) sluggish growth in global oil demand (by 0.5%) due to weak economic activity in the industrial countries and declines in oil consumption in the FSU and Eastern Europe; and (iii) increases in oil supplies from OPEC sources (by 1.1%). Crude oil prices declined to an average of $15.60/bbl in the first quarter of this year. The decline from the fourth quarter 1991 average of $17.99/bbl is within the range expected for the seasonal decline at this time of the year-especially considering the mild northern hemisphere winter. Prices increased to an average of $17.70/bbl and $18.40/bbl in the second and third quarters of 1992, respectively; an increase which also seems consistent with year-to-year changes at this time. There was some tightening of prospective supplies in light of declines in supplies from Russia and expected seasonal increase in oil demand. There was also some credence given to the view that Saudi Arabia had shifted its policy to support higher prices. SHORT-TERM PRICE PROJECTIONS (1992-93). Pricesfu in 1992 but downwardpressures in 1993. Based on futures prices for oil, our indicator price is expected to increase to $18.40/bbl in the second half of 1992, by comparison with a first half average of $16.70/bbl. The expectation of increases in the near-term price of oil appears to be based on the following: (i) a seasonal increase in the global demand for oil; (ii) the UN embargo on Iraqi crude oil will remain in effect; and (iii) increases in non- OPEC oil supplies will be more than offset by declines in the FSU and the United States. The oil market is perceived to remain fairly tight during this period and highly susceptible to shocks. The increase in the precautionary demand for stocks to hedge against supply disruptions during the fourth quarter, when demand is seasonally high, is likely to put upward pressure on spot and nearby futures prices. Prices are expected to decline to an average of $17.30/bbl in 1993. We expect global oil demand to increase by roughly 0.9% (or 45 mtoe). However, we expect the increase in oil demand, in response to economic recovery in the OECD and continuing growth in developing countries, to be more than offset by increases in supplies from OPEC and non-OPEC sources. Looking at possible shocks during this period, it is reasonably likely that the sanctions on Iraq's exports will be removed. By the latter half of 1993, Iraq's output capacity could increase to above 2.5 mb/d (up from 1.5 mb/d in mid-1992). Other OPEC members such as Saudi Arabia and Iran could well be reluctant to yield market share to accommodate Iraq. In that case, prices could come under severe downward pressure. MEDIUM-TERM PRICE PROJECTIONS (1993-95). Mediwn-ten prices to remain flat. Over the period 1993-95, nominal prices are expected to increase by 2% p.a. However, in real terms, they are expected to remain unchanged. The impact on prices of economic recovery and growth, particularly in the United States, and further declines in supplies from the FSU should be offset by increases in supplies from OPEC sources. Following a period of modest increase over the period 1990-93 (0.7% p.a.), oil demand is expected to increase by roughly 1% p.a. over the period 1993-95 due largely to a pickup in global economic activity. However, increases in oil demand in OECD Europe will be sluggish compared with the early 1990s, despite solid economic growth. Improvements in the efficiency of oil use, increases in energy taxes (on environmental grounds), and concerns over energy security matters should contribute to the slowdown in oil demand in Europe. In the OECD, oil demand in the transport 24 sector should provide the most dynamic growth, although other end-uses, particularly for power generation, should also contribute. Oil demand in the FSU and Eastern Europe can be expected to decline due to sluggish economic growth, higher end-use oil prices, and reduced availabilities of oil supplies to the domestic markets. Energy-saving structural changes and improvements in energy efficiency will reduce the demand for all forms of energy including oil. In the developing countries, we expect oil demand to increase in all sectors. Robust economic growth (projected at 5% p.a.) and rising per capita income should increase the demand for public and private transportation and induce sharp increases in transport fuel demand. In the residential and commercial sectors, the shift from traditional fuels to oil products as well as the expected increase in oil demand for electricity generation is likely to contribute to increases in oil use. The expected increase in the demand for oil for power generation (about 3.2% p.a.) is based upon the projected robust growth in electricity demand, limits to nuclear generation, and insufficient availability of gas for power generation over this period. However, the overall rate of growth in oil demand over the period 1991- 2005 is likely to be lower (about 3.2% p.a.) than the roughly 4% p.a. achieved during the 1985-91 period. With a rising oil import bill, a shift to market-oriented pricing including reduction of subsidies should suppress growth in oil demand. LONG-TERM PRICE PROJECTIONS (To 2005). 7ghter market to force prices to rise to 2000, then cyclical downturn to 2005. Over the longer run, between 1995 and 2000, we expect real prices to increase at 3% p.a. and then decline to 2005. This forecast assumes that global oil demand will increase at 1.5% p.a. over the period 1995-2000. In the OECD, oil demand is expected to increase at a rate of 0.7% p.a. as the effects of increases in energy taxes and efficiency improvements are more than offset by the income effect. The increase in transport demand and the substitution of petroleum products for coal in Japan should support increases in oil use in the power-generation sector. During this period, the bulk of the incremental oil consumption (69%) is expected to occur in the developing countries; the FSU and Eastern Europe will contribute marginally to the increase. Oil production in non-OPEC regions (including the FSU) is expected to increase marginally, by 14 mtoe, mainly in developing countries and in the FSU. FSU oil production should recover in response to increased investments in exploration and development and increases in diomestic prices. OECD oil production is likely to decline, however. The expected increases in global oil demand, by 259 mtoe over the 1995-2000 period, and modest increases in non-OPEC supplies will put pressure on oil-producing capacity. Based on the expected increases in OPEC output capacity to 1,700 mtoe, an increase in OPEC output to 1,480 mtoe to meet the increased demand translates into an OPEC capacity utilization rate of over 87%. Under this tight market balance, prices should increase in real terms. Beyond the year 2000, we expect a cyclical downturn in prices. OPEC output capacity is expected to reach 2,000 mtoe by the year 2005, in response to the investment incentives provided by the higher prices in the late 1990s. The FSU is also expected to increase oil output as a result of increased foreign investment and the application of improved technologies. Developing countries' consumption growth should slow due to improvements in the efficiency of oil use and increased availability of natural gas. Increased dependence on Middle East oil in the late 1990s will accelerate policies to reduce this vulnerability and further reduce demand growth. A slowdown in population growth in the developing countries, aging population in the industrial countries, and the development of alternative energy technologies such as electric cars and gas-based substitutes for oil in the transport sector should moderate the growth in global oil demand considerably. 25 Uncertainties in the Petroleum Market Outlook There is always a high degree of uncertainty with regard to the outlook for the global oil market and for international energy prices. This uncertainty is reflected in the wide range of forecasts of global oil demand, non-OPEC crude oil supplies, OPEC capacity expansion plans, and OPEC's policy responses to changes in market conditions. The dramatic turn of events in the FSU has widened the range of oil market uncertainty. Underlying this uncertainty are the different views on the oil industries' capital needs and its ability to generate these investments, as well as the impact of uncertain environmental policies. A review of the oil industry's investment needs is summarized in Annex I. Uncertainties about these factors is reflected in the wide range of oil price forecasts. For example, the Energy Modeling Forum (EMF) poll responses show oil price forecasts in the range of $15-35/bbl for the year 2000 and $20-55/bbl in 2010 (in constant 1990$).' SENSIIVITY ANALYSIs. A key assumption underlying the short-term (late 1992 and 1993) oil price forecasts is that the embargo on Iraqi crude oil exports will continue. Over the long run, the base case demand and price forecasts rest upon the following key assumptions: (i) GDP growth in the OECD and developing countries will be 2.8% p.a. and 5% p.a. respectively, through the year 2000; (ii) OPEC output capacity will increase to about 1,700 mtoe (34 mb/d) by 2000 and to around 2,000 mtoe (40 mb/d) by 2005; (iii) non-OPEC oil output (excluding the FSU) will increase to 1,475 mtoe (29.50 mb/d) by 2000 and remain around this level to 2005; and (iv) FSU output will continue to decline up to the mid-1990s but increase steadily thereafter to reach 550 mtoe (11 mb/d) by 2005. However, different rates of change in any of these long-term assumptions would give different forecasts to our baseline forecasts. To see the impact of changes in the baseline assumptions on crude oil prices, various model simulations were carried out. The results of this sensitivity analysis are presented in Table 17. REMOVAL OF UN SANCTIONS ON IRAQI CRUDE OL EXPORTs. A scenario is explored wherein the removal of UN sanctions would lead to Iraq's crude exports reaching about 125 mtoe (2.5 mb/d) in the latter part of 1993. Roughly, an unexpected increase in Iraqi exports of this magnitude would result in a $3/bbl (17%) decline in price in the second half of 1992 and about a $4/bbl (22%) decline in 1993. UNCERTAINTIES IN OIL DEMAND FORECAsrs. Oil demand in the western economies (excluding the FSU and China) has been forecast in the range 2,600-3,550 mtoe (52-71 mb/d) by the year 2000 (within a 70% confidence interval). Some estimates of oil demand are as high as 4,000 mtoe (80 mb/d) in 2000. The largest discrepancies are observed in demand forecasts for the developing countries where the relationships between demand and the variables that affect demand are not as clearly known as for the industrial countries. Energy Modeling Forum, "International Oil Supplies and Demands, Summary Report," April 1992. 26 Table 17: Model Simulation Price Trends Under Alternative Assumptions 1995 2000 2005 ($/bbl) Base Case 15.6 18.0 17.0 Economic Growth High 16.0 18.7 17.8 Low 15.2 17.1 15.9 Income Elasticity High 16.0 20.3 19.5 Low 15.1 16.0 14.9 Reduction in OECD Demand 15.1 17.1 15.4 Non-OPEC Supplies High 15.2 17.2 15.3 Low 16.1 21.1 20.0 FSU Oil Production High 13.5 14.9 14.0 Low 16.5 19.9 18.7 Source: World Bank, International Economics Department. Differences in demand forecasts are due to different assumptions about: (i) economic growth prospects; (ii) prices; (iii) income and price elasticities; (iv) lagged effects of oil prices; and (v) efficiency improvements.' Under an assumption of higher world economic growth (from 2.8% p.a. to 3.2% p.a.), the market-clearing price of oil (in real terms) would increase by about 2.5% in the medium term and about 5% in the long term. However, a corresponding decline in the economic growth assumption (down to about 2.5%) would translate into a larger decline in price-about 3% down in the medium term (1995) and about 8% lower in the long term (2005). We estimate the impact on world oil prices of alternative assumptions about the income elasticity of oil demand to be as follows: under an assumed higher income elasticity of oil demand to 1.4 (the base case assumes a value of unity), prices trend about 3% higher in the medium term and 15% higher in the long term. Alternatively, reducing the income elasticity to 0.6 results in the price rending 4 Energy Modeling Forum, *International Oil Supplies and Demands, Summary Report," April 1992. Models that generate high demand growth usually assume one or a combination of the following assumptions: (i) rapid economic growth; (ii) high income elasticities of demand; (iii) lagged effect on consumption of lower oil prices in the mid-1980s; and (iv) little or no energy efficiency improvements. Alternatively, assumptions behind slow demand growth forecasts include: (i) low economic growth and income elasticities; (ii) little or no effect of past prices; and (iii) substantial efficiency improvements. 27 3% and 14% lower in the medium and long run, respectively. Optimistic assumptions with regard to efficiency improvements can be reflected in lower income elasticity estimates. IMPACT OF ENVIRONMENTAL AND ENERGY POLICY CHANGES. While the impact of policies such as use of carbon taxes to achieve CO reduction is ambiguous because of offsetting own- price and substitution effects, we expect that OECD policies to reduce dependence on oil imports will have an impact on prices. We estimate that while environmental policies (carbon tax) should adversely affect oil demand through the price effect, the substitution effect should lead to an increase in oil demand. This is based on the assumption that, given the higher carbon content of coal, its price will rise relative to prices of oil and gas and induce shifts out of coal into oil and gas in electricity generation and industrial uses. Oil demand should increase more in regions where the availability of natural gas is limited. Concerns over the environment are likely to have a greater impact on the supplies of petroleum products and the refinery industry (see Annex II). The impact on prices of more aggressive energy policies than we have assumed to reduce OECD oil dependence will depend upon how rapidly these policies are implemented (and on how much reduction is intended). A policy that targets a quick reduction in demand should have a greater impact on prices and higher adjustment costs than one which encourages gradual adjustment. We estimate that the impact of a gradual reduction in oil demand (1-2 mb/d in the medium term and 3-4 mb/d in the long term) would to be lower oil prices by 4-6% in the medium term and 8-10% in the long run, ceteris paribus. UNCERTAINTY OF SUPPLIES FROM NON-OPEC souRcEs. Surprisingly, a review of the literature shows that the forecast range of non-OPEC oil supplies is not as large as the range of uncertainty for global oil demand-surprising, given the high level of uncertainty which must be attached to FSU output. Under the same oil price assumption, the forecasts in EMF models range between 1,200 mtoe and 1,700 mtoe (24-34 mb/d) in the year 2000 (excluding the FSU). Non-OPEC supply forecasts under various oil price scenarios, by the US DOE and other forecasting sources, are presented in Table 18. The relatively small range in non-OPEC supply forecasts is due mainly to the use of uniform estimates of oil reserves-despite the uncertainty with regard to the resource base. The variation is caused by differences in fiscal regimes and regulatory practices assumed by different analysts, as well as differences in price assumptions. In general, higher taxes on exploration and development activities yielded lower profiles for non-OPEC supplies. Most sources forecast non-OPEC oil production to decline steadily through the turn of the century and more sharply beyond that. There appears to be a consensus that US oil production will decline steadily over the forecast period. The US DOE base-case estimates are for oil production capacity in the United States to decline to 425 mtoe by 2000 and to 415 mtoe by 2010 (from 484 mtoe in 1990), despite expected increases in price to $26.40/bbl and $33.40/bbl in 2000 and 2010, respectively.5 The production declines in regions outside OPEC are based on several assumptions including: (i) rapid depletion of reserves; (ii) high costs of reserve replacements; and (iii) reduced possibilities of new finds in areas that possess the supporting infrastructure to expand drilling activity. An assessment of the impact of the range of non-OPEC supply forecasts (1,200-1,700 mtoe, or 24-34 mb/d) from various industry sources shows that an increase in oil supplies from non- US Department of Energy, International Energy Agency, 'International Energy Outlook 1992.' 28 Table 18: Non-OPEC Supply Forecasts, 2000 and 2010 (mb/d) &/ Forecasters 2000 2010 IEO 92 31.2 (26.4) b/ 27.4 (33.40) b/ IEO 91 27.5 (26.4) 24.6 (33.40) Canada c/ 26.1 (23.2) 25.0 (27.00) DRI 26.6 (27.2) 26.9 (35.70) County Natwest d/ 30.0 (20.0) - - a/ Excludes the FSU and China. b/ Figures in parentheses are assumed prices of crude oil in 1990 dollars per barrel. c/ Estimates from Canada include only not exports from the former centrally planned economics, and not consumption/production estimates. For purposes of comparability, the consumption/production estimates underlying the net export projections (derived from the International Energy Outlook 1989) are added to the Canadian estimates. d/ West Texas Intermediate oil price. The US refiner acquisition cost of imported crude oil used as the world oil price for the IE092 and other projections listed here runs about $2-3/bbl less than West Texas Intermediate. Source: IE092-International Energy Outlook 1992, US Department of Energy/International Energy Agency- 0484(92); IE091-International Energy Outlook 1991, US Department of Energy/International Energy Agency-0484(91); Canada-Canadian Energy Supply and Demand 1990-2010, National Energy Board, June 1991; DRI-DRI/McGraw-Hill, International Oil Bulletin, Fall 1991; County NatWest USA, Oil Market Outlook, January 1992. OPEC sources to 1,700 mtoe (34 mb/d) would result in about an 11% decline in the long-run price of oil. An assumed lower production from non-OPEC of 1,200 mtoe (24 mb/d) results in a sharper increase in oil price, by about 18%, since it would lead to sharp increases in the utilization of oil production capacity. However, the impact on prices would be lessened beyond the year 2000 because of lagged demand and supply (OPEC) side responses to higher prices in the late 1990s. SUPPLY PROSPECTS FROM THE FSU. The recent dramatic turn of events in the FSU has obviously widened the band of uncertainty about the world oil supply outlook. In the medium term, the US DOE forecasts a decline in FSU and Eastern European oil production capacity to a range of 365-525 mtoe-down from 535 mtoe in 1991, and in the long term the forecasts are in the range of 425-585 mtoe in 2000 and 495-705 mtoe in 2010. Petroleum Economics Limited has forecast FSU oil production to decline to 510 mtoe by 1995 (from a figure of 520 mtoe in 1991) but expects production to increase at about 2% p.a. beyond 1995 to reach 620 mtoe in 2005. Under various scenarios for the speed of adjustment to economic reforms, Studiefonbundet Naringsliv Och Samballe forecasts FSU production to decline to 380-390 mtoe in 1995 (from 515 mtoe in 1991). The range of forecast in 2005 is 500-550 mtoe.7 6 Petroleum Economics Limited, "World Long-Term Oil and Energy Outlook to 2005," 1991. Studiefonbundet Naringsliv Och Samballe, "Economic Reform and Energy in the CIS," 1992. 29 We have put together the following two scenarios about the development of the FSU oil sector in response to economic and political changes and have assessed their impact on oil prices. (a) CONTINUED DISRUPTION: It is conceivable that prolonged political and economic uncertainties and the perceived high risks of investing will lead to prolonged disruptions in output. In this case, output should decline steadily to 400 mtoe by 1995 and to 300 mtoe by the turn of the century (roughly 240 mtoe lower than the base-case projection for 2000). (b) RAPID RECOVERY: If efforts to reform prove successful and economic and political problems are relatively short-lived, the constraints on foreign technology imports and investment should be reduced. In this case, FSU oil production could increase to 625 mtoe by 1995 and could reach 700 mtoe by the year 2000. In assessing the impact of different outlooks for supplies from the FSU, we find that the price risks associated with unexpected changes in FSU oil supplies could be asymmetric. Medium-term impact on prices are likely to be greater when FSU output increases sharply than would be the case if output declined sharply. Declines in FSU output are likely to be offset by increases from other producers. However, in a situation of a higher than anticipated output, other producers could be reluctant to reduce output below prevailing levels, thereby creating a glut in the market. In the case of sharply lower FSU output, assumed under the disruption case, prices would trend only 6-10% higher than in the base case over the long run. In a situation of higher than expected increases in FSU oil supplies, prices would average 16-21 % lower than the base case. However, where increases in FSU output were accompanied by offsetting supply adjustments by OPEC producers (say by mothballing of capacities) or increases in demand, the price decline would be substantially moderated. The expected tightening of the market balance in the late 1990s is the key factor underlying our forecasted increase in the real price of oil over the 1995-2000 period. To test the robustness of these results, we also carried out a model simulation which assumes flat prices (in real terms) over the forecast period. We have found that if oil prices are held constant at $16.50/bbl (the average price expected in 1992 in constant 1990 US dollars) over the period 1992-2000, the projected demand and supply responses exert strong pressures on OPEC's oil production capacity. We estimate that in this situation, global oil demand would be roughly 100 mtoe (2 mb/d) higher and non-OPEC supply would be roughly 50 mtoe (1 mb/d) lower than assumed in the base case by the year 2000. This would lead to an increase in OPEC's capacity utilization rates to roughly 95% thereby exerting strong upward pressure on prices. Alternatively, at a utilization rate of around 81%, it is estimated that OPEC's output capacity would need to expand to roughly 1,850 mtoe (37 mb/d) by 2000 to meet oil demand growth generated by flat oil prices. These estimates are outside the range of industry estimates (or at the very high end) for OPEC's output capacity. Uncertain capacity expansion prospects for Saudi Arabia (estimated in the range of 9-12.5 mb/d) account for more than two thirds of OPEC's output capacity forecast range of 31-36 mb/d by the turn of the century (within a 70% confidence interval). 30 ANNEX I: Investment Needs of the International Oil Industry Oil industry demand for capital is expected to increase sharply. British Petroleum (BP) has assessed the investment needs to be roughly $1 trillion to develop about 20 mb/d of additional capacity believed to be needed over the next decade or so. The capital needs are much higher when the costs of environmental compliance and revitalization of the FSU oil sector are included. Total investment estimates range between $1,350 and $1,850 billion over the next ten years. These are broken down as follows: (a) Between $200 and $250 billion in OPEC countries to: (i) increase output capacity; (ii) reconstruct Kuwaiti and Iraqi oil production facilities; (iii) maintain existing and future capacities; and (iv) implement new technologies (especially enhanced recovery). (b) Between $750 and $1,000 billion to expand output capacities in the non-OPEC regions. The bulk of the capital needs (about two thirds) are for the oil sector in the FSU. (c) Between $50 and $100 billion to expand and upgrade refining capacity (mainly in the non-OECD countries). (d) Between $200 and $250 billion for expanding oil transportation (mainly new tankers, pipelines, etc.) and distribution infrastructure, mainly in the developing countries where the systems are very rudimentary. (e) Between $150 and $250 billion for environmental compliance. Roughly 75% is expected to be spent in the United States in accordance with the Clean Air Act. For other OECD countries, there is a large refinery component included. In developing countries, these include costs for diesel desulfurization and for reducing lead in gasoline. The significant capital needs have changed the financial aspects of the oil industry. The industry's ability to generate capital will be crucial in shaping its future. With rising external debt, many cash-strapped OPEC members are offering less stringent terms to foreign participants. Total OPEC official debt increased from around $151 billion in 1982 to roughly $225 billion in 1990 (Table Al). In an effort to reduce debt, several OPEC producers are now trying to attract equity partnerships in oil exploration and development activity. Majors with high debt-to- equity ratios (Chevron, Exxon, Mobil and mainly BP) are also attempting to reduce their debt overhang by selling assets. In the United States, where the oil business has traditionally been financed by equity, capital shortages have resulted in sharp declines in drilling activity. Equity capital is very scarce. Many financial institutions are now reluctant to lend for oil exploration and development on the basis of reserve volumes, except to companies that have diversified cash flows to repay debt. The less integrated independent oil companies have suffered the most. They are now resorting to other sources of funding including acquisition financing, as well as selling assets. The inability of governments to raise capital internally and to attract new technologies for the expansion of the oil/gas industry has also provided impetus towards privatization. Witness the developments in several developing countries, especially in Latin American countries such as Argentina and Chile, where large parts of industry, including the oil/gas sectors, have been privatized. 31 As capital markets become more competitive, potential investors will become more discriminating. An important determinant in investment decisions is likely to be the economic and political climate of the country. Factors such as currency convertability, repatriation of funds, control over domestic resources, and the legal and institutional framework will play a key role. We believe that given the pressures described, many more developing countries will become more open to foreign investment in this sector. Table Al: Official Debt of OPEC Countries, 1982 and 1990 1982 1990 a/ ($ million) Algeria 17,728 26,067 Ecuador 7,705 11,311 Gabon 1,000 3,176 Indonesia 26,305 53,111 Iran 8,350 8,575 Iraq 5,312 21,529 Kuwait 9,513 6,870 Libya 3,937 1,800 Nigeria 12,954 32,832 Qatar 806 916 Saudi Arabia 15,372 19,133 United Arab Emirates 9,652 7,197 Venezuela 32,153 33,144 Total OPEC 150,787 225,661 a/ 1990 or the latest year for which data are available. Figures for Iraq and Kuwait were assessed before the war. Source: Shearson Lehman Brothers. 32 ANNEX I: World Refming Outlook Changes in the demand for petroleum products have important implications for their supplies and, therefore, for the refining industry. During the early to mid-1980s, world refinery capacity declined sharply-in line with the decline in the demand for oil. Over the period 1979-85 world oil demand declined by about 320 mtoe (mainly in the OECD), while refinery capacity in the OECD declined by about 465 mtoe (Table A2). The decline was a response to the reduced profitability of refinery operations. In the FSU and Eastern Europe, however, oil demand was essentially unchanged during this period at around 560 mtoe (11.20 mb/d) and refinery capacity increased steadily from 660 mtoe (13.20 mb/d) in 1979 to around 750 mtoe (15 mb/d) in 1985. Since the oil price collapse in 1986, however, world refinery capacity has increased in response to the increase in the world demand for oil (see Table A2). By the end of 1991, world refinery capacity had increased by 90 mtoe, despite the loss of Kuwaiti capacity of roughly 40 mtoe. Almost the entire increase occurred in the developing countries where demand growth was the most robust. Since 1990, significant investment plans to expand refinery capacity have been announced worldwide. Based on this information, world refinery capacity is anticipated to increase by about 225 mtoe (4.50 mb/d) over the 1990-95 period. There is, of course, greater uncertainty about what will happen in the late 1990s. Between 1996 and 2000, world refinery capacity is expected to increase by 185 mtoe or 3.70 mb/d (based on firm industry investment plans) and by 330 mtoe or 6.60 mb/d (based on all plans). The increases in refining capacity are expected to take place mainly in the developing countries-although increases are also expected in Japan. OECD In the United States, refinery capacity is likely to decline over the forecast period from its current capacity of around 15.6 mb/d. The declines should be achieved by mothballing smaller units and closing Chevrons' (30-year old) Port Arther refinery (with 0.325 mb/d capacity). While US distillation capacity is expected to decline, significant upgrading plans are underway. In compliance with the Clean Air Legislation, more cokers and desulphurization plants are coming on line. In Western Europe, increased capacity of between 0.16 and 0.30 mb/d is expected by 2000, with most of the increase taking place in Germany. In Japan, the joint-venture project with Saudi Aramco and other capacity expansion and demothballing projects are expected to add up to 1.20 mb/d of extra capacity. Low estimates are around 0.70 mb/d. Japan's current refinery capacity is about 4.55 mb/d. Asia-Pacific In the developing countries, the Asia-Pacific region is likely to be the front runner in building new refinery capacity. Upgrading and expansion plans are underway in several countries including Thailand (1.50-2.60 mb/d), India (0.20-0.32 mb/d), Taiwan, China (0.20 mb/d), the Philippines (0.20-0.30 mb/d), Indonesia (0.125 mb/d), and Malaysia (0.15-0.23 mb/d). Several countries, including Indonesia and Malaysia, are pulling together joint-venture schemes to support their expansion plans. 33 Table A2: Refinery Capacity and Utilization Rate by Economic Regions, 1979, 1985, and 1991 -1979- -1985- 1991 Capacity Utilization Capacity Utilization Capacity Utilization Rate Rate Rate (mtoe p.a.) (nto. p.a.) % (mto p.a.) % OECD 2,355 75 1,890 71 1,865 82 North America 1,020 83 875 77 880 85 Western Europe 1,015 67 745 67 720 77 Pacific 320 77 270 65 265 85 Non-OECD 1,630 80 1,800 78 1,915 76 OPEC 3,985 77 3,690 74 3,780 78 Note: Refinery capacity relates to distillation capacity and not to other refinery processes. Source: Petroleum Economics Ltd. Latin America and de Caribbean In Latin America and the Caribbean, between 1 mb/d and 2 mb/d of capacity expansion is expected over the next five years. Venezuela tops the list with plans to expand capacity by 0.40 mb/d, followed by Mexico (0.30 mb/d), Brazil (0.25 mb/d), and Colombia (0.075 mb/d). In line with the shift in the rest of the world, oil consumption in Latin America is moving towards lighter products. The increase in the demand for gasoline by roughly 3.7% p.a. over the past ten years has increased gasoline's share to 35% (up from 29% in the early 1980s). Over the same period, the shares of diesel oil and fuel oil fell from 27% for both to 25% and 23%, respectively. Plans also call for a substantial upgrading of the region's refinery configuration to supply products to the US market. However, the completion of these plans is contingent upon the success with which investment capital can be secured. Restrictions on foreign participation and competition with capital for upstream investments, as well as other energy sector investments (such as for power plants), would limit investments in the refinery industry. However, the shift towards privatization and more liberal policies towards foreign participation are likely to alleviate the capital shortages. Middle East In the Middle East, significant capacity increases are expected in Kuwait (0.80 mb/d), Iran (0.50 mb/d), and the UAE (0.18 mb/d). Kuwait's increase mainly reflects the reconstruction of its refineries damaged during the Middle East conflict. Kuwait's pre-invasion capacity was around 0.82 mb/d. Current output (mid-May, 1992) of the most heavily damaged Magnae-al-Ahamadi refinery is around one third of its pre-war capacity of about 0.27 mb/d. Most of the refinery capacity restored so far supplies the domestic market. It is expected that capacity of about 0.65 mb/d will be restored by the end of 1993. Iraq's plans to raise capacity to about 0.4 mb/d also reflect reconstruction efforts. Prior to the war, Iraq's capacity was around 0.65 mb/d, but war damage reduced it to between 0.10 and 0.20 34 mbld. Post-war plans are to increase capacity to 0.80 mbld by 1995. Iran is expected to be the most active in the region. The completion of the Arak refinery and the Bander Abbas refineries would add roughly 0.4 mb/d by the end of 1992 and another 0.20 mb/d by 1995. At that time, Iran's capacity would reach 1.4 mb/d. While not much capacity expansion is planned in Saudi Arabia, the kingdom is planning a major shift in the product mix, as well as an improvement in the quality of the products. Therefore, most investments are geared to modernization and upgrading of existing refineries. Plans also include reducing the share of fuel oil exports (from the current 50%) while increasing gasoline output. This involves putting in additional reformation capacity. The trend towards the lighter end of the barrel is common worldwide. The demand for lighter products is increasing faster than that for heavier products because fuel oil faces greater competition from low-priced natural gas for power generation. Plans also call for reducing the sulphur content of gas oil to below 0.5% for exports to the Far East market. Current Saudi modification plans are focused on the Ras Tanura plants and will be extended to Yanbu and the Jubail refineries. Saudi plans to increase its processing capacity for lighter products are evident in its oil capacity expansion plans which favor the development of fields yielding light and sweet crude oil. In 1991, the capacity to produce light Arab crude increased by up to 0.4 mb/d. It is expected that most of the capacity expansion plans, 0.8-0.9 mb/d (an increase from 8.9 mb/d in 1991 to 10 mb/d by 1995), will be for lighter grade. Hawiyah (part of the large on-shore Ghawar structure with current output of 2.2 mb/d) is expected to add roughly 0.6 mb/d of 34* API Arab light, while 0.17 mb/d of very light and sweet crude of 490 API is expected to be available from the Hawtch field. The state-owned Saudi Aramco also plans to downsize its offshore fields producing Arab heavy (270 API) and Arab medium (31- API). Planned increases in Africa are fairly limited. Plans are envisaged only in Nigeria (0.1 mb/d), Egypt (0.05 mb/d), Libya (0.04 mb/d), and Tunisia (0.03 mb/d). Nigeria is considering sites for two export refineries with a total capacity of 0.1 mb/d. Impact of Environmental Concerns Environmental concerns are pushing refiners towards improving the quality of their products. Most countries now have less tolerance for polluting fuels. While quality improvement plans are well under way in several industrial countries, large investments for this purpose are also planned in the developing countries. The EC has agreed to adopt sulphur limits for petroleum products. Rules call for a cut in diesel sulphur (measured by weight) to a maximum of 0.2% by end-September 1994 and to 0.05% in 1996. In the United States, plans call for a cut in diesel sulphur to 0.05% by end-September 1993 (from 0.25%). In the Pacific Rim, where gas oil accounts for about 30% of oil consumption (of 13.4 mb/d), there are plans for a sharp reduction in sulphur--to as low as 0.05% in some plants in Singapore, the Republic of Korea, and Thailand by 1995. The current sulphur limit in these countries is roughly 0.5%. In other countries such as the Philippines and Malaysia, new standards mandate a cut in sulphur to 0.5% by 1995 (from 1%). Lead reduction in gasoline is also planned in many countries, noticeably Pakistan, Singapore and Malaysia. Lead is one of the major and most dangerous pollutants from gasoline. 35 The significant investment needs to expand capacity and for environmental compliance could restrict the realization of these plans. Conventional plant costs for the capacity expansion plans (by 4.2-4.8 mb/d) in the Asia-Pacific region are around $50 billion; the costs for environmental compliance add considerably to these costs. In the United States, costs to meet the Clean Air Legislation requirements amount to around $40 billion for the US refining industry (although these estimates are somewhat speculative). Since environmental compliance will involve the revamping of many front-end refineries, the failure of plants to generate investments for this purpose would result in their closures. Industry estimates are that between 0.5 and 2 mb/d of refining capacity may be shut down because of either inability or unwillingness to comply with environmental standards. Up to 0.6 mb/d of capacity shutdown is expected in California. These reductions will be achieved by phasing out older plants where compliance costs are very high. While stricter environmental standards should lead to a more modern and efficient refinery industry in the long run, there are concerns that the higher costs could increase the barriers to entry and reduce competition in the US refining industry. Other concerns are that the closing of US refineries would increase dependence on imported products and hence increase US vulnerability to supply-side shocks. 36 Table A3: Liquid Fuets * Production By Nain Countries and Economic Regions Actual Projected Growth Rates o/ Averages Countries/ ----*-------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 ---------------------------------------------------------------------------------- -------------------------------------------- ----------------------------(Million Tons of Oil Equivalent)-------------------------- -------(% p.a.)------ High-Income 825 911 939 925 970 1,004 1,033 1,049 1,034 1,060 1.8 0.7 1.0 OECD 603 701 765 779 788 791 795 794 735 708 2.1 1.8 -0.7 United States 508 478 440 445 439 429 419 409 387 374 0.4 -0.2 -1.2 Canada 68 81 89 89 90 91 92 93 95 95 3.2 0.5 0.5 Germany 8 6 5 5 5 5 5 5 5 5 -1.5 -1.6 -0.7 United Kingdom 0 83 93 92 99 106 106 121 88 79 41.9 45.5 -1.1 Non-OECD 222 210 178 146 182 213 239 255 290 353 0.7 -2.9 6.5 United Arab Emirates 40 83 97 111 115 112 108 105 126 150 1.1 2.2 LMICs 1,501 2,184 2,167 2,193 2,175 2,168 2,173 2,186 2,464 2,683 4.0 0.8 1.5 Africa 62 120 139 149 148 148 146 146 152 146 11.9 0.9 -0.2 Nigeria 53 96 80 91 90 88 85 83 97 100 11.2 -1.3 0.7 Americas 269 301 360 380 374 375 375 377 424 463 1.7 2.1 1.5 Mexico 24 105 145 150 153 155 158 160 173 186 10.5 12.6 1.6 Venezuela 191 119 108 120 112 110 106 103 131 151 -2.8 -3.7 1.6 Asia & Pacific 82 211 268 273 286 286 289 292 297 340 347.0 5.6 1.6 Europe 384 634 596 541 519 511 504 500 520 550 4.6 2.6 0.1 Middle East & North Africa 705 918 799 841 849 845 857 871 1,008 1,177 3.2 -2.1 2.4 Saudi Arabia 197 501 342 419 412 404 380 362 431 519 4.5 -1.7 1.6 Iran 196 102 150 164 165 162 156 151 176 205 0.7 -5.8 1.6 Iraq 78 115 99 14 25 30 70 100 117 158 1.9 -0.1 19.0 Libya 148 84 63 69 71 70 67 65 76 90 3.4 -5.6 1.9 world 2,326 3,094 3,106 3,122 3,145 3,172 3,206 3,235 3,474 3,744 3.2 0.8 1.3 ------------------------------------------------------------------------------------------------------------------------------ a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: United Nations Energy Statistics (actual); World Bank, International Economics Department (projected). Table A: Liquid Fuels - Apparent Consumption By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 -- --------------------------(Million Tons of Oil Equivalent)-------------------------- -------( p.a.)------ High-Income 1,597 1,855 1,786 1,791 1,811 1,829 1,847 1,862 1,931 1,955 2.1 -0.4 0.6 OEC 1,54 1,797 1,737 1,740 1,759 1,777 1,793 1,807 1,869 1,884 2.1 -0.5 0.6 United States 673 792 793 776 781 788 795 802 831 835 1.5 0.2 0.5 Canada 71 88 78 74 75 76 77 77 79 79 1.6 -0.7 0.4 Sermany 132 152 135 137 139 143 143 144 149 150 3.0 -0.7 0.6 United Kingdom 101 84 83 83 84 85 85 86 89 89 0.5 -2.1 0.5 France 96 111 88 92 94 96 97 98 101 103 2.6 -1.9 0.8 Italy 92 99 92 95 98 99 101 102 105 106 2.8 -0.8 0.8 LNICs 680 1,185 1,320 1,329 1,336 1,343 1,359 1,373 1,564 1,789 5.2 3.1 2.2 Africa 29 41 50 52 53 54 56 57 64 70 4.7 2.6 2.4 Americas 147 229 255 262 270 277 285 293 336 386 3.9 2.4 2.8 Mexico 25 61 81 82 84 86 88 89 99 110 7.3 6.8 2.1 o Brazi 27 53 62 64 66 68 70 72 85 98 5.8 3.1 3.4 Asia & Pacific 104 218 310 325 341 354 368 382 464 530 8.0 4.7 3.6 China, People's Rep. 31 88 110 115 121 126 132 138 171 194 11.1 5.4 3.8 India 19 33 56 59 62 66 69 73 94 120 6.6 5.3 5.2 Korea, Rep. of 9 26 47 51 54 56 58 59 68 74 14.4 6.7 2.7 Europe 340 581 530 503 480 460 446 430 440 515 4.8 2.4 0.3 World 2,277 3,039 3,106 3,122 3,145 3,172 3,207 3,235 3,484 3,744 3.2 0.9 1.3 --------------------------------------------------------------------------------------------------------------------------------------- a/ Least sqares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: United Nations Energy Statistics (actual); World Iank, International Economics Department (projected). TabLe AS: Liquid Fuets - Gross Exports ly Main Countries and Economic Regions ActuaL Projected Grouth Rates a/ Averages Countries/ .........--1991- Economies 1%9-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 ---------------- ition Tons of Oil Equivalent)- **--- p.O.)--****-* Nigh-Income 392 506 584 561 592 618 646 678 682 700 4.2 1.4 1.6 OEC 155 246 351 353 356 358 360 362 350 330 6.6 4.9 -0.5 United Kingdom 17 58 59 58 64 70 80 84 60 so 10.6 11.4 -1.0 Mon-OECD 238 263 233 208 236 260 286 316 332 370 1.9 -2.1 4.2 United Arab Emirates 40 81 89 101 102 103 106 109 116 138 0.5 2.3 Kuwait 145 87 50 10 39 70 96 117 136 157 -2.7 -5.5 21.7 LMICs 1,130 1,438 1,328 1.328 1,306 1,308 1,305 1,311 1,425 1,589 2.8 -0.9 1.4 Africa 62 114 127 133 129 127 127 125 125 120 11.1 0.3 -0.7 Nigeria 50 90 78 84 80 78 78 77 87 90 10.7 -1.8 0.5 Americes 251 205 207 221 213 210 209 205 230 225 -0.6 -0.7 0.2 Mexico 3 44 66 68 69 70 70 71 74 75 18.5 27.6 0.7 Venezuela 178 100 91 103 95 92 90 86 110 122 -3.3 -4.4 1.2 Asia & Pacific 41 91 105 109 112 115 117 120 100 86 7.5 4.9 -1.7 -0.8 Europe 105 193 210 180 173 168 160 162 170 160 5.6 4.2 -0.8 Middle East & North Africa 672 836 679 685 679 688 692 699 805 963 2.5 -3.1 2.5 Saudi Arabia 180 472 302 370 362 337 336 320 376 450 4.2 -2.3 2.5 Iran 179 73 116 127 127 123 120 119 145 175 -0.4 -7.5 2.3 Iraq 75 107 84 - - 40 50 75 90 125 1.5 -0.7 Libya 147 80 55 61 62 61 60 60 66 77 2.9 -5.0 1.7 Morld 1,523 1,946 1,912 1,889 1,898 1,926 1,951 1,989 2,107 2,280 3.2 -0.3 1.4 ---------------------------------------------------------------------------------------------------------------------- --------- a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: United Nations Energy Statistics (actual); World Bank, International Economics Department (projected). Table A6: Liquid Fuets - Gross Imports By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ - ----------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 - --------------------------(MiLLion Tons of Oil Equivalent)-------------------------- -------(% p.a.)------ High-Income 1,180 1,480 1,431 1,427 1,432 1.443 1,460 1.491 1,582 1,595 3.3 -0.6 0.8 OECD 1,128 1,366 1,327 1,315 1,326 1,344 1,357 1,375 1,484 1,507 3.0 -0.7 1.0 United States 173 350 396 401 407 413 419 424 450 492 5.0 1.8 1.5 Germany 134 160 136 138 139 140 141 142 150 156 3.2 -0.6 0.9 NetherLands 65 78 80 80 81 82 82 83 85 86 3.6 0.7 0.5 France 105 126 98 99 100 100 101 102 105 107 2.6 -1.8 0.6 Italy 114 114 109 110 110 111 112 112 115 117 2.4 -1.2 0.5 Non-OECD 51 114 104 112 106 99 103 116 98 88 7.5 1.5 -1.7 Singapore 20 39 58 64 65 66 66 67 69 72 7.7 4.8 1.0 LNICs 311 451 480 464 468 483 491 498 526 686 3.5 1.2 2.8 Africa 30 36 38 35 34 34 36 36 37 44 3.4 0.6 1.6 0 Americas 129 136 102 105 109 112 117 121 128 148 0.2 -2.7 2.5 Brazit 19 47 43 44 45 46 48 49 55 62 5.7 1.6 2.6 Asia & Pacific 65 101 147 161 167 183 196 210 267 276 5.6 3.4 3.8 Korea, Rep. of 10 27 144 45 47 48 49 51 61 70 14.7 7.0 3.0 Europe 51 120 144 142 134 117 102 92 92 110 7.7 3.8 -1.8 World 1,491 1,931 1,911 1,891 1,794 1,926 1,951 1,989 2,107 2,280 3.3 -0.2 1.4 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: United Nations Energy Statistics (actuaL); World Bank, International Economics Department (projected). Table A7: OPEC CrMd 14Ioleum - Avags 1is, a/ 1960.1 (AcSaal), 1992-2005 (Projcad) Curent 1990 Coan~ - -5 MUV b/ G-7 CM cl 1960 1.5 7.2 9.4 1961 1.5 7.1 9.2 1962 1.4 6.5 8.4 1963 1.4 6.6 8.1 1964 1.3 6.1 7.3 1965 1.3 6.0 7.1 1966 1.3 5.8 6.9 1967 1.3 5.8 6.7 1968 1.3 5.8 6.5 1969 1.3 5.5 6.2 1970 1.3 5.2 5.9 1971 1.7 6.4 7.2 1972 1.9 6.6 7.3 1973 2.7 8.1 9.1 1974 11.2 27.5 34.0 1975 10.9 24.1 29.4 1976 11.7 25.5 30.6 1977 12.8 25.4 30.3 1978 12.9 22.3 24.9 1979 18.6 28.3 33.9 1980 30.5 42.4 49.7 1981 34.3 47.5 55.5 1982 31.0 43.6 50.8 1983 28.1 40.4 45.3 1984 27.5 40.4 44.5 1985 26.7 38.9 42.3 1986 13.6 16.8 17.9 1987 17.2 19.4 20.0 1988 13.6 14.3 14.7 1989 16.3 17.2 17.7 1990 21.2 21.2 21.2 1991 17.3 17.0 16.5 1992 17.6 16.5 15.8 1993 17.3 15.7 15.0 1994 17.4 15.5 14.7 1995 18.0 15.6 14.8 2000 24.7 18.0 16.8 2005 26.8 17.0 15.3 a/ For the priod 1960-73, pritesfer to Saudi Auaba 34-34.9 AP, f.o.b. Ra Ta=ur; for the following year dy am averg. OPEC cpot pries w ~ by *bei w ~petive expart vouns. bl Deflated by G-5 M~ ~factuing Unid Value (MUV) Index. o/ Deaed by 0-7 Comnor Pdit Index (CP). sourMeU: ki ~ 1 c Lad., oiln 20 l ll~ u,EnWuorhBnk, Inkenon~ Econoedes 41  Coal Summary * Since early 1991 when crude oil prices returned to the levels prevailing before the Gulf conflict, thermal coal export prices have been declining. The current f.o.b. price of just over $40/tOn for the US marker thermal coal represents a decline in real terms to its lowest level since the price series started in 1977. * For the period up to 1995, thermal coal prices are expected to increase somewhat in nominal terms (enough to maintain constancy in real terms) as economic activity picks up and marginal coal capacities are trimmed. Over the long term, thermal coal prices are forecast at about the f.o.b. costs of the major coal exporters. The relatively low prices during the first half of the 1990s should dampen investment in capacity expansion to restore market balance over the longer term. * During 1985-89, world coal consumption increased at 2.2% p.a. compared with 2.9% p.a. growth during 1980-85. However, in 1990 world consumption increased only 0.2%, mainly because of sharp downturns in the FSU and Eastern Europe. Although demand growth for coal has slowed in recent years, the latest figures represent robust demand growth in view of the adverse economic conditions and heightened environmental concerns. Coal Prices, a/ 1975-2005 CS/ton) 90 s0 Constant 1990 $ by 70 s0 5% 50 00 40 A Current $ \ ,- 1% 1 30 . . . . . . A L t p * i 1975 1980 1985 1990 1995 2000 2005 a/ US, 12,000 btu/Ib, <1% sulfur, 12% ash, fob b/ Deflated by G-5 MUV Index 42 * Coal's share in world primary energy consumption increased for most of the 1980s, but recent data show that the share has stabilized and we expect it to decline slightly over the forecast period. World coal consumption is projected to increase at 1.8% p.a. over the 1991-2005 period, while world thermal coal consumption is forecast to grow at 2.5% p.a. in this period. * The coal export capacity of major coal-exporting countries is expected to increase by 50 million tons over the 1991-95 period, a large enough quantity to keep supply slightly in excess of demand. Future production increases should come principally from the traditional coal-producing countries such as Australia, China, India, South Africa, and the United States. Most European countries and Japan are expected to phase down their coal industries due to high costs, while Colombia, Indonesia, and Venezuela are likely to become more important coal exporters. Demand Outlook Over the second half of the 1980s, OECD's coal consumption growth rate steadily decelerated from a 3.6% increase in 1987 to a 0.3% decrease in 1990. A large part of this decline is attributable to the slowdown in economic activity-from 3.4% growth in GDP in 1987 to 2.3% growth in 1990. However, the decline also appears to have been the result of changes in relative fuel prices and expectations of more stringent environmental restrictions, as discussed below. In the non-OECD countries, coal consumption grew at much faster rates, by 8.1% in 1987 and 3.3% in 1988, but declined by 1.7% in 1990, largely due to sharp declines in the FSU and Eastern Europe. For China, the world's largest coal producer and consumer, consumption continued to grow through 1990, as it did in other developing countries such as India and South Africa. Coal losing market share, mostly to nateul gas. The much lower petroleum prices since 1986 clearly have had an impact on fuel choices, favoring oil and natural gas in place of coal. Coal's share in primary energy consumption had been increasing since the first oil price shock, but after the 1986 collapse of petroleum prices its share started to decline, both in the OECD and developing countries (see Table 1). Many EC member countries have shown a preference for natural gas, available from nearby areas (e.g., the North Sea, the Netherlands, the FSU, and Algeria), rather than coal, especially as some countries (the United Kingdom and Germany) have reduced subsidies to domestic coal producers. The proposed EC carbon tax has also been a factor. Japan, however, has continued to rely heavily on imported coal in its plans for power generation capacities, partly because imported natural gas is much more costly in Japan than in Europe (see Table 3). In Eastern Europe and the FSU, both coal and petroleum have been losing ground to natural gas. Growth in thermal coal consnption for electricity generation increases. By far the largest and fastest-growing market for coal is in thermal electricity generation. The growth rate of world coal consumption for electricity generation accelerated from 3.3% p.a. during 1980-86 to 4.6% p.a. during 1986-89. The growth rate accelerated in all major country groups over this period: in the OECD from 2.3% to 3%; in developing countries (including China) from 7.9% to 9.5%; in the FSU and Eastern Europe from 1.2% to 2.2%. In terms of volume, the bulk of the increases took place in the traditionally large coal consumers, i.e., China, the United States, and India, while the most rapid rates of increase were in the industrializing developing countries in Asia, particularly in recent years. The post-1986 acceleration in thermal coal consumption for power generation in the OECD is explained mostly by increased demand for electricity, because as Table 2 shows, coal's share 43 Table 1: Trends in Fuel Shares, by Region, 1980-90 1980 1985 1987 1989 1990 (%) OECD Coal 20.4 22.4 21.7 21.4 21.4 Oil 49.8 43.4 43.6 43.3 42.9 Gas 20.1 19.3 18.9 19.4 19.5 Others 9.6 14.9 15.8 16.0 16.3 Developing Countries a/ Coal 37.6 40.6 40.9 40.6 39.9 Oil 40.8 35.1 33.5 33.5 36.6 Gas 9.0 10.8 11.5 12.5 12.6 Other§ 12.6 13.5 14.1 13.4 10.9 Eastern Europe & the FSU Coal 36.7 33.3 32.6 30.5 29.1 Oil 35.3 30.6 29.6 29.3 28.5 Gas 24.7 31.2 32.6 34.6 36.7 Others 3.3 4.9 5.2 5.6 5.7 a/ Defined as non-OECD countries excluding the FSU and Eastern Europe. Sources: International Energy Agency (IEA), Coal Information 1992, and Energy Statistics and Balances of Non-OECD Countries 1988/89, Paris, OECD. in total thermal power generation in the OECD declined slightly in recent years. In developing countries, coal's share continued to increase through 1989, while in Eastern Europe and the FSU an abundance of natural gas in the FSU has allowed steady increases in its share to the detriment of oil and coal. Natural gas-fired power generation to grow more rapidly than coal-fired. Between 1982 and 1989, OECD's coal-fired power generating capacity (including multi-fuel units that can use coal) increased at 0.8% p.a. and its share of total thermal power capacity also increased slightly. Thus, the bulk of the increase in coal-fired power generation was achieved through enhanced capacity utilization of coal-fired units. According to a recent OECD member governments' survey, coal-fired power capacity is expected to grow at 1.1% p.a. between 1989 and the year 2000, oil-fired units should decline marginally, and natural gas-fired capacity should increase most rapidly at 3.5% p.a. Natural gas is available to many OECD countries on attractive terms either from domestic production (for Canada, United States, United Kingdom, Norway, and the Netherlands) or through imports (for most other countries in Western Europe). Among the industrial countries, Western Europe is leading the way in terms of the speed with which natural gas use for power generation is expected to increase (3.9% p.a.), followed by North America (3.4% p.a.) and the Pacific region (3%). 44 Table 2: Fuel Shares in Thermal Power Generation, by Region, 1980-90 1980 1985 1987 1989 1990 OECD Coal 58.0 69.1 69.9 68.1 68.7 Oil 25.3 14.4 14.2 16.2 15.4 Gas 16.7 16.5 15.9 15.7 15.9 Developing Countries a/ Coal 49.5 49.7 50.9 52.8 52.9 Oil 33.1 25.6 22.2 18.6 19.0 Gas 17.4 24.7 26.9 28.6 28.1 Eastern Europe & the FSU Coal 50.0 44.6 45.0 43.9 43.1 Oil 28.0 21.2 17.8 12.6 13.4 Gas 22.0 34.2 37.2 43.5 43.5 a/ Non-OECD countries excluding the FSU and Eastern Europe. Source: See Table 1. Relative fuel prices vary within OECD. Table 3 shows fuel prices delivered to electric utilities in four major industrial countries. At the relative fuel prices prevailing in the United States in 1991, coal is hardly more advantageous than natural gas even without environmental considerations-that is, according to the conventional rule of thumb that coal costs should not be more than one half the costs of oil or natural gas per unit of heat value in order to be competitive in the thermal power market. In other countries, coal's cost disadvantage is far more apparent. In Italy, for example, where domestic prices reflect international prices more closely than in other European countries, natural gas appears to be highly attractive compared with coal. The advantage coal had during 1980-85 is no longer apparent. In Japan and Germany, fuel prices are distorted by subsidies; the prices in Table 3 therefore do not reflect their true costs to these countries. If natural gas prices in these countries reflected the true import costs and coal prices were to come down to a level similar to that in Italy, coal would be highly competitive against natural gas. This is the reason why Japan still has an ambitious plan to expand coal-fired power capacity. Germany strongly leans to natural gas rather than imported coal, but commitments to import coal from Poland are likely to be maintained. Developing country coal consumption growth strong. Rapid increases in coal consumption in developing countries have been seen in both thermal and metallurgical coals. Coal-fired thermal power generation expanded rapidly between 1980 and 1989: in India (at 11.4% p.a.), China (8% p.a.), Indonesia (from zero in 1980 to 3.4 million tons in 1989), Thailand (20.1% p.a.), South Africa (4% p.a.), the Republic of Korea (17% p.a.), Hong Kong (from practically zero in 1980 to 9.9 million tons in 1989), Israel, and Taiwan, China (15.3% p.a.). Metallurgical coal consumption for steel production also increased rapidly in the Republic of Korea (9.6% p.a.) and Brazil (8.1%). We expect these countries to continue to provide the main source of growth for coal demand in developing countries. 45 Table 3: Fuel Prices Delivered to Electric Utilities in Major Industrial Countries, 1980-91 1980 1985 1987 1989 1990 1991 -(US$/ton of oil equivalent) United States Thermal Coal 57.1 68.5 63.2 59.8 60.4 60.1 Heavy Fuel Oil 178.2 177.3 124.3 119.5 139.5 102.1 Natural Gas 97.0 151.3 98.5 104.0 102.3 94.4 Japan Thermal Coal 120.8 105.2 151.6 170.1 168.8 :176.1 Heavy Fuel Oil 260.7 233.1 179.0 190.3 215.8 n.a Natural Gas 217.4 214.6 151.3 164.0 185.7 n.a. Germany, Fed. Rep. of Thermal Coal 148.3 130.3 192.7 191.1 224.3 215.7 Heavy Fuel Oil 203.6 184.5 131.5 129.9 149.2 149.2 Natural Gas 140.8 164.0 165.7 143.0 175.7 n.a. Italy Thermal Coal 81.9 82.0 74.5 83.6 96.8 94.8 Heavy Fuel Oil 188.7 179.7 116.7 105.4 140.2 148.1 Natural Gas 192.2 167.4 111.6 101.3 129.9 134.3 Source: IEA/OECD, Enerr Prices and Taxes, various issues. Supply Outlook Coal resources plentful and mqjor producers plan capacity increases. Between 1980 and 1989, world coal production increased steadily at an average annual rate of 2.8%. However, in 1990 world production declined 0.8%, led by sharp falls in the FSU and Eastern Europe (by 9.1%) and in Western Europe (by 4.6%). Production losses in these countries are a part of the ongoing structural adjustments in the industry-although of a different nature for Eastern and Western Europe-and are likely to continue for several more years at least. During the 1980s, production increases were achieved mostly in the traditional coal-producing countries such as Australia, China, India, South Africa, and the United States, and in newcomers such as Colombia, Indonesia, and Venezuela. World coal resources are ample to meet expected demand increases. The major coal- producing countries mentioned above will continue to provide most of future incremental production. Most firm or planned capacity expansion projects are located in these countries. Projects considered firm and expected to be completed between 1991 and 1995 amount to 50 million tons of annual capacity. Australia alone accounts for more than one half (28.6 million tons) of this total. The largest single project is the 10 million ton/year La Loma project in Colombia. Indonesia has two large projects totaling 10 million tons/year. 46 Over the longer term, many new projects, either extensions of exiting mines or greenfield projects, could be brought into production. Most are located in countries that are already major producers and exporters. Projects geared mainly for the export market amount to 187 million tons, of which Australia accounts for 118 million tons. Indonesia, Colombia, and Venezuela each could add 10-15 million tons. As political conditions in South Africa improve, more export projects could be added to that country's current plans. Most new projects in the United States are geared to domestic markets; the only large project considered for the export market is in Alaska, which has obvious transportation advantages for Asian markets but faces serious environmental constraints. Exporters' relative costs consistent with capacity expansion plans. Competition among the major coal exporters for world markets will be determined by their relative cost competitiveness. Table 4 summarizes the results of a recent study of coal costs for the main coal exporters in supplying the two main markets-Western Europe and East Asia. These are intended to show the typical (average) cost configuration; it should be kept in mind that a significant proportion of mines may have costs below or above the levels shown. In the European market, the United States and Colombia are roughly at par, although Australia is not far behind. Australia has a clear advantage in East Asian markets. The sub- bituminous coal from the US Midwest faces tough competition in the Asian market from Australian coal; it is therefore unlikely that the abundant low-sulphur coal from the US Midwest will find its way into the export market much beyond the current low levels. South Africa can export to both Europe and Asia. However, South African coal is of relatively low quality and hence its cost advantage is not as large as what the figures suggest. Nevertheless, it is not difficult to see that South Africa will play a greater role as a coal exporter in the long term provided the political conditions improve enough to attract more investment. The planned capacity increases summarized above are roughly consistent with the current relative cost structure. Australia's coal production has apparently benefitted from its new labor practices, with production increasing 9.5% in 1990 and 8.1% in 1991. Partly to offset appreciation of the Australian dollar vis-a-vis the US dollar since the mid-1980s, Australian coal exporters have been under considerable pressure to remain profitable. Measures taken to reduce costs have included lowering rail freights and royalties, in addition to adopting new labor practices. At current export prices, some marginal producers in the United States are reportedly hard-pressed to remain in business, and industry consolidation (closures, mergers, and takeovers) has been reported among US coal producers. US exports benefitted from US dollar depreciation in the second half of the 1980s, but competition from Colombia, South Africa, and Venezuela is likely to keep export volumes from growing significantly over the remainder of this decade. Over the longer term, only the US coal industry has a resource base large enough to supply a big volume to the European market. Most production for domestic market in China and India. In China and India, the two largest coal producers within the developing world, the coal industry is heavily regulated. About 65% of China's coal production comes from publicly-owned mines and their prices to domestic users are set well below economic costs to the country. Coal India Limited (CL), a state-owned enterprise, controls 90% of India's coal production. Only in recent years have CL's prices to domestic users been allowed to increase to the level of CIL's costs, assuming efficient operation. In both of these countries, coal serves as the predominant source of energy, so as to minimize imports of oil. Thus, domestic needs have taken precedence over exports and export volumes have been relatively small compared with total production. China has been known to occasionally divert export coal to domestic uses. It is believed that China is the least-cost supplier to Asian markets and has the potential to become a major supplier to the 47 Table 4: Representative Costs of Thermal Coal Production in Major Producing Countries United South States Australia Africa Colombia Under- Under- ground Surface ground Surface Surface Surface (1990 US$/ton) Mining Cost 22.0 28.1 27.5 12.2 9.3 25.0 Capital Charge a/ 1.0 1.2 6.3 6.6 1.0 12.7 Inland Trans. 20.0 12.5 6.0 10.0 10.0 4.0 Lading Cost 2.5 1.4 3.7 3.5 2.3 3.0 Total f.o.b. cost 45.5 43.2 43.5 32.3 22.6 44.7 Ocean Freight to Japan 14.8 12.8 8.5 7.2 9.7 N.A. to Europe 6.7 8.5 12.5 11.6 8.7 6.7 Total c.i.f. Cost c.i.f. Japan 60.3 56.0 52.0 39.5 32.3 N.A. c.i.f. Europe 52.2 51.7 56.0 43.9 31.3 51.4 a/ Assuming 10% rate of return on investment. Source: IEA, Coal Information 1991, OECD, Paris, 1991. region. However, rapid increases in domestic consumption, partly stimulated by subsidies, have prevented it from becoming a large exporter. India's coal is of relatively low quality and not suitable for export without extensive preparation. In a number of countries where coal has been mined for a long time, the remaining resources are becoming scarce and increasingly costly to mine. These countries include Germany, the United Kingdom, most other European countries, Japan, and the Republic of Korea. Almost all of these countries have plans to phase out subsidies to domestic coal production, so that their production levels are expected to decline over time. The system of contracts and subsidies currently in force in Germany is likely to slow down the decline in its coal production, while lignite production in the former East Germany may be phased out quickly for environmental reasons. Privatization of electric utilities in the United Kingdom should speed up the substitution of imported coal and natural gas for domestic coal in that country. Limited expansion of FSU coal production foreseen. The FSU and Eastern Europe are significant coal producers and coal resources in the FSU are believed to be large. They are located in three major fields-Moscow Basin, Donbass Basin north of the Black Sea, and the Kuzbass/Yakut fields in western Siberia and Kuznetsk and Kansk-Achinsk Basins in southern Siberia. The Siberian coal fields 48 have high-quality metallurgical coal and there are high-capacity rail links to Asian market ports. These characteristics have attracted Japanese investments, which explains the current export volume to Japan, and Japan has shown an interest in further investments in an effort to diversify its supply sources. For the domestic market and exports to Europe, coal's position in the FSU's energy supplies depends on its competitiveness with the region's vast natural gas resources. European markets clearly prefer natural gas to coal. At the same time, the FSU's coal resources and infrastructure requirements to serve the European markets put it at a significant disadvantage to natural gas. Therefore, the FSU's coal sector is not likely to grow significantly either in the domestic or export market. Polish industry needs to become more efficient. Poland has been an important coal producer and exporter for a long time but still has sizable high-quality reserves. However, good-quality hard coals are mined from deep underground mines and production costs have been rising rapidly. With the shift to a market-oriented system, the Polish coal industry suffered losses partly because of the existence of inefficient mines. The ongoing industry restructuring is likely to result in permanent closures of marginal hard coal mines and most brown coal mines, and sharp declines in production over the near term. But Poland needs coal for its long-term energy supplies, so investments in the industry and appropriate changes in its management and regulatory environments will be required to meet these needs economically. Other East European countries rely heavily on brown coal production for electric power generation and industrial and domestic uses. Because of the high sulphur content of brown coal, its production and consumption is likely to be phased out in these countries. Price Outlook Since early 1991 when crude oil prices began to turn downward, thermal coal prices in international markets have also been declining, albeit mildly. In terms of the US export prices of the marker thermal coal, the total decline from the last quarter of 1990 to the second quarter of 1992 amounted to $1.75/ton or 4.2%. The fall in prices has been caused by weak economic conditions, stronger competition from oil and natural gas, and the re-emergence of South Africa as a major coal exporter. Although the market is currently burdened with excess supply capacity, production has been restrained so as not to cause a large buildup of stocks. At just over $40/ton currently for the US marker thermal coal, prices are at their lowest level in real terms since the price series commenced (in 1977). No increase in real prices in medium term, small increase in long term. Over the period to 1995, thermal coal prices are expected to increase somewhat in nominal terms as the industrial economies recover from recession and marginal coal capacities are removed from the market. At current prices, the industry faces consolidation and restructuring; investments for capacity expansion are also being restrained. This is likely to relieve some of the excess supply pressure but not by enough to result in price increases in real terms. New export coal projects expected to come on-stream over the 1991-95 period amount to 50 million tons, or about 16% of world exports. Given that world import demand is expected to increase by 14.5% over this period, the market balance is not expected to improve significantly, if at all. Over the long term, thermal coal prices are forecast at about the f.o.b. costs of the major coal exporters (see Table 4). For example, the price for the US benchmark thermal coal is forecast at US$43-44/ton for the years 2000 and 1990 in 1990 constant dollars, in line with US costs shown in Table 4. The long-term price forecasts imply small increases in real terms for thermal coal export prices. These forecast increases can be justified on several grounds. First, relatively low prices during the first half of 49 the 1990s should discourage investments in coal capacity expansion and hence result in a more balanced market towards the year 2000. Next, petroleum prices are expected to rise in real terms in the second half of the 1990s, and this expectation keeps coal as a viable energy option. Finally, production costs of coal are expected to rise moderately over the long term as production moves to more difficult locations and opportunities for efficiency improvements become harder to find. The long-term forecasts in this report are virtually the same as those made two years ago. Coal and the Enviroment Coal use generates environmentally harmful side-effects throughout its life cycle, from mining to waste disposal. The most serious environmental impact arises from its combustion, which discharges sulfur dioxide, nitrogen oxide, carbon dioxide, and various particulates. Reasonably cost- effective technology exists for removing sulphur dioxide, nitrogen oxide, and particulates from the effluent gas. Recently, attention has been focused on the possible "greenhouse" effect of the carbon dioxide emissions from coal burning. This aspect is much more difficult to deal with because no effective way of reducing coal's carbon dioxide emissions is known. Carbon taxes becoming more popular. Coal discharges about twice the carbon dioxide emissions of natural gas, per unit of energy produced, and about 23% more than oil. Thus, greenhouse considerations have led to a preference for natural gas or oil. On May 26, 1992, the EC Council of Ministers failed to reach agreement on a proposed energy/carbon tax. The main point of contention appears to have been the issue of conditionality, i.e., whether the EC should adopt the proposal irrespective of whether the United States and Japan adopt similar measures. Four European countries (Finland, the Netherlands, Norway, and Sweden) already have carbon taxes. Nine other industrial countries, including France, Germany, Japan, and Italy, are planning to impose carbon taxes. Thus, the likelihood of some form of carbon tax coming into force in the EC remains high. This is one reason why EC members are opting for natural gas for a large part of their future utility plans. Most industrial countries have adopted stringent emission standards for sulfur dioxide, nitrogen oxide and particulates. However, a large proportion of coal users in industrial countries lack effective measures for reducing the discharge of these pollutants, and the situation in developing countries, the FSU, and Eastern Europe is far worse than it is in industrial countries. 50 Table Al: Solid Fuels - Production By Main Countries and Economic Regions ------------------------------------------------------------------------------------------------------------------------------ Actual Projected Growth Rates a/ ----------------------------- ---------------------------------------------------- -------------------------- Averages Countries/ 1991- Economies 1969-71 1979-81 1990 b/ 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 ------------------------------------------------------------------------------------------------------------------------------ --------------------------(Million Tons of Oil Equivalent)-------------------------- -------(% p.a.)------ High-Income 723.2 773.9 894.6 852.6 855.5 859.5 860.5 863.5 883.5 921.5 0.8 1.5 0.6 OECD 720.2 772.3 893.1 851.1 854.0 858.0 859.0 862.0 882.0 920.0 0.8 1.5 0.6 United States 363.3 430.9 537.9 515.7 525.0 535.0 540.0 545.0 580.0 615.0 2.0 2.2 1.3 Germany 154.0 141.5 130.0 104.0 100.0 97.0 95.0 95.0 80.0 70.0 -0.7 -0.2 -2.8 United Kingdom 88.5 72.8 52.3 54.0 50.0 45.0 40.0 35.0 20.0 15.0 -3.0 -2.3 -8.7 Australia 32.1 57.8 106.9 112.4 115.0 118.0 121.0 124.0 140.0 160.0 7.0 6.0 2.6 LMICs 753.0 1,051.5 1,409.9 1,404.0 1,405.3 1,431.1 1,473.9 1,511.4 1,691.5 1,886.5 3.5 3.4 2.1 Africa 36.6 69.1 108.1 108.6 111.3 116.6 116.9 118.4 134.5 154.5 5.3 6.5 2.5 South Africa 33.7 66.1 104.0 104.5 107.0 112.0 112.0 113.0 128.0 147.0 5.6 6.7 2.5 Americas 6.1 10.5 24.1 26.4 27.0 27.5 28.0 35.0 49.0 65.0 5.3 6.7 6.6 Asia & Pacific 231.3 414.9 652.7 695.0 717.0 737.0 759.0 768.0 868.0 977.0 6.3 5.7 2.5 U. China, People's Rep. 165.7 305.5 480.0 505.0 520.0 535.0 550.0 553.0 618.0 690.0 6.8 5.8 2.3 India 36.3 65.3 112.7 130.0 133.0 136.0 139.0 142.0 160.0 177.0 5.3 6.7 2.2 Europe 478.3 555.7 625.0 574.0 550.0 550.0 570.0 590.0 640.0 690.0 1.6 1.3 1.3 World 1,476.2 1,825.4 2,304.5 2,256.6 2,260.8 2,290.6 2,334.4 2,374.9 2,575.0 2,808.0 2.2 2.6 1.6 ---------------------------------------- -------------------------------------------------------------------------------------- a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: United Nations Energy Statistics (actual); World Sank, International Economics Department (projected). Table A2: Solid Fuels - Apparent Consumption By Main countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 b/ 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 - --------------------------(Million Tons of Oil Equivalent)-------------------------- -------(% p.a.)------ High-Income 755.0 790.4 856.0 835.0 842.0 851.0 874.0 896.0 953.0 1,009.0 0.8 1.5 1.4 OECD 751.9 786.3 850.0 829.0 836.0 845.0 868.0 890.0 947.0 1,003.0 0.8 1.5 1.4 United States 325.1 370.9 458.0 449.0 455.0 460.0 470.0 480.0 497.0 515.0 1.9 2.2 1.0 United Kingdom 90.9 71.6 62.0 63.0 60.0 57.0 54.0 50.0 50.0 48.0 -2.5 -1.6 -1.9 Germany 157.8 140.7 135.0 112.0 113.0 114.0 114.0 115.0 105.0 100.0 -0.5 0.0 -0.8 LMICs 733.5 1,023.6 1,321.9 1,318.1 1,320.2 1,338.5 1,381.7 1,414.3 1,573.8 1,736.2 3.5 3.4 2.0 Africa 35.9 51.6 70.3 72.4 74.5 77.7 79.8 83.0 94.0 105.0 3.7 4.0 2.7 South Africa 32.7 48.4 65.0 67.0 69.0 72.0 74.0 77.0 87.0 98.0 3.9 4.1 2.8 Americas 8.6 14.8 24.0 25.0 26.0 27.0 28.0 30.0 36.0 43.0 5.0 5.8 3.9 Ln Asia & Pacific 230.1 419.1 639.6 681.1 702.7 722.3 743.8 752.6 855.0 967.2 6.4 5.9 2.5 China, People's Rep. 164.5 303.8 470.4 494.9 509.6 524.3 539.0 540.8 598.2 659.0 6.8 5.8 2.1 India 35.6 65.4 111.6 128.7 131.7 134.6 137.6 140.6 158.4 175.2 5.3 6.6 2.2 Europe 457.5 535.9 588.0 539.6 517.0 511.5 530.1 548.7 588.8 621.0 1.7 1.4 1.0 World 1,488.5 1,814.0 2,177.9 2,153.1 2,162.2 2,189.5 2,255.7 2,310.3 2,526.8 2,745.2 2.2 2.5 1.8 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: United Nations Energy Statistics (actual); World Bank, International Economics Department (projected). Table A3: Solid Fuels - Gross Exports By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 b/ 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 ---------------------------(MiLlion Tons of Oil Equivalent)------------------------- --------(% p.a.)------ High-Income 77.9 113.0 169.7 183.0 180.0 181.0 183.0 183.0 202.1 227.1 3.5 4.0 1.6 OECD 77.9 113.0 169.7 183.0 180.0 181.0 183.0 183.0 202.1 227.1 3.5 4.0 1.6 United States 37.0 49.6 67.0 69.0 70.0 70.0 70.0 70.0 77.2 87.7 2.6 2.7 1.7 Australia 11.3 27.3 67.7 78.0 78.0 80.0 83.0 85.0 99.9 117.4 13.4 8.9 3.0 LMICs 42.1 62.3 112.7 109.8 134.8 147.6 149.6 158.6 204.7 239.7 4.3 5.0 5.7 Africa 1.6 18.6 29.1 31.3 38.4 40.4 38.4 36.4 39.5 46.5 14.8 22.7 2.9 South Africa 1.1 18.2 30.0 31.0 38.0 40.0 38.0 36.0 39.1 46.1 18.9 24.4 2.9 Americas 0.0 0.1 11.4 12.0 12.0 14.2 14.3 18.9 35.0 42.0 31.6 38.0 9.4 Colombia 0.0 0.1 9.5 10.0 10.0 12.0 12.0 16.5 22.0 27.0 44.1 7.4 Asia & Pacific 1.8 3.8 15.2 16.5 18.4 19.5 21.0 24.0 32.0 40.0 6.5 Europe 38.6 39.7 57.0 50.0 66.0 73.5 75.9 79.3 98.2 111.2 2.3 1.6 5.9 World 120.0 175.2 282.4 292.8 314.8 328.6 332.6 341.6 406.7 466.8 3.8 4.4 3.4 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: United Nations Energy Statistics (actual); World Bank, International Economics Department (projected). Table A: Solid Fuels - Gross Imports By Main Countries and Economic Regions --------------------------------------------------------------------------------------------------------------------------------------- Actual Projected Growth Rates a/ Averages Countries/ ---------------- 1991- Economies 1969-71 1979-81 1990 b/ 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 ---------------------------------------------------------------------------------------------------------------------------- ------- ---------------------(Million Tons of OiL Equivalent)-------------------------- -------(% p.a.)------ Nigh-Income 104.1 147.2 191.5 211.9 211.9 219.2 226.9 240.0 276.2 318.5 3.3 3.2 3.0 OECD 104.0 144.4 171.5 190.9 190.4 197.2 202.9 215.0 246.2 283.5 3.1 2.9 2.9 Japan 33.0 48.6 70.0 75.9 74.0 76.0 79.0 82.0 95.0 110.0 7.9 4.0 2.7 Germany 14.7 14.5 14.7 12.2 13.0 17.0 19.0 20.0 25.0 30.0 -1.2 -1.1 6.6 Belgium-Luxembourg 7.8 9.1 10.1 9.8 9.9 10.2 10.4 10.7 12.2 13.7 0.9 1.0 2.4 France 10.8 20.2 13.7 14.8 15.0 15.2 15.4 15.7 17.5 19.3 0.4 -0.8 1.9 Italy 8.1 11.1 13.9 13.4 13.7 14.0 14.3 14.6 16.5 18.5 2.6 3.6 2.3 LMICs 23.5 37.1 72.8 72.4 73.8 76.6 79.0 82.5 107.6 132.5 5.2 6.4 4.4 Africa 0.9 0.4 2.7 2.7 2.8 2.8 2.9 2.9 3.6 4.5 -4.3 -1.7 3.7 Americas 2.6 4.7 9.5 9.7 10.0 10.3 10.6 11.0 14.0 17.0 6.7 7.7 4.1 Asia & Pacific 1.4 8.3 26.7 27.0 28.0 28.5 29.5 30.6 43.0 55.0 12.9 21.9 5.2 Europe 18.1 22.5 33.9 33.0 33.0 35.0 36.0 38.0 47.0 56.0 3.5 3.2 3.8 World 127.7 184.2 264.3 284.3 285.7 295.8 305.9 322.5 383.8 451.0 3.7 3.9 3.4 --------------------------------------------------------------------------------------------------------------------------------------- a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: United Nations Energy Statistics (actual); World Bank, International Economics Department (projected). Table AS: Thermal Coal - Prices, 1977-91 (Atual), 1992-2005 (ProjeCted (ti/sn United States a/ - Autlia b/ Currfnt $ - 1990 Con~ant $- Curr~n $ - 1990 Con~at $-- G-S MUVo/ G-7 CPI d/ G-S MUV O/ 0-7 Cd/ 1977e/ 33.4 66.3 79.0 29.0 57.6 68.6 1978 39.6 68.4 76.4 28.6 49.4 55.2 1979 35.4 54.0 64.6 29.7 45.3 54.2 1980 43.1 59.9 70.3 39.4 54.7 64.3 1981 56.5 78.2 91.4 52.8 73.1 85.4 1982 52.2 73.4 85.5 54.8 77.0 89.8 1983 44.5 64.0 71.8 38.2 54.9 61.6 1984 48.6 71.4 78.6 31.0 45.6 50.1 1985 46.6 67.9 73.9 33.8 49.3 53.6 1986 43.9 54.3 57.7 31.1 38.4 40.8 1987 36.2 40.1 42.1 27.5 31.0 32.0 1988 37.1 38.9 40.1 34.9 36.6 37.7 1989 40.5 42.8 43.9 38.0 40.1 41.2 1990 41.8 41.8 41.8 39.7 39.7 39.7 1991 41.5 40.7 39.5 39.7 38.9 37.8 1992 40.4 37.9 36.2 39.0 36.6 34.9 1993 42.0 38.0 36.3 41.0 37.1 35.5 1994 43.0 38.2 36.3 42.0 37.3 35.5 1995 45.0 38.9 37.0 44.0 38.1 36.1 2000 60.0 43.7 40.9 57.0 41.5 38.8 2005 68.0 43.1 38.7 65.0 41.2 37.0 al 12,000 btulb, <1% Iufur, 12% amh, f.o.b. plors, Hampton oa, Nodolk, United Stats. b/ 12,000 bu/lb, <1% ulfur, 14 amh, f.o.b. piers, Newcade/PortaKsd~ la, AuraUi. c/ Deftated by G-5 M~afacturing Unit Valu (MUV) Index. d/ Deted by G-7 Conumer Price Index (CP). e/ May-Decmber 1977. Sourcos: C mALWeJ ad Coal Weak International, varous immes (actual); World Bak, Inornational Econo~ica Department 55 Copper Summary * Among base metals, copper has continued to enjoy relatively high prices over the past several years, supported by a series of supply disruptions at a time of sustained and substantial demand growth. Copper prices rarely fell below $1/lb over the 1988-92 period and lately rose to $1. 10/lb on fears of supply disruptions and a surge in demand. * Over the period to 1996, the market balance is expected to turn to a moderate surplus as supplies increase. Prices are likely to decline. The constraints of smelter capacity should continue to be binding through early 1995, limiting the size of the surplus, which in any case is likely to be a fragile one easily wiped out if industrial economies recover strongly. * If the historical pattern of the investment cycle is repeated, copper prices are likely to trend upwards in the latter part of the 1990s, followed by declines over the 2000-2005 period. Long- term copper prices are forecast to be roughly at a level sufficient to bring on-stream new projects currently envisaged for the 1990s and beyond. Copper Prices, a/ 1950-2005 Cs/ton) 9,000 7,000 Constant 1990 $ b/ 5,000 5,000 4,000 3,000 2,000- Current $ 1,000 I 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 0/ LME high grade/Grade A cathodes and wirebare b/ Deflated by G-5 MUN Index * World consumption of refined copper is forecast to be relatively robust with an average growth rate of 2% p.a. over the 1991-2005 period. As before, industrializing developing countries are expected to be the main growth market for copper. The copper intensity of industrial production 56 in industrial countries should decline at a slower rate than over the past two decades owing to lower energy prices and faster expansion of the capital goods sector. Demand in the FSU and Eastern Europe is expected to remain depressed in the near term but it should grow quickly from the mid-1990s as these economies require huge investments in infrastructure and equipment. A tally of announced projects indicates that over the 1992-95 period, world mine and refined copper capacities will expand by as much as 1 million tons each. Most of these increases will take place in Latin America, North America, and Asia. For the 1995-2005 period, probable and potential new projects amounting to 2.3 million tons of copper are envisaged, mostly in developing countries whose investment climate should improve as they move away from state monopoly of mineral resource exploitation. Demand Outlook Recent copper consumption growth mirrors growth of industridl production. Despite economic recession, preliminary statistics show that refined copper consumption in the world excluding the centrally planned economies (CPEs) (former and present)' increased 1.4% in 1991, following increases of 5.1% and 1.5% in 1989 and 1990, respectively. Most of the increase in 1991 (and the preceding years as well) came from the LMICs, with an 8.1% increase, while consumption in the high- income countries remained largely unchanged. Refined copper consumption in the CPEs (former and present) declined 14.4% in 1990, the latest year for complete data, resulting in a 1.9% decline in world consumption in that year. Among industrial and developing countries, there have been clear regional disparities in consumption growth. In Western Europe, substantial demand increases in Germany were offset by decreases in other countries so that the total for Western Europe for 1991 was unchanged. US copper consumption declined 4.3% in 1991, while Japan's consumption increased 2.3%. The above changes in copper consumption closely reflect the performance of industrial production in these countries. For example, US industrial production declined 2% in 1991, while Japan's and Germany's industrial production increased 2.3% and 3.2%, respectively. The industrializing developing countries in East Asia again provided the most sparkle in the copper market (Taiwan, China led the region with a 50% increase in 1991). The latest consumption figures are broadly consistent with the post-1985 trend that suggests a revival of copper demand growth in relation to economic activity. Because of its superior electrical conductivity, copper's main use has been in electrical wires. Its other important uses have been in plumbing tubes, in various hardware in the form of alloys with other metals, and in automobile radiators. As electricity has found uses in a diverse range of processes and products, so has copper. According to 1990 US data, about 72% of refined copper consumption was related to electrical uses, 15% was in construction, 5% in machinery, 4% in transportation, 1% in ordinance, and 3% in other uses. The construction industry is an important end- user of electrical wires; copper used in construction is estimated at 40% of total US consumption. Over the years, copper's market share in electrical uses in the United States has been increasing, despite significant market penetration by aluminum in long-distance electrical cables and by optical fibers in telecommunications. The predominance of electrical uses of copper can also be seen in other industrial and developing countries. In LMICs, the shares of the construction and transportation sectors have been low because copper tubes in household plumbing are too costly in relation to income and because the 1These include the FSU, Eastem Europe, China, Cuba, and the Democratic People's Republic of Korea. 57 automobile manufacturing industry has been small. The situation has started to change, however. In industrializing developing countries, copper demand typically increases rapidly as a result of rapid increases in exports of copper-containing parts and finished products. Then as income levels increase, more copper is demanded for the domestic market in consumer durables and construction. Both Taiwan, China and the Republic of Korea are good examples of this pattern. Effect of energy/capitol complementarily on copper demand. Although copper is an important industrial raw material, its share in total manufacturing value-added is small. Thus, the elasticity of copper demand with respect to its own price has been estimated to be relatively small, usually not exceeding 0.1-0.2 in the long run, and highly unstable. The conventional copper demand models that explain copper consumption in terms of industrial production and prices of copper and its substitutes thus could be leaving out other important factors. In particular, the energy price increases in the 1970s and early 1980s seem to have had a significant impact on copper consumption through energy-capital complementarily in the short run. In many cases, energy saving also meant metals saving. For example, high electricity prices reduced electricity demand, which in turn reduced the demand for power cables and building wires. The converse of this is the revival of copper demand following the post-1986 decline in energy prices. Increases in copper demand in the second half of the 1980s can be attributed in large part to increased manufacturing output and capital investment, which can be partly explained by energy- capital complementarily. The increases can also be attributed to wider use of electronic devices in most types of products, and because few new opportunities for replacement of copper with other materials have been found in recent years. Substantial long-term demand growth expected. This report forecasts the long-term demand for copper using a copper demand model that has as explanatory variables industrial production, relative prices of copper to energy, aluminum, and labor, and trend changes in technology. Based on the demand model and the base-case assumptions about industrial production growth, exchange rates, and inflation in the major copper-consuming countries, world demand for refined copper is projected to increase at 2% p.a. over the 1991-2005 period. The current economic difficulties in the FSU and:Eastern Europe, and the long time it is likely to take for these countries to resume stronger economic growth have led to reductions in the forecasts of copper demand growth rates for these countries from those previously anticipated. Nevertheless, the world demand growth projected for the 1990s and beyond is substantial and would require large capacity expansions of both mines and smelters. To show the sensitivity of the demand forecasts to assumptions made about the growth of industrial production and relative prices, Table 1 presents simulation results under different assumptions, that is, the levels of industrial production and prices of copper and other production inputs are 10% higher than the base case for the entire forecast period and for all countries. It is apparent from Table 1 that a higher industrial production level would have the most profound impact on copper demand. Slight changes in the world economic growth scenario, say a 4% p.a. growth rate instead of 3%, could mean an additional copper demand of 80-100,000 tons. The effects of changes in oil prices and prices of copper and its substitutes are relatively minor compared with that of industrial activity. 58 Copper intensity ofproduction forecast to fall in developing countries. Over the forecast period, the base-case forecast of refined copper consumption implies a 19.4% decline in the copper intensity of industrial production (copper consumption per unit of industrial output) for a sample of five large industrial economies (France, Germany, Italy, Japan, and the United States) and an 8.8% decline for a sample of four developing economies (Brazil, India, Republic of Korea, and Mexico). These reductions in copper intensity represent a change from the historical pattern-over the 1973-90 period, for example, copper intensity declined by 22.8% in the five industrial countries and increased by 35.9% in the four developing countries. We believe that the rate of decline in copper intensity in the industrial economies should slow slightly due to lower energy prices and the prospect of large and increasing demand for investment in infrastructure and equipment. For the four developing countries, the historical copper intensity is heavily influenced by a rapid increase in copper use followed by a sharp slowdown in recent years in the Republic of Korea. This pattern resulted from the rapid expansion of certain copper-intensive industrial sectors which after reaching their potential suddenly reduced their copper consumption growth rate. A similar pattern is probably being repeated in Taiwan, China. In the years ahead, it is likely that copper consumption growth in industrializing Asia will slow to substantially below the growth rate of industrial production. For the LMICs as a whole, copper consumption increased only at 3.1% p.a. during 1970-90, while GDP grew at 4.5%. For the period 1991-2005, the growth rate of copper consumption for this group is projected to slow only slightly to 2.8%. Figure 1 plots past and projected copper intensities for the representative countries in the two income groups. Slow recovery of FSU and Eastern European demand. Over the next few years, the FSU and Eastern Europe should continue to experience the low levels of copper demand seen in 1991 as they transform to market-oriented economies. The FSU has been the third largest consumer of copper in the world after the United States and Japan. Declines in copper consumption in recent years led to sharp increases in exports; exports in 1990 amounted to 176,800 tons, mostly in low-grade cathodes. The most serious problem these countries face is outdated industrial machinery and infrastructure that require retrofitting and modernization in order to compete with the rest of the world. This means huge capital investments, mostly from foreign sources. Once this transformation process begins, copper demand should rise. But a return to the levels of a few years ago is likely to take a long time because these levels were inflated by inefficiencies in raw materials use and large consumption for military purposes. Table 1: Sensitivity of Copper Demand to Economic Assumptions a/ 1992 1995 2000 -( %) Industrial Production 7.2 8.4 8.6 Copper Price -0.3 -1.1 -1.3 Oil Price -0.4 -1.3 -1.5 Aluminum Price 0.1 0.4 0.5 a/ Percentage change in copper demand as a result of a 10% increase in industrial production, copper price, oil price, or aluminum price over that assumed in the base case. Source: World Bank, International Economics Department. 59 Figure 1: Copper Intensity Trends (Per Unit of Industrial Production) 2.8 2.6 S2.4- 2.2 2- 1.6 .1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 5 hdust. Comtria 4 Devd Comiis Technological innovations should continue to play a major role in shaping the copper industry although in an unpredictable manner. The latest development in the telecommunications market is the discovery by AT&T that copper wires can pack as many lines of communication as optical fibers. Thus, it now appears that the telecommunications market for copper is at least not completely lost to optical fibers. However, the race to find a room-temperature, superconducting material seems to have stalled and its final outcome in relation to copper use is highly uncertain. At least for the moment copper does not face serious challenges from new technologies. Supply Outlook MINE PRODUCTION. Production expansion during 1987-91 despite big supply disruptions. Between 1987 and 1991, mine production of copper in the West (in the form of copper concentrates and leach output) increased by about 822,000 tons (a 2.9% p.a. increase). The net increase in mine production would have been much larger, by more than 500,000 tons, had the Bougainville mine in Papua New Guinea not shut down and Zambia and Zaire maintained production at their 1987 levels. Chile and the United States contributed most to the increase, each with a net increase of more than 400,000 tons, while Australia, Indonesia and Portugal were also large contributors. Several large projects that came on-stream during this period account for most of the increases; for example, the new Escondida mine, the Chuquicamata sulphide expansion and a new SxEw plant in Chile; the Bingham Canyon expansion and the Morenci SxEw plant in the United States; the new Neves Corbo mine in Portugal; the Ok Tedi mine in Papua New Guinea; and the Ertsberg mine expansion in Indonesia. Production disruptions of various kinds were a major factor in shaping the market during the 1987-91 period. In addition to the problems in Papua New Guinea, Zambia, and Zaire, there were production losses due to strikes (in Canada, Chile, Mexico, and Peru to name only the most disruptive ones) civil unrest (most seriously in Peru), accidents (for example, rock-bursts in Chile's El Teniente mine), and adverse weather conditions (drought in Papua New Guinea that inhibited barge transportation). 60 Since the collapse of communism, the copper mining industry in the FSU and Eastern Europe has been undergoing radical changes in terms of ownership, management, and pricing. The impact of the changes has been felt in world markets in the form of greater availability of low-grade copper from these countries. In the FSU, copper ore is found mainly in the Urals, Caucasus, Kola Peninsula, Kazakhstan, Georgia, Northern and Central Siberia, and Uzbekistan; it is likely that the FSU has abundant undiscovered resources of copper including some high-grade ore bodies and valuable by- products. Poland and former Yugoslavia are significant producers of copper. Poland's underground copper mines have relatively rich ores with significant by-products; proven reserves are large enough to support production for a long time. The FSU and Poland both experienced declines by 110,000 tons in copper mine production (or by 11% and 25%, respectively) during the 1987-90 transitional period. Both suffered from lack of supplies, environmental problems, and poor maintenance. Significant recent discoveries of copper deposits, together with other metals, are likely to maintain Yugoslavia's production at its historical levels. Long-term mine capacity changes uncertain. Table 2 shows the mine capacity forecasts to the year 2005. Copper mine projects already being implemented or in the planning stage give fairly good indications of likely capacity changes to 1995, but capacity forecasts for the years 2000 and 2005 should be considered highly conjectural. A number of large and small projects in the pipeline should come on-stream by 1995, with an expected net addition of about 600,000 tons to western world capacity. The large ones among them include new projects at Chuqui Mansa Mina, La Candelaria, and Quebrada Blanca SxEw, and expansions at Escondida, Los Blonces (all in Chile), Sar Cheshmeh (Iran), and Mt. Isa (Australia). Between 1995 and 2000, the net increase in mine capacity is expected to be relatively small because only a few large projects have a reasonable chance of being realized, for example, Chuqui Norte SxEw, and the Bingham Canyon and Escondida expansions. (We assume that the Bougainville mine will restart production at least partially by the year 2000). However, the forecasts here may turn out to be too conservative, since the high copper prices of recent years have attracted a great deal of interest in new copper mining projects--the list of probable and potential new mine projects at different stages of investigation around the world keeps growing. The total mine capacity of these projects presently amounts to more than 2 million tons of copper. These projects are only partially reflected in the forecasts given in Table 2 but more of them could come on-stream by 2005 if copper prices are favorable and the growth in copper demand remains strong. At the same time, the industry faces much more stringent environmental constraints in both industrial and developing countries than before, and the difficulties experienced during the 1980s have taught it to adopt lofty economic feasibility criteria. So if copper prices decline to the relatively low levels forecast here for the mid-1990s, a good number of these projects are likely to be put on hold beyond the year 2000. Over the long term, however, most of them should be realized if demand growth is to be met and the depletion of existing mines is to be compensated. REFINED COPPER PRODUCTION. In the 1980s industrial country producers cut costs and increased output. Between 1987 and 1991, the western economies' refined copper production increased by 924,000 tons, which slightly exceeded increases in refined copper consumption over the same period. Some 86% of the increase was accounted for by the United States, Chile, and Australia. Most developing countries experienced production stagnation or decline. Between 1987 and 1990, the FSU and Eastern Europe suffered substantial losses in refined copper production; however, Poland's production recovered strongly in 1991. Refined copper production increased slightly more than mine production largely due to an increase in secondary copper production (scrap recovery). 61 Table 2: Western World Copper Mine Capacities, 1990-2005 --Production- Mine Capacity- - 1990 1991 1992 1995 2000 2005 -----('000 tons Cu in Concentrates and Leach Output)-- - Africa Zambia 496 412 360 340 300 250 Zaire 356 292 120 250 320 400 Other Africa 280 285 280 280 200 200 Asia & Pacific Australia 327 311 315 350 380 420 Papua New Guinea 170 205 190 172 250 250 Philippines 182 144 150 175 160 150 Indonesia 170 200 290 300 400 550 Other Asia 212 238 235 315 305 305 America United States 1,587 1,635 1,720 1,750 1,670 1,800 Canada 802 777 770 700 650 680 Chile 1,588 1,814 1,855 2,260 2,570 2,850 Peru 317 381 380 390 370 410 Mexico 291 267 310 363 365 370 Brazil 37 37 37 15 125 150 Europe Portugal 160 165 170 145 105 80 Yugoslavia 119 113 60 70 70 70 Other Europe 122 120 110 90 70 50 Total 7,216 7,396 7,352 7,965 8,310 8,985 Source: World Bank, International Economics Department. The average direct cash cose of producing refined copper in the western economies is estimated at about C62/lb, while the total cost is estimated at C86/lb. Most of the world's major copper- producing countries have direct cash costs of less than 080/lb. At today's copper prices, almost all copper producers can operate profitably, and those that have brought costs down in recent years (see below) are making good profits. In terms of direct cash costs, Finland, Peru, Philippines, Sweden, and Zaire are relatively high-cost producers among the significant copper-producing countries; low-cost producers include Chile, Indonesia, Portugal, Zambia, and the United States. The ranking on a total cost basis is not much different, except that developing countries tend to shift to the higher-cost group because of higher investment costs and high debt service. 2 Direct cash cost is defined as the sum of mining, milling, smelting and refining costs, plus cash outlays for freight and marketing. Total cost is the sum of direct cash cost, depreciation and interest charges, royalties, taxes and duties other than income or profit taxes, all indirect cash and non-cash costs for administration and overhead, research and exploration, and extraordinary items such as strike costs and pension shortfalls. 62 Throughout the 1980s, copper producers in industrial countries achieved remarkable cost savings in response to adverse market conditions. In sharp contrast, however, producers in developing countries, many of them state-owned companies, failed to make similar adjustments to enhance productive efficiency. In order to maintain competitiveness, developing country producers have relied heavily on exchange rate adjustments. Table 3 shows estimates of direct cash costs for major copper-producing countries in both nominal and real terms.3 Although most developing countries were cost-competitive in 1990 in nominal dollar terms compared with the United States, it is clear that many of them would have become high-cost producers had their exchange rates been adjusted to compensate only for inflation (though depreciation in excess of inflation was probably justified in some of the countries because their exchange rates were already overvalued). In a matter of less than a decade, US copper producers have been transformed from high- to low-cost suppliers, while increasing production. If the currencies of developing countries appreciate, as in the case of Chile in recent years, their competitive position in world markets could be seriously jeopardized. The contrast between developing and industrial country copper producers has indeed been a major feature of the industry's adjustments in the 1980s. US copper producers led the industrial country group. A variety of methods were used by US producers to cut costs, for example, labor costs were reduced by cuts in wages and benefits, by downsizing, and by work force training; open-pit mine costs were reduced by introducing in-pit crushers and conveyors to replace costly hauling by trucks; metal recovery was increased and costs reduced by leaching of waste dumps, heap leaching of low-grade ore, in-situ leaching, and SxEw production; and concentration mills were improved with column flotation and semi-autogenous grinding. Developing country producen failed to improve efficiency. The record of efficiency improvements in the 1980s among the developing-country producers falls far short of that of industrial countries, and the experience at Codelco, Chile's state-owned mining company and the world's largest copper producer, illustrates some of the basic problems common to most state-owned mining entities in developing countries. Faced with a sharp deterioration in ore grade, Codelco embarked on ambitious investment programs in 1984 and again in 1988 to increase production and reduce costs. The programs contained many of the cost-saving measures adopted in industrial countries, such as mobile in-pit crushers and conveyors and SxEw production. But implementation of the programs was hampered by government diversion of funding to other purposes and frequent management changes. As a result, the programs fell short of attaining the desired objectives. Furthermore, other basic issues such as increasing labor efficiency and work force downsizing have not been seriously addressed. Codelco is widely considered to be relatively well managed as a state-owned company. At the other extreme, the performance of ZCCM in Zambia and Gecamines in Zaire, both state-owned copper producers, has been abysmal. They failed not only to improve productivity, but also to maintain existing production capacity, resulting in sharp declines in output. In these countries, almost all of the profits from the copper operation were diverted to other uses, and the copper mines and related infrastructure were allowed to deteriorate. Records of other state-owned companies (in Peru and Mexico) fall somewhere in between. Recently, Mexico privatized its two state-owned copper mining companies. Chile has passed a law that allows Codelco to form joint ventures with the private sector, domestic and foreign, in developing new mining properties owned by Codelco. Indonesia and Papua New Guinea have allowed foreign participation and 3 Real cash costs are defined as the constant local-currency operating costs of mining, milling, and smelting/refining per unit of refined copper, using local currency deflators and converting into US dollars at January 1990 exchange rates. 63 their copper mining operations have been run efficiently. This partly explains why Indonesia had the lowest cash costs among major copper-producing countries in 1990 (see Table 3). Recently, concerns have been raised about the adequacy of smelter and refinery capacities over the near term. These concerns reflect the current high capacity utilization rates (the average capacity utilization rate of Japanese smelters in 1991 was 94.3%) and the consequent rise in treatment and refining charges. The western economies' refined copper production capacity is expected to increase by 1.3 million tons of copper between 1991 and 1995. Of this total, approximately 370,000 tons should be coming from SxEw/hydromet production that does not require the smelting and refining process. Primary smelter and refinery capacities are expected to increase by about 950,000 tons, sufficient to process the expected increase in mine production shown in Table 2. If these anticipated increases materialize, concern about near-term smelter/refinery capacity would be unwarranted. We assume that investments in smelter and refinery capacities will be adequate over the longer term (to the year 2005) on the grounds that mining, smelting and refining are normally a part of the same investment decision. The dominant smelter technology today is flash smelting and its variants which have higher productivity and better environmental characteristics than the reverberatory furnace technology of the 1960s. However, reverberatory furnaces remain the dominant technology in Latin America, Africa, and Australia. Reverberatory furnaces require costly scrubbing of furnace off-gas to control sulfur dioxide emission to acceptable levels. Such controls have not yet been strongly implemented or enforced in developing countries. Sulfuric acid, mostly used for fertilizer production, is a valuable by-product of environmental controls to reduce sulfur dioxide emissions, and the cost of emissions control can be significantly reduced if a ready market for the by-product exists. This is a serious problem for smelters located in remote areas of developing countries. Table 3: Direct Cash Costs of Copper Production in Selected Countries, 1982-90 Chile Mexico Peru Zaire Zambia Indonesia Philippines Canada US (nominal C/lb) 1982 41.3 33.6 57.2 49.3 62.5 48.5 60.4 63.7 75.0 1984 34.4 38.7 54.4 26.4 42.2 43.5 54.1 50.4 66.2 1986 33.0 34.1 51.7 42.5 31.9 35.3 48.1 49.0 57.2 1988 41.4 48.5 51.7 54.1 59.9 40.9 64.6 58.8 55.3 1990 51.6 51.1 81.5 82.2 47.6 44.7 80.0 73.6 60.6 - (constant C/lb) a/ 1982 41.3 33.6 57.2 49.3 62.5 48.5 60.4 63.7 75.0 1984 36.9 32.5 63.9 61.6 63.3 51.1 41.9 49.2 63.8 1986 40.3 36.1 52.1 90.7 49.7 48.8 39.1 49.6 57.1 1988 50.9 40.0 33.8 103.4 51.1 59.3 43.9 49.1 51.7 1990 56.3 35.5 29.3 163.3 36.1 59.3 51.4 57.5 52.1 a/ In constant 1982 local currencies at 1982 exchange rates, using wholesale prices as deflators. Source: Brook Hunt Associates and World Bank, International Economics Department. 64 The current situation and prospects for the major copper-producing countries are summarized below. UNrTED STATES. US mine production continued to expand in 1991 due to expansions at a number of mines and SxEw facilities. Over the 1987-91 period, US production achieved a remarkable 400,000 ton net increase without a single greenfield project. In the next five years (1991-96), the net increment is expected to be only 150,000 tons, largely through further expansions at Bingham Canyon, Mission, Morenci, and Ray mines. The long-term outlook for US production is not promising because of the apparent difficulties in developing new large-scale mines. CHLE. Although Chile's mine production increased sharply in 1991 due to the large Escondida mine coming on-stream, other copper mines continued to have production problems despite Codelco's large investments in increasing production capacity to compensate for declining ore grade. El Teniente production continued to slide and the Chuquicamata open-pit mine suffered a decline in production for the first time in 1991, a decline which is considered irreversible. Because of these difficulties, Codelco has started to look for new opportunities at other Codelco-owned mining properties, such as Chuquicamata Norte SxEw and Mansa Mina. Recently, the government passed a law that allows Codelco to form joint ventures with private entities, including multinational mining companies, m developing new mines on state-owned properties. This measure is expected to attract a great deal of interest from foreign companies and probably change the nature of state-owned mining in Chile. In fact, the private sector has provided the largest increments to Chile's copper production in recent years, and several large projects financed to some extent by foreign companies are expected to be the main source of incremental production in Chile for the remainder of 1990s. PERU. Peru's mine production in 1991 benefited from a relative easing of production disruptions. However, Peru's economic and political problems remain unresolved and new investments in capacity expansion have not been forthcoming, thus preventing Peru from realizing its substantial potential as a copper producer. ZARE AND ZAMBIA. The economic and political problems in these countries further worsened and sharply reduced copper production in 1991, particularly in Zaire. As the paralysis in Zaire continues, the prospect of returning to normal production slips further away as it becomes more difficult to rehabilitate ill-maintained mines. The new government in Zambia announced its intention to improve ZCCM operations by allowing foreign participation or by privatizing ZCCM, but it has a long way to go to create conditions favorable for the large foreign investments Zambia needs. These countries are endowed with high-grade ore bodies that have kept costs low despite production inefficiencies. The decline in ore grade has been relatively minor during the past ten years. Other Countries. CmNA plans to allow greater freedom for foreign participation in the development of its base-metal mining industry. The copper ores discovered so far have been mostly low grade, requiring advanced technology such as in-situ SxEw. Expansion of the Dexing mine, as well as development of two other medium-sized mines, is progressing on schedule. The Grasberg deposit in INDoNEsIA's Erstberg complex came on-stream in 1990 and, as a result, Indonesia became a major low- cost copper producer. Proven reserves at the site continue to increase. 65 Price Outlook Despite the economic slowdown in major industrial economies over the past year or so, copper prices have remained firm. Among the base metals, only the copper price has not experienced a major decline from the peak reached in 1989. This unique strength is largely attributable to supply-side factors. Of late, sharp reductions in supplies from Zaire and Zambia, due to political and economic problems, have been the most important contributing factor to this price strength. At the same time, smelter capacity constraints and consequently higher treatment and refining charges have meant relatively high prices for refined copper compared with prices of copper concentrates. For the exporters of copper concentrates, the net back from concentrate exports was substantially reduced, by as much as 015/lb, between 1989 and 1991. Another important factor in keeping copper prices firm has been sporadic but substantial import demand from China. Also, world commercial stock levels have increased during this period, although not excessively. Figure 2 shows commercial stocks in terms of weeks' consumption since 1987. The current stock level, roughly at five weeks' consumption, is considered normal. The long-term price forecasts for copper have been made on the basis of our econometric model, given assumptions about industrial production growth, exchange rates, and interest rates. Unanticipated supply disruptions have not been included. The forecasts of constant dollar prices are virtually identical to those made two years ago. Uncedainties on demand side in medium term. Table 4 shows the base-case forecasts of supply and demand balance and the implied stock changes for refined copper. Forecasts over the short to medium term indicate the likelihood of moderate excess supplies and a buildup of stocks. The excess supplies arise because of capacity expansions due to come on-stream in the near term, at a time when demand growth is expected to be relatively weak. However, the size of this imbalance is not expected to be overwhelming and therefore its impact on copper prices is likely to be mild. In the case of stronger-than-expected economic activity, the excess supplies could be easily wiped out and copper prices Figure 2: Commercial Stocks of Copper (As of End of Period) 5.5 5-I 4.5- C 0 4- 3.5 1987 1988 1989 1990 1991 Feb 1992 66 Table 4: World Supply/Demand Balance and Stock Changes Actual Estimate Projected 1990 1991 1992 1993 1994 1995 2000 (milion tons of copper content) Mine Production 9.03 9.08 9.08 9.26 9.46 10.03 10.41 Smelter Losses 0.34 0.34 0.34 0.35 0.36 0.38 0.39 Primary Smeltes Production 8.63 8.74 8.74 8.91 9.10 9.65 10.02 Scrap 1.31 1.28 1.33 1.35 1.37 1.39 1.45 Electro-refined 9.94 10.02 10.07 10.26 10.47 11.04 11.47 SxEw 0.80 0.81 0.81 0.88 1.06 1.19 1.50 Refined Production 10.74 10.83 10.88 11.14 11.53 12.23 12.97 Demand 10.79 10.73 10.79 11.05 11.29 11.54 12.89 Stock Changes 0.06 0.25 0.10 0.09 0.25 0.69 0.08 Source: World Bank, International Economics Department. could move upward. Given that the strength of economic recovery in the industrial countries remains highly uncertain at this moment, it seems fair to say that, over the short to medium term, the main source of uncertainty for copper prices lies on the demand side rather than on the supply side. Supply factors more important in long term. Over the longer term, the market outlook hinges to a large degree on the response of the copper mining industry to the recent relatively high level of prices. One important feature of the 1987-92 period of relatively high prices is the fact that few large- scale greenfield projects have been initiated. Such projects as have been touted in the industrial countries have met environmental constraints. Political and economic disarray in some developing countries have precluded them from potential new investments for copper mining-e.g., most of Africa, Peru, and Papua New Guinea. Thus, new large-scale greenfield projects are considered only for a few developing countries such as Chile and Indonesia. However, important changes in the investment climate have started to take place in some developing countries and in the former CPEs. An increasing number of developing countries have taken steps to reduce state intervention in mining and to phase down the role of state-owned mining companies in favor of private ownership and foreign participation. These changes lend support to the view that over the long term, supply should become more responsive to price signals than in the past and adequate supplies are likely to be forthcoming in response to higher prices. Given the market balance projections for the medium term, the market trough is likely to occur at around mid-1990; the 1995-96 period is likely to see relatively low prices, probably in the 085-90/lb range. This means another period of relatively higher prices towards the late 1990s, followed by a downturn in copper prices towards the year 2005. This cyclical pattern, which would also be implied by extrapolation of the historical pattern, is highly conjectural. The pattern should also depend on the business cycle, which has not been explicitly built into the long-term forecasts. For the long term, it is reasonable to expect the price level to average out to the long-term marginal cost. The average forecast price for the 2000-2005 period is within the range C83-90/lb (in 67 constant 1990 dollars), at which most of the new greenfield projects currently being studied are likely to be economically viable. The cash costs of new sulphide projects and extensions of existing mines expected to come on-stream in the 1990s are estimated in the range of C46-50/1b; and on a total cost basis the range is 080-85/lb. In the recent past, production costs of new mines tended to be lower than those of existing mines, thus reducing the overall costs, for example Chile's Escondida, Indonesia's Grasberg, and Portugal's Neves Corbo mines, and various SxEw projects. However, a number of new greenfield projects envisaged for the 1990s and beyond have somewhat less favorable cost characteristics in terms of ore grade than those of the recent past. In addition, heightened environmental concerns could add to the costs of new and existing mines and smelters. These potential cost increases, mostly among producers in developing countries, could be substantially moderated by efficiency improvements. The long-term price forecasts assume that costs increase only marginally in the 1990s. 68 Table Al: Copper Ore - Production By Main Countries and Economic Regions ------------------------------------------------------------------------------------------------------------------------------- Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 ------------------------------------------------------------------------------------------------------------------------------- --- -----------------------------------('000 Tons) ---------------------------------- --------- p.a.)------ High-Income 2,494 2,549 2,869 2,868 2,931 2,977 2,948 2,904 2,787 2,983 1.0 -0.4 0.3 OECD 2,462 2,546 2,855 2,854 2,918 2,964 2,935 2,891 2,774 2,970 1.0 -0.4 0.3 United States 1,447 1,388 1,587 1.634 1,720 1,770 1.750 1,750 1,670 1,800 0.5 -0.7 0.7 Canada 595 681 802 777 770 745 720 700 650 680 2.0 0.3 -0.9 Australia 155 237 327 311 315 350 360 350 380 420 4.1 2.1 2.2 LMICs 3,762 5,383 6,179 6,346 6,190 6,365 6,700 6,980 7,585 8,280 3.3 2.4 1.9 Africa 1,277 1,340 1,116 987 760 785 840 870 820 810 0.7 -0.7 -1.4 Zambia 685 591 496 412 360 350 350 340 300 250 -0.9 -2.0 -3.5 Zaire 386 455 356 292 120 150 200 250 320 400 1.9 0.2 2.3 Americas 984 1,619 2,237 2,503 2,590 2,660 2,805 3,023 3,405 3,780 4.0 4.6 3.0 Chile 694 1,071 1,588 1,814 1,855 1,900 2,050 2,260 2,570 2,850 4.0 4.4 3.3 Mexico 63 171 291 267 310 330 345 363 365 370 6.9 8.5 2.4 Peru 204 364 318 381 380 385 385 390 370 410 3.2 3.3 0.5 Asia & Pacific 295 800 1,224 1,267 1,390 1,470 1,585 1,627 1,840 2,040 8.5 5.5 3.5 China, People's Rep. 107 165 360 370 370 430 500 510 570 630 5.3 6.0 3.9 Indonesia - 59 170 212 290 290 300 300 400 550 - - 7.0 Philippines 163 302 182 147 150 160 170 175 160 150 5.3 0.1 0.1 Papua New Guinea - 161 170 205 190 180 175 172 250 250 - - 1.4 Europe 1,202 1,615 1,614 1,589 1,450 1,450 1,470 1,460 1,520 1,650 3.1 1.2 0.3 World 6,256 7,932 9,048 9,214 9,121 9,342 9,648 9,884 10,372 11,263 2.5 1.4 1.4 ------------------------------------------------------------------------------------------------------------------------------- a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: World Bureau of Metal Statistics, Metal Statistics (actual); World Bank, International Economics Department (projected). Table A2: Copper Ore - Gross Exports By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 - ------------------------------------(000 Tons)----- ----------------------------- -------(% p.a.)------ High-Income 261 550 720 669 685 682 669 660 627 660 6.5 3.9 -0.1 OECD 240 550 720 669 685 682 669 660 627 660 7.5 4.3 -0.1 Canada 170 294 349 320 323 313 302 294 273 286 6.2 1.9 -0.8 United States 21 101 258 253 267 274 271 271 259 279 17.9 18.1 0.7 LMICs 297 858 971 1,228 1,405 1,353 1,434 1,461 1,630 1.775 8.1 4.4 2.7 Africa 19 75 42 40 38 38 40 40 43 43 6.5 3.8 0.5 Zaire - 33 17 13 10 10 12 12 15 15 1.0 Americas 108 225 354 603 645 636 722 859 920 1,015 7.6 7.3 3.8 Chile 52 112 257 508 540 510 580 700 750 830 9.5 7.7 3.6 Peru 39 25 26 29 30 31 32 34 40 50 1.3 0.0 4.0 Asia & Pacific 149 539 380 399 547 507 502 409 537 627 8.8 2.2 3.3 Indonesia - 54 83 115 280 280 290 210 260 350 Philippines 149 296 103 98 50 20 10 0 0 0 2.9 -4.1 - Papua New Guinea - 161 156 160 190 180 175 172 250 250 3.2 Europe 17 11 195 186 175 172 170 153 130 90 -5.1 Portugal - 1 177 176 165 162 160 143 120 80 -5.5 World 558 1,408 1,691 1,897 2,090 2,035 2,103 2,121 2,257 2,435 7.4 4.2 1.8 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: World Bureau of Metal Statistics, Metal Statistics (actual); World Bank, International Economics Department (projected). Table A3: Copper Ore - Gross laports By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 --- ------------------------------------('000 Tons) ---------------------------------- ---------(% p.a.)-------- High-Income 544 1,169 1,378 1,395 1,417 1,471 1,548 1,686 1,812 1,986 7.0 2.7 2.6 OECD 544 1,135 1,375 1,394 1,415 1,469 1,546 1,684 1,810 1,984 6.9 2.5 2.6 United States 31 39 154 62 62 61 88 171 290 273 -1.6 -3.5 11.2 Japan 387 843 880 930 928 928 928 928 885 851 7.4 2.1 -0.6 Germany 74 141 122 131 125 130 130 135 145 160 6.1 1.0 1.4 LMICs 12 141 184 203 219 214 222 341 399 485 16.2 16.4 6.4 Asia & Pacific 4 94 150 173 189 184 191 308 360 440 24.3 6.9 Korea, Rep. of 3 72 92 107 104 104 104 104 109 109 22.8 0.2 Europe 5 44 17 12 12 12 13 13 15 17 7.8 3.9 2.5 World 556 1,310 1,562 1,598 1,636 1,685 1,770 2,027 2,211 2,471 7.8 3.8 3.2 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: World Bureau of Metal Statistics, Metal Statistics (actual); World Bank, International Economics Department (projected). Table A4: Copper Blister - Production By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 --- ----------------------------- ------('000 Tons)------ --------%---------------------------(% p.a.)-------- High-Income 3,106 3,401 4,057 4,082 4,126 4,216 4,231 4,316 4,356 4,456 1.9 0.7 0.6 OECD 3,103 3,373 4,045 4,067 4,110 4,200 4,215 4,300 4,340 4,440 1.9 0.6 0.6 United States 1,497 1,261 1,463 1,450 1,470 1,510 1,520 1,600 1,650 1,740 0.0 -1.0 1.3 Canada 439 445 525 546 530 530 530 530 540 550 1.1 0.3 0.1 Japan 589 930 1,041 1,085 1,090 1,090 1,090 1,090 1,040 1,000 4.8 1.6 -0.6 Australia 132 173 202 207 240 290 295 300 330 370 3.5 1.4 4.2 LMICs 3,458 4,668 5,427 5,197 5,124 5,324 5,464 5,869 6,370 7,030 3.0 2.4 2.2 Africa 1,270 1,269 1,033 901 664 694 644 689 650 640 0.6 -0.9 -2.4 Zambia 677 592 461 409 390 370 270 275 260 220 -0.9 -2.1 -4.3 Zaire 385 421 338 268 50 100 150 190 190 240 1.6 0.2 -0.8 Americas 878 1,363 1,912 1,888 1,970 2,060 2,130 2,230 2,560 2,880 3.6 4.3 3.1 Chile 640 952 1,329 1,296 1,300 1,380 1,450 1,550 1,790 2,010 3.3 3.9 3.2 Peru 171 333 261 265 300 310 310 310 320 350 3.1 3.4 2.0 Asia & Pacific 126 299 805 938 980 1,090 1,210 1,450 1,580 1,760 8.9 10.6 4.6 China, People's Rep. 98 178 425 430 460 520 580 580 650 750 6.2 7.6 4.1 Philippines - - 154 168 160 200 200 200 200 230 2.3 Korea, Rep. of 5 74 161 186 180 180 180 180 190 190 20.3 19.3 0.2 Europe 1,183 1,737 1,582 1,470 1,510 1,480 1,480 1,500 1,580 1,750 3.4 1.4 1.3 World 6,564 8,069 9,483 9,279 9,250 9,540 9,695 10,185 10,726 11,486 2.5 1.6 1.5 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: World Bureau of Metal Statistics, Metal Statistics (actual); World Bank, International Economics Department (projected). Table A5: Copper Blister - Gross Exports By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 --------------------------------------------------------------------------------------------------------------------------------------- ---- -----------------------------------('000 Tons)----- ---------------------------- --------(% p.a.)-------- High-Income 44 90 65 76 70 70 72 72 75 80 6.1 1.5 0.4 OECD 44 90 65 76 70 70 72 72 75 80 6.2 1.6 0.4 Finland - 12 25 25 25 25 25 25 25 25 0.0 LMICs 735 663 567 460 382 417 448 465 459 489 -1.1 -1.1 0.4 Africa 409 371 271 172 94 124 154 164 144 154 -0.5 -1.5 -0.8 Namibia 28 42 30 30 30 30 30 30 - - 1.0 - Zaire 193 272 216 120 40 70 100 110 120 130 2.3 0.4 0.6 Americas 320 291 296 279 280 285 287 295 310 330 -1.5 -0.6 1.2 ChiLe 183 163 151 92 95 100 100 100 105 110 -1.8 -0.3 1.3 Peru 130 123 93 101 95 95 95 100 100 100 -1.5 -2.1 -0.1 -. Europe 5 1 4 9 8 8 7 6 5 5 -4.1 World 779 753 631 536 452 487 520 537 534 569 -0.6 -0.8 0.4 --------------------------------------------------------------------------------------------------------------------------------------- a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: World Bureau of Metal Statistics, Metal Statistics (actuaL); World Bank, International Economics Department (projected). Table A6: Copper Blister - Gross Imports By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 --- -----------------------------------('000 Tons) ---------------------------------- --------(% p.a)------- High-Income 736 533 455 447 410 437 465 475 475 480 -1.8 -2.4 0.5 OECD 735 533 455 447 410 437 465 475 475 480 -1.8 -2.3 0.5 United States 187 64 84 94 95 92 90 90 90 80 -6.5 -3.6 -1.1 United Kingdom 40 67 64 32 35 35 40 40 40 40 -0.1 1.8 1.6 Japan 136 80 27 33 30 30 35 35 35 40 - -6.9 1.4 LMICs 29 95 103 56 41 47 49 57 57 78 11.0 5.4 2.4 Americas 1 4 3 1 1 1 1 2 2 5 - - 12.2 Asia & Pacific 10 49 48 55 40 45 45 50 50 65 16.7 5.6 1.2 Europe 17 42 52 - - 1 3 5 5 8 10.2 5.3 - World 764 627 557 503 451 484 514 532 532 558 -0.9 -1.4 0.7 452 487 520 537 534 569 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: World Bureau of Metal Statistics, Metal Statistics (actual); World Bank, International Economics Department (projected). Table A7: Copper Refined - Production By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 --- ------------------------------------('000 Tons)---------------------------------- --------(% p.a.)------ High-Income 4,497 4,911 5,327 5,359 5,457 5,548 5,628 5,794 5,925 6,365 1.2 0.4 1.2 OECD 4,493 4,881 5,299 5,348 5,445 5,535 5,615 5,780 5,910 6,350 1.2 0.3 1.2 United States 1,941 1,886 2,017 1,995 2,050 2,030 2,070 2,200 2,300 2,500 -0.2 -0.7 1.6 Canada 459 460 516 538 530 540 540 540 580 650 1.0 0.0 1.4 Japan 683 1,016 1,008 1,076 1,080 1,080 1,080 1,080 1,030 1,020 4.5 1.2 -0.4 Australia 149 182 274 280 310 350 360 360 400 500 3.4 1.8 4.2 Germany 498 433 533 522 520 520 520 560 560 600 0.8 0.1 1.0 United Kingdom 194 140 122 70 55 55 55 50 50 30 -2.3 -1.9 -5.9 LMICs 2,953 4,417 5,413 5,472 5,427 5,591 5,905 6,434 6,969 7,875 4.1 2.9 2.6 Africa 869 870 813 734 650 675 700 710 720 725 171.0 -0.6 -0.1 Zaire 194 133 173 140 80 120 170 210 260 300 0.9 0.7 5.6 Zambia 573 578 479 428 400 380 350 320 280 240 0.1 -1.5 -4.0 Americas 569 1,131 1,694 1,793 1,785 1,816 1,945 2,122 2,433 2,900 6.3 5.9 3.5 Chile 462 789 1,192 1,238 1,200 1,200 1,300 1,400 1,680 1,900 5.5 4.9 3.1 Mexico 54 90 152 154 190 210 240 280 310 350 5.1 4.7 6.0 Peru 34 218 182 244 240 250 248 270 270 300 9.1 10.6 1.5 Asia & Pacific 157 429 999 1,075 1,202 1,300 1,420 1,732 1,866 2,100 8.6 9.0 4.9 China, People's Rep. 127 295 490 510 550 590 650 700 780 900 6.3 5.9 4.1 Korea, Rep. of 6 89 187 201 200 200 200 210 210 210 20.7 0.3 Philippines - - 126 116 122 160 160 162 166 190 3.6 Europe 1,350 1,984 1,907 1,870 1,790 1,800 1,840 1,870 1,950 2,150 3.6 1.7 1.0 FSU 1,082 1,307 1,260 1,210 1,200 1,220 1,230 1,250 1,350 1,550 2.4 0.6 1.8 World 7,450 9,328 10,740 10,831 10,884 11,139 11,533 12,228 12,894 14,240 2.5 1.5 2.0 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: World Bureau of Metal Statistics, Metal Statistics (actual); World Bank, International Economics Department (projected). Table A8: Copper Refined - Apparent Consumption By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ -------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 --- -----------------------------------('000 Tons)----- ----------------------------- -------(% p.a.)------ High-Income 5,509 6,497 7,311 7,415 7,511 7,649 7,791 7,931 8,625 9,349 19.1 1.1 1.7 OECD 5,499 6,407 7,006 6,975 7,066 7,192 7,321 7,444 8,040 8,661 1.7 0.9 1.6 United States 1,881 2,021 2,150 2,058 2,140 2,170 2,200 2,230 2,310 2,400 0.9 0.7 1.1 Japan 811 1,248 1,577 1,613 1,600 1,640 1,680 1,710 1,940 2,180 5.0 2.6 2.2 Germany 841 934 1,028 995 990 1,005 1,020 1,040 1,120 1,200 1.5 0.5 1.3 France 336 407 478 481 480 487 496 505 545 585 1.9 1.1 1.4 Italy 261 369 475 470 480 490 500 510 560 610 3.0 2.5 1.9 Non-OECD 9 90 305 440 445 457 470 487 585 688 17.4 17.7 3.2 LMICs 1,812 3,141 3,478 3,313 3,280 3,403 3,494 3,604 4,269 4,891 4.2 3.1 2.8 Africa 44 97 96 97 97 98 99 101 111 122 4.5 3.0 1.7 Americas 196 463 372 386 383 392 402 416 497 597 5.7 3.5 3.2 Mexico 60 123 120 118 123 127 132 140 185 240 5.8 3.9 5.2 Brazil 75 215 137 145 142 147 150 155 180 210 7.4 3.2 2.7 Asia & Pacific 258 579 1,364 1,390 1,460 1,513 1,554 1,597 1,872 2,183 7.1 7.4 3.3 China, People's Rep. 180 367 512 530 560 580 590 600 680 780 5.8 4.5 2.8 India 54 73 135 145 150 155 160 166 197 230 2.7 4.8 3.4 Korea, Rep. of 8 104 324 344 360 374 389 405 495 590 25.3 20.9 3.9 Europe 1,300 1,987 1,646 1,440 1,340 1,400 1,440 1,490 1,790 1,990 3.0 1.6 2.3 World 7,321 9,637 10,789 10,728 10,790 11,052 11,286 11,536 12,895 14,241 2.5 1.7 2.0 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: World Bureau of Metal Statistics, Metal Statistics (actual); World Bank, International Economics Department (projected). Table A9: Copper Refined - Gross Exports By Main Countries and Economic Regions ------------------------------------------------------------------------------------------------------------------------------- Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 ------------------------------------------------------------------------------------------------------------------------------- --- -----------------------------------(000 Tons) ---------------------------------- --------(% p.a.)------ High-Income 1,046 1,026 1,331 1,415 1,383 1,440 1,395 1,403 1,553 1,690 0.1 -0.2 1.3 OECD 1,045 1,026 1,290 1,372 1,340 1,390 1,350 1,360 1,510 1,640 0.0 -0.4 1.3 Canada 246 263 333 375 380 390 400 380 440 500 1.5 0.4 2.1 United States 189 50 213 271 230 220 210 230 200 180 -7.4 -6.3 -2.9 Australia 38 51 150 174 180 190 190 200 270 360 7.3 3.8 5.3 LMICs 1,486 2,165 2,597 2,591 2,656 2,786 2,901 2,909 3,385 3,705 3.7 2.4 2.6 Africa 826 789 664 594 548 579 600 585 635 640 0.5 -1.0 0.5 Zambia 578 599 460 389 390 380 350 300 280 240 0.1 -1.7 -3.4 Zaire 187 125 141 128 80 120 170 200 260 300 0.7 0.5 6.3 South Africa 62 65 63 77 78 79 80 85 95 100 5.5 1.0 1.9 Americas 465 960 1,271 1,357 1,382 1,422 1,503 1,559 1,930 2,155 6.2 5.8 3.4 ChiLe 434 756 1,134 1,122 1,150 1,180 1,260 1,300 1,640 1,840 5.9 5.3 3.6 Peru 30 205 137 208 203 212 212 230 240 270 8.5 10.5 1.9 Asia & Pacific 1 3 167 135 136 185 188 185 200 230 3.9 Philippines - - 116 94 95 140 140 140 145 170 4.3 Europe 194 414 483 505 590 600 610 580 620 680 5.8 0.8 2.1 World 2,532 3,191 3,929 4,006 4,039 4,226 4,296 4,312 4,938 5,395 2.3 1.5 2.1 ------------------------------------------------------------------------------------------------------------------------------- a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: World Bureau of Metal Statistics, Metal Statistics (actuaL); World Bank, International Economics Department (projected). Table A10: Copper Refined - Gross Imports By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 --- --------------------------- --------('000 Tons)---- --------(----------------------%-------( p.a.)-------- High-Income 2,198 2,648 3,358 3,413 3,375 3,525 3,578 3,540 4,090 4,400 1.8 1.2 1.8 OECD 2,193 2,595 3,030 2,940 2,899 3,038 3,080 3,027 3,477 3,680 1.5 0.7 1.6 United States 129 343 287 295 320 360 340 260 210 80 5.6 5.3 -8.9 Japan 171 258 618 624 580 620 660 690 970 1,220 6.3 5.3 4.9 France 313 382 430 437 430 430 440 450 480 510 1.8 0.7 1.1 Germany 502 512 614 557 550 565 580 580 640 680 1.2 -0.5 1.4 Italy 254 343 412 407 400 405 415 425 455 505 2.8 1.9 1.6 Non-OECD 5 54 328 473 476 487 498 513 613 720 21.6 19.3 3.0 LMICs 363 670 671 670 663 678 714 715 884 1,037 4.2 3.9 3.2 Americas 90 231 40 30 30 30 32 35 45 60 2.9 -1.8 5.1 Asia & Pacific 111 211 373 425 432 446 479 485 619 732 8.6 9.1 4.0 India 47 52 92 60 102 107 112 118 149 182 2.7 5.1 8.2 Korea, Rep. of 3 17 136 144 160 174 189 195 285 380 22.0 7.2 China, People's Rep. 56 123 40 114 60 50 60 50 50 20 12.0 4.2 -11.7 Europe 156 223 211 165 150 150 150 140 160 180 1.6 1.3 0.6 World 2,561 3,318 4,028 4,083 4,038 4,203 4,292 4,255 4,974 5,437 2.2 1.7 2.1 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: World Bureau of Metal Statistics, Metal Statistics (actual); World Bank, International Economics Department (projected). Table All: Copper - Prices, a/ 1950-91 (Actual) and 1992-2005 (Pr~od Curret $ 1990 Consant $ G-S MUVb/ G-7 CPI o/ Actual 1950 493 3,019 3,976 1951 607 3,224 4,418 1952 715 3,623 4,990 1953 664 3,462 4,595 1954 686 3,658 4,676 1955 968 5,064 6,576 1956 906 4,576 6,024 1957 605 2,993 3,936 1958 545 2,651 3,480 1959 655 3,231 4,191 1960 677 3,271 4,250 1961 633 3,007 3,883 1962 644 2,999 3,842 1963 646 3,067 3,734 1964 968 4,517 5,454 1965 1,290 5,975 7,041 1966 1,530 6,847 8,078 1967 1,138 5,034 5,857 1968 1,241 5,543 6,214 1969 1,466 6,212 7,035 1970 1,413 5,634 6,391 1971 1,080 4,086 4,576 1972 1,071 3,718 4,113 1973 1,786 5,352 6,001 1974 2,059 5,064 6,247 1975 1,237 2,737 3,341 1976 1,401 3,057 3,662 1977 1,310 2,602 3,098 1978 1,367 2,360 2,638 1979 1,985 3,025 3,620 1980 2,182 3,032 3,558 1981 1,742 2,411 2,818 1982 1,480 2,081 2,425 1983 1,592 2,290 2,567 1984 1,379 2,027 2,230 1985 1,417 2,066 2,246 1986 1,374 1,699 1,804 1987 1,783 2,006 2,072 1988 2,602 2,730 2,810 1989 2,848 3,009 3,090 1990 2,662 2,662 2,662 1991 2,339 2,292 2,225 Projcted 1992 2,315 2,173 2,073 1993 2,380 2,153 2,058 1994 2,100 1,865 1,774 1995 1,970 1,704 1,618 2000 2,710 1,974 1,847 2005 2,900 1,840 1,652 a/ London Metal Exhange, cash, elecolt wirbar up to 1981; 1982-lumn 1986, high-grade cathodca, mininwn 99.9935% cu; m Juy 1986 onward, Grad A cathodos and wireba. bl Deflated by G-5 Manufacturing Unit Valuts (MUV) Index. c/ Deflated by G-7 Coneme Price Index (CPI). Sources: E and Motals Week (actual; World Bank, International Economics 79 Tin Summary * After reaching a peak of over US$10/kg in April 1989, tin prices turned sharply downward to below US$6/kg by mid-1990 and remained low until the beginning of 1992. This decline resulted mainly from large production increases in 1988 and 1989, followed by a sharp fall in consumption in 1990. * Prices are expected to rise in real terms through 1993 to a peak in 1995. After that the long-term outlook for real prices is slow decline (to US$6. 10/kg in constant 1990 dollar terms by the year 2005) because of the potential for large production increases at low cost, particularly from Brazil and China. Tin Prices al, 1955-2005 (cents/kg) 2, 500 Constant 1990 $ b/ 2, 000- I A I II 1/000 500Current 1955 1.. 1955 197 197 10 1935 190 1 995 2000 2005 a/ Straits C1950-84); KLTM C1985-2005), m1n. 99.85% sn b/ Deflated by G-5 MLN Index * World consumption of refined tin increased to a remarkable degree over the 1983-89 period but turned downward with the recession in 1990 and 1991. Substitution of other materials for tin may be close to its limit and tin's remaining market share appears to be relatively secure. In the long term, tin demand is expected to increase, albeit slowly. * The size of the production changes in the last few years has shown that the price elasticity of tin supplies can be substantial, even in the short term. With prices at levels substantially lower than in the International Tin Agreement days, marginal producers in Malaysia and elsewhere wound down their operations. Their share in world production will be taken over by low-cost producers in Brazil and China. 80 Demand Outlook Tin gains market strength in 1980s. Over the 1983-89 period, tin consumption increased at an average annual rate of 2.9% (Table 1), remarkable for a metal previously deemed to have been following a path of long-term decline. In the 1970s, consumption in major tin-consuming OECD countries, except for Japan, declined sharply-at 1.1% p.a. for the OECD as a whole. Consumption growth in non-OECD high-income countries and LMICs over the same period was not sufficient to offset the decline in the OECD countries. Although the six-year recovery period is not long enough to draw any firm conclusions about its long-term implications, the increases have been unmistakable in that they included most tin-consuming countries and end-use sectors. World consumption of tin declined in the latest recession (1990-92). However, the sharpest decline was concentrated in Eastern European countries where economy-wide industry restructuring has been in progress, while elsewhere, tin consumption seems to have held up very well for a recession period. The image of tin as a high-priced, raw material in decline has changed. Among the end-use sectors, tinplate production accounted for about 31% of tin consumption in major industrial countries in 1987, followed by 30% for solders, 15% for chemicals, and 24% for others (brass/bronze, babbitt, white metal and other alloys, pewter, and miscellaneous others). According to the International Tin Council (ITC) statistics for the major industrial economies (G-5 countries), consumption by the tinplate sector suffered the biggest loss over the 1970-85 period with a 4.4% p.a. decline, followed by a 3.7% decline for brass/bronze, while solder and chemical uses scored small increases. Because of the ITC's dissolution in late 1985, updates of the data to 1991 are not available. However, estimates by industry sources indicate that the tinplate and solder uses gained significantly, while chemicals and others at least held steady. Lower prices improve prospects in long term. The main factor that distinguishes the periods before and after 1986 is the collapse of the ITC in late 1985 and the consequent sharp decline of international tin prices. The contribution of lower tin prices to the increase in tin consumption in the short term probably was relatively small, because it takes time to adjust inputs to a change in relative prices. Over the long term, however, tin demand is likely to be stimulated by lower tin prices, through substitution of tin for other materials and through technological changes induced by the lower prices. For most of the past 25 years, tin consumption per unit of industrial output has declined in two major ways, namely, through substitution of aluminum and plastics for tin cans in the packaging market, and through technological innovations that progressively lead to thinner coatings of tin in tinplate making. Since tin prices were maintained at high levels for a long time, tin-saving activities seemed like a permanent feature of tin-using industries. However, given the highly competitive nature of the packaging market and other tin-using industries, we believe that lower tin prices will have a significant positive impact on long-term tin consumption. About 30% of tin consumption in high-income countries goes to tinplate making. Until recently, tinplate had been steadily losing market share in the packaging market, to aluminum in beverage cans and to plastics and paper in other packaging markets. This trend seems to have been reversed in the second half of the 1980s at least in Japan, the EC, and the United States. Much of the increase in tinplate use has been in food packaging, which is the main market for tinplate, and in general packaging of liquids such as paints, motor oils, and aerosol sprays. The increased demand for tinplate for these latter purposes mostly reflects increased industrial production and income. The return of tin cans in the beverage can market reflects the renewed competitiveness of tin cans; tin cans have become economic vis-a-vis aluminum cans because of lower tin prices and reductions in the cost of manufacturing tin cans. 81 Table 1: Tin-Consumption, 1961-91 (Actual) and 2005 (Projected) Annual Average Growth Rate Actual (3-Year Averages) Projected 1960- 1970- 1980- 1960- 1990- Country/Region 1961-63 1969-71 1979-81 89-91 2005 1970 1980 1990 1990 2005 ('000 ton ) (% p.a.) High Income 141 154 138 140 157 0.9 -1.0 0.1 0.0 0.8 OECD 141 153 136 130 142 0.8 -1.1 -0.4 -0.3 0.6 OECD Europe 63 62 53 53 56 -0.1 -1.7 0.2 -0.5 0.3 France 11 11 10 8 8 0.3 -1.3 -1.8 -0.9 0.0 Germany 16 18 18 22 24 1.2 0.0 1.8 1.0 0.6 Italy 6 7 5 6 6 2.1 -2.7 0.8 0.1 0.2 United Kingdom 21 17 11 10 10 -2.2 -4.6 -0.6 -2.5 0.0 Japan 15 27 31 35 41 5.9 1.5 1.1 2.8 1.1 Australia 4 4 3 2 2 0.0 -1.8 -6.2 -2.7 0.8 Canada 5 5 5 3 4 0.0 0.0 -3.2 -1.1 1.2 United States 54 55 45 37 39 0.2 -2.1 -1.9 -1.3 0.3 Non-OECD 0 1 2 10 15 - 11.6 17.0 - 3.0 LMICs 57 66 76 90 105 1.5 1.5 1.7 1.6 1.0 Americas 10 12 15 18 21 2.5 2.0 1.9 2.1 1.0 Asia & Pacific 18 19 18 37 44 0.4 -0.7 7.6 2.3 1.2 China 11 13 11 18 21 1.4 -1.7 5.0 1.6 1.0 India 5 5 3 3 4 -0.7 -5.4 1.2 -1.7 1.9 Korea, Rep. of 0 0 2 7 10 0.0 19.6 13.3 0.0 2.4 Europe 26 32 41 32 37 2.3 2.3 -2.1 0.7 1.0 FSU 18 17 25 21 25 -0.8 3.9 -1.7 0.5 1.2 Middle East/North Africa 1 0 1 1 1 0.0 0.0 1.3 0.4 -0.8 Africa 2 2 2 2 2 0.0 0.0 0.3 0.1 -0.2 World 198 219 215 230 262 1.0 -0.2 0.7 0.5 0.9 Sources: Metallgesellachaft AG, Metail Staistik, various issues; World Bureau of Metal Statistics, Metal StatiWics, various issues. The demand outlook for tin in the tinplate sector appears to be brighter than before because both tinplate demand and tin intensity of tinplate are likely to follow a pattern different from the past. Tinplate's share in the packaging market has stabilized at a low level after years of substitution away from tin cans, and, with tinplate's enhanced cost competitiveness, there appears to be little downside potential. LMICs hold the greatest potential for increased demand through growth in the market for food cans. The tin intensity of tinplate is likely to decline further, but the speed of decline should slow down. 82 Outlook vares among end uses. In many OECD countries, tin in solder is the largest single end use of tin, accounting for about one third of total tin consumption. The main user of tin solder is the electronics industry which has experienced rapid production increases, particularly in East Asia. However, consumption of tin solder has increased at a far slower rate than that of electronics production because of technological innovations replacing soldered electronic components with surface-mounted units. This process of substitution is expected to continue and, therefore, the demand outlook for tin solder in this sector is not bright. Another important end use for tin solder is in plumbing where substantial demand increases are expected as the tin content of tin/lead solder increases because of concerns over lead poisoning in water supplies. The most rapid increases in tin consumption have taken place in the chemical uses of tin, for example, in the manufacture of PVC as stabilizers, pesticides, paints, and catalysts. Sharply lower tin prices have helped lower prices of tin chemicals and hence increase their use. It is expected that this end-use sector will continue to provide the most rapidly expanding outlet for tin. Consumption of tin in other uses includes tinning of copperware, electroplating, white- metals, bearings, bronzes, pewter and other alloys. Individually, these end uses have experienced significant changes over the years, but collectively their consumption has remained more or less constant. Supply Outlook Price elasticity of supply evident. Two prominent features of tin supply behavior in recent years have been the speed and extent of the supply changes in response to tin price changes since the mid-1980s, and the emergence of Brazil and China as the world's leading producers. World tin mine production decreased from 236,000 tons in 1980 to 187,000 tons in 1987 (Table 2), following the collapse of tin prices in 1986. Prices recovered in 1987 and were quite strong by 1989 (see Table Al). Mine production of tin-in-concentrate also recovered and had reached 210,000 tons by 1990. In addition to new deposits tapped in Brazil and China, many of the old facilities that had been closed down in the wake of the ITC's collapse were brought back into production in Malaysia, Indonesia, Thailand, and Bolivia, among others. The experience illustrates that supplies can be quite price-elastic even in the short term because tin mining often involves little capital equipment, particularly in Southeast Asia. Brad dominates world production. In the 1988-90 period, Brazil was the world's largest producer of tin-in-concentrate, ahead of Malaysia. China has also become an important exporter of tin concentrates. Since Brazil can support even larger production increases at low cost, what happens in Brazil will in effect dictate the future of the world tin market. After the collapse of the ITC, seven major tin-producing countries (Australia, Bolivia, Indonesia, Malaysia, Nigeria, Thailand, and Zaire) formed the Association of Tin Producing Countries (ATPC). Brazil and China were included only as observers. ATPC's main function has been to implement Supply Rationalization Schemes (SRS), or export quotas. These efforts have been moderately successful in that most of the members complied with their quotas. However, as tin prices recovered in 1988-89, most producers responded to the higher prices with increased production, in particular, the "garimpeiros" (independent miners) of Brazil, whose production was uncontrollable, even by the Brazilian government. In 1991, however, Brazil's tin-in-concentrate production declined sharply, mainly because of the Brazilian government's concern that indiscriminate mining activities in the Amazon region were environmentally harmful but also possibly because of the realization by the largest Brazilian tin exporting company that large production increases might not be in its own interest. 83 Table 2: Tin-Mine Production (Metal Content), 1980-1991 (Actual) and 1995-2005 (Projected) Growth Rate 1980 1985 1987 1990 1991 1995 2000 2005 1980-91 1991-200 5 ('000 tons of tin-in-concentrate) -(% p.a.)- High Income 17.8 15.1 17.9 15.6 11.8 13.5 13.5 13.0 -3.7 0.7 Australia 11.6 6.4 7.7 7.4 5.7 6.0 6.5 7.0 -6.3 1.5 Canada 0.2 0.1 3.4 2.8 3.0 3.0 3.0 2.5 27.9 -1.3 United Kingdom 3.3 5.2 4.1 3.4 1.1 2.5 2.0 1.5 -9.5 2.2 Other 2.7 3.4 2.7 2.0 2.0 2.0 2.0 2.0 -2.7 0.0 LMICs 217.9 182.9 169.0 194.6 175.4 201.0 216.0 234.5 -2.0 2.1 Americas 36.0 47.2 44.3 61.6 52.7 65.0 72.0 84.0 3.5 3.4 Bolivia 27.5 16.1 8.1 17.3 16.8 18.0 20.0 22.0 -4.4 1.9 Brazil 6.9 26.4 30.4 39.1 29.3 40.0 45.0 55.0 14.0 4.6 Other 1.6 4.7 5.8 5.2 6.6 7.0 7.0 7.0 13.7 0.4 Asia & Pacific 152.9 110.1 102.4 112.9 103.0 113.5 119.0 123.5 -3.5 1.3 China 16.0 20.0 28.0 35.8 36.0 40.0 45.0 50.0 7.7 2.4 Indonesia 32.5 21.8 26.2 31.7 30.1 30.0 28.0 27.0 -0.7 -0.8 Malaysia 61.4 36.9 30.4 28.5 20.7 25.0 27.0 27.0 -9.4 1.9 Thailand 33.7 16.6 14.8 14.6 14.2 15.5 16.0 16.5 -7.6 1.1 Other 9.3 14.8 3.0 2.3 2.0 3.0 3.0 3.0 -13.0 2.9 Europe 16.5 16.4 15.6 14.7 14.4 16.5 19.0 21.0 -1.2 2.7 FSU 16.0 16.0 15.0 13.0 11.0 13.0 15.0 16.0 -3.3 2.7 Other 0.5 0.4 0.6 1.7 3.4 3.5 4.0 5.0 19.0 2.8 Middle East/North Africa 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Africa 12.5 9.2 6.7 5.4 5.3 6.0 6.0 6.0 -7.5 0.9 World 235.7 198.0 186.9 210.2 187.2 214.5 229.5 247.5 -2.1 2.0 Sources: Metallgesellschaft AG, Metall Statistik 1980-1990; World Bureau of Metal Statistics, Metal Statisics, June 1992; World Bank, International Economics Department (projected). 84 Potential for production increases substantial. Table 2 provides forecasts of mine production of tin in terms of tin-in-concentrate. A brief description of our expectations for the most important tin-producing countries is given below: BoLIVIA. Bolivia's tin industry has been restructured, partly with World Bank assistance. Since Bolivia's tin production comes mostly from underground mines, it had one of the highest production costs and its production level had been steadily declining until recently. Current plans call for production increases, to almost double the present level by 2000. Reorganization of COMIBOL and assistance given to small- and medium-scale mines have reduced costs, but it is not clear to what extent the planned increases will be adversely affected by the lower tin prices expected in the 1990s. Although not yet proven, Bolivia has the potential to find low-cost tin resources in areas close to Brazil's Rondonian tin province. BRAzIL. It is estimated that Brazil produced more than 52,800 tons of tin-in-concentrate in 1989, of which approximately 56% was produced by garimpeiros. Discovery in 1987 of the high-grade Ariquemes deposit in Rondonia has made garimpeiros' "pick and shovel" mining profitable. The recent discovery of the Suracananus deposit in the Roraima province, probably the largest in Brazil, could invite another run by garimpeiros unless the government enforces strict controls. It is highly likely that other large deposits will be found. From known deposits alone, Brazil's production could easily reach more than 60,000 tons of tin-in-concentrate in a few years. Currently, Brazil has the lowest production cost in the world. CHINA. Recently, China emerged as an important exporter of tin, with production estimated at 36,000 tons of tin-in-concentrate and tin metals in 1990. China's tin resources are believed to be large enough to support substantial production increases. SOUrHEAsr AsIA. Characteristics of tin mining are similar in Malaysia, Indonesia, and Thailand. Gravel pumps and dredges are the two dominant mining techniques used in Southeast Asia. The long history of tin mining in these countries has resulted in declines in ore grade and cost increases. Thus, as tin prices declined in 1991, production in Malaysia declined dramatically. In the 1990s, these countries' production is likely to stabilize at approximately the level in 1987. Price Outlook Slow price recovery as stocks and production decline. After the tin price peaked at over US$10/kg in April 1989, reported tin stocks increased from 43,000 tons in March 1989 to 58,700 tons at the end of September 1990 and tin prices declined to below US$6/kg for the July-September period. The price decline and stock increase were caused mainly by sharp production increases. LME and Penang exchange stocks began to decline from March 1991 onwards, but prices remained at US$5.50/kg or lower during most of 1991. The Kuala Lumpur market price declined to a bottom of US$5.35/kg in December 1991. With the sharp decline in production during 1991 and signs that stocks were edging down, tin prices have been recovering slowly during 1992. We expect tin prices to increase slowly in the 1992-95 period, based on the assumption that ATPC, with Brazil's cooperation, will be moderately successful in managing potential excess supplies so that stocks decrease. 85 Over the longer term, the market balance will depend critically on the extent of Brazil's production expansion. Under the assumption that Brazil's production, which declined sharply in 1991 and apparently in 1992 as well, increases only gradually from 29,000 tons of tin-in-concentrate in 1991 to 55,000 tons by 2005, prices are expected to decline slowly to US$6.10/kg in 1990 dollars by the year 2005. 86 Table Al: Tin - Prices, 1950-91 (Actual) and 1992-2005 (Projected) 1990 Constant $ ----Current - --- 5 MUV al- --- -7 CPI b/-- LM c/ Kuala Lunpur d/ LME c/ Knala Lumpur d/ LME c/ Kuala Lumpur d1 Actual 1950 206 1,258 1,657 1951 298 1,581 2,166 1952 266 1,347 1,855 1953 201 1,050 1,394 1954 198 1,056 1,350 1955 204 1,070 1,389 1956 217 1,097 1,443 1957 208 1,029 1,353 1958 203 200 985 970 1,294 1,274 1959 216 214 1,057 1,058 1,385 1,372 1960 220 213 1,061 1,028 1,379 1,335 1961 245 241 1,162 1,147 1,501 1,481 1962 247 242 1,151 1,126 1,474 1,443 1963 251 246 1,190 1,168 1,449 1,422 1964 341 335 1,590 1,561 1,920 1,885 1965 389 380 1,804 1,758 2,126 2,C72 1966 357 349 1,599 1,560 1,887 1,840 1967 333 324 1,474 1,434 1,714 1,669 1968 313 306 1,396 1,365 1,565 1,530 1969 343 338 1,453 1,433 1,645 1,623 1970 367 359 1,464 1,432 1,661 1,625 1971 350 342 1,324 1,294 1,483 1,449 1972 377 368 1,309 1,276 1,448 1,411 1973 483 472 1,447 1,414 1,622 1,586 1974 820 784 2,017 1,929 2,488 2,379 1975 687 669 1,520 1,480 1,855 1,807 1976 758 747 1,654 1,630 1,982 1,953 1977 1,076 1,071 2,138 2,128 2,545 2,533 1978 1,291 1,252 2,228 2,160 2,491 2,415 1979 1,546 1,482 2,356 2,259 2,819 2,703 1980 1,678 1,644 2,331 2,284 2,736 2,681 1981 1,416 1,406 1,959 1,946 2,290 2,275 1982 1,283 1,295 1,803 1,820 2,101 2,121 1983 1,299 1,303 1,868 1,875 2,095 2,102 1984 1,227 1,246 1,804 1,830 1,985 2,01.4 1985 1,195 a/ 1,154 1,741 1,682 1,893 1,829 1986 n.a. 616 n.a. 762 n.a. 809 1987 n.a. 669 n.a. 753 n.a. 778 1988 n.a. 705 n.a. 740 n.s. 762 1989 n.a. 853 n.a. 902 n.a. 926 1990 599 G/ 609 599 609 599 609 1991 560 548 548 537 532 521 Projete 1992 615 577 551 1993 660 597 571 1994 720 640 608 1995 780 675 640 2000 851 620 580 2005 961 610 547 a! Deflated by G-5 Manufacturing Unit Value (MUV) Index. b/ Deflated by 0-7 Consumer Price Index (CPI). c/ London Metal Exchange (UME), standard grade, minimu 99.75% so, settlement price. d/ For 1950-1984, Malaysian, Straits, minimum 99.85% so, official selling price, ex-amelter, Penang; beginning 1985, Kuala Lumpur In Market (KLTM), settlement price. el During October 24, 1985 through June 1990, tin trading at the UIE was suspended. Sources: International Tin Council, Tin Statistics, and Metals Week (actual); World Bank, International Economics Department (projected). 87 Nickel Swnmary * The outlook for nickel can be described as a fundamentally optimistic metal compared with other base metals because of the demand for its indispensable properties in end products such as stainless steel. * Nickel production capacity worldwide is expected to increase by at least 100,000 tons over the next five years, with expected major expansions in Australia, Brazil, China, Cuba, and Colombia. Such capacity expansion would allow a 2.1% p.a. production growth over this period. * The nickel price is expected to decline in 1992 (by about 17% in constant terms), stabilize during 1993-94 period, and rise in constant terms during the 1996-2000 period (by 2% p.a.). The long- term forecast (2000-2005) for the nickel price is for it to be essentially flat in constant terms. Nickel Prices a/, 1950-2005 ($/tonD 16,000 14,000 12,000 Constant 1990 $ b/ 10,000 - B, 000 - 6,000 4,000 / V Current $ 2,000 1950 1955 1960 1955 1970 1975 1960 1985 1990 1995 2000 2005 a/ Cathodes, Canadian (1950-79), LME (1990-2005) b/ Def Iated by CG5 MUV Index 88 Nickel Production Apparent Nickel Consumption Actual and Forecasts, 1961-2005 Actual and Forecasts, 1961-2005 1200 1200 1000 1000 600 Goo- 00 800 00 9 200a 200- 0 1981 1985 1989 1973 1977 1981 1985 1989 1993 1997 2001 2005 Yea 11 1965 1969 1973 1977 1981 195 1189 1993 1997 2001 2005 Yea Demand Outlook RECENT DEVELOPMENTS. Because of its desirable characteristics such as high resistance to corrosion and high tensile strength at elevated temperatures, nickel is used primarily as a metal industry additive in the manufacture of stainless steel, alloy steel, high- and/or special-performance alloys, nonferrous alloys, and nickel-based plating. A relatively new area of application is in rechargeable electrical batteries. Currently, about 60% of total nickel consumption goes to stainless steel, 10% is used in alloy steel, 15% in other nickel-based alloys, and the rest in coating and foundry uses. Nickel metals are divided into two classes based on their nickel content and physical appearance. Class I nickel products have a nickel content of at least 99% (e.g., electrolytic cathodes have 99.9% nickel and carbonyl pellets have 99.7%). Products of this type can be used directly for almost any application. Class II metals are suitable for limited applications (e.g., stainless steel and alloy steel). Their nickel content is between 20% and 96%. The majority of class II products are ferronickels with a nickel content of 25-45%. Less common products include nickel oxide sinter (76-90%) and nickel sand (20-25%), mainly used as catalysts in batteries, fuel cells, and insecticides. Stainless steel has been by far the largest consumer of nickel. Figure 1 plots stainless steel production and primary nickel consumption of the OECD countries during the period 1983-91. Primary nickel consumption has closely followed stainless steel production. For example, stainless steel production increased 14% in 1984 and nickel consumption increased by 12%. In 1989, stainless steel production dropped 4% and nickel consumption fell 2.2%. However, the link between stainless steel production and primary nickel consumption is complicated by two factors: (i) not all stainless steel contains nickel (nickel containing (austenitic) stainless steel accounts for about 75% of total stainless steel production); and (ii) not all nickel used in stainless steel is primary nickel (a considerable amount of nickel comes from ferronickel materials and secondary sources such as stainless steel scrap). This means that stainless steel producers have substitution possibilities when the price of primary nickel is high. 89 F[pr.1 Stainless Steel vs. Nickel, 1983-91 OECD Co=eMes 2900, ISO 0 a -170 30 2200 OWO 1800 I. '110 5 ~1400 . 110 18.1 184.1 198.1 1915.1 1987.1 1988.1 189.1 1990.1 1991.1 1183.3 1984.3 18.3 19863 187.3 1888.3 18.9 19031991.3 Yew.OMtW Indeed, due to the nickel price boom in the late 1980s, the ratio of primary nickel consumption to stainless steel production declined from about 8% in 1983 to about 6% in 1991. The average annual growth rate for stainless steel production over the nine-year period was 6.2% and for nickel it was 4.1%. Table 1 presents apparent nickel consumption data for the past two decades. Because of the close association between nickel use and the high-tech durable goods industries, it is not surprising to find that the OECD countries have been the dominating nickel consumer, accounting for about 70% of world consumption. Among the OECD countries, Germany, Japan, and the United States are by far the three largest consumers, with about 65% of total OECD nickel consumption. Nickel consumption in the LMICs has been relatively small but has been increasing rapidly. During the 1970-90 period, the average annual growth rate of consumption for this group of countries was 3.2%, compared with 1.5% for OECD countries. Within the LMIC group, the Asia and Pacific region posted the fastest growth (5.8% p. a.) during the 20-year period. OurLooK. Long-term potendial for nickel demand solid. The outlook for nickel demand is generally upbeat because of its indispensability in products that have strong growth potential. The global push for environmental protection should boost nickel demand in the form of stainless steel and high nickel alloys used in flue gas desulphurization in power plants. Additionally, recent technological innovations in nickel-based rechargeable batteries for use in electric cars and trucks present enormous potential. However, the sharp increase in nickel prices in the 1988-89 period resulted in declines in nickel consumption in the past two years (about -1% each year). The expected slow economic recovery in OECD countries should keep world nickel consumption growth flat or negative in the near term (1992-93). During the 1994-2000 period, the projected healthier growth in industrial production of OECD countries and the strong showing in some LMICs (e.g., China and India) would be likely to push the growth rate of world nickel consumption to above 2% p.a. Over the long term (2000-2005), growth in nickel consumption should be sustained at around 2% p.a.-a better performance than other base metals. 90 Table 1: Nickel Apparent Consumption, by Main Countries and Economic Regions, 1969-90 ('000 tons of metal content) Countries 1969-71 1979-81 1983-85 1988 1989 1990 High-Income 407 497 505 602 595 601 OECD 406 494 499 582 577 582 United States 140 145 142 135 127 125 Germany 43 79 72 91 89 89 France 33 37 35 40 40 45 United Kingdom 34 27 24 33 30 33 Sweden 19 20 18 20 18 19 Italy 18 25 27 29 31 27 Japan 88 121 132 162 163 159 LMICs 129 211 237 243 247 234 Asia and Pacific 22 30 39 51 62 66 China, People's Rep. of 19 19 19 28 27 28 Korea, Rep. of - 3 3 6 18 24 Europe 101 161 177 165 157 144 FSU 87 130 141 130 120 115 World 535 709 742 845 842 835 Source: World Bank, International Economics Department. Supply Prospects PRODUCTION CHARACTEPiSTICS. Table 2 shows the trends in nickel production by main countries and economic regions over the past 20 years. World nickel production grew at 2.1% p.a. over the 1970-90 period, with most of the growth taking place in the LMICs (4.1% p.a.); the high-income countries registered only 0.5% p.a. growth. Among the OECD countries, Australia and Canada have been the two dominant suppliers of nickel, from mining to refining. For nickel metal production, Japan depends entirely on imports of nickel ore and concentrates from Australia, Indonesia, New Caledonia, and the Philippines. The FSU is the world's largest producer of nickel. Newcomers to the ranks of nickel producers include Brazil and China, where production growth has exceeded 5% p.a. over the past 20 years. The nickel industry is characterized by significant vertical integration. Most nickel mining companies also produce finished ferronickels and/or nickel metals. Further downstream, however, integration is rare. Inco, the Canadian producer, has a significant presence in nickel metal fabrication. Outokumpu, the Finnish producer, also has downstream operations such as stainless steel production. 91 Table 2: Nickel Production, by Main Countries and Economic Regions, 1969-90 ('000 tons of metal content) Countries 1969-71 1979-81 1983-85 1988 1989 1990 High-Income 358 377 356 411 397 402 OECD 358 377 356 400 387 392 Canada 162 115 110 146 130 127 France 10 8 6 10 10 10 United Kingdom 35 21 21 28 26 27 Finland 4 11 15 16 13 17 Australia 5 39 40 42 43 47 Japan 87 103 89 101 106 102 LMICs 197 327 379 439 460 450 Africa 14 32 33 46 49 47 South Africa 9 18 21 27 30 28 Zimbabwe 5 14 12 18 19 19 Americas 20 43 67 83 88 82 Cuba 18 20 18 24 27 21 Brazil 2 2 12 13 14 13 Dominican Republic - 20 23 29 31 29 Asia and Pacific 27 66 60 68 68 65 China, People's Rep. of - 11 18 26 26 28 Europe 135 187 219 242 255 257 FSU 123 165 194 215 225 230 Greece 8 13 15 13 16 16 World 555 704 735 850 857 853 Source: World Bank, International Economics Department. Nickel is mined from either sulphide or oxide ores. Nickel sulphide ores usually occur in hard rock, vein-type underground mines, and also contain by-product copper, cobalt, and precious metals. Oxide deposits occur in laterites formed by the weathering of such rocks as peridotites. Sulphide ores are mined in Australia, Canada, China, Finland, South Africa, the FSU, and Zimbabwe. Laterite ores are mined in Australia, Brazil, Cuba, Dominican Republic, Greece, Indonesia, New Caledonia, Philippines, and the FSU. Sulphide ores are mined by either surface or underground mining. Nickel concentrates from sulphide ores are smelted and refined to nickel metal containing 90% or more nickel. There are almost as many metallurgical processes developed and commercially used as there are nickel recovery plants. 92 Laterites are usually mined by surface mining. A major problem associated with laterites is the high moisture content (up to 30%), which adds weight and significantly reduces equipment efficiency. Unlike sulphide ores, oxide ores cannot be readily upgraded to higher grade concentrates for refining. Therefore, metallurgical recovery is more costly for the oxide ores. However, this is compensated somewhat by the much lower costs of open-pit mining. Laterites are processed mostly to ferronickel containing 20-40% of nickel. Identified world nickel reserves with an average ore grade of 1% or better amount to 143 million tons of nickel. About 60% of the total is in laterites and 40% is in sulphides. World reserves of lower grade nickel are large. There are extensive deep-sea deposits of nickel in manganese crusts and nodules covering large areas of the ocean floor, particularly in the Pacific Ocean. Table 3 presents estimates of nickel reserves and reserve bases of major producing countries. Given the current rate of extraction, the available reserves would last for 50-100 years. By-products and co-products of nickel-such as cobalt, copper, and precious metals-are significant factors in nickel recovery. For example, Canada's Sudbury nickel mine yields nearly three- quarters of a pound of copper for each pound of nickel produced, together with large quantities of platinum, gold, silver, and cobalt. The choice of the metallurgical process to recover nickel usually is highly influenced by the presence of by-products. Because of the wide range of variations described above, production costs may vary considerably, depending on the kinds of deposits and metallurgical processes. Table 4 presents estimates of nickel production costs for major nickel operations. Costs of energy and labor and by-product credits significantly affect direct cash costs. Under the assumption that crude oil cost is $15/bbl, the share of energy in the total cost of nickel production is 15% for sulphide ores and 45% for laterite ores. Some laterite operations have made efforts to decrease their reliance on oil by converting to coal, natural gas, and hydroelectric power. The cost of labor accounts for 50% of the total cost for sulphide operations and 20% for laterites, because sulfides are more labor intensive due to underground mining. Materials used in nickel recovery include ammonia, naphtha, acid, sulfur, electrodes and other materials. They represent on average about 35% of the total production cost for both sulfides and laterites. Revenues from co-products and by-products play a more important role in most sulphide operations than for laterites. SUPPLY OUTLOOK. Sustaining 2% p.a. growth after 1995 is doubtfid. Nickel production capacity worldwide is expected to increase by about 100,000 tons over the next five years, with major expansions in Australia, Brazil, China, Cuba and Colombia. In Australia, annual nickel production capacity currently stands at 69,000 tons. However, the development of the Mt. Keith and Yakabindie deposit will push the output of refined nickel as high as 130,000 tons per year (tpy). Several Brazilian nickel producers have announced intentions to expand, and about 20,000 tpy of additional capacity is expected there. China's largest nickel producer, Jinchuan, will double its capacity from 30,000 tpy to 60,000 tpy when a new smelter comes onstream in 1994. Cuba has plans to continue development of its Las Camariocas project, which could have 30,000 tpy of capacity by the mid-1990s. In Colombia, Cero Matoso is to expand refined nickel production capacity by 18,000 tpy by 1993. If all plans come to fruition, this would imply a total of 2.1% p.a. production growth over the next five years, which is approximately the same rate as in the recent past. 93 Table 3: World Nickel Mine Production, Reserves, and Reserve Base ('000 tons of nickel content) Mine Production Reserve 1990 1991 Reserve a/ Base b/ Albania 9.0 9.1 181.4 181.4 Australia 69.8 72.6 2,176.8 6,802.5 Botswana 25.0 27.2 476.2 907.0 Bradl 20.9 22.7 665.7 4,262.9 Canada 201.8 204.1 6,167.6 13,605.0 China 25.0 25.4 725.6 907.0 Colombia 17.0 17.2 557.8 739.2 Cuba 38.4 36.3 18,140.0 22,675.0 Dominican Republic 32.7 32.7 453.5 680.3 Finland 10.5 10.9 79.8 99.8 FSU 258.5 258.5 6,621.1 7,346.7 Former Yugoslavia 1.0 0.9 157.8 204.1 Greece 18.1 18.1 453.5 907.0 Indonesia 58.0 59.0 3,201.7 12,698.0 New Caledonia 88.0 90.7 4,535.0 14,965.5 Philippines 12.0 12.7 408.2 10,974.7 South Africa 36.3 36.3 2,539.6 2,630.3 United States 0.3 7.6 22.7 2,539.6 Zimbabwe 7.5 8.2 77.1 99.8 Other 4.5 4.5 0.0 7,890.9 World Total 934.4 954.5 47,641.1 111,116.6 a/ That part of the reserve base that could be economically extracted or produced at the time of determination. b/ That part of an identified resource that meets specified minimum physical and chemical criteria related to current mining and production practices. Source: United States Bureau of Mines, Mineral Industry Surveys (Nickel in 1991). However, almost all projects discussed above are incremental expansions to existing facilities because the nickel price necessary to meet the stringent financial criteria for investing in a greenfield project is about $13,000/ton and capacity additions can be justified at about one half of that price. The only announced greenfield project, the Mt. Keith and Yakabindie project, is still in the stage of feasibility study. Thus, it is questionable that production growth worldwide can be sustained at 2% p.a. in the mid to long term (1995-2005) when incremental expansion possibilities are exhausted. 94 Table 4: Nickel Production Costs (1990 US$/Ib) Producer Country Ore Type a/ Mlbs b/ Cl c/ C2 d/ C3 e/ Cerro Matoso Colombia L 40 1.32 2.02 3.11 Namew Lake Canada S 16 1.54 2.21 2.33 Inco Canada S 330 1.84 2.28 2.84 Codemin Brazil L 14 1.89 2.47 2.71 Selebi-Phikwe Botswana S 40 1.92 2.09 5.09 Greenvale Australia L 32 1.95 2.50 2.58 Falconbridge Canada S 58 1.99 2.89 3.32 PT Inco Indonesia L 60 2.04 2.60 2.91 Falcondo Dominican Republic L 60 2.25 2.41 2.60 Bindura Zimbabwe S 24 2.36 2.41 2.53 Enonkoski Finland S 12 2.37 2.73 2.85 Western Mining Australia S 95 2.46 3.02 3.19 SLN New Caledonia L 85 2.76 3.08 3.28 PT Aneka Tambang Indonesia L 74 3.14 3.39 3.54 Redstone Canada S 4 3.39 3.74 3.89 Glenbrook United States L 8 4.72 5.22 5.42 Total/Weighted Average 952 2.16 2.61 3.09 a/ Ore types include sulphide (S) and laterite (L) ores. b/ Production capacity in million pounds of nickel content. c/ Direct cash costs (Cl) which include costs incurred in mining, milling, ore freight, on-site administration expenses, smelting, refining, intermediate and final freight, marketing, and by-product credit. d/ Operating costs (C2) which includes direct cash costs (Cl) and depreciation. e/ Fully allocated costs which includes operating costs (C2), indirect costs, and interest charges. Indirect costs include the portion of corporate overheads attributable to the operation, research and exploration attributable to operation, royalties, and other extraordinary costs. Source: Brook Hunt & Associates Limited, Nickel Mine Cost Study (1982-95), 1990 Edition. The most important issue for nickel market analysts is the future development of the FSU nickel industry. The FSU nickel industry was re-organized in 1989 when the nickel complexes at Norilsk (in eastern Siberia), Pechenga and Monchegorsk (on the Kola Peninsula), and the Krasnoyarsk platinum plant (in eastern Siberia) were grouped together as the Norilsk Nikel Kombinate (NNK). The new organization controls some 80% of total FSU nickel production (300,000 tons in 1991). Other important nickel deposits are in the central Urals and the Ukraine. About 85% of FSU nickel is contained in sulphide ore and 15% in laterites. At present, NNK's largest complex is at Norilsk, which produces 200,000 tpy of nickel in concentrates. About 30,000 tons are shipped to Pechenga for smelting, with the balance smelted at Norilsk. Nickel was traditionally treated as a strategic material in the FSU because of its use in military weaponry. The statistics on nickel production and consumption were kept secret. Following the political and economic transition in the FSU, military applications of nickel have been greatly reduced, and the surplus nickel has been channeled to the international market in order to generate 95 foreign exchange. In 1991, the FSU exported 120,000 tons of nickel, more than double the exports of 55,000 tons of four years ago. One school of thought believes that the combination of ever-pressing hard currency needs and declining domestic consumption will result in an unabated outflow of nickel, even exceeding the high of 1991. However, another school of thought, citing the depletion of the stockpiled nickel, transportation difficulties, inability to secure raw materials and spare parts, pollution-forced shutdowns, long-delayed repairs and modernization, the confusing maze of export taxes and licenses and foreign exchange requirements, believes that FSU exports of nickel will definitely decline from the 1991 level. Price Prospects Prices to stabilize before resuming growth after 1995. The nickel market is by far the smallest of the base metals markets in terms of tonnages. In addition, the cost of nickel is only a small part of the total cost of downstream products such as stainless steel. Thus, the price elasticity of demand for nickel is relatively small, and a minor imbalance of nickel demand and supply can often cause large price fluctuations. Historically, market prices for virtually all nickel products were related to Inco's prices for electrolytic cathodes. After 1980, however, Inco gradually lost its position as the price setter with the number of nickel producers worldwide increasing significantly and exports from Cuba and the FSU expanding. Subsequently, spot nickel prices on the LME became the market benchmark, although they have been very volatile. Figure 2 plots the price volatility (calculated as the five-year standard deviation of the first differences in price) of nickel, in both current and constant terms, over the 1950-91 period. Before 1980, the price volatility was moderate (averaging 5% both in current and constant terms). However, after 1980, volatility increased significantly, surpassing 15% in the early 1980s and exceeding 40% in some years in the late 1980s. Figre 2 Nickel Price Volatility, 1953-91 60% 50 40% 20% 10 Cr.mle 1955 1957 1981 19 5 19 ' 19 73 '1977 1981 195 igiA 1 .55 1959 19.. 19I7 1971 1975 1979 1983 1987 1991 Yea Nickel price movements (LME cash price) in the late 1980s were characterized by large shocks, both up and down. After declining 40% in constant terms in 1986, nickel prices doubled in 1988 (a 97% increase in constant terms and a 104% increase in current terms), to reach a historical high. The high prices in 1988 were caused by strong demand, low inventories, and the threat of disruptions to 96 production by strikes and technical problems at several major mines. However, prices fell 46% in constant terms in 1990 and 11% in 1991, because of weakening demand and greatly increased FSU exports to European markets. Nickel prices are expected to decline further in 1992 (by about 17% in constant terms) before stabilizing in 1993-94. The causes for the decline remain the poor economic outlook for the OECD countries and expectations of continued large exports from the FSU. The steady and robust economic growth assumed for the second half of the 1990s would push up the demand for stainless steel and other nickel-containing durable goods. Due to the low prices in the early 1990s, capacity expansions, particularly greenfield projects, during the second half of the 1990s may fall behind demand increases. Thus, nickel prices are forecast to increase at an average annual rate of 2% in real terms over the 1995- 2000 period. For the long term (2000-2005), nickel prices are forecast at the level of production costs of the potential new mines likely to be brought into production by that time. This assumes that the market will be approximately balanced over the long term. 97 Table Al: Niwkul - Pies, al 1950-91 (Acaa) and 1992-2005(Peojod Cunem $ - 1990 COama - -5MUb/ 0-7 CPI o/ 1950 988 6,049 7,965 1951 1,191 6,323 8,664 1952 1,246 6,312 8,692 1953 1,321 6,86 9,139 1954 1,334 7,112 9,092 1955 1,422 7,439 9,660 1956 1,437 7,260 9,557 1957 1,631 8,070 10,614 1958 1,631 7,934 10,418 1959 1,631 8,047 10,438 1960 1,631 7,882 10,241 1961 1,711 8,127 10,496 1962 1,762 8,202 10,510 1963 1,742 8,268 10,067 1964 1,742 8,126 9,812 1965 1,735 8,036 9,471 1966 1,739 7,784 9,184 1967 1,936 8,563 9,962 1968 2,075 9,266 10,388 1969 2,363 10,015 11,341 1970 2,846 11,34 12,873 1971 2,932 11,092 12,424 1972 3,080 10,691 11,828 1973 3,373 10,108 11,334 1974 3,825 9,407 11,605 1975 4,570 10,110 12,342 1976 4,974 10,852 12,999 1977 5,203 10,335 12,303 1978 4,610 7,957 8,196 1979 5,986 9,122 10,914 1980 6,519 9,058 10,631 1981 5,953 8,238 9,630 1982 4,838 6,799 7,925 1983 4,673 6,721 7,536 1984 4,752 6,984 7,685 1985 4,899 7,142 7,764 1986 3,881 4,798 5,097 1987 4,872 5,484 5,664 1988 13,778 14,457 14,879 1989 13,308 14,061 14,437 1990 8,864 8,864 8,864 1991 8,156 7,993 7,759 1992 7,200 6,759 6,449 1993 7,600 6,376 6,572 1994 8,100 7,195 6,843 1995 8,700 7,527 7,144 2000 11,500 8,377 7,837 2005 13,500 8,566 7,690 a/ For 1950-79, C.,dia Oclytk cdu MaWumn 99.9% 9; tom 1980 oward, Lod~mMal Zwange cathodos, 99.8% ui. b/ Deflated by G-5 M,-faturiug Unik Vau» (MUV) Index. e/ Doflated by 0-7 Co~m-o-Pic Index (CM). Sourect: MehaLWk (acta); World Bank, aer to.al bt.an-an DSa&M (p~ojuwd). 98  Aluminum and Bauxite Summary * Aluminum consumption growth has slowed over the past three decades from an annual growth rate of 9.1% in the 1960s, to 4.8% in the 1970s, and to a mere 1.6% in the 1980s. However, in the period 1992 to 2005, world consumption is expected to increase at a slightly faster annual rate of 2.2%, with the share of LMICs rising from 31% in 1991 to 36% in 2005. * As aluminum prices and consumption growth are expected to remain relatively low (by historical standards) in the 1992-2005 period, aluminum smelting capacity growth should be slow. Most investments in new smelting capacity are likely to take place in Australia, Canada, the Middle East, South America, and southern Africa where both bauxite and low-cost energy are available. The ratio of exports to production in aluminum is expected to rise substantially, reflecting the ongoing global restructuring of the industry. * Prices of aluminum and bauxite are expected to increase in the period 1992-94, as the market balance adjusts from the gross surplus in 1991, then to peak around 1995 at relatively low real prices. The greatest source of uncertainty regarding the course of aluminum prices in the medium term is the level of FSU exports to international markets, which will depend on the 1pace of domestic demand recovery and the renovation of antiquated smelting facilities. Over the long term (2000 and 2005), aluminum prices are projected at levels that cover the costs of production of new capacities, estimated at about $1,400/ton in constant 1990 dollars. Aluminum Prices a/, 1955-2005 C$/ton) 3,000 Constant 1990 $ b/ 2,500 2,100 A/ 1,500iI % 1 / I 1 1/ I 'II 'II I.I 1,000 V Current $ I\vif 0 0 . I . I . .. I .. . .I. .. . 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 a/ UnaIloyed prinary ingots. LME C1979-2005).other transactions C1958-78) b/ Deflated by G-5 MUV Index 99 Projected Aluminum Smelting Capacity, Projected Aluminum Consumption, by Region by 1 4or OECD Countries 25 2 20 2 I 5 15 1191 1995 2000 2005 1%90 1995 2000 2005 WO a aus= mA t t#dSe Sw L a,v; WO-- EstV Demand Outlook From 1960 to 1970, world consumption of aluminum grew at an annual rate of 9.1%. In the 1970s the annual growth rate was reduced by one half to 4.1%. In the 1980s, consumption grew at only 1.6% p.a. The fast growth in consumption of the earlier period was due to end-users replacing other metals with aluminum because of its declining real prices and special characteristics-i.e., aluminum is light and has a density about one third that of steel; it conducts electricity and heat well; it is resistant to corrosion from exposure to air, water, and many foodstuffs and liquids; it has some decorative properties that make it suitable for interior and exterior architectural applications; and finally, although it is a strong material, it is easy to work with. With the recession of 1975 and the increase in energy prices (which increased the costs of aluminum production), the boom years came to an end. However, consumption of aluminum has continued to increase in several important end-use sectors, namely, packaging, construction, and transportation. Total aluminum consumption can be divided into primary and secondary components. The ratio of secondary (recycled) to total consumption has increased from 19% in 1960 to 29% in 1990. The main reason for the increase in recycling is the major energy savings associated with the use of recycled aluminum. Per capita consumption of aluminum in the United States, the largest consumer, increased from 10.8 kg in 1960 to 27 kg in 1990. In Japan and Germany, the amount of aluminum consumed per head of population also reached similar levels, 26.6 kg and 25.7 kg, respectively. Data on consumption by end-uses are available for only a few industrial countries. Table 1 shows consumption by main end-use sectors for the United States, Japan, and four major Western 100 European countries for selected years over the past two decades. Packaging is the most important consumer of aluminum in the United Staes, followed by transportation and construction. In Japan and Western Europe, transportation and constuction are the two most important end-uses. Consumption of aluminuzm for packaging slowed in 1980s. In the United States, growth in the consumption of aluminum in packaging slowed down sharply in the 1980s-from 8.5% p.a. in the 1970s to 3.6% p.a. Similarly, in Japan, the growth of aluminum use in packaging slowed from 21% p.a. in the 1970s to 5.8% p.a. in the 1980s, In Europe, however, the growth rate was sustained at around 3.8% over the whole period. The slowdown in the growth of aluminum consumption in packaging in the United States and Japan is attributed mainly to three factors: (i) aluminum cans are much lighter today than previously; (ii) aluminum packaging has not been as successful in the food sector as in the beverage sector; and (iii) competition from "newer" materials, e.g., plastics, has been increasing. Demand for primary aluminum will be affected by the characteristics and prices of substitutes as well as levels of GDP or industrial production and the price of aluminum. Technological advances have made plastics a major competitor, especially in automobile production. However, the difficulty of recycling plastics has limited their use in this area. In the food industry, technological progress is improving the resistance of aluminum to corrosion so that is becoming more likely to replace steel in packaging. LMICs lead growth in primaiy aluninum consumption. World consumption of primary aluminum is expected to increase over the period from 1991 to 2005 at an average annual rate of 2.2% (Table AS). Aluminum consumption should grow faster in LMICs (3.4% p.a.) than in high-income countries (1.7% p.a.) mainly because income growth rates are expected to be higher in the former group. Consumption in Japan and non-OECD high-income countries should grow at a faster rate than in other high-income countries partly because of their higher expected economic growth rates and partly because of expected shifts of major aluminum-using industrial activities (particularly, electrical engineering) to these countries from other high-income countries. Among the LMICs, consumption in Asia should grow faster than in other regions (4.3% p.a.) mainly due to continued fast growth there (see Table A8). The intensity of use of aluminum in these countries has increased because they have devoted a large share of their GDP to construction and capital equipment, which are metal-intensive activities. However, as per capita incomes continue to rise and basic needs are satisfied, an increasing share of their expenditures will be expected to shift toward services such as medical care and education. The intensity of use of metals should thus decline. Uncertainties in Eastern Europe and the FSU. One important source of uncertainty regarding world aluminum consumption growth is the future consumption trend in Eastern European countries and the FSU. Consumption in these countries stagnated in the 1980s, and indeed declined in the early 1990s. A number of competing influences are now at work. Tending to decrease the intensity of use of aluminum is the emphasis on product quality and more efficient materials management as well as the decline in production of military hardware, which has been an important end-user of aluminum. On the other hand, these changes should help to stimulate economic recovery and growth, although how rapidly consumption recovers is uncertain. 101 Table 1: Total Aluminum Consumption by Main Sectors, Selected Major Industrial Countries, 1970-90 1970 1974 1976 1980 1985 1986 1987 1988 1989 1990 ('000 tons) United States Transportation 734 1,169 1,163 1,042 1,364 1,372 1,500 1,536 1,448 1,320 Electrical engineering 574 780 555 612 642 616 620 671 663 594 Construction 1,006 1,363 1,221 1,165 1,381 1,432 1,441 1,316 1,294 1,208 Packaging 665 1,027 1,166 1,512 1,863 1,929 2,052 2,036 2,112 2,157 Other 1,080 1,411 1,188 1,115 1,134 1,195 1,247 1,105 876 740 Exports of semi-manufactures a/ 527 428 379 995 546 413 569 786 1,060 1,131 Total 4,5386 6,178 5,782 6,441 6,930 6,958 7,430 7,450 7,452 7,151 Japan Transportation 257 325 371 582 763 785 834 930 1,039 1,140 Electrical engineering 156 163 178 226 155 165 237 238 244 254 Construction 298 546 668 732 701 733 819 897 902 934 Packaging 20 83 104 134 177 202 223 266 270 310 Other 409 472 523 553 632 602 746 901 925 964 Exports of semi-manufactures a/ 53 34 77 82 258 231 194 122 123 169 Total 1,193 1,623 1,921 2,309 2,686 2,724 3,054 3,365 3,502 3,771 Western Europe b/ Transportation 611 617 667 741 766 841 902 923 782 764 Electrical engineering 259 265 244 277 237 232 232 241 249 286 Construction 234 376 409 486 438 487 539 570 687 761 Packaging 178 250 238 258 291 298 301 294 379 381 Other 641 858 848 826 856 887 887 926 936 954 Exports of semi-manufactures 300 493 570 759 1,048 1,062 1,130 1,240 1,310 1,403 Total 2,223 2,859 2,976 3,347 3,636 3,807 3,991 4,194 4,342 4,549 a/ Exports of semi-manufactures are not broken down into various end-uses but are included because they represent a component of domestic consumption. b Historical end-use data are available only for the Federal Republic of Germany, France, Italy, and the United Kingdom. The ratio of D.R. Germany's primary consumptionto that of these four countries averaged only 8% in the 1980-1990 period. Note: Total aluminum consumption includes both primary and secondary aluminum. Thus, country totals in this Table will be greater than primary consumption data shown elsewhere. Sources: Metallgesellachaft AG, Metal Statistics (various issues). Outlook in packaging and transportation rmains strong in OECD. Projections by end- use for the United States, Japan, and four major industrial countries in Europe are presented in Table 2. In the United States, packaging is expected to remain the most important consumer of the metal, due to improvements in its technical properties that make it more resistant to corrosive materials. However, the fastest growth of aluminum use is expected to occur in the transportation sector as the automobile industry is likely to increase its use of aluminum to improve the fuel efficiency of its products. In Europe and Japan, the transportation sector has been the most important consumer, but the relative importance of the packaging sector will increase as marketed beverage consumption increases rapidly. Consumption of aluminum in the transportation sector in the United States is expected to increase on average by 3% p.a. from 1990 to 2005. The use of aluminum is expected to reduce the midsize car's overall weight by about 300 kg. This would cut gasoline consumption by 1 gallon for each 100 miles driven. In Western Europe and Japan, expected growth rates in transportation are 3.3% and 2.2%, respectively. In the construction sector, the consumption of aluminum is expected to increase over the long term by 1.8% p.a. in the United States, by 3.2% in Japan and by 1.7% p.a. in Western Europe. Due to aluminum's bulkiness and poorer electrical conductivity than copper, its use in the electrical engineering sector is expected to experience a virtual standstill in the United States, and indeed to decrease in Western Europe and Japan. Supply Outlook Reflecting the decline in the growth rate of consumption in the 1980s, investments in aluminum smelting capacity slowed in the 1980s, from an annual increase of 2.3% in the 1970s to 1.5% in the 1980s. Following the upturn in demand in 1987, capacity utilization improved, and in 1989, 1990 and the first half of 1991, smelters were producing at close to 100% of name-plate capacity in the western world. However, capacity utilization declined in Eastern Europe and the FSU, and the slowdown in the growth of consumption in 1990 led to a decline in the capacity utilization rate in the western world in the second half of 1991 and early 1992, despite a decline in smelting capacity in this period. Table 3 shows projected worldwide primary aluminum smelting capacity in the period 1991-2005. Expansion of capacity depends on the expected price of aluminum, the expected growth in demand, and the expected trend in costs. The most important factor determining the location of a new investment for aluminum smelting capacity is cost and availability of an energy source. The abundance of bauxite also has an important influence. For these reasons, most investments in new aluminum smelting capacity in the 1990s are expected to take place in Australia, Canada, the Middle East, South America, and southern Africa. Little change in geographic spread ofsmelting capacity. Worldwide aluminum capacity is expected to increase from 20.3 million tpy in 1991 to 24.3 million tpy in the year 2005. In 1991, the LMICs accounted for 42% of world aluminum smelting capacity. This share is expected to remain unchanged through 2005. On the one hand, within this group the share of Europe, which includes all of Eastern Europe and the FSU, is expected to decline as industrial restructuring takes place. On the other hand, there will be significant capacity increases in all other LMICs. In particular, large increases in capacity are expected in Venezuela and possibly Chile and Trinidad and Tobago in the Americas; China and India in Asia; South Africa in Sub-Sahara; and possibly in Algeria, Iran, and Qatar in the Middle East and North Africa. 103 Table 2: Aluminum Consumption by End-Uses, Major OECD Countrics, 1990-2005 Actual Prjcted- Growth Rate 1990 1995 2000 2005 1990-2005 ('000 tons) (%) United States Transportation 1,320 1,550 1,800 2,050 3.0 Electrical engineering 594 600 600 600 0.1 Construction 1208 1,320 1,450 1,580 1.8 Packaging 2157 2,300 2,500 2,720 1.6 Other 740 800 950 1,050 2.4 Exports of semi-manufactures 1131 1,000 1,000 1,100 -0.2 Total 7151 7,570 8,300 9,100 1.6 Japan Transportation 1,140 1,200 1,380 1,590 2.2 Electrical engineering 254 250 245 245 -0.2 Construction 934 1,100 1,290 1,470 3.1 Packaging 310 420 570 680 5.4 Other 964 1,000 1,130 1,250 1.7 Exports of semi-manufactures 169 170 200 230 2.1 Total 3,771 4,140 4,815 5,465 2.5 Western Europe a/ Transportation 764 950 1,150 1,225 3.2 Electrical engineering 286 270 260 260 -0.6 Construction 761 795 850 980 1.7 Packaging 381 430 540 650 3.6 Other 954 1,025 1,090 1170 1.4 Exports of semi-manufactures 1,403 1,450 1,570 1,700 1.3 Total 4,549 4,920 5,460 5,985 1.8 a/ Historical end-use data are available only for the Federal Republic of Germany, France, Italy and the United Kingdom. The ratio of D. R. of Germany's primary consumption to that of these four countries averaged only 8% in the 1980-90 period. Note: Total aluminum consumption includes both primary and secondary aluminum. Sources: MetallgesellMcaat u (actual); World Bank, International Economics Department (projections). Among the high-income countries, significant capacity increases are expected in Australia, Canada and Norway. Unless political and social conflicts delay investments, aluminum capacity in Canada is expected to reach 2.8 million tpy by the year 2005. Australia, which expanded its capacity significantly in the last decade, has a range of further possibilities. Australia benefits from its large reserves of high-grade bauxite. Table 4 provides the projected pattern of increases in alumina refinery capacity from 1991 to the year 2005. World alumina refinery capacity is expected to grow from 41.4 million tons in 1991 104 Table 3: Aluminum Smeling Capacity, 1991-2005 Growth Rate Actual Projected 1991 Share of World Countries/Economies 1991 1995 2000 2005 -2005 1991 2005 ('000 tc) : (% p..) (%) (%) High Icome 11,845 11,677 12,452 13,382 0.9 58.3 58.0 OECD 11,599 11,437 12,212 12,962 0.3 57.1 56.8 OECD Europe 3,454 3,340 3,572 4,012 1.1 17.0 16.6 Germany 720 720 720 720 0.0 3.5 3.6 France 456 440 440 440 -0.3 2.2 2.2 Italy 196 127 127 127 -3.1 1.0 0.6 United Kingdom 307 273 200 200 -3.0 1.5 1.4 Netherlands 282 282 282 362 1.8 1.4 1.4 Norway 880 880 930 1,280 2.7 4.3 4.4 Spain 355 355 355 355 0.0 1.7 1.8 Other OECD Europe 258 258 468 528 5.2 1.3 1.3 Japan 35 35 35 35 0.0 0.2 0.2 Canada 2,347 2,300 2,640 2,750 1.1 11.5 11.4 United States 4,101 4,100 4,100 4,100 0.0 20.2 20.3 Australia 1,397 1,397 1,600 1,800 1.8 6.9 6.9 New Zealand 265 265 265 265 0.0 1.3 1.3 Non-OECD High-Iacome 246 240 240 420 3.9 1.2 1.2 MICs 8,486 8,473 9,521 10,892 1.8 41.7 42.0 Americas 2,443 2,417 2,786 3,450 2.5 12.0 12.0 Argentina 165 165 165 165 0.0 0.8 0.8 Brazil 1,202 1,206 1,206 1,206 0.0 5.9 6.0 Venezuela 974 974 1,200 1,450 2.9 4.8 4.8 Other Americas 102 72 215 629 13.9 0.5 0.4 Asia & Pacific 1,686 1,686 1,893 2,305 2.3 8.3 8.4 China 850 850 350 1,050 1.5 4.2 4.2 India 604 604 811 910 3.0 3.0 3.0 Indonesia 225 225 225 225 0.0 1.1 1.1 Other Asia/Pacific 7 7 7 120 22.5 0.0 0.0 Europe 3,134 3,207 3,299 3,379 0.5 15.4 15.9 Greece 155 155 155 155 0.0 0.8 0.8 Romania 178 247 240 240 2.2 0.9 1.2 Yugoslavia 350 350 350 350 0.0 1.7 1.7 PSU 2,200 2,200 2,300 2,300 0.3 10.8 10.9 Other Europe 251 255 254 334 2.1 1.2 1.3 Middle East & North Africa 760 700 580 905 1.3 3.7 3.5 Egypt 180 180 180 180 0.0 0.9 0.9 Bahrain 460 460 460 460 0.0 2.3 2.3 Other Middle Eat/North Africa 120 60 240 265 5.8 0.6 0.3 Africa 463 463 663 853 4.5 2.3 2.3 Ghana 200 200 200 200 0.0 1.0 1.0 South Africa 175 175 375 565 8.7 0.9 0.9 Other Africa 88 88 88 88 0.0 0.4 0.4 World Total 20,331 20,150 21,973 24,274 1.3 100.0 100.0 Sources: World Bank, International Economics Department; based on information provided by industry sources, including, in particular, Anthony Bird Associates. 105 Table 4: Alumina Refinery Capacity, 1991-2005 Actual Prjected-- Growth Rate --Share in World- 1991 1995 2000 2005 1991-2005 1991 2005 ('000 lpy)- - (% p.S.) -(%)------- High Income 21,495 22,590 23,435 24,635 0.98 51.9 45.9 OBCD 21,495 22,345 23,195 24,395 0.91 51.9 45.4 OECD Europe 3,680 3,670 3,670 3,670 -0.02 8.9 6.8 Germany 640 630 630 630 -0.11 1.5 1.2 France 310 310 310 310 0.00 0.7 0.6 Italy 730 730 730 730 0.00 1.8 1.4 Spain 1,000 1,000 1,000 1,000 0.00 2.4 1.9 Ireland 1,000 1,000 1,000 1,000 0.00 2.4 1.9 Japan 180 180 130 180 0.00 0.4 0.3 Canada 1,015 1,015 1,015 1,015 0.00 2.4 1.9 United States 5,030 5,030 5,030 5,030 0.00 12.1 9.4 Australia 11,590 12,450 13,300 14,500 1.61 28.0 27.0 Non-OECD High-Income 0 245 240 240 105.53 0.0 0.4 LMICAs 19,939 20,395 25,100 29,050 2.72 48.1 54.1 Americas 8,731 9,590 11,320 14,580 3.73 21.1 27.2 Brazil 1,665 2,065 3,200 4530 7.41 4.0 8.4 Jamaica 3,450 3,925 4,500 5,450 3.32 8.3 10.2 Surinams 1,616 1,600 1,600 1,600 -0.07 3.9 3.0 Venezuels 2,000 2,000 2,020 3,000 2.94 4.8 5.6 Asia & Pacific 3,275 3,330 4,855 5,375 3.60 7.9 10.0 China 1,700 1,700 1,720 1,720 0.08 4.1 3.2 India 1,575 1,630 3,135 3,655 6.20 3.8 6.8 Europe 7,263 7,305 8,195 8,365 1.01 17.5 15.6 Greece 600 600 1,300 1,300 5.68 1.4 2.4 Turkey 160 175 200 200 1.61 0.4 0.4 Czechoslovakia 209 209 209 209 0.00 0.5 0.4 Hungary 848 875 900 900 0.43 2.0 1.7 Romania 410 410 400 400 -0.18 1.0 0.7 Yugoslavia 1,036 1,036 1,086 1,086 0.34 2.5 2.0 FSU 4,000 4,000 4,100 4,270 0.47 9.7 8.0 Middle East/North Africa 0 0 0 0 0.00 0.0 0.0 Africa 670 670 730 730 0.61 1.6 1.4 Guinea 670 670 730 730 0.61 1.6 1.4 World Total 41,434 43,485 48,535 53,635 1.37 100.0 100.0 Source: World Bank, International Economics Department; based on information provided by industry sources, in particular, Anthony Bird Associates and James F. King. 106 to 53.7 million tons in the year 2005. The share of the high-income countries will decrease from 52% in 1991 to 46% in 2005. LMICs are expected to increase their share from 48% to 54%. Alumina refinery growth in Latin America, India, Greece, Australia. The bulk of the increase in alumina refinery capacity is expected to occur in bauxite-producing countries in Latin America, with projects located in Brazil, Jamaica, and Venezuela. Latin America's alumina capacity is expected to increase on average by 3.7% p.a. over the projection period. During this period, Asia's capacity is expected to increase by 3.6% p.a. due to some major investments in India. Greece is the only other LMIC that is expected to experience a significant increase in alumina capacity in the period. Among the high-income countries, Australia is expected to increase its alumina capacity significantly. Significant LMIC bauxite capacity increases. Table 5 shows our forecasts of bauxite producing capacity by countries and regions. Worldwide, capacity is expected to increase from 113 million tons in 1991 to 145 million tons in 2005-about 1.9% p.a. The share of LMICs will remain at the same level in 2005 as in 1991, around 64%. Among the high-income countries, significant increases in bauxite capacity are expected to take place only in Australia. Significant bauxite capacity increases are expected to materialize in a number of LMICs, however: China and India in Asia; Guinea in Africa; several countries in Europe, including possibly Greece, Turkey, Hungary, Yugoslavia, and the FSU; and several Latin American/Caribbean countries such as Brazil, Guyana, Jamaica, Suriname, and Venezuela. Aluminum's real costs decreased as components "indexed." Aluminum production costs have been increasing in current US dollar terms, as can be seen from Table 6. The average total cost is estimated to have increased by 14% from 1984 to 1991--or an average annual increase of 1.9%, far lower than the average rate of inflation in the G-5 countries of about 6% p.a. over the same period (refer to the MUV index shown in Table 6). Thus, the cost of production of aluminum decreased substantially in real terms in this period. One important reason for this decrease was the fact that electricity prices hardly increased in current dollar terms over this period as most aluminum producers benefitted from discounted electricity prices. Another reason for the fall in the real cost of producing aluminum was that the price of alumina, another important cost component for aluminum, increased only moderately-again by far less than the inflation represented by the MUV index. In fact, it is increasingly prevalent to "index" the alumina and electricity prices to be paid by aluminum producers to the price of aluminum. The LME price of aluminum rose by only 4% in current dollar terms over the period 1984-91. Furthermore, in some cases, even wages, or labor compensation, were "conceded" by labor unions at the time of depressed aluminum prices. Because of this indexing feature of aluminum cost components, when aluminum prices decline, costs of producing aluminum also decline. This indexing of cost components could be said to be at least partly responsible for the unexpectedly modest scale of production cutbacks implemented during 1991 when the aluminum market suffered a significant oversupply problem, with the LME stocks steadily mounting to record levels. Trade Outlook Exports of bauxite worldwide are expected to increase at only 0.2% p.a. from 1991 to 2005, compared with the average rate of 0.5% p.a. in the 1970-90 period (see Table A2). This rate of increase is far less than that for production of bauxite, indicating that the trend for bauxite to be transformed to alumina in bauxite-producing countries will continue. In Latin America and the 107 Table 5: Bauxite Producing Capacity, 1991-2005 Actual Prjected-- Growth Rate --are in World- 1991 1995 2000 2005 1990-2005 1991 2005 -('000 py)-- (% p.a.) %) High Income 41,105 43,100 47,100 52,100 1.71 36.5 35.8 OECD 41,105 43,100 47,100 52,100 1.71 36.5 35.8 OECD Europe 5 0 0 0 - 0.0 0.0 United States 100 100 100 100 0.00 0.1 0.1 Australia 41,000 43,000 47,000 52,000 1.71 36.4 35.8 Non-OECD High-Income 0 0 0 0 0.00 0.0 0.0 IMICs 71,400 74,395 82,204 93,304 1.93 63.5 64.2 Americas 29,030 29,520 32,900 37,700 1.88 25.8 25.9 Brazil 10,500 10,800 11,700 13,000 1.54 9.3 8.9 Guyana 2,200 2,000 2,300 2,500 0.92 2.0 1.7 Jamaica 11,610 12,000 13,300 15,000 1.85 10.3 10.3 Suriname 3,200 3,200 3,600 4,200 1.96 2.8 2.9 Other America 1,520 1,520 2,000 3,000 4.98 1.4 2.1 Asia & Pacific 10,790 10,900 11,600 13,400 1.56 9.6 9.2 China 4,200 4,300 4,700 5,100 1.40 3.7 3.5 India 4,810 4,800 4,800 6,000 1.59 4.3 4.1 Indonesia 1,400 1,400 1,700 1,900 2.21 1.2 1.3 Other Asia & Pacific 380 400 400 400 0.37 0.3 0.3 Europe 12,855 14,900 17,904 20,904 3.53 11.4 14.4 Greece 2,225 2,550 4,100 5,000 5.95 2.0 3.4 Turkey 480 700 800 1,000 5.38 0.4 0.7 Hungary 2,050 2,450 2,800 3,000 2.76 1.8 2.1 Romania 200 200 204 204 0.14 0.2 0.1 Yugoslavia 2,550 3,000 3,000 3,100 1.40 2.3 2.1 FSU 5,350 6,000 7,000 8,600 3.45 4.8 5.9 Middle East/North Africa 0 0 0 0 0.00 0.0 0.0 Africa 13,725 19,075 19,800 21,300 0.92 16.6 14.6 Ghana 325 325 400 400 1.49 0.3 0.3 Guinea 17,100 17,400 17,900 19,200 0.83 15.2 13.2 Other Africa 1,300 1,350 1,500 1,700 1.93 1.2 1.2 World Total 112,505 11,7495 129,304 145,404 1.85 100.0 100.0 Sources: The World Bank, lIternational EconomicsDepartment; based on inormation available from industry sources, in particular, lames F. King. 108 Table 6: Estimated Aluminum Production Costs, 1982-91 a/ Average Average Average Average LME MUV Marginal Total Electricity Alumina Aluminum Index Year Cost Cost Price Price b/ Price ($/ton) - (mills/KWh) - -($/ton)------- 1990= 100 1982 1,279 N.A. 20.3 206 991 71.1 1983 1,257 N.A. 20.0 212 1,440 69.5 1984 1,140 1,400 18.2 204 1,251 68.1 1985 1,052 1,390 16.2 180 1,041 68.6 1986 1,010 1,365 16.8 158 1,150 80.9 1987 1,043 1,395 18.4 156 1,565 88.8 1988 1,166 1,533 20.7 198 2,551 95.3 1989 1,336 1,680 20.5 258 1,951 94.7 1990 1,404 1,705 20.8 273 1,639 100.0 1991 1,254 1,598 18.6 228 1,302 102.1 a/ All costs and prices in this table are measured in current US dollars. b/ C.i.f. import price. Sources: Anthony Bird Associates, Aluminum Annual Review 1992, March 1992; World Bank, International Economics Department. Caribbean, where significant increases in bauxite production are expected, bauxite exports are expected to decrease. Big increase in LMICs' share of alumina expoitr. Exports of alumina are expected to increase worldwide at an annual rate of 1.9%. This is comparable to the 1.8% p.a. increase projected for alumina production, indicating that a marginally rising share of alumina will be exported without being transformed into aluminum in the country of origin. The Latin America and the Caribbean region is representative of the pattern of an increasing portion of alumina produced being exported in the form of alumina. The share of LMICs in world exports of alumina should increase from 31% in 1991 to 44% by 2005. The high-income countries should remain the major importers of alumina, with Canada and the United States being the major importers, and Australia should continue to be the leading exporter (see Tables AS and A6). Faster aluminum trade growth expected. Exports of aluminum, in contrast, are expected to expand worldwide at an annual rate of 3.9%, which is faster than the expected growth of aluminum production. The shares of some LMICs (e.g., Bahrain and Venezuela) should rise as their production increases. Canada's exports are expected to increase at an average annual rate of 3%. The share of LMICs will grow at over 4% p.a., with exports from Latin America and the Caribbean growing at an average annual rate of about 5.7%. Since Japan does not plan to increase its production, it is expected to remain the largest importer of aluminum, with imports growing at an annual rate of about 2.4% (see Tables A9 and A10). 109 Price Outlook Like most primary commodities, aluminum prices have been highly volatile in recent years. Since the London Metal Exchange (LME) price was established in 1979, it has fluctuated significantly. Statistical tests carried out at the World Bank support the notion of the aluminum market being "efficient."' The volatility of the price is seen to be due mostly to fluctuations in industrial production, as well as to changes in aluminum production capacity. In addition, the "indexing" of some cost components in aluminum production may have contributed to the price volatility, as discussed above. Stocks rise with recession and increased FSU sales. Aluminum prices peaked in 1988, when the LME price averaged US$2,551/ton. But with the slowdown of industrial production growth, particularly in the US economy, prices declined in the 1989-91 period until the LME price reached a low monthly average of less than US$1,100/ton in December 1991. LME stocks of aluminum, which were as low as 57,000 tons at the end of 1989, rose steadily in 1990 and 1991 and in the first half of 1992, reaching a recent peak of over 1.3 million tons in April 1992. In addition to the worldwide recessionary tendency, one important factor fueling this surplus has been the rising sales of aluminum by the FSU. Hard-pressed for hard currency revenues and with the collapse of domestic consumption due to the industrial confusion associated with radical political changes, the FSU increased aluminum exports to western markets, selling out of stocks and increasingly out of current production. Price outlook remains weak for aluminum and bauxite. Aluminum prices are expected to increase in the 1992-95 period as the size of the market surplus is reduced and a balance between demand and supply is restored; the LME price is likely to rise from US$1,302/ton in 1991 to US$1,850/ton in 1995 in current dollar terms. After peaking around that level in 1995, aluminum prices are expected to enter the declining phase of another price cycle. The recent decline in aluminum prices has discouraged investments in smelter capacity. Many previously announced projects have been delayed. It is clear from Table 6 that the total costs were higher than LME prices in 1990 and 1991, and thus, any new smelter will be unable to cover full costs. The expected pick-up in aluminum prices should encourage some of the low-cost smelting projects to go ahead. Price increases, however, are likely to be only modest, given adequate stocks and the short-term supply response. As capacity expansions come on-stream, prices are expected to remain relatively weak (by historical standards) in real terms through 1995. For the longer term, the trend in prices is primarily determined by estimated production costs. Thus, our projections for aluminum prices for the 2000-2005 period are set at slightly above expected average total costs. We expect the average total cost of producing aluminum to decline gradually over the projection period, from US$1,565/ton in 1991 (in 1990 dollars) to US$1,400/ton by 2005-a declining trend of 0.7% p.a. Thus, we expect the price of aluminum (in 1990 constant dollars) to be at about US$1,465/ton in the year 2000 and US$1,435/ton in the year 2005. Unlike aluminum, bauxite is not traded on exchanges such as the LME or COMEX. Bauxite prices vary widely depending on the sources of supply, as quality and transport convenience differ from producer to producer. Spot prices also differ substantially from contract prices. It is estimated that roughly 75% of internationally-traded bauxite is priced on a contract basis, with the rest transacted I See L. Hobeika, "On the Efficiency of the Aluminum Market," presented at the Metal Bulletin 6th International Aluminum Conference in Singapore, Nov. 11-13, 1990. 110 on a spot basis. Naturally, contract prices tend to be more stable than spot prices. For the purpose of this report, the US import price is used as the indicator price for bauxite (see Table A 12). The price for a ton of bauxite in current dollar terms is expected to increase from US$34/ton in 1991 to $41/ton by 2000 and to $47/ton by 2005. These prices imply a slight increase in real terms to the year 2000, and then a slight decline to the year 2005, in line with the forecast for aluminum prices. Bauxite-exporting countries, such as Jamaica, Guyana, and Guinea, depend heavily on bauxite exports for their foreign exchange earnings. These countries cannot easily decrease their production whenever there is a slowdown in the major industrial economies. As long as the price of bauxite remains above their own estimated costs, they will continue to produce it. Bauxite prices, therefore, are likely to average at close to their recent low levels in real terms over the forecast period. 111 Table Al: Bauxite - Production By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 --------------------------------------(000 Tons)----- ----------------------------- -------(% p.a.)------ High-Income 15,203 30,291 42,376 41,052 41,055 42,055 42,055 43,055 47,105 52,105 8.4 3.9 1.7 OECD 15,203 30,291 42,376 41,052 41,055 42,055 42,055 43,055 47,105 52,105 8.4 3.9 1.7 France 3,010 1,897 490 0 0 0 0 0 0 0 -3.8 -7.1 0.0 Australia 9,970 26,734 41,391 41,000 41,000 42,000 42,000 43,000 47,000 52,000 20.6 5.7 1.7 LMICs 44,818 59,750 71,149 71,272 69,375 70,725 72,875 74,125 81,504 91,204 3.2 1.6 1.8 Africa 3,344 14,245 19,357 18,674 18,225 18,625 19,075 19,075 19,800 21,300 10.2 8.5 0.9 Guinea 2,526 13,371 17,524 17,054 16,700 17,000 17,300 17,400 17,900 19,200 11.0 9.6 0.9 Americas 24,703 23,706 26,360 28,969 28,000 28,600 28,600 29,520 32,900 37,700 0.9 -1.1 1.9 Jamaica 11,684 11,696 10,937 11,609 11,600 11,700 11,700 12,000 13,300 15,000 0.0 -2.9 1.8 Suriname 6,325 4,590 3,267 3,136 3,100 3,200 3,300 3,300 3,300 3,300 -1.4 -4.3 0.4 Guyana 4,319 2,934 1,424 2,204 2,000 2,000 2,000 2,000 2,300 2,500 -2.1 -5.1 0.9 Brazil 476 3,486 9,876 10,500 10,000 10,500 10,600 10,800 11,700 13,000 17.6 17.2 1.5 Venezuela 0 0 771 1,514 1,300 1,200 1,500 1,520 2,000 3,000 - - 5.0 Asia & Pacific 4,021 5,391 11,084 10,794 10,770 10,800 10,850 10,900 11,600 13,400 5.1 4.6 1.6 India 1,326 1,887 5,277 4,811 4,800 4,800 4,800 4,800 4,800 6,000 6.8 6.0 1.6 Indonesia 1,131 1,168 1,206 1,406 1,400 1,400 1,400 1,400 1,700 1,900 1.0 -2.8 2.2 Europe 12,750 16,408 14,348 12,835 12,380 12,700 14,350 14,900 17,904 20,904 2.4 0.4 3.5 Yugoslavia 2,062 3,133 2,952 2,542 2,500 2,500 3,000 3,000 3,000 3,100 3.5 3.0 1.4 Greece 2,367 3,014 2,504 2,224 2,200 2,400 2,500 2,550 4,100 5,000 2.7 -0.4 6.0 World 60,021 90,041 113,525 112,324 110,430 112,400 114,900 117,500 129,300 145,400 4.5 2.3 1.9 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: MetaLLgeseLtschaft, Metal Statistics, UNCTAD (actual); World Bank, International Economics Department (projected). Table A2: Bauxite - Gross Exports By Main Countries and Economic Regions ------------------------------------------------------------------------------------------------------------------------------- Actual Projected Growth Rates al Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 ------------------------------------------------------------------------------------------------------------------------------- --- ------------------------ -----------(000 Tons)---- --------(----------------------%-------- p.a.)-------- High-Income 3,904 6.735 5,344 5,459 5,270 6,175 6,100 5,900 9,900 12,100 7.9 -0.9 5.9 OECD 3,904 6,735 5,344 5,459 5,300 6,175 6,100 5,900 9,900 12,100 7.9 -0.9 5.9 Australia 3,705 6,540 5,200 5,300 5,200 6,050 6,000 5,750 9,550 11,600 14.4 -1.1 5.8 LNICs 23,565 29,036 30,082 31,642 28,300 27,250 27,600 28,160 25,750 25,270 1.8 0.8 -1.6 Africa 1,595 11,293 16,048 16,767 16,438 15,420 15,830 15,830 15,850 16,400 15.9 11.8 -0.2 Guinea 777 10,459 14,300 15,486 15,090 15,100 15,510 15,510 15,500 16,000 23.0 15.8 0.2 Americas 16,095 12,787 10,907 11,710 9,800 8,525 8,250 9,000 7,000 4,050 -1.6 -3.0 -7.3 Jamaica 7,722 5,987 4,169 4,170 3,000 2,000 1,500 1,500 1,400 600 -2.5 -4.7 -12.9 Suriname 3,565 1,707 0 0 0 0 0 0 0 0 Guyana 2,810 1,559 1,254 2,200 2,000 2,000 2,000 2,000 2,300 2,500 -1.7 -4.3 0.9 Asia & Pacific 2,057 2,113 3,300 3,050 2,450 2,400 2,400 2,300 2,100 2,300 0.1 -2.2 -2.0 Indonesia 988 1,053 1,240 1,400 1,400 1,400 1,400 1,400 1,400 1,400 0.3 -3.4 0.0 Europe 3,819 2,844 1,500 700 700 870 1,280 1,300 1,200 2,900 -2.0 -4.5 10.7 World 27,469 35,771 35,426 36,530 33,600 33,450 33,700 34,100 35,700 37,400 2.4 0.5 0.2 ------------------------------------------------------------------------------------------------------------------------------- a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: NetailgeseLlachaft, Metal Statistics, UNCTAD (actual); World Bank, International Economics Department (projected). Table A3: Bauxite - Gross Iports By Main Countries and Economic Regions -------------------------------------------------------------------------------------------------------------------------------------- Actual Projected Growth Rates a/ -------------------------------------- ---------------------------------------------------- -------------------------- Averages Countries/ ---------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 -------------------------------------------------------------------------------------------------------------------------------------- ---- -----------------------------------(000 Tons)----- ----------------------------- --------(% p.a.)------ High-Income 24,880 30,961 28,989 29,600 24,525 24,600 24,800 25,000 25,000 25,000 1.5 -0.7 -1.2 OECD 24,758 30,756 28,953 29,600 24,010 24,010 24,200 24,200 24,150 24,150 1.5 -0.7 -1.4 United States 13,584 14,043 13,816 14,150 13,350 13,350 13,430 13,430 13,380 13,380 -0.3 -2.0 -0.4 Canada 2,416 2,785 2,311 2,910 2,680 2,700 2,720 2,730 2,720 2,720 1.2 -0.4 -0.5 Germany 2,754 4,034 3,077 3,200 1,715 1,730 1,700 1,720 1,700 1,700 2.4 -0.5 -4.4 France 491 1,689 1,389 1,620 830 830 830 830 830 830 9.5 2.6 -4.7 Italy 621 2,095 1,681 2,015 1,875 1,900 1,915 1,940 1,930 1,930 6.3 2.6 -0.3 Japan 3,817 4,886 2,302 2,385 480 500 480 480 480 480 2.1 -4.3 -10.8 LMICs 2,447 4,588 8,763 8,000 9,000 9,050 9,100 9,330 11,000 12,750 9.3 7.6 3.4 Europe 2,373 4,391 6,232 6,000 5,970 6,115 5,650 6,355 6,430 6,200 7.8 4.9 0.2 World 27,327 35,549 37,752 37,410 33,500 33,575 33,900 34,125 35,735 37,500 2.6 0.7 0.0 -------------------------------------------------------------------------------------------------------------------------------------- a/ Least squares trend for historical periods (1961-90); end-point for projected Periods (1991-2005). b/ Estimate. Sources: Metaligesellschaft, Metal Statistics, UNCTAD (actual); World Bank, International Economics Department (projected). Table A4: Alumina - Production By Main Countries and Economic Regions Actual Projected Growth Rates/a Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 --------------------------------------('000 Tons)----------------------------- ------------(% p.a.)--------- High-Income 13,070 21,007 23,248 23,720 21,620 21,730 22,561 22,591 23,440 24,640 NA 2.3 0.3 OECD 13,028 20,954 23,248 23,720 21,420 21,530 22,315 22,345 23,195 24,395 NA 2.3 0.2 United States 6,089 6,623 5,430 5,430 5,000 5,000 5,030 5,030 5,030 5,030 NA -1.8 -0.5 Canada 1,083 1,078 1,087 1,087 1,000 1,000 1,015 1,015 1,015 1,015 NA -0.2 -0.5 Japan 1,317 1,886 890 890 180 180 180 180 180 180 NA -3.7 -10.8 Australia 2,265 7,247 11,231 11,703 11,590 11,700 12,420 12,450 13,300 14,500 NA 7.2 1.5 LMICs 8,222 12,954 18,382 18,205 19,060 19,810 20,568 20,869 25,100 29,050 NA 3.8 3.4 Americas 3,138 4,409 7,458 7,781 8,600 9,300 9,590 9,590 11,300 14,580 NA 3.0 4.6 Jamaica 1,625 2,340 2,869 3,000 3,300 3,800 3,925 3,925 4,500 5,450 NA -0.6 4.4 Suriname 1,091 1,333 1,531 1,531 1,550 1,550 1,600 1,600 1,600 1,600 NA 1.1 0.3 Brazil 124 487 1,653 1,700 1,850 2,050 2,065 2,065 3,200 4,530 NA 13.8 7.3 Europe 3,872 6,468 7,305 6,805 6,610 6,660 7,003 7,304 8,195 8,365 NA 3.6 1.5 World 21,291 33,961 41,630 41,925 40,680 41,540 43,129 43,460 48,540 53,690 NA 2.9 1.8 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: MetalLgeselLschaft, Metal Statistics, UNCTAD (actual); World Bank, International Economics Department (projected). Table AS: Alumina (Metal Contents) - Gross Exports By Main Countries and Economic Regions -------------------------------------------------------------------------------------------------------------------------------------- Actual Projected Growth Rates/a Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 --------------------------------- ----(000 Tons)---------------------------- -------------(% p.a.)--------- High-Income 1,784 4,555 6,757 7,138 6,848 6,893 7,175 7,191 7,028 7,482 11.9 5.6 0.3 OECD 1,784 4,555 6,757 7,138 6,848 6,893 7,176 7,191 7,028 7,482 11.9 5.6 0.3 United States 533 403 550 400 300 300 200 200 0 0 4.0 -1.4 -60.2 Australia 902 3,352 4,150 4,679 6,024 6,069 6,348 6,369 6,496 6,951 6.7 2.9 LNICs 2,041 3,000 2,980 3,200 3,495 3,633 3,796 3,843 4,770 5,887 4.8 1.0 4.5 Africa 304 342 300 303 315 315 365 365 426 426 0.7 -0.4 2.5 Guinea 304 342 300 303 315 315 365 365 426 426 0.7 -0.4 2.5 Americas 1,431 1,903 1,840 1,971 2,229 2,430 2,550 2,597 2,895 3,943 4.8 0.3 5.1 Jamaica 790 1,173 800 1,050 1,155 1,330 1,374 1,374 1,575 1,907 3.5 -1.3 4.4 Suriname 485 642 800 811 822 822 848 848 848 848 2.0 0.3 Europe 303 718 775 880 863 800 784 785 1,125 1,182 8.9 3.9 2.1 World 3,824 7,556 9,737 10,337 10,343 10,526 10,971 11,034 11,797 13,369 8.2 3.7 1.9 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: MetaLigesellschaft, Metal Statistics, UNCTAD (actual); World Bank, International Economics Department (projected). Table A6: Alumina (MetaL Contents) - Gross Imports By Main Countries and Economic Regions Actual Projected Growth Rates/a Averages Countries/ 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 ---- ---------------------------- ------('000 Tons)---------------------------- ------------ Mp.a.)--------- High-Income 3,047 5,490 6,238 6,718 7,353 7,722 7,870 8,161 8,640 9,738 7.0 2.8 2.7 OECD 3,040 5,411 5,998 6,346 7,194 7,571 7,769 8,066 8,555 9,282 6.8 2.6 2.8 United States 1,111 2,059 2,050 2,182 2,412 2,412 2,394 2,394 2,394 2,394 11.0 2.3 0.7 Canada 472 495 1,000 1,256 1,330 1,462 1,550 1,739 2,061 2,165 3.8 5.2 4.0 Japan 178 359 30 30 30 30 30 30 30 30 7.2 -13.6 0.0 LMICs 717 1,880 2,015 2,518 2,413 2,248 2,610 2,716 3,380 3,962 11.7 5.2 3.3 Africa 164 320 405 419 418 418 418 424 697 927 8.4 3.8 5.8 Americas 62 466 312 808 702 546 575 575 640 640 16.5 10.6 -1.7 Europe 463 768 454 642 583 506 502 476 432 483 9.0 0.1 -2.0 orld 3,765 7,370 8,254 9,236 9,767 9,970 10,480 10,877 12,020 13,701 7.8 3.4 2.9 -4 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: MetaLLgesellschaft, Metal Statistics, UNCTAD (actual); World Bank, International Economics Department (projected). Table A7: Primary Aluminun - Production By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 --- ------------------------------- ----('000 Tons)---- --------------------------------------(% p.a.)- High-Income 7,493 10,572 10,747 11,036 10,827 11,132 11,449 11,677 12,452 13,382 3.4 1.0 1.4 OECD 7,468 10,475 10,573 10,797 10,587 10,892 11,209 11,437 12,212 12,962 3.3 1.0 1.3 United States 3,536 4,566 4,048 4,121 4,100 4,100 4,100 4,100 4,100 4,100 2.0 -0.3 0.0 Canada 981 1,019 1,567 1,822 1,860 2,000 2,100 2,300 2,640 2,750 2.8 3.1 3.0 Germany 390 754 720 710 710 710 720 720 720 720 4.9 2.0 0.1 France 379 421 326 286 240 380 420 440 440 440 0.2 -1.1 3.1 Australia 185 318 1,233 1,229 1,230 1,250 1,390 1,390 1,600 1,800 13.4 11.1 2.8 LMICs 2,781 5,080 7,290 7,428 7,351 7,384 7,881 8,078 9,521 10,890 6.6 4.8 2.8 Africa 172 319 422 432 431 431 433 438 663 853 7.9 3.6 5.0 Americas 169 758 1,789 1,998 2,048 2,078 2,130 2,130 2,786 3,450 16.7 13.7 4.0 Brazil 60 252 931 1,140 1,160 1,180 1,200 1,200 1,200 1,200 15.7 15.2 0.4 Venezuela 19 282 594 610 640 650 670 670 1,200 1,450 21.0 6.4 Asia & Pacific 308 583 1,499 1,545 1,545 1,552 1,578 1,603 1,893 2,303 9.4 8.5 2.9 co China, People's Rep. 137 353 850 850 850 850 850 850 950 1,050 9.3 9.5 1.5 India 157 203 433 504 504 504 510 510 700 900 7.6 4.6 4.2 Europe 2,129 3,154 3,130 2,997 2,867 2,867 3,090 3,207 3,299 3,379 4.2 1.8 0.9 Middle East & North Africa 3 265 451 456 460 456 650 700 880 905 NA NA 5.0 Bahrain 3 131 212 208 208 208 400 450 450 450 NA NA 5.7 World 10,274 15,652 18,037 18,464 18,178 18,516 19,330 19,755 21,973 24,272 4.4 2.3 2.0 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: MetallgeselLschaft, Metal Statistics & World Bureau of Metal Statistics (actual); World Bank, International Economics Department (projected). Table A8: Primary Aluminum - Apparent Consumption By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 ---- --------------------------- ------('000 Tons)----- ------------------------------ --------(X p.a.)------ High-Income 7,735 10,696 12,269 12,384 12,528 12,904 13,303 13,081 14,438 15,635 3.8 1.7 1.7 OECD 7,679 10,549 11,994 11,984 12,109 12,464 12,833 12,601 13,813 14,845 3.7 1.6 1.5 United States 3,707 4,561 4,325 4,236 4,280 4,350 4,450 4,400 4,600 4,731 2.3 0.3 0.8 Germany 976 1,344 1,379 1,423 1,432 1,465 1,500 1,435 1,535 1,620 4.0 1.7 0.9 France 386 579 721 708 710 752 785 765 829 888 4.1 2.8 1.6 Japan 897 1,671 2,414 2,443 2,470 2,570 2,665 2,650 3,090 3,496 8.4 3.4 2.6 LMICs 2,530 4,620 5,651 5,431 5,510 5,777 6,149 6,293 7,535 8,637 6.0 3.8 3.4 Americas 200 544 733 708 711 782 858 839 1,030 1,196 9.3 6.4 3.8 Asia & Pacific 416 1,025 1,842 1,884 1,930 2,042 2,195 2,256 2,795 3,390 9.8 6.9 4.3 Europe 1,829 2,824 2,638 2,436 2,455 2,522 2,644 2,729 3,157 3,414 4.1 1.9 2.4 Middle East & North Africa 22 118 316 262 264 278 297 311 382 451 13.9 4.0 World 10,265 15,316 17,998 17,815 18,038 18,681 19,453 19,374 21,973 24,272 4.4 2.3 2.2 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: MetalLgeselLschaft, Metal Statistics & World Bureau of Metal Statistics (actuaL); World Bank, International Economics Department (projected). Table A9: Primary Auamin - Gross Exports By Nain Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 ..........(000 Tons)--------- --------( p.a.)-------- High-Income 1,998 3,177 5,579 5,924 5,930 6,534 7,118 7,171 8,501 9,853 4.8 5.3 3.7 OECD 1,998 3,145 5,406 5,808 5,820 6,414 6,988 7,021 8,301 9,553 4.6 5.0 3.6 Canada 791 687 1,253 1,470 1,515 1,638 1,735 1,921 2,202 2,212 2.3 3.3 3.0 United States 261 381 798 902 900 900 950 950 1,000 1,000 2.9 6.9 0.7 Netherlands 72 359 362 332 330 325 325 325 370 400 13.7 5.1 1.3 France 154 175 135 128 130 140 150 145 170 200 -0.9 -1.4 3.2 Germany 62 250 314 290 280 290 310 290 350 400 8.3 7.2 2.3 Australia 62 82 910 951 920 970 1,100 1,130 1,350 1,500 16.3 3.3 LNICs 802 1,458 2,951 3,200 3,300 3,275 3,658 3,493 4,920 5,660 9.2 6.3 4.2 Africa 149 195 290 306 310 340 375 335 590 715 6.1 3.5 6.2 Ghana 111 151 150 166 165 165 165 165 165 165 -0.6 0.0 Americas 54 296 1,092 1,150 1,337 1,300 1,270 1,270 2,100 2,500 19.6 5.7 Europe 587 806 1,127 1,103 993 1,010 1,240 1,180 1,280 1,390 6.3 2.3 1.7 World 2,800 4,636 5,530 9,124 9,230 9,810 10,775 10,665 13,420 15,510 5.8 5.6 3.9 --------------------------------------------------------------------------------------------------------------------------------------- a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: MetaUlgesellschaft, Metal Statistics, UNCTAD (actual); World Bank, International Economics Department (projected). Table AI0: Primary ALuminumx - Gross Imports By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 ---- -----------------------------------('000 Tons)----- ----------------------------- --------(% p.a.)------ High-Income 2,343 3,725 7,446 7,631 7,895 8,343 9,230 8,985 11,265 12,975 6.8 6.2 3.9 OECD 2,312 3,611 7,099 7,205 7,510 7,915 8,745 8,470 10,435 12,180 6.6 6.0 3.8 United States 417 563 976 1,039 1,195 1,550 2,135 1,920 3,200 4,040 4.5 5.5 10.2 Germany 498 582 899 1,007 953 960 975 930 1,060 1,170 5.6 3.3 1.1 France 164 316 541 552 565 445 440 390 465 540 8.4 5.4 -0.2 Netherlands 45 182 190 183 170 180 180 190 190 190 10.9 6.3 0.3 Japan 266 929 2,652 2,830 2,790 2,845 2,975 2,990 3,495 3,960 18.7 12.8 2.4 LMICs 380 896 1,286 1,318 1,335 1,465 1,555 1,685 2,150 2,540 9.1 6.1 4.8 Asia & Pacific 65 322 601 708 735 775 830 880 1,175 1,395 14.4 12.7 5.0 Europe 224 455 608 536 525 605 645 705 975 1,150 10.9 4.9 5.6 , World 2,723 4,621 8,732 8,950 9,230 9,810 10,785 10,670 13,410 15,515 7.1 6.2 4.0 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: Metatigeselischaft, Metal Statistics & World Bureau of Metal Statistics (actual); World Bank, International Economics Department (projected). Table All: Primary Aluminum - Prices, 1958-91 (Actual) and 1992-2005 (Projected) 1990 Con~an 8 u---c rrent -4[-5 MUV a/l--7 CPI bl Other ara Other European Other Euroea LME el Trasactiona d/ IUE c/ Transactione d/ IJMB cl Tranactiona d/ Actual 1958 490 2,383 3,129 1959 486 2,397 3,109 1960 500 2,416 3,139 1961 470 2,233 2,883 1962 435 2,026 2,595 1963 444 2,108 2,566 1964 493 2,300 2,777 1965 489 2,265 2,669 1966 479 2,144 2,529 1967 484 2,141 2,491 1968 457 2,041 2,288 1969 581 2,462 2,788 1970 540 2,153 2,442 1971 435 1,646 1,843 1972 432 1,500 1,659 1973 663 1,987 2,228 1974 944 2,322 2,864 1975 690 1,526 1,863 1976 862 1,881 2,253 1977 991 1,969 2,343 1978 1,045 1,804 2,017 1979 1,603 1,520 2,442 2,316 2,923 2,772 1980 1,775 1,730 2,466 2,404 2,894 2,821 1981 1,263 1,336 1,748 1,849 2,043 2,161 1982 992 1,061 1,394 1,491 1,625 1,738 1983 1,439 1,495 2,069 2,150 2,320 2,411 1984 1,251 1,371 1,839 2,015 2,023 2,217 1985 1,041 1,112 1,517 1,621 1,649 1,762 1986 1,150 1,261 1,422 1,559 1,510 1,656 1987 1,565 1,608 1,762 1,810 1,820 1,870 1988 2,551 2,545 2,677 2,670 2,755 2,748 1989 1,951 2,040 2,062 2,155 2,117 2,213 1990 1,639 1,721 1,639 1,721 1,639 1,721 1991 1,302 1,390 1,275 1,361 1,239 1,322 Proiected 1992 1,260 1,183 1,129 1993 1,450 1,312 1,254 1994 1,650 1,466 1,394 1995 1,850 1,600 1,519 2000 2,011 1,465 1,370 2005 2,261 1,435 1,288 al Deflated by G-5 Manufacturing Unit Valuc (MV) Index. bl DefIld by 0-7 Conumer Price Index (CP). c/ London Metal Exrhange.(IME), unalloyed ia ots, average bidlaakd, a published by the Metala Week. Uni November 1988, standard grade, måni- 99.5% al; thereafter, gh gradå, minimm 99.7% a, cash price. d/ Certain other tranactione, US åhipments to Europe, Minimum 99.5%, in-warehouse, EC duty paid, c.i.f. Europe. Thit quotation has been publisbed regularly by the MetalB since December 1975. Sources: Meallgesellschaft AG, Metal Statistict, Metal Bulletin & Metals Week (actual); World Bank, International Economics Department (projected). 122 Table A12: Bauxite - Prices, a/ 1975-1991 (Actual) and 1992-2005 (Projected) s/ton ) Current S - 1990 Constant S- 0-5 MUV b/ G-7 CPI c/ Actual 1975 22 49 60 1976 25 5 66 1977 30 60 72 1978 31 54 60 1979 31 47 56 1980 32 44 52 1981 35 48 57 1982 40 57 66 1983 37 54 60 1984 37 54 60 1985 36 52 57 1986 34 42 44 1987 30 33 34 1988 30 32 33 1989 34 36 37 1990 34 34 34 1991 34 33 32 Project d 1992 32 30 29 1993 33 30 29 1994 34 30 29 1995 36 31 30 2000 41 30 28 2005 47 30 27 a/ US import price, c.i.f. US ports. b/ Deflated by G-5 Manufacturing Unit Value (MUV) Index. c/ Deflated by 0-7 Consumer Price Index (CPI). Sources: US Bureau of Mines, Minerals Yearbook: Bauxite and Alumina, various issues (actual); World Bank, International Economics Department (projected). 123  · 단―!--[’· Projected Iron Ore Consumption, by Region 700- 600- r 400- 500- MMM#M E 300- 200- i00-- 1990 1995 2000 2005 High Income M LMIC Asia LMIC Arnertca M LMIC Europe Demand Outlook Broadly, there are three kinds of iron ore, distinguished by their physical appearance and purity. The most common is called "sinter feed." About 60% of iron ore consumed by the steel industry undergoes a process known as "sintering" prior to smelting. Sintering is normally carried out in steel mills. It is an agglomeration process, which involves roasting of iron ore concentrates (sinter feed or fines) in specially-built machines to produce a self-fluxing sinter. High sinter ratios (up to 70% of blast furnace burden is sinter) are common in the Japanese and European steel mills but not in United States and Canada where pellets, the second type of iron ore, form the main ore burden in blast ffirnaces. Pelletization is generally undertaken as a part of iron ore mining and treatment activities. The third kind of iron ore is called directly-reduced iron (DRI) or sponge iron. It is used mainly in mini-mills as a substitute for steel scrap. However, the DRI technology has not gained much popularity as yet, currently accounting for no more than 1 % of ferrous input. Since DRI is mostly produced through a gas-based direct reduction route, countries with abundant natural gas may have the comparative advantage in its production. Global environmental concerns favor direct smelting technologies that require DRI as the feed. Improvements in DRI technology, such as the production of hot briquetted iron (HBI), have made sponge iron a more stable product, suitable for shipment. Iron ore demand may be directly derived from crude steel production, but the ratio of iron ore consumption per ton of steel production may vary over time or across regions depending on the use of steel scrap. Table I presents estimates of this ratio for the 1980s (iron ore consumption is defined as production plus imports minus exports). High-income countries have a lower ratio than LARCs because they use more steel scrap (mini-mills use 100% steel scrap and in the United States contribute 25% of 125 Table 1: Ratio of Iron Or. Consumption to Steel Production, by Region, 1980-90 Regions 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 High-Income 64.4 61.1 59.5 51.7 58.1 59.6 59.5 60.1 57.8 60.2 59.8 OECD 64.6 61.3 59.6 51.6 58.1 59.7 59.6 60.1 57.8 60.3 60.0 IMICa 78.4 80.1 77.8 75.1 75.4 77.0 76.8 75.5 76.6 77.9 79.0 Americas 82.9 73.2 87.5 81.0 78.3 83.3 88.1 90.8 96.7 95.8 112.0 Asia 130.0 136.9 114.4 110.9 113.9 114.3 116.2 110.6 106.6 107.2 105.3 Europe 71.5 70.1 71.7 70.1 69.5 70.1 67.2 65.4 67.9 68.5 67.5 World 70.5 69.4 68.4 63.1 66.4 68.0 68.4 68.1 67.3 69.1 69.4 Source: World Bank, International Economics Department. steel production). In the Asia and Pacific region of the LMIC group, however, the ratio exceeded 100% for all 11 years because of the low availability of steel scrap. About 30% of global steel production comes from scrap. Slow growth in second half of the 1990s, stronger after 2000. During the 1962-73 period, world steel production grew at 5.4% p.a., while iron ore consumption increased by 5% p.a.. During the 1974-90 period, however, the growth rate of steel production dropped to 0.2% p.a., and ironore consumption growth fell to 0.6% p.a. Due to declines in crude steel output in the high-income countries, and in Eastern Europe and the FSU, world iron ore consumption declined 1.6% in 1990 and 4.5% in 1991. Only the Asia and Pacific region registered positive growth in these two years. World steel production is projected to grow at 1.4% p.a. over the 1991-2005 period (see the steel chapter). This implies a 1.3% p.a. growth in world iron ore demand. As for steel production, the demand outlook for iron ore is sharply different between high income and LMICs (see Table 2). Since global steel production is not expected to recover fully until the mid-1990s, world iron ore consumption is expected to decline slightly (-0.14% p.a.) over the near term (1990-95). The outlook for the long-term is somewhat brighter. Steady growth in steel production in the 2000-05 period should translate into solid growth in iron ore consumption in all regions. By then, steel industries in Eastern Europe and the FSU are likely to have emerged from the current transition and to require increasing amounts of steel and iron ore. However, by 2000-05 Latin America is expected to have reached a saturation point in steel production and hence in iron ore consumption with their competitive position in the world steel market deteriorating. Supply Outlook Iron ore is an abundant resource. Table 3 shows mine production (in gross weight), reserves, and reserve base estimates for major iron ore-producing countries. At the current level of world iron ore production, the existing economically recoverable iron ore reserves can last for another 100 years. Major iron ore producers (ranked by average production in 1990-91) are: the FSU, Brazil, Australia, China, United States, India, and Canada. Iron ore mined in the FSU, China, and the United States is of low quality (in terms of iron content, at about 30%) and therefore its production is less 126 Table 2: Projected Growth Rates of Iron Ore Consumption, 1990-2005 1990-95 1995-2000 2000-2005 1990-2005 (% p.a.) High-Income -0.57 0.81 0.95 0.40 OECD -0.61 0.76 0.92 0.36 United States -2.37 -0.03 -0.04 -0.81 Japan -0.12 0.99 1.19 0.69 Germany -0.92 1.35 1.36 0.60 France -0.15 0.85 0.79 0.50 United Kingdom -1.75 0.24 0.12 -0.46 Italy -0.83 1.55 1.56 0.76 LMICs 0.18 2.41 2.16 1.58 Americas 4.18 0.78 0.47 1.81 Asia 4.63 3.37 2.81 3.60 China, People's Republic 3.84 3.00 2.53 3.12 Korea, Republic of 6.81 3.19 2.74 4.25 Europe -3.53 1.75 1.76 -0.01 World -0.14 1.75 1.68 1.10 Source: World Bank, International Economics Department. economically viable than in Australia, South Africa, India, and Brazil which have high-quality ore reserves (around 60% iron content). In terms of mine capacity and reserves, Australia and Brazil are and should continue to be the two dominant producers. After reaching a record level in 1989, world iron ore production fell 5% in the recent two years. Declines took place mostly in the FSU (-19%) and Liberia (-250%), in both cases because of abnormal situations (economic transition in the FSU and civil war in Liberia). Australia registered a robust growth of 6% in the two years, while Brazil's production remained flat. Other countries more or less managed to maintain their productions levels. Capacity expansions in near term. Owing to relatively favorable iron ore prices during the 1989-91 period, major iron ore producers have undergone significant capacity expansions. Most of them are in export-oriented mines in Australia and West Africa. Table 4 lists the expected iron ore capacity expansions for the near term (1991-95)-West Africa has four major projects at various stages of development, and about 30 million tpy of capacity is expected be brought on-stream; Brazil's CVRD is spending $130-150 million dollars to increase the capacity of the Carajas mine (in Northern Brazil) by 15 million tpy; Venezuela's CVG Ferrominera is concentrating on increasing its DRI production capacity in response to strong domestic and export demand; Australia is the most active country in terms of mine capacity expansion, focusing on both increasing the productivity of existing mines and expanding capacities within existing operations (no significant greenfield projects have been announced); LKAB of Sweden is expected to implement a $600 million capital expenditure program to renovate the Kiruna mine and expand pelletizing capacity; India has decided to go ahead with its third major iron ore mine in order to supply the DRI plant in Andhra Pradesh; and Iran and Pakistan have plans to exploit new iron ore mines in the 1990s. 127 Table 3: World Mine Production, Reserves, and Reserve Bane, by Major Producers, 1990 and 1991 Crude Ore Iron Content Reserve Reserve Mine Production eggyv a/ Base b/ Reserves 1990 1991 1991 1991 1991 1991 (million tons) United States 56.4 53.7 16,100 25,200 3,800 6,000 Australia 112.0 115.0 16,000 28,100 10,200 17,900 Brazil 154.4 155.0 11,100 17,300 6,500 10,100 Canada 36.4 38.0 11,900 25,500 4,600 10,000 China 108.0 116.0 9,000 9,000 3,500 3,500 France 8.7 8.0 2,200 2,200 900 900 India 53.7 56.0 5,400 12,100 3,300 6,300 Liberia 4.1 0.5 900 1,600 500 800 Mauritania 11.4 11.0 400 700 200 300 South Africa 30.3 31.0 4,000 9,300 2,500 5,900 Sweden 19.9 20.0 3,000 4,600 1,600 2,400 FSU 236.2 216.0 63,700 78,000 23,500 29,000 Venezuela 20.4 21.0 2,000 3,300 1,200 1,700 Other 67.4 65.0 5,400 12,500 2,300 6,300 World Total 919.3 906.2 151,100 229,400 64,600 101,100 al That part of the reserve base that could be economically extracted or produced at the time of determination. b/ That part of an identified resource that meets specified minimum physical and chemical criteria related to current mining and production practices. Source: United States Bureau of Mines, Mineral Industry Surveys (Iron Ore in 1991). Production to expand from 1993. World iron ore production is expected to continue declining in 1992 (-4.9% p.a.), before recovering in 1993 (1.4% p.a.) and 1994 (2.7%). The projected average growth rate is 1.3% p.a. over the 1991-2005 period. The high-income countries as a group should see their iron ore output drop significantly in 1992 and 1993. Most countries in this group are expected to lose ground in iron ore production compared with their 1990 levels; only Australia is expected to gain with a 7 million tons' additional to its annual production by 2000 and 15 million tons by 2005. The LMICs should accomplish a 1.9% p.a. average annual growth rate in iron ore production during the 1991-2005 period. The centers of this growth will be in Latin America and Asia. Brazil should utilize its high-quality iron ore reserves and strong competitive position to expand production at 4.7% p.a. over the forecast period. Iron ore production in Asia should be fueled by the region's strong growth in steel production. China is likely to seek new high-grade iron ore mines (one possible site is in the Hainan island). China's production is expected to grow at 3.3% p.a. during 1991-2005. Economic transition in Eastern Europe and the FSU will greatly alter the historical pattern of production and consumption; the FSU will no longer be the exclusive iron ore supplier in the region. We expect that iron ore production in the FSU will not be able to regain its pre-transition level during the 1991-2005 period. 128 Table 4: Expected Iron Ore Capacity Expansions, to 1995 Annual Production Country Supplier Mine Capacity (Mill. Tons) Guinea Nimco Nimba 7.5-9.0 Gabon Somifer Mekambo 10.0 Senegal Miferso Faleme 3.0-12.0 Mauritania SNIM M'haoudat 6.0 Brazil CVRD Carajas 15.0 Brazil MBR Itabirito 3.4 Brazil Samarco Alegria 4.0 Venezuela CVG DRI facility 5.0 Australia BHP Mt. Whaleback 3.0 Australia BHP Yandicoogina 5.0 Australia BHP McCamery's Monster 5.0 Australia BIP Goldsworthy ? Australia Hamersley Marandoo 10.0 Australia Hamersley Channar 2.5-7.0 Australia Robe River Robe River 7.0 India NMDC DRI facility ? Source: Drewry Shipping Consultants Ltd, NGrowth Prospects for Iron Ore and Coking Coal," 1991. Rapid growth using DRI/HBI technology. World production of DRI/HBI is expected to grow rapidly because: (i) mini-mills with the EAF technology will greatly increase their share of world steel output; and (ii) the supply of high quality steel scrap will be rather limited. Total DRI/HBI production is foreseen to climb to 34-45 million tpy (from the current 18 million tpy) by the end of this century, equivalent to a rise from 1% to 2% in its share of total world ferrous input. 7),ade Outlook More than 40% of the world's iron ore demand is met by imports, of which 90% is transported by sea. About 20% of the iron ore trade takes the form of pellets, 10% are lumps, and the rest are fines for sintering or direct blast furnace feed. Table 5 compares world seaborne trade in different types of cargos. Iron ore ranks second in terms of tonnage and ton-miles to crude oil, which illustrates the enormous size of the world steel industry and the fact that the leading steel producers in Europe and Japan have to import virtually all of their iron ore requirements amounting to 65% of the total iron ore trade in 1990. Another 14% went to Asian countries other than Japan. Long-term contmeti dominate market. Iron ore is generally traded on a long-term contract basis; only marginal tonnages enter the "spot" market. Brazil's CVRD, for example, has several 15-year contracts extending to 1999 covering deliveries of more than 10 million tpy to the Japanese 129 Table 5: World Seaborne Trade Volumes, 1975-90 Tonnages Crude Oil Iron Other (Mills.) Oil Products Ore Coal Grain Cargo Total 1975 1,263 233 292 127 137 995 3,047 1980 1,320 276 314 188 198 1,310 3,606 1985 871 288 321 272 181 1,360 3,293 1990 1,190 336 347 342 192 1,570 3,977 Tor-Miles Crude Oil Iron Other (Bills.) Oil Products Ore Coal Grain Cargo 1975 8,885 845 1,504 618 734 2,810 15,396 1980 8,219 1,020 1,651 957 1,087 3,720 16,654 1985 4,007 1,150 1,702 1,473 1,004 3,750 13,086 1990 6,261 1,560 1,978 1,849 1,073 4,400 17,121 Source: The TEX Report, Feb. 6, 1992. market. Contracts are usually negotiated between steel producers or their intermediaries and iron ore suppliers, and they can cover as little as one year or as long as the anticipated "life" of a particular mine. Contracts typically fix the quantities to be traded in each year (or allow relatively narrow upper/lower bounds), but leave the price open for "annual renegotiation." Steel producers have strong incentives to secure long-term supplies of iron ore. Besides long-term contracts, they also make significant investments (as partners) in iron ore mines known as "captive mines." Examples of "captive mines" are investments of Germany's Thyssen AG in mines in Liberia and Brazil, Italy's ILVA SpA in Brazil and Canada, and China's CMIEC in Australia. Major iron ore importers are large integrated steel mills in Europe and East Asia (Japan and Korea). They have a fairly well established pattern of supply based on medium- and long-term contracts with major iron ore suppliers. In the European market, contracts are usually written between iron ore suppliers and individual steel companies. An exception is Germany where two trading agents handle all the purchasing and transportation businesses. In Japan, the six large integrated steel companies tend to cooperate with each other in iron ore trading activities. Other major importers are integrated steel mills in China, the Republic of Korea, and Turkey, and some smaller, DRI-based mini-mills in Malaysia, Indonesia, and the Middle East. Major iron ore exporters are located in Australia, Brazil, Canada, India, and in African countries. Since the 1970s, Australia and Brazil have become the leaders in contract negotiations with iron ore importers by virtue of their ability to supply large quantities of high-grade ore at relatively low prices. Shipping costs important determinant of trade patterns. Importers are usually responsible for transportation of the iron ore because the majority of contracts are on f.o.b. terms. Importers can use this arrangement as a means of controlling inventories and final import costs. Some long-term contracts involving long-distance shipping are on c&f terms. A notable example is Australian exports to Europe because of the long distance involved. Such an arrangement can work to the advantage of both exporters and importers because exporters can cut a better deal with shipping companies due to 130 Table 6: Iron Ore Freight Costs (as of 1991) (US$/dwct a/) Ship Size (dwt) b/ 65,000 120,000 180,000 Sweden To North West Europe Operating +Voyage c/ 2.92 2.40 2.12 Operating +Voyage+ Capital d/ 5.35/5.44 4.33/4.42 3.89/3.97 Brazil To North West Europe Operating +Voyage 5.37 4.23 3.74 Operating +Voyage+ Capital 10.99/11.19 8.58/8.79 7.60/7.78 West Australia To North West Europe Operating+Voyage 9.51 7.56 5.93 Operating +Voyage+ Capital 16.60/16.85 13.14/13.41 11.47/11.73 a/ dwct: deadweight cargo tons. b/ dwt: deadweight tons. c/ Voyage cost includes bunker costs, port charges and canal tolls. d/ Newbuilding/secondhand. Source: Drewry Shipping Consultants Ltd. their ability to place large orders. An innovative use of f.o.b. and/or c&f type contracts is the "freight sharing" system adopted between Japanese steel mills and Brazilian iron ore suppliers where the differences in freight costs for shipments between Japan-Brazil and Europe-Brazil are shared by both parties. Table 6 shows freight costs for different iron ore suppliers to Europe. The main factors affecting shipping costs are: (i) size of ship; (ii) age of ship; and (iii) shipping distance. Sweden has the lowest shipping cost to the European markets, and f.o.b. prices of its iron ore are normally set at a premium. Shipping costs can be prohibitively high for particular iron ore suppliers (e.g., as high as 50% of the f.o.b. price for Australian iron ore in the Europe market). Table 7 presents recent data on iron ore trade from Australia, Brazil, and Canada to major destinations. Due to the high shipping costs, only 19% of Australia's exports goes to Europe while 80% goes to Japan and the rest of Asia. However, because the "freight sharing" system between Japan and Brazil, Brazil's iron ore exports to Japan and Europe are much more balanced than are Australia's. Canada has been concentrating mostly in the US and European markets. Strong import growth in LMIC Asia. The current structure of world iron ore trade is likely to persist. The high-income countries will continue to be net importers and the LMICs net exporters. Because of the projected near flat growth in iron ore production, iron ore imports will continue to grow faster than consumption in the high-income countries. The two most important iron ore importers, Japan and Germany, should see their imports grow at 1.2% p.a. and 1.3% p.a., respectively, over the 1991-2005 period. Australia and Brazil should further strengthen their position as major exporters with 1.9% p.a. and 2.9% p.a. growth, respectively, over 1991-2005. Gross iron ore imports of the LMICs are expected to increase at 2.5% p.a. The strongest growth is expected in Asia. Thus, the share of LMICs in world gross imports is expected to increase from 20% in 1991 to 24% in 2005. 131 Table 7: Iron Ore Exports of Major Suppliers (million tons, gross weight) EC-12 US Japan Asia al Total Australia 1986 14.5 - 47.3 16.6 79.5 1990 18.9 0.1 52.2 24.4 96.2 Brazil 1986 37.1 3.8 28.3 7.3 92.3 1990 48.4 4.1 31.6 13.4 114.3 Canada 1986 17.9 9.5 2.1 - 31.0 1990 14.5 9.2 1.8 - 27.0 a/ Includes China, Republic of Korea, and other Asian economies. Source: UNCTAD Trust "Iron Ore Statistics 1991." Changes in Eastern Europe and FSU Imde. Future developments in the FSU iron ore industry remain highly uncertain. However, the current consensus seems to be that: (i) Eastern Europe and the FSU will experience a sizable reduction in crude steel production and iron ore consumption; (ii) Eastern European steel mills will look for supply sources other than the FSU, because FSU iron ore has lower quality and the past favorable deals no longer exist; and (iii) the FSU will have to restructure its iron ore industry significantly. Price Prospects Prices for iron ore are determined by annual negotiations at specific times of the year. In Europe, contracts run on the calendar year basis and price talks are normally conducted in November or December. In Japan, contracts are on a fiscal year basis (i.e., April-March), and thus negotiations over price do not normally commence until January or February. Prces set during annual negotiations. Negotiations usually take place in the two major markets independently (i.e., Europe and Japan). In both markets, there is a dominant supplier whose price agreement with an importer serves as the reference for subsequent negotiations. In the European market, the reference price is the c&f Rotterdam price of Brazil's CVRD fines. This reference price, negotiated between Brazil's CVRD and German steel mills, takes into account expectations of market conditions, exchange rates, and ocean freight rates. Once the benchmark price is known, other suppliers to the European market adjust their f.o.b. prices to enable them to compete on a c&f Rotterdam basis. Australia has the transportation advantage in the Pacific region, particularly in the large Japan market. Producers like Hamersley and BHP are the market leaders. A unique feature of price negotiations in the Japanese market is that both parties agree on a percentage change in the reference f.o.b. price, which is then applied to all import brands. 132 There have been close interactions between the two markets, particularly over the past six to seven years. Table 8 shows the settlement prices of the past 13 years for the two major brands of iron ore. As can be seen the differences between columns (4) and (1) have been small and almost unchanged over the past six years. This implies that iron ore from Hamersley in the Japanese market and CVRD in the European market are treated as essentially identical (the small gaps in prices are due to slight differences in purity). However, even under the freight sharing system between the Tapanese steel mills and Brazilian iron ore suppliers, the differences between CVRD and Hamersley f.o.b. prices (columns (1) and (2)) are relatively large, and the differences between CVRD f.o.b. prices in the European and Japanese markets (columns (4) and (2)) are even larger. Negotiations in the European market have tended to be delayed in recent years, with the Japanese steel mills able to set the tone for the market. With better organization among themselves, Japanese buyers have been able to obtain price concessions from Australian iron ore suppliers. Table 9 shows that in the past 12 years the European market set the price seven times and the Japanese market, five times. The CVRD f.o.b. price of Standard Sinter Feed (64.5% purity ores from Itabira and other southern mines, or Southern System) in the European market (column (4) in Table 8) is the World Bank's indicator price for iron ore. Figure 1 shows the historical movements of the indicator price for iron ore and the World Bank Steel Price (WBSP) index. Iron ore prices move closely with steel prices but with a lag. During the 1970s, the lag seems to have been about two years. In the early 1980s, the lag was reduced to 1.5 years, and most recently, to about 1.25 years. Such lagged behavior is reasonable because price negotiations have been conducted one year ahead of time and expectations formation has been somewhat adaptive. Nevertheless, it appears that the adaptive process has become quicker in recent years. The steel industry in the high-income countries suffered losses in the 1990-91 period as the WBSP index dropped by 11% and 3.5% in constant dollar terms over these two years. This translated into a decline by 8% in constant dollar terms for the iron ore contract price in 1992. Iron ore prices to decline then rise after 1995. Given that steel prices are forecast to decline at an average rate of -0.8% p.a. in constant dollar terms over the 1991-2005 period, iron ore prices also are projected to fall, at -1.4% p.a. over the same period (see Table A4). This is equivalent to a 2.1% p.a. increase in current dollar terms--slower than the expected inflation rate. In the medium term, the expected upturn in the global steel market should lead to constant dollar increases in iron ore prices in 1995-96. 133 Table 8: Price Premiums of Major Iron Ore Brands, 1980-92 Japanese Market European Market Price Differentials (1) (2) (3) (4) (4)() ()-(2) (4)-(2) Hamersley CVRD Hamersley CVRD * (US$/metric ton of metal content) 1980 27.16 25.00 39.60 28.10 0.94 2.16 3.10 1981 29.23 26.88 38.60 28.10 -1.13 2.35 1.22 1982 34.25 30.53 40.40 32.50 -1.75 3.72 1.97 1983 29.97 27.04 34.90 29.00 -0.97 2.93 1.96 1984 26.25 23.89 32.90 26.15 -0.10 2.36 2.26 1985 26.62 24.26 34.30 26.56 -0.06 2.36 2.30 1986 25.56 23.29 32.40 26.26 0.70 2.27 2.97 1987 24.28 21.89 29.35 24.50 0.22 2.39 2.61 1988 23.31 20.90 31.35 23.50 0.19 2.41 2.60 1989 26.34 23.61 35.30 26.56 0.22 2.73 2.95 1990 30.54 27.38 41.47 30.80 0.26 3.16 3.42 1991 32.97 29.56 44.77 33.25 0.28 3.41 3.69 1992 31.35 28.11 42.57 31.62 0.27 3.24 3.51 * Price per Fe 1% c.i.f. Rotterdam. Source: The TEX Report - 'Iron Ore Manual." Figur 1 Iron Ore/Steel Price Movements (in Currait Terms) 240 45 220 40 200 0 Iran Ore -35 180 -30 0 10 140 120 2 1000 80" 1970 1974 1978 1982li 1988 '19;i 1994 199i(200 1972 1976 1980 1984 1988 82 1IN 2000 2004 Yea 134 Table 9: Iron Ore Market Price Setting Process in Japanese and European Markets, 1981-92 Japanese Market European Market Date Setter Change Date Setter Change 1981 02/26/81 CVRD 7.5% 02/15/81 Venezuela/Belgium 6.1% 1982 03/26/82 Newman 17.2% 02/05/82 CVRD/Germany 15.7% 1983 03/28/83 CVRD -11.4% 03/08/83 Carol/Germany -11.2% 1984 01/20/84 CVRD -11.6% 12/07/83 QCM/Germany -8.5% 1985 01/31/85 MMTC 0.0% 12/07/84 QCM.Carol/Germany 0.0% 1986 02/14/86 MMTC -1.9% 12/03/85 QCM.Carol/Germany -1.1% 1987 02/20/87 Newman -5.0% 03/05/87 QCM/Holland -9.3% 1988 12/22/87 Hamersley -4.0% 12/24/87 Hamersley/Britain 8.6% 1989 12/14/88 Hamersley 13.0% 12/19/88 CVRD/Germany 13.0% 1990 01/24/90 Hamersley 16.0% 01/27/90 CVRD/Germany 16.0% 1991 01/30/91 Hamersley 7.9% 01/31/91 CVRD/Germany 7.9% 1992 12/17/91 Hamersley -4.9% 12/17/91 CVRD/Germany -4.9% Note: Producers belong to the following countries: CVRD (Brazil); MMTC (India); Hamernley & Newman (Australia); QCM & Carol Lake (Canada). Source: The TEX Report, May 6, 1992. 135 Table Al: Iron Ore (Metal Contents) - Production By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ --------- ------ 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 ---- -----------------------------------('000 Tons)----- ------------------------------ -------(% p.a.)------ High-Income 165,002 173,991 148,713 153,134 133,114 132,018 135,488 134,588 136,647 140,179 0.5 -1.6 -0.6 OECD 164,909 173,974 148,713 153,134 133,060 131,964 135,455 134,561 136,650 140,241 0.5 -1.6 -0.6 United States 52,052 48,769 35,650 35,088 31,310 30,729 32,060 31,543 31,216 30,745 -1.8 -3.3 -0.9 Canada 26,396 33,211 22,628 18,811 20,004 19,767 19,986 19,824 19,588 19,520 1.5 -1.2 0.3 France 17,523 8,339 2,650 2,270 2,793 2,812 2,978 2,858 2,495 2,180 -7.1 -10.1 -0.3 Sweden 20,700 16,296 12,901 10,093 9,930 9,942 10,269 10,326 10,980 11,702 -1.4 -3.3 1.1 Australia 32,417 55,489 70,560 70,210 64,902 66,165 68,098 69,449 76,920 84,457 10.9 2.1 1.3 LMICs 248,667 347,653 412,058 386,245 380,579 388,994 399,562 408,955 456,465 503,613 3.5 2.0 1.9 Africa 34,762 35,801 30,286 17,559 31,554 31,581 31,867 31,940 32,662 33,521 3.1 -0.8 4.6 South Africa 5,686 17,770 19,689 10,092 12,048 12,275 12,860 13,136 14,899 16,701 7.5 6.0 3.6 Liberia 15,557 11,344 2,667 668 916 954 1,061 1,961 3,682 3,446 0.7 -5.3 11.7 Mauritania 5,640 5,790 7,009 6,798 5,103 5,169 5,350 5,431 5,969 6,525 5.6 0.3 -0.3 Americas 49,784 87,208 128,693 127,499 124,707 129,077 134,450 138,750 154,243 168,894 5.6 3.8 2.0 Brazil 21,864 62,627 102,659 72,659 95,974 99,139 103,356 106,629 123,954 141,133 9.8 6.4 4.7 Asia & Pacific 50,757 81,533 98,972 101,285 110,344 115,043 119,848 124,371 146,215 166,458 4.0 2.7 3.5 China, People's Rep. 23,400 51,417 59,378 61,566 64,638 67,174 69,418 71,842 83,170 93,749 5.5 3.7 3.0 India 19,686 25,619 34,369 39,719 39,696 41,576 43,808 45,634 54,785 63,194 4.1 2.3 3.3 Europe 110,549 140,248 148,232 123,911 119,729 120,057 121,434 122,950 131,326 140,410 2.4 1.2 0.9 World 413,670 521,643 560,771 539,378 513,693 521,012 535,049 543,543 593,113 643,791 2.5 0.8 1.3 --------------------------------------------------------------------------------------------------------------------------------------- a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: UNCTAD (actual); World Bank, International Economics Department (projected). Table A2: Iron Ore (Metal Contents) - Gross Exports By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 --- -----------------------------------(000 Tons)----- ----------------------------- -------(% p.a.)------ High-Income 77,291 99,814 94,130 78,486 76,438 76,753 78,419 78,939 83,152 87,728 3.3 0.2 0.8 OECD 77,291 99,814 94,130 78,486 76,438 76,753 78,419 78,939 83,152 87,728 3.3 0.2 0.8 Canada 22,738 26,812 17,101 15,060 14,320 14,150 14,307 14,191 14,022 13,973 0.8 -1.5 -0.5 France 5,750 2,609 1,033 1,172 1,128 1,236 1,344 1,320 1,186 1,060 -7.3 -10.0 -0.7 Sweden 17,521 13,423 10,679 8,214 7,786 7,795 8,051 8,097 8,609 9,175 -1.4 -3.1 0.8 Australia 25,710 48,925 60,578 52,698 52,778 53,805 55,377 56,475 62,551 68,680 2.6 1.9 LMICs 105,525 137,017 153,174 132,206 131,331 133,598 137,659 140,169 154,687 169,441 3.6 1.4 1.8 Africa 27,135 29,133 21,057 21,053 20,780 20,798 20,986 21,034 21,510 22,076 3.2 -1.5 0.3 Liberia 12,993 12,902 2,607 568 848 883 983 1,816 3,409 3,191 1.5 -4.1 12.3 Americas 41,383 66,678 87,303 74,391 74,835 77,109 80,421 82,814 95,786 108,706 5.0 3.0 2.7 Brazil 17,807 50,861 72,292 63,142 64,016 66,126 68,939 71,123 82,678 94,137 10.4 6.0 2.9 Europe 20,163 24,361 22,025 13,326 17,183 16,822 16,670 16,356 15,400 14,127 3.1 0.3 0.4 World 182,816 236,831 247,304 210,691 207,769 210,351 216,078 219,109 237,839 257,169 3.5 0.9 1.4 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: UNCTAD (actual); World Bank, International Economics Department (projected). Table A3: Iron Ore (Metal Contents) - Gross Imports By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 --------------------------------------('000 Tons)------------------------------------------------ p.a.)-------- High-Income 159,701 184,103 175,149 167,488 164,964 166,076 170,739 172,127 183,569 195,771 2.5 -0.4 1.1 OECD 159,636 181,883 169,162 161,295 159,229 160,158 164,748 165,949 176,523 187,957 2.4 -0.6 1.1 United States 25,929 18,008 13,026 11,542 11,382 11,292 11,762 11,702 11,968 12,085 -3.3 -5.8 0.3 Germany 26,761 30,345 26,219 26,288 24,821 25,136 26,257 26,630 29,174 31,708 1.2 -0.7 1.3 France 5,306 10,675 11,297 10,892 10,902 10,973 11,258 11,339 12,116 12,774 7.6 2.7 1.1 United Kingdom 11,492 8,338 12,146 11,154 10,953 11,063 11,127 11,251 11,661 11,901 0.1 -0.6 0.5 Italy 6,534 9,721 9,356 9,121 8,818 8,609 8,799 9,077 10,043 11,039 4.6 1.5 1.4 Japan 61,283 77,572 70,915 70,090 67,986 68,854 70,528 71,273 76,691 82,929 5.4 -0.2 1.2 LMICs 20,626 44,297 46,038 43,203 42,805 44,275 45,339 46,981 54,270 61,398 6.2 4.8 2.5 Americas 800 1,633 2,151 2,211 2,383 2,506 2,602 2,717 2,894 3,017 6.9 5.8 2.2 Asia & Pacific 23 10,265 25,740 28,104 29,212 30,763 32,001 33,552 40,666 47,375 38.9 3.7 China, People's Rep. - 2,807 8,288 8,750 8,983 9,421 9,722 10,155 12,082 13,903 3.3 Korea, Rep. of 23 5,466 13,926 15,572 17,397 18,219 18,953 19,788 23,775 27,624 35.3 4.1 Europe 19,799 32,376 17,358 14,698 14,052 14,276 14,468 14,859 16,610 18,474 3.5 0.8 1.6 World 180,327 228,401 221,187 210,691 207,769 210,351 216,078 219,109 237,839 257,169 3.1 0.5 1.4 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: UNCTAD (actual); World Bank, International Economics Department (projected). Table A4: hon Oro - Prices, al 1958-91 (Actual), 1992-2005 (Projectd) Currmt $ 1990 Cosant 3 - G-SMUVc/ G-7 CPI d/ 1958 14.4 69.9 91.8 1959 12.2 60.3 78.3 1960 11.4 55.2 71.7 1961 11.0 52.4 67.6 1962 11.0 51.3 65.7 1963 11.0 52.3 63.7 1964 11.0 51.4 62.0 1965 10.2 47.4 55.9 1966 e/ 9.5 42.6 50.3 1967 e/ 8.9 39.2 45.6 1968 8.8 39.2 43.9 1969 9.8 41.7 47.2 1970 9.8 39.2 44.5 1971 10.3 39.1 43.8 1972 10.3 35.9 39.7 1973 10.3 31.0 34.7 1974 13.4 32.9 40.6 1975 17.3 38.3 46.7 1976 22.7 49.6 59.4 1977 23.1 45.8 54.5 1978 21.5 37.0 41.4 1979 23.5 35.8 42.9 1980 28.1 39.0 45.8 1981 28.1 38.9 45.4 1982 32.5 45.7 53.2 1983 29.0 41.7 46.8 1984 26.2 38.4 42.3 1985 26.6 38.7 42.1 1986 26.3 32.5 34.5 1987 24.5 27.6 28.5 1988 23.5 24.7 25.4 1989 26.5 28.0 28.8 1990 30.8 30.8 30.8 1991 33.3 32.6 31.6 Proiccted 1992 31.6 29.7 28.3 1993 30.3 27.4 26.2 1994 29.7 26.4 25.1 1995 32.1 27.8 26.4 2000 38.3 27.9 26.1 2005 42.7 27.1 24.3 a/ Br-zitian ainter fed, Companhia Vale do Rio Doce (CVRD). For 1958-65, Itabira 68% fe in dry weight, f.o.b. priGe. for European market; 1966-74, coract prico to Germany (Federal Republk) fr 64.2% fe; 1975-85, dandard ier foed 64% fe; be 1986 Souhem Symtm (Itabira and oder outhern mines) 64% fe; 1988-89, 64.2% fe; from 1990 onw~ ,64.3% fe. b/ Prices are hown in US dollarm per metric ton of irn co~tet (fe), equivalent to US cents per fe unit (1%). c/ Dofated by G-5 Mamfacturing Unit Valu (MUV) Index. d/ Deflated by G-7 Conme Price Index (CP). e/ Eimate. Sources: Companhia Val. do Rio Doco (&~tual); Word Ban]4 IMeratonal Economica Department 139 Steel Summary * Steel consumption in high-income countries declined 7% in 1991 because of the global economic slowdown. Full recovery is not expected until the mid-1990s, and the average annual growth rate for the 1991-2005 period is forecast at only 0.7%. * Steel consumption in the LMICs fell 8% in the past two years, mainly because of the sharp declines in the Eastern European countries and the FSU republics, but LMIC growth over the 1991-2005 period is projected at 2.1% p.a.. * The steel industry worldwide has entered an era of restructuring. Total capacity in high-income countries declined by 120 million tons during the 1980-91 period and a further 20 million tons reduction is expected before the year 2000. In contrast, steel producing capacity increased by 110 million tons in LMICs during this period, and a further 30 million tons increase is expected by 2000. * The average growth rate of steel production in high-income countries is projected to be 0.7% p.a. over the 1991-2005 period, identical to the consumption growth of this group. For LMICs, steel production should grow at a much faster pace (2% p.a.). The traditional net-trade pattern in steel is expected to hold where by high-income countries are basically net exporters and LMICs are net importers. * The World Bank steel price index in real terms is expected to decline in 1992 and to hold relatively constant over the forecast period, with slight upswings in the mid-1990s and around the year 2000. Steel Prices al, 1969-2005 The World Bank Steel Price Index 1o- ISO- 14o Constant Current <120- 80- Constant g lo 40. Current 901961 9 1 973 197i 19841 1985 1989 9i 197i 200 200id5 1971 1975 1979 1983 1957 1991 1995 1999 2003 140 Demand Outlook HIGH-INCOME couNTRIs. Sluggish growth overall, some dedine. Steel consumption in high-income countries is affected by the ups and downs of general economic activity and the continued decline in the share of steel-using sectors in GDP. Over the past 20 years steel consumption has been notably volatile, reacting strongly to the two oil price shocks in the 1970s and the boom in industrial activity in 1988. The current economic slowdown in high-income countries has not spared steel consumption, which declined by 7% in 1991. Steel consumption in high-income countries has also shown a slow or negative long-term growth since the 1970s, because of: (i) relative sluggishness in the construction and infrastructure sectors; (ii) the shift from heavy industries to information industries and services; and (iii) substitution by lighter and stronger materials in the automobile industry due to higher energy prices. Table 1 shows the change in the composition of materials used in a passenger car between 1985 and 1990. Steel lost about 10% of its share of the total weight of a car, or about 185 kg. Based on present indications, steel consumption in high-income countries in 1992 is expected to decline again, by about 2%. For the 1991-2005 period, the average annual growth rate is forecast at 0.7% p.a. for the 1991-2005 period-from 354 million tons in 1991 to 390 million tons in 2005. The United States is expected to register a slight decline (-0.3% p.a.) during the period, while other major OECD countries should show slight increases ranging from 0.02% p.a. (France) to 0.77% p.a. (Italy). Japan, which has been the leader in steel consumption over the past two decades, is expected to lose its momentum and behave much like other OECD countries in the forecast period because of intensified competition in the exports of manufactured goods and the forecast slower growth of industrial output. Low- AND MIDDLE-INCOME COuNTRIEs. GDP steel intensity mostly rising. Steel consumption in the LMICs is also affected by the fluctuations in general economic activity and by the changing shares of steel-using sectors in GDP. Global economic downturns in 1980-82 and 1990-91 made dents in steel consumption of this group of countries, with declines of 2% and 8% in the respective periods. Table 1: Materials Used in a Car, 1985 and 1990 1985- -1990- (kg) (% (kg) (% Steel 822.2 75.6 637.0 65.0 Aluminum 32.6 3.0 53.9 5.5 Rubber 54.4 5.0 53.9 5.5 Glass 34.8 3.2 34.3 3.5 Cardboard/Fibers 27.2 2.5 78.4 8.0 Mastics/Paints 29.4 2.7 24.5 2.5 Plastic Materials 87.0 8.0 98.0 10.0 Total 1,087.5 100.0 980.0 100.0 Source: Ferruzzi Group Economic Research Department, March 1990. 141 Nevertheless, many LMICs are in the early stage of industrialization and so require large quantities of steel for infrastructure, plant and equipment, and consumer durables. A useful measure for looking at steel consumption is to rank countries according to "steel intensity," defined as crude steel consumption per unit of GDP. Table 2 shows steel intensities (historical and projected) for selected countries at different stages of industrialization. OECD countries generally have lower steel intensities than the LMICs, and are trending downward. Trends in steel intensity in the LMICs are mostly upwards (Brazil is somewhat of an exception) although we are projecting declining intensities in countries such as China, India, and the Republic of Korea. Eastern Europe and the FSU outlook uncertain. Eastern Europe and the FSU account for nearly 50% of crude steel consumption of the LMIC group, and both have been major steel producers and consumers. However, over the last two years, their steel consumption fell-8% in 1990 and 20.3% in 1991--because of the widespread slowdown in manufacturing output. Forecasts of steel consumption for these countries are extremely uncertain. Upon completion of the transition from planned to market economies, economic growth in this region could accelerate to a faster rate than under the old system and replacement of outdated infrastructure and increased demand for consumer durables could lead to a strong increase in the demand for steel (particularly for high quality steel products). For this reason, the long- term steel demand of Eastern Europe and the FSU is projected to grow at rates similar to other developing countries. Asia continues strong. In Latin America, there are signs of recovery after a long period of government policies that strongly curtailed steel consumption. For example, Brazil is expected to have a 10.5% increase in steel consumption in 1992, and an average annual growth of 1.6% p.a. during the forecast period. Asia, the high growth region in steel consumption over the past 20 years, is the only region in the world that has done well despite the current global economic slowdown. During the 1990- 91 period, steel consumption in the region grew 12%; the Republic of Korea posted an impressive 23% Table 2: Steel Intensities in Major Markets, 1975-2005 Historical Forecast-- 1975 1980 1985 1990 1995 2000 2005 -(kg/'o00 1985 US$) a/ Brazil 59.8 63.9 53.8 31.3 38.3 33.4 28.7 China, People's Rep. of 119.6 151.3 248.9 274.7 271.8 221.0 179.3 France 36.9 31.8 28.1 22.5 20.0 18.3 16.9 Germany 62.6 57.4 63.9 38.6 37.9 36.2 33.5 India 58.5 71.0 67.3 107.1 128.9 88.1 81.2 Italy 55.1 61.9 51.2 39.0 34.6 31.6 29.8 Japan 89.6 77.9 54.2 48.8 39.8 39.0 37.1 Korea, Rep. of 92.4 102.2 121.7 133.4 105.9 75.9 50.7 United Kingdom 58.3 31.4 31.3 25.0 20.6 18.3 16.7 United States 48.6 44.6 26.6 27.5 23.7 20.8 18.1 a/ Steel intensity is defined as the ratio of crude steel consumption to GDP in constant terms. Source: World Bank, International Economics Department. 142 gain. China's steel consumption forecasts have been revised downwards from our previous projection, because the country has switched the focus of steel consumption from simply pursuing high tonnages to emphasizing a better balance in different steel products. However, China is still expected to have an average annual growth rate of 2.9% p.a. during the 1991-2005 period, to reach 100 million tons by 2000. The Republic of Korea is forecast to keep its strong growth momentum in the first half of the 1990s with an average 5% p.a. increase, and gradually slow to 3.5% p.a. growth for the remainder of the forecast period. With reform of the steel sector, India has positioned itself to embrace strong growth in domestic steel demand-at an estimated 2.8% p.a. for the forecast period. Combining steel consumption forecasts from high-income countries and LMICs, the total world steel consumption is expected to decline by 1.4% in 1992 before recovering in 1993. The average annual rate of growth during the 1991-2005 period is expected to be 1.42% p.a., and the total tonnage of steel consumption would be around 830 million tons in 2000 and 895 million tons in 2005. Ironically, world total gross capacity for steel production in 1991 is 970 million tons, more than enough to cover the projected consumption growth over the next 14 years. Supply Outlook According to 1990 statistics, the world's major steel-producing countries were the FSU (154 million tons), Japan (110 million tons), United States (89 million tons), China (66 million tons), Germany (44 million tons), Republic of Korea (26 million tons), and Brazil (20 million tons). The high- income countries and LMICs each produce about one half of total crude steel. However, the prospects for steel production in these two groups of countries are distinctively different. In the high-income countries, excess steel-producing capacity has been a serious problem for most of the past 20 years despite efforts to cut capacity. Table 3 presents the capacity utilization rates of major steel-producing countries, capacity utilization defined as the ratio of production to gross producing capacity. High-income countries had relatively low capacity utilization rates until very recent years, when they increased to levels comparable with the LMICs. Steel industries in the LMICs, on the other hand, have been expanding in order to meet domestic demand, and capacity utilization rates have been generally higher. In particular, capacity utilization rates in Asia reached abnormal levels in 1990-91 (i.e., 86% is considered normal and economically viable according industry standard). Eastern Europe and the FSU previously had relatively high capacity utilization rates, but since the collapse of the centrally planned economic systems in those countries, capacity utilization rates have plunged. Capacity reductions in OECD countres continue. Table 4 shows recent changes in steelmaking capacities. Total capacity in high-income countries declined 20.6% during the 1980-91 period. About 8 million tons of capacity was cut during 1990-91 alone, largely due to reductions in the United States, Japan, and Italy. In the United States, further closures of integrated steel mills are likely during the next few years. However, those closures will be partly offset by increases in mini-mills using electric arc furnace (EAF) technology. Overall, US capacity is expected to decline by about 4 million tons before the year 2000. In Japan, the major integrated steel producers have already reduced their capacity considerably over past three years. Additional reductions by about 5 million tons are likely to follow in the 1990s. In the EC, Italy, Spain, and the United Kingdom have reduced capacity, and in Germany, particularly in the eastern part of the country, further plant closures are expected. Capacities in other EC countries are likely to remain constant. Total EC capacity is expected to decline by 9-13 million tons during this decade. 143 Table 3: Capacity Utilization Rates for Major Steel-Producing Countries, 1980-91 Regions/Countries 1980 1985 1990 1991 High-Income 62.0 61.8 71.7 70.0 OECD 62.0 61.8 71.9 70.0 United States 66.2 59.8 72.9 66.3 Japan 70.2 65.2 78.7 78.6 Germany 66.1 79.5 78.9 75.8 Italy 67.1 66.6 81.8 88.1 France 71.3 65.8 78.6 76.5 United Kingdom 40.1 65.5 80.4 75.6 LMICs 86.8 85.1 83.8 77.5 Americas 76.2 70.8 66.8 67.4 Brazil 84.9 82.8 71.1 78.2 Asia and Pacific 85.7 88.1 95.9 98.2 China, People's Rep. of 87.0 96.4 105.2 109.3 Korea, Rep. of 91.5 93.6 103.5 101.8 India 76.4 76.7 76.9 80.1 Europe 89.3 87.9 82.9 69.9 FSU 86.9 86.8 81.4 69.6 World Total 70.9 71.3 77.3 73.5 Source: World Bank, International Economics Department, estimates (derived from Paine Webber and IISI statistics). LMICs expand capacity. In contrast, the LMICs registered a capacity increase of 26% during the 1980-91 period. During the 1991-2000 period, these countries are likely to achieve considerable capacity expansion (although many financial and political uncertainties remain): in Latin America, Brazil could add 7-8 millions tons of steelmaking capacity by 1995 if its privatization program proves successful; Argentina is expected to increase capacity by about 1.1 million tons by the mid-1990s; a new integrated plant based on the direct reduction technology could add 1 million tons to Venezuela's total capacity; in Africa, 3-4 million tons of additional capacity is expected in Egypt, Algeria, and Nigeria; in the Asia and Pacific region, China could continue its rapid expansion and add 15-25 million tons before the end of this century; India is expected to add 7 million tons; the Republic of Korea increased capacity by 3.2 million tons in 1991 and another 5 million tons are in the pipeline; and Indonesia is expected to almost triple its capacity from the current 3 million tons level to 8 million tons by the year 2000. However, the FSU and Eastern Europe are likely to experience a decline in total capacity of 6-9 million tons during this decade because most steel mills in these countries are economically and technologically uncompetitive. 144 Table 4: Gross Steel-Making Capacities, 1980-91 Regions/Countries 1980 1985 1988 1989 1990 1991 (million tons) High-Income 649.0 598.2 543.3 537.5 536.3 527.9 OECD 641.0 587.1 528.1 521.2 518.5 510.4 United States 153.2 133.9 121.8 121.9 121.9 119.4 Japan 158.8 161.6 146.0 144.9 140.2 139.6 Germany 77.4 60.8 56.4 56.2 55.8 55.6 Italy 39.5 35.9 31.7 31.2 31.2 28.4 France 32.5 28.6 24.4 24.3 24.2 24.1 United Kingdom 28.1 24.0 22.2 22.2 22.2 21.8 Belgium 19.6 14.7 13.4 13.2 13.2 13.2 Netherlands 8.5 8.0 8.0 8.0 8.0 8.0 Luxembourg 6.4 5.5 5.5 5.5 5.5 5.5 LMICs 361.8 410.8 441.4 454.6 460.7 469.8 Africa 10.1 11.0 11.9 11.8 11.7 9.5 South Africa 10.1 10.5 10.7 10.6 10.6 10.6 Americas 38.2 51.0 53.8 56.0 57.8 58.4 Brazil 18.1 24.7 27.9 28.9 28.9 28.9 Mexico 10.1 11.6 10.7 11.7 13.4 13.6 Argentina 4.9 5.3 5.1 5.1 5.1 5.4 Venezuela 2.9 5.5 5.7 5.7 5.6 5.6 Asia and Pacific 73.2 93.3 112.1 119.2 123.4 129.5 China, People's Rep. of 42.7 48.6 56.4 61.4 63.1 64.5 Korea, Rep. of 9.4 14.5 20.1 21.8 22.4 25.6 India 12.5 15.6 17.7 17.9 19.5 20.5 Indonesia 1.3 2.4 2.5 2.6 2.8 2.8 Pakistan 0.5 1.7 1.7 1.7 1.7 2.1 Middle East 5.5 8.7 10.5 11.4 11.4 15.9 Iran 1.6 2.3 2.3 2.9 2.9 5.3 Egypt 1.6 2.3 2.3 2.9 2.9 5.3 Algeria 1.0 2.1 2.2 2.2 2.2 2.2 Saudi Arabia 0.0 1.2 1.7 2.0 2.0 2.0 Europe 234.8 246.8 253.1 256.2 256.4 256.5 FSU 170.3 178.1 185.3 188.4 189.7 190.5 World Total 1,010.8 1,009.1 984.7 992.1 997.1 997.7 Source: Paine Webber, World Steel Dynamics, CapacitV Monitor (various issues); and World Bank, International Economics Department, estimates. 145 NEW TECHNOLOGY AND CHANGING CosT STRUCTUREs. Labor costs fall but totalpre-tax costs rise. Besides closing inefficient plants and raising capacity utilization rates, major steel producers have also introduced technological innovations to enhance efficiency in steel production. Such innovations include mini-mill technology, alternative iron-making technologies (e.g., the direct smelting process), continuous casting, thin slab casting, and computerized production planning and management. Table 5 shows the current cost structure of major steel-producing countries. Comparisons with the cost table in our previous report, Price Prospects for Major Primary Commodities, 1990, show an overall improvement in terms of man hours per ton of output, with Brazil making the largest gains from 14 to 10.9 (no updated information on the FSU and China is available). Relative pre-tax cost relationships among these countries have changed quite significantly. In comparable 1991 dollars, Germany's pre-tax costs increased by $100/ton which made it the highest cost producer. Brazil's cost advantage has been eroded from two years ago with a $78 increase in pre-tax costs. The United States is the only country that has managed to reduce its pre-tax costs (by $31), largely through the depreciation of the dollar. In 1991, fixed costs accounted for about 30% of pre-tax costs for EC steel producers, 40% for Japanese producers, and about 20% for US producers. Cost advantages of mini-mills for high-income countries. The Republic of Korea and Brazil had the lowest operating costs because of their relatively low labor costs. However, they do not possess a significant total pre-tax cost advantage over the high-income countries because of high capital depreciation charges for newer facilities. For other developing countries that plan large integrated steel mills, capital costs can be prohibitively high. However, the recent development of mini-mill technologies gives these countries an alternative route to increase their steelmaking capacity. The core of a mini-mill is the EAF, which uses scrap or directly-reduced iron as feedstock. Modern mini-mills have a yearly capacity of 0.5-1.0 million tons, compared with 2-3 million tons for integrated steel mills. Mini-mills have a significant cost advantage because of the lower initial investment. Of the total cost of producing a ton of liquid steel in the United States, the material costs account for 59.3% for EAF and 43.7% for basic oxygen furnace (BOF); energy accounts for 19% for EAF and 16.6% for BOF; labor cost shares are about the same; but capital costs (depreciation and interest) amount to 11.6% for EAF and 24.7% for BOF. Moreover, the smaller size of mini-mills gives them more flexibility in adapting to technical changes. Mini-mills traditionally have specialized in lower-valued long products such as wire rod and beams, rather than the higher-valued, flat-rolled products. However, the emergence of new technology such as thin-slab casting has made it possible for mini-mills to enter the flat-rolled products market. Table 6 presents a cost structures of mini-mills for three typical regions. It can be seen that mini-mills have a cost advantage over integrated steel mills in developed countries, but it is not clear whether that cost advantage holds in developing countries where electricity is costly and steel scrap has to be imported. The United States is currently the leading country in terms of the number of operating mini-mills and related technological advances. About 50 mini-mills produced 25% of total US steel output in 1991, and this share is projected to increase to 40% by the year 2000. Turkey also has a notable mini-mill sector, where about 15 mini-mills produce more than one-half of the country's steel output. Brazil, the Republic of Korea, and Mexico are other countries where mini-mills have found success. In Venezuela, directly-reduced iron is used to feed EAFs, to take advantage of this country's high quality iron ore and natural gas. 146 Table 5: Cost Structure of Integrated Steel mills in Major Steel-Producing Countries a/ US Japan Germany France UK Korea Brazil Operating rate(%) 90 90 90 90 90 90 90 Raw materials 142 141 162 150 156 153 150 Coking Coal 38 40 45 42 43 44 60 Iron Ore 66 66 80 71 75 67 43 Scrap 38 35 37 37 38 42 47 Other material 174 189 190 171 185 156 186 Labor cost 151 146 178 153 124 67 82 Cost/hour ($) 28.5 27.5 33.0 28.8 22.5 10.0 7.5 Man hour/ton (hour) 5.3 5.3 5.4 5.3 5.5 6.7 10.9 Operating cost 467 476 530 474 465 376 418 Financial cost 41 90 55 51 26 125 130 Depreciation 26 70 44 37 25 110 80 Interest 15 20 11 14 1 15 50 Pretax cost 508 566 585 525 491 501 548 Cost thru. process Coke 108 112 133 122 117 109 124 Blast furnace 145 145 173 155 152 141 128 Liquid steel 203 202 229 210 209 193 192 Slabs 246 235 271 247 248 222 224 Hot rolling 319 310 260 322 322 271 285 Cold rolling 467 476 530 474 465 376 418 a/ Unit is $/ton of cold-rolled sheet shipped as of December 1991 unless otherwise noted. Source: Paine Webber, World Steel Dynamics, Cost Monitor, January 13, 1992. PRIVATIZATION. Gains from privatizadon require supponing policies. The world steel industry has entered an era of restructuring. In particular, privatization of state-owned steel companies in developing countries has progressed significantly. Mexico has sold two government-owned companies, Sicartsa and AHMSA, to private interests. In Brazil, the state-owned integrated steel company, Usiminas, has been sold to a group of companies. An other smaller plant, Cosinor, was bought by German interests. The Argentine government sold SOMISA in 1992. In India, the government-owned steel company SAIL will be 20% privatized. Privatization of the steel sector in these countries is motivated by: (i) a re-orientation of the role of the state away from activities that can be more effectively managed by the private sector; (ii) the need to reduce public debt; (iii) a desire to modernize the industry through private-sector investments; and (iv) the strengthening of the domestic capital market. Despite widespread enthusiasm for privatization, some questions remain. One concern is that certain privatization practices lack clearly defined objectives, or that the objectives defined are in 147 Table 6: Cost Structure of Mini Steel mills, 1991 Newly Developing Industrializing Developed Economies Economies Economies (:/ton of liquid steel) Materials 137.15 137.15 109.58 Electricity 43.86 27.80 19.76 Electrode 10.04 10.04 10.04 Labor 2.06 3.43 14.39 Maintenance 6.90 6.90 6.90 Interest 20.71 17.25 14.02 Depreciation 17.25 17.25 17.25 Total cost 237.96 219.81 191.94 Source: "International Trends in Steel Mini-Mills," World Bank, Industry and Energy Department Working Paper, No. 52, Dec. 1991. conflict. An other concern is that privatization tends to enhance industry concentration and oligopoly power, discourage competition, and increase domestic prices. A study conducted by International Economics Department staff suggests: (i) that industry concentration raises domestic prices above the competitive level; (ii) that the degree of domestic price distortion depends positively on the degree of market segregation (defined by the elasticity of substitution and trade protection mechanisms) between domestic products and imports; and thus (iii) that the incentives for trade protection in a concentrated industry are far greater than in a competitive industry. In light of these findings, it is important to recognize that industry competitiveness and trade liberalization are indispensable parts of the privatization process, it will not be successful without them. DEVELOPMENTS iN THE FSU. Huge steel industry faces formidable pmblems. The dissolution of the FSU will have major implications for the development of the world steel market. The FSU is the world's largest steel producer with 20% of total steelmaking capacity. The iron and steel industry in the FSU was almost completely closed to the outside world. All iron ore requirements were met from domestic production of which 44% came from Russia, 46% from the Ukraine, and 10% from Kazakhstan. Steel is produced mostly in Russia (58%) and the Ukraine (34%). The FSU steel industry is beset by: (i) production inefficiency due to general disrepair of facilities and long disregard for technological modernization; (ii) shortages of energy supplies; (iii) disastrous environmental pollution, disproportionally large relative to the level of output; and (iv) an unbalanced product mix, with too many low quality steel products and too few high quality products such as cold-rolled products, pipes and tubes in demand by the oil and gas industry. The outlook depends on how these problems are dealt with. World production and tade patterns generaly unchanged. For the rest of the world, the recent pattern of steel production growth is likely to persist in the 1990s. Steel production in high- income countries is expected to decline 1.7% in 1992, stay flat in 1993, and increase by 2.9% in 1994. 148 The average growth rate of steel production in high-income countries is projected at 0.7% p.a. over the 1991-2005 period. Among OECD countries, US steel production will decline slightly over the 1991-2005 period. Germany's production is expected to grow at 1% p.a., due to her new technologies and better access to Eastern Europe. Japan will continue to be the largest steel producer among high-income countries, with 0.7% p.a. production growth. Most capacity expansions will take place in the LMICs, where excess demand for steel still prevails. However, these countries face: (i) scarcity of capital investment, (ii) shortage of energy and raw materials, and (iii) strong global environmental pressure. Therefore, they are not likely to increase steel production sufficiently to change the global pattern of steel trade. Rather, we expect steel production in these countries to grow in line with consumption; both production and consumption are forecast to grow at slightly over 2% p.a. over the 1991-2005 period. After successful privatization, Brazil could utilize its abundant high-grade iron ore resources to strengthen her position as a major steel exporter. China should to continue to expand production but at a rate slower than previously forecast due to adjustments towards high quality products. Steel production in the Republic of Korea should continue to grow fast in the near term (1991-95); the growth rate will slow down somewhat in the long term. India's steel industry could also benefit from domestic market liberalizations. Eastern Europe and the FSU will continue to lose ground in steel production in the next one to two years, but it is conceivable that steel industries in this region could start regaining ground three to four years from now. Trade Prospects International trade in steel has become increasingly important. Exports accounted for 28% of total crude steel production in 1990, compared with 19.7% in 1970 and 25.6% in 1980. The trade flow is primarily from high-income countries to LMICs (see Table 7), although some LMICs (e.g., Brazil, Turkey, and former Yugoslavia) have exported significant quantities of low quality steel to high- income countries. Table 7 shows the historical data and forecasts (derived from production and consumption forecasts) for net trade. The traditional trade pattern is expected to persist during the 1991-2005 period, with high-income countries being net exporters and LMICs being the net importers. Individually, Japan, Germany, and Brazil are likely to be major exporters of steel, and India, China, and the United States, major importers. Price Outlook Complex price structure for steel. Unlike other metals such as copper, steel prices are not driven by "spot" trading on the LME. In fact, no steel product is traded on a commodity exchange. Steel is mostly traded directly between producers and consumers and pricing on international or domestic markets can be complex and confusing. The relationship between international and domestic prices is far from being transparent because of the complicated trade restrictions in place. Depending on the domestic market structure (oligopolistic, competitive, oligopsonistic, etc), there can be more than one price for a product. For example, in the United States the same cold- rolled sheet can have a list price, a real price, a normal three-month spot price, a distress spot price, a 149 Table 7: Net Trade of Steel, by Main Countries and Regions, 1969-2005 1969-71 1979-81 1989-91 1995 2000 2005 ('000tons) High-Income 1,629 32,456 10,390 15,106 14,575 16,887 OECD 1,742 3,826 16,376 19,543 18,988 21,258 United -1,239 -1,696 -12,986 -13,970 -11,617 -9,499 Japan 24,952 32,171 13,251 17,455 19,847 18,618 Germany -1,988 -436 3,642 2,878 4,330 5,323 France 79 3,149 1,457 2,894 3,252 3,581 Italy -2,335 131 1,215 930 623 371 United Kingdom 2,767 -1,328 -2,805 -2,475 -2,572 -2,421 LMICs -1,962 -44,237 -17,121 -15,106 -14,575 -16,887 Americas -6,647 -6,189 11,893 16,508 17,465 17,760 Brazil -898 114 11,407 14,062 16,673 19,417 Asia and Pacific -8,638 -1,768 -21,893 -23,781 -23,529 -24,601 China, People's Rep. of -389 -662 -6,489 -8,588 -8,794 -10,307 Korea, Rep. of -845 1,947 2,693 4,139 4,447 6,016 India -43 -2,542 -6,263 -7,260 -9,037 -11,133 Europe 4,498 -221 4,896 1,780 1,466 1,261 FSU 5,468 -2,458 -1,624 -655 -1,338 -1,998 World -3,394 -11,781 -6,731 0 0 0 Source: World Bank, International Economics Department. primary price, a regular price, and a special deal price. The list price is often not the actual price but rather a reference for negotiating discounts to buyers. Japan has two major pricing systems. The big buyer price is the discounted price given to large buyers such as the automobile industry; the dealer price is used mostly by Japanese mini-mills to sell their products to the residual market. However, the oligopolistic nature of the Japanese steel industry has resulted in domestic prices being almost twice as high as export prices. The EC steel pricing system is similar to that of the United States. Prices vary moderately from country to country within the EC. Given the substantial intra-EC trade in steel, intra-EC export prices can serve as a good indicator of the actual price level in the EC. Three different export prices are generally considered as representative of the international market. They are: the spot export price, f.o.b., Antwerp; the German extra-EC export price, f.o.b.; and the Japanese export price, f.o.b., to non-US destinations. The German extra-EC export prices and the Japanese export prices are contract prices announced monthly. In a declining market, these prices often lag behind the Antwerp spot prices which change from week to week. Since most steel available at the Antwerp spot market comes from Germany, the Antwerp and German export prices are closely related. Since Japan was the largest steel exporter during the 1980s and its export prices were the lowest among the three, we take Japanese prices as the indicator prices of international steel trade. 150 High volatility in export prices. Japanese export prices for eight carbon steel products are used to calculate the World Bank steel price (WBSP) index. Products included are cold-rolled sheet, hot-rolled sheet, galvanized sheet, merchant bar, plate, section, wire rod, and re-enforcing bar. The weights are shares of each product in the total output of Japan, the United States, and Germany during the 1984-86 period. The WBSP index in constant terms has been declining from the peak reached in 1974; it has fluctuated widely both in current and constant dollar terms, particularly during 1973-77 and 1988-91. The coefficient of variation (c.v.) of the 5-year moving average of current dollar prices exceeds 20% in 5 of the 18 years (1973-91), with an average of 15%. The c.v. for constant dollar prices averages 11% for the 18-year period. The high volatility of steel prices reflects the highly competitive nature of the international steel market. Large fixed costs and excess capacity are predominant features of the steel industry. When domestic demand is weak, steel producers tend to maintain production levels and channel the excess supplies to exports. Since demand tends to weaken in countries simultaneously, exports can increase significantly, forcing export prices to plummet. The opposite can happen in the case of an economic upturn. 1990-91 decrease to continue until upturn in 1994. Table 8 presents the historical statistics and forecasts for the WBSP index and the associated Japanese export prices. Steel prices were riding a roller coaster from 1987 to 1989. The general metals price boom in 1988-89 pushed the WBSP up by a cumulative 38.3% in current dollar terms in the two-year period. Construction-related products such as merchant bars and re-enforcing bars enjoyed the largest price increases-by 45-50% in two years. Auto-industry related products such as cold-rolled and galvanized sheets came in second with 30-40% increases. Prices of general purpose products such as hot-rolled sheets, plates and wire rods increased 25-30%. During the 1990-91 period, the economic slowdown pushed steel product prices down acrosss the board. For example, galvanized sheet prices fell by 23.3% in two years, prices for sections and cold- rolled sheets by 16.3%, and the WBSP index by 14.5%. Steel prices are expected to decline further in 1992, with falls ranging from 25.5% for sections to 4.7% for plates, resulting an aggregate 12.3% drop in the WBSP index. For 1993, continued declines in steel prices are expected, although only marginally (-2.8% for the WBSP index). The strong growth expected in the mid-1990s should contribute to an upturn in steel prices; the WBSP index is projected to increase by 11.4% in 1994. Among individual products, merchant bars, sections, re- enforcing bars and galvanized sheets are likely to enjoy higher price increases because they suffered larger losses previously. As can be seen from Figure 1, the projected long-term trend for the WBSP index over the 1995-2005 period is almost flat in constant terms. 151 Table 8: Steel Product Prices, a/ 1969-1991 (Actual) and 1992-2005 (Projected) World Bank CR HR Galv. Merchant Wire Year Steel Index b/ Sheets Sheets Shets Bars Plates Sections Rod Rebar (1990=100) (current $/ton) 1969 26.5 126.9 107.5 155.9 108.0 130.4 99.3 1970 30.5 154.2 134.2 169.3 106.0 137.1 129.0 1971 27.9 135.9 112.2 169.4 105.9 127.8 116.5 1972 29.8 147.7 110.6 173.0 130.8 129.2 129.3 1973 45.5 233.3 165.8 255.0 211.3 195.4 183.8 1974 66.2 299.2 245.0 353.3 363.3 269.2 325.4 1975 52.1 228.8 216.3 296.0 237.9 227.5 248.3 1976 54.1 260.6 221.1 328.4 217.5 227.9 227.5 1977 53.1 256.3 217.5 311.3 216.7 229.5 228.3 1978 68.2 314.1 276.3 412.8 290.4 308.0 284.2 1979 76.3 365.0 298.3 448.7 340.0 332.9 315.8 1980 79.1 380.3 323.1 472.8 342.5 341.7 357.5 351.7 330.8 1981 81.7 398.7 326.8 503.7 328.8 357.1 383.3 390.0 313.3 1982 70.4 367.9 281.7 425.0 258.3 309.2 333.3 353.3 242.5 1983 66.5 360.4 269.2 419.2 245.8 286.3 285.0 311.7 222.5 1984 70.2 377.1 284.7 455.0 266.4 305.1 293.3 308.8 233.3 1985 60.4 325.4 245.0 369.6 229.6 266.7 246.7 272.9 225.4 1986 61.5 324.6 268.8 383.7 229.2 285.0 247.9 238.8 219.6 1987 72.0 386.3 323.3 454.2 244.2 360.8 330.8 248.3 202.9 1988 94.1 501.2 395.8 625.8 370.4 432.5 433.3 313.8 262.5 1989 105.7 550.4 441.9 735.0 416.3 478.8 441.7 350.8 341.7 1990 100.0 511.7 411.3 634.2 412.9 458.8 438.3 362.9 364.2 1991 98.6 504.2 408.3 627.1 385.0 469.2 404.2 383.3 367.1 Pro*ete 1992 89.6 481.5 381.5 566.8 312.5 460.0 322.0 364.0 305.0 1993 90.5 486.4 385.4 572.5 315.7 464.7 325.3 367.7 308.1 1994 101.4 518.9 420.3 645.4 396.2 482.9 416.0 394.5 377.8 1995 102.6 524.9 425.2 652.9 400.9 488.5 420.8 399.1 382.2 2000 124.0 634.2 513.6 788.8 484.3 590.1 508.4 482.2 461.8 2005 142.9 731.1 592.1 909.4 558.3 680.4 586.1 555.9 532.3 a/ Steel prices are f.o.b. Japan excluding shipments to China and the United States. All products items refer to base size: cold- rolloYf coil/sheet, hot-rolled coil/sheet, galvanized iron sheet, merchant bars, medium plates, H-shape sections, wire rod, concrete reinforcing bars (rebars). b/ Composite steel price index is a weighted average price index (1990-100) based on available data for products given in the table. Weights used are shares of apparent consumption of each product in the Federal Republic of Germany, Japan and the United States during the period 1984-86. Source: Japan Metal Bulletin; Commodities Research Unit LTD, CRU Metal Monitor (actual); World Bank, International Economics Department (projected). 152 Table Al: Steel (Crude Equiv.) - Production By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 ---- ------------------------------ ----('000 Tons)---------------------------------- ---------(% p.a.)------ High-Income 388,478 412,081 383,782 369,572 363,375 362,936 373,562 373,559 388,983 407,453 1.0 -0.8 0.7 OECD 387,770 407,753 372,638 357,271 352,086 351,344 361,834 361,530 375,487 392,689 0.9 -0.9 0.7 United States 118,907 111,586 88,900 79,203 77,970 76,746 80,036 78,980 78,870 78,224 -1.1 -2.5 -0.1 Germany 54,430 51,096 44,000 42,169 39,748 39,934 41,763 42,015 44,942 47,975 0.1 -1.4 0.9 France 23,042 22,598 19,015 18,434 18,419 18,392 18,893 18,875 19,693 20,392 -0.1 -1.7 0.7 United Kingdom 26,316 16,105 17,841 16,474 16,149 16,182 16,296 16,344 16,539 16,580 -1.9 -2.5 0.0 Italy 17,052 25,176 25,510 25,007 24,135 23,378 23,920 24,477 26,444 28,549 3.3 1.2 0.9 Japan 88,015 108,273 110,331 109,649 106,173 106,680 109,402 109,664 115,214 122,369 4.1 0.2 0.8 LMICs 201,205 311,199 385,923 364,162 360,180 369,613 378,932 389,489 439,291 488,140 4.3 3.3 2.1 Americas 13,108 27,928 38,587 39,341 42,325 44,152 45,891 47,546 49,442 50,616 7.3 6.2 1.8 Brazil 5,437 14,152 20,567 22,617 25,088 25,725 26,389 27,008 30,183 33,463 8.9 8.0 2.8 Asia & Pacific 27,732 61,994 118,383 127,071 131,853 137,759 143,468 149,209 176,573 202,045 7.5 7.6 3.3 China, People's Rep. 18,333 35,736 66,349 70,436 72,188 75,104 77,600 80,401 93,398 105,556 7.0 6.6 2.9 Ln Korea, Rep. of 442 8,974 23,125 26,001 28,999 30,129 31,380 32,498 38,124 43,506 23.7 21.8 3.7 India 6,311 10,135 14,963 16,394 17,063 17,441 18,062 18,487 20,578 22,933 3.8 4.4 2.4 Europe 154,481 208,638 212,671 179,301 171,126 172,474 175,007 178,274 194,581 212,563 3.0 1.7 1.2 FSU 115,613 148,512 154,414 132,666 128,527 128,033 129,924 132,277 144,113 157,150 2.7 1.4 1.2 World 589,682 723,280 769,704 733,734 723,555 732,549 752,494 763,048 828,274 895,592 2.4 1.0 1.4 --------------------------------------------------------------------------------------------------------------------------------------- a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: International Iron and SteeL Institute (actual); World Bank, International Economics Department (projected). Table A2: SteeL (Crude Equiv.) - Apparent Consumption By Main Countries and Economic Regions Actual Projected Growth Rates a/ Averages Countries/ ----------------- 1991- Economies 1969-71 1979-81 1990 1991 b/ 1992 1993 1994 1995 2000 2005 1961-90 1970-90 2005 --------------------------------------('000 Tons)---------------------------------------- --------(% p.a.)-------- High-Income 372,269 379,625 379,101 354,277 346,394 347,138 357,562 358,452 374,409 390,565 0.9 -0.7 0.7 OECD 369,583 368,848 358,128 341,631 330,772 331,285 341,327 341,987 356,499 371,431 0.7 -0.9 0.6 United States 131,216 127,682 102,480 91,083 90,629 91,768 94,371 92,950 90,486 87,723 -0.5 -1.9 -0.3 Germany 56,418 51,532 39,550 38,536 37,548 37,938 38,671 39,137 40,612 42,652 -0.4 -2.1 0.7 France 22,252 19,449 18,076 16,775 16,038 16,025 15,990 15,981 16,441 16,811 -0.2 -2.3 0.0 United Kingdom 23,549 17,433 16,690 15,321 14,957 14,895 15,458 15,414 15,916 16,209 -1.4 -2.5 0.4 Italy 19,387 23,866 28,532 27,799 25,614 25,889 26,673 26,952 29,016 30,970 2.6 1.2 0.8 Japan 63,063 76,102 99,032 95,826 90,607 91,159 92,587 92,209 95,367 103,751 4.0 1.2 0.6 LMICs 220,807 355,436 403,565 379,457 377,161 385,411 394,932 404,595 453,866 505,027 4.3 3.0 2.0 Americas 19,755 34,117 27,313 27,774 29,846 30,522 30,928 31,038 31,977 32,856 4.3 1.3 1.2 Brazil 6,335 13,048 10,197 11,197 12,440 12,594 12,853 12,946 13,509 14,047 5.8 2.2 1.6 Asia & Pacific 36,370 79,062 138,188 150,728 156,484 162,143 167,389 172,990 200,103 226,646 7.3 7.2 2.9 China, People's Rep. 21,422 42,356 68,832 77,688 81,373 84,386 86,090 88,989 102,192 115,863 7.4 6.9 2.9 Korea, Rep. of 1,287 7,027 21,650 22,999 24,421 25,791 27,105 28,359 33,677 37,490 17.7 14.1 3.5 India 6,714 12,677 21,700 23,019 23,672 24,344 25,036 25,747 29,616 34,066 4.4 5.6 2.8 Europe 149,983 210,848 207,656 175,090 169,068 170,253 173,345 176,494 193,115 211,301 3.1 1.6 1.3 FSU 110,145 150,970 156,403 134,134 128,875 128,232 130,561 132,932 145,451 159,149 3.0 1.6 1.2 World 593,076 735,061 782,666 733,734 723,555 732,549 752,494 763,048 828,274 895,592 2.4 1.0 1.4 a/ Least squares trend for historical periods (1961-90); end-point for projected periods (1991-2005). b/ Estimate. Sources: International Iron and Steel Institute (actual); World Bank, International Economics Department (projected).  ・目織’・ Zinc hices a/, 1950-2M5 C$/ton) 3,500 3,000 2,500 Constant 1990 $ bl 2,000 500 Current $ ------------ 0 1950 1955 1960 1965 1970 1975 1980 19B5 1990 1995 2000 2005 a/ LME high/special high grade b/ Deflated by G-5 MIN Index Demand Outlook New uses for lead boost demand. Since about two thirds of lead consumption is in automotive batteries, the demand outlook for lead depends critically on automobile production and the number of registered vehicles. However, promising new uses of lead have been found and their importance has been growing in the past several years. These include large-capacity industrial batteries for power storage and industrial vehicles and applications in nuclear safeguards. In major industrial and developing countries representing 70% of world consumption, lead consumption in batteries and shot/ammunition increased over the 1985-90 period at 4.7% p.a. All other uses (cable sheathing, rolled and extruded products, alloys, pigments and other compounds, and gasoline additives) experienced declines. Total lead consumption in these countries increased at 2.7% p.a. during this period. Strong demand for zinc in galvanizing. Zinc's main market (47 % in 1990) is in galvanizing steel products for corrosion protection; rolled zinc sheet and strip, brass, and zinc diecastings are also important uses. The galvanizing market increased the fastest over the 1985-90 period in the major industrial and developing countries, at 4.7% p.a. compared with 2.8% p.a. growth for total zinc consumption. Hence, the share of the galvanizing market increased, while shares of other uses either declined slightly or remained flat. Automobile changesfavor both lead and zinc demand. Between 1986 and 1991, world lead and zinc demand received moderate boosts from developments in the automobile industry. In the case of lead, downsizing of batteries had more or less run its course by the mid-1980s and therefore changes 156 in automobile production and stocks led to corresponding changes in lead demand for batteries. Furthermore, the trend to greater use of electronic equipment in cars required more powerful batteries. In the case of zinc, the greater quality competition among automobile manufacturers led to more corrosion protection and therefore greater demand for galvanized steel. Modest growth in lead market expected. The industrial economies' demand for original equipment and replacement batteries for automobiles is expected to grow at a slightly lower rate in the 1991-2005 period than over the preceding 15 years. In most of the high-income countries, the automobile stock is reaching saturation level in terms of per capita ownership, pollution emission, and parking space. In most developing countries, however, automobile ownership is in its infancy and will increase quickly as per capita incomes grow. Lead consumption for batteries is expected to continue to increase at a robust pace in the forecast period, although not as fast as in the 1985-90 period. Environmental constraints will exert greater pressure on most other uses of lead, particularly for gasoline additives in LMICs. On the other hand, continued expansion of industrial applications and technological improvements in lead acid batteries are expected to allow total lead consumption to grow at moderate rates. We project that world lead consumption should grow at around 1.3% p.a. during the 1991-2005 period. Intense research and development has been directed at improving batteries capacity and efficiency with a view to eventual use in electric motor vehicles. The latest development involves a battery which is a composite of different metals, use of which considerably shortens the recharging time and extends the peak output and range. This development may signal the introduction of viable electric cars earlier than previously expected. Environmental concerns continue to restrict some lead uses. Restrictions on lead use for environmental reasons are intensifying in both the high-income and developing countries. Four bills relating to lead use are under consideration in the United States. These laws, if enacted, will further restrict lead use in soders, alloys, and chemicals, and encourage recycling of lead batteries. In many developing countries, the use of lead in gasoline, paints, and soders is still prevalent. These uses will come under strong pressure which will moderate demand growth rates in these countries. Zinc outlook moderate but vulnerable. World zinc consumption is expected to increase at 1.4% p.a. in the 1991-2005 period, a rate slightly higher than in the 1970-91 period. Zinc demand is highly sensitive to fluctuations in industrial activity and vulnerable to substitution by other materials in the medium term. It should be kept in mind, therefore, that zinc consumption could experience wide fluctuations over the forecast period and its long-run consumption growth could be significantly lower if prices remain high over an extended period. Supply Outlook Lead and zinc are produced mostly as joint products; about 60% of the market economies' lead and zinc output is derived from ores containing both lead and zinc, often together with silver and copper, and sometimes gold. In 1987, for each ton of zinc produced, the market economies' lead and zinc mines produced as joint products 0.42 tons of lead, 1.20 kg of silver and 0.11 tons of copper. Because of this joint production, decisions about investment in lead and zinc mining are complicated by the different market outlooks for the various metals vis-a-vis the metal content of the deposits. With the poor market prospects for lead over the last 15 years, deposits consisting mostly of lead were rarely considered economic. However, lead deposits with significant silver content have been attractive, in which case lead 157 Table 1: Cash Production Costs for Lead and Zinc 1987 1989 1990 --($/ton) Lead United States 412 340 423 Mexico 293 364 483 Peru 426 437 540 Australia 421 421 540 Canada 485 567 628 Zinc United States 745 1,131 946 Mexico 553 1,019 959 Peru 673 1,120 1,023 Australia 686 1,129 1,032 Canada 639 1,133 1,039 Source: Brook Hunt and Associates Ltd. is treated as a by-product. It has, therefore, become rare to develop lead-only mines; they are confined mostly to Morocco, South Africa, and the United States. Over the years, lead and zinc mining has gradually shifted to deposits that contain more zinc relative to lead, unless otherwise justified by the presence of silver or gold. Lead production has been almost "involuntary." Lead and zinc production costs have increased substantially since the mid-1980s largely due to exchange rate movements and increases in smelter charges. Table 1 shows estimates of cash production costs for primary lead and zinc for the major producing countries. The United States emerged as the lowest cost producer of lead and zinc, thanks to the depreciation of the US dollar and the start-up of the large low-cost Red Dog mine in Alaska. The changes in costs reveal their dependence on market prices, as mine costs of lead and zinc are often allocated between the two on the basis of their revenue shares.Furthermore, labor costs and smelter charges closely follow changes in product prices and hence in industry profits. Costs expected to faull as prices remain low. Today's low market prices are expected to reverse the recent increases in lead and zinc costs. It is estimated that lead and zinc costs will decline by about 10% over the 1990-95 period, for the existing mines as well as for new mines expected to come on-stream during the period. On a total cost basis that includes capital charges, the average cost of production by 1995 for the operating and expected new mines is estimated at $600/ton for lead and $1,100/ton for zinc. About 90% of all lead and zinc production expected by that time will have total costs at or below $730/ton for lead and $1,320/ton for zinc. Potential for new mine and smeking capacity. During the 1991-96 period, lead mine projects expected to be completed total 159,000 tons, most of it in conjunction with zinc. In 1991, there was a large gross addition to lead smelting capacity-93,000 tons in primary smelting and 119,500 tons 158 in secondary smelting. By the mid-1990s, ten lead smelters are expected to come on-stream, with a total capacity of 202,000 tons. World zinc mine capacity was pared down significantly in 1991--permanent and temporary closures exceeded additions by 141,500 tons. However, effective zinc mine capacity as of the end of 1991 stood at 13% above actual 1991 production. During the period 1991-96, zinc mine projects that are considered firm add up to 628,000 tons. Zinc smelter capacity increased by a wide margin in 1991; net smelter capacity additions amounted to 175,000 tons in 1991, and the largest for a single year since 1984. New zinc smelters and expansions to existing ones expected to come on-stream by 1996 total 328,000 tons, including a 100,000 ton smelter in China. Over the long term, to the year 2000 and beyond, the lead and zinc industry has no shortage of potential new projects that could be developed and brought into production. Price Outlook Market depressed by recession. Lead and zinc prices dropped to record low levels in constant dollar terms in the recent past. The decline was caused by weak demand, particularly in the automobile sector, and increased supplies. The excess of supplies over demand was exacerbated by increased exports from the FSU and Eastern Europe, estimated at 128,000 tons for lead metal and 143,000 tons for zinc over a two-year period (1990-91). Prior to 1990, the FSU and Eastern Europe had been net importers of these metals. Over the same two-year period, total commercial stocks of lead in the market economies increased by 81,000 tons and zinc by 65,000 tons. At 5.5 to 6.5 weeks' consumption, the current stock levels are high compared with those in 1987-88, but not unusually high. However, most of the increased stock converged on the LME warehouses. Fluctuations in the exchange stocks have immediate market impact. Over the short term, lead and zinc prices are not likely to deviate much from the low levels of the recent past. This expectation is based on the assessment that the current market imbalance (excess supplies over demand) is likely to persist for the next two to three years. Net exports from the FSU and Eastern Europe are likely to continue at levels only slightly lower than in 1991. The likelihood of demand increases absorbing the anticipated increases in supplies is small. Real gains from mid-1990s. The market balance is expected to improve by the mid- 1990s, as demand increases catch up with production capacity. Lead and zinc prices, therefore, are projected to gain in real terms by 1995. Assuming that net exports of lead and zinc from the FSU and Eastern Europe will decline to zero by 1996 and that the anticipated increases in smelter capacities all materialize without any closures of existing ones, the market will achieve exact supply/demand balance if demand grows at 1% p.a. for lead and at 1.9% p.a. for zinc over the 1991-96 period. If one half of the gross additions to smelter capacities are offset by closures of old smelters, the anticipated smelter capacity increases can support demand increases of 0.6% p.a. for lead and 1.3% for zinc. We expect lead and zinc demand to grow at 1.0-1.5% p.a. during the 1990s. Thus, based on the available information, one must conclude that market balance will be restored by or before 1996, probably earlier for lead than for zinc. Production costs to fall in long term. For the long term, lead and zinc prices are on average expected to closely reflect the costs of production. Although new greenfield lead and zinc projects have gradually moved to more hostile environments in recent years (e.g., the Red Dog mine in Alaska), their total costs of production have tended to be lower than the existing facilities. For example, the total production cost of the Red Dog mine in 1991 is estimated at $840/ton for zinc and $570/ton for lead. The 159 Cayeli copper/zinc project in Turkey is expected to have total costs in the neighborhood of $950/ton for zinc in 1995. A similar cost structure is expected for the Iscaycruz project in Peru. Thus, new projects to come on-stream in the latter half of the 1990s should have favorable cost structures in relation to existing mines in order to justify project viability in a more competitive environment. We forecast that lead and zinc prices for the years 2000 and 2005 should average about $600/ton for lead and $1, 100/ton for zinc in constant 1990 dollars. 160 Table Ai: Lcad - Prices, 1950-91 (Actual) and 1992-2005 (Projected) ($/ton)______________ 1990 Constant i -C urrent $ -5 MUV a/- - G-7 CPI b/-- US LMdi us ci LM d/ US ci LME d/ Acisi 1950 293 293 1,794 1,794 2,363 2,363 1951 386 446 2,050 2,369 2,809 3,246 1952 363 372 1,839 1,885 2,533 2,596 1953 297 252 1,548 1,314 2,055 1,744 1954 310 265 1,653 1,413 2,113 1,806 1955 334 292 1,748 1,528 2,269 1,984 1956 353 321 1,783 1,621 2,347 2,134 1957 323 266 1,598 1,316 2,101 1,731 1958 267 201 1,299 978 1,705 1,284 1959 269 195 1,327 962 1,721 1,248 1960 263 198 1,271 957 1,651 1,243 1961 240 176 1,140 836 1,472 1,080 1962 212 154 987 717 1,265 919 1963 245 174 1,163 826 1,416 1,006 1964 300 278 1,400 1,297 1,690 1,566 1965 353 318 1,635 1,473 1,927 1,736 1966 333 262 1,490 1,173 1,758 1,383 1967 309 229 1,367 1,013 1,590 1,179 1968 291 240 1,300 1,072 1,457 1,202 1969 328 289 1,390 1,225 1,574 1,387 1970 344 304 1,372 1,212 1,556 1,375 1971 305 254 1,154 961 1,292 1,076 1972 331 302 1,149 1,048 1,271 1,160 1973 359 430 1,076 1,289 1,206 1,445 1974 497 593 1,222 1,458 1,508 1,799 1975 474 417 1,048 923 1,280 1,126 1976 509 445 1,111 971 1,330 1,163 1977 677 617 1,345 1,226 1,601 1,459 1978 742 661 1,281 1,141 1,432 1,276 1979 1,160 1,208 1,768 1,841 2,115 2,203 1980 936 906 1,301 1,259 1,526 1,477 1981 805 727 1,114 1,006 1,302 1,176 1982 562 546 790 767 921 894 1983 478 425 687 612 771 686 1984 563 444 827 652 910 718 1985 420 391 612 570 666 620 1986 486 406 601 502 638 533 1987 792 597 891 672 921 694 1988 819 656 859 688 884 708 1989 868 673 917 711 942 730 1990 1,015 811 1,015 811 1,015 811 1991 738 558 723 546 702 530 Proiected 1992 565 530 506 1993 600 543 519 1994 620 551 524 1995 690 597 567 2000 900 656 613 2005 850 539 484 a/ Dellated by G-5 Manufacturing Unit Value (MUV) Index. b/ Deflated by G-7 Connuner Price Index (CPI). c/ Pig, common grade, domeatic producer price, New York. d/ Soft pig, rofined, 99.97% pb, London Metal Exchange. Sources: E and Minine Joral. Metalleselshaft Metal Statisics and Mea Week (actua); Worid Bank, International 161 Table A2: Zin - Prices. 1950-91 (Atual) and 1992-2005 (oectd 19 Comsat† -Curreat $- - -5 MUV al- -0-7 CP bl- US C/ LMEd US c/ LME d/ US C LME d/ 1950 317 328 1,941 2,009 2,556 2,645 1951 408 473 2,167 2,512 2,969 3,443 1952 369 412 1,870 2,088 2,575 2,875 1953 250 207 1,303 1,079 1,730 1,433 1954 247 216 1,317 1,152 1,684 1,472 1955 282 250 1,476 1,308 1,916 1,698 1956 308 269 1,556 1,359 2,048 1,789 1957 262 225 1,296 1,113 1,705 1,464 1958 238 182 1,158 885 1,520 1,162 1959 263 226 1,297 1,115 1,683 1,446 1960 297 247 1,435 1,193 1,864 1,551 1961 265 214 1,259 1,017 1,626 1,313 1962 267 185 1,244 862 1,593 1,104 1963 276 212 1,311 1,007 1,595 1,225 1964 310 324 1,446 1,512 1,746 1,825 1965 331 311 1,533 1,440 1,807 1,698 1966 331 282 1,481 1,262 1,748 1,489 1967 316 273 1,398 1,208 1,626 1,405 1968 309 262 1,380 1,170 1,547 1,312 1969 333 287 1,411 1,216 1,598 1,377 1970 349 295 1,391 1,176 1,578 1,334 1971 356 309 1,347 1,169 1,508 1,309 1972 391 377 1,357 1,309 1,502 1,448 1973 455 851 1,363 2,550 1,529 2,860 1974 793 1,239 1,950 3,047 2,406 3,759 1975 859 743 1,900 1,643 2,320 2,006 1976 823 712 1,796 1,554 2,151 1,861 1977 758 591 1,506 1,174 1,792 1,397 1978 683 593 1,179 1,024 1,318 1,144 1979 822 742 1,253 1,131 1,499 1,353 1980 825 761 1,146 1,057 1,345 1,241 1981 982 846 1,359 1,171 1,588 1,368 1982 848 745 1,192 1,047 1,389 1,221 1983 912 764 1,312 1,099 1,471 1,233 1984 1,072 922 1,575 1,355 1,734 1,491 1985 890 733 1,297 1,141 1,410 1,241 1936 838 754 1,036 932 1,101 990 1987 924 799 1,040 899 1,074 929 1988 1,327 1,242 1,392 1,303 1,433 1,341 1989 1,808 1,659 1,910 1,753 1,961 1,800 1990 1,644 1,513 1,644 1,513 1,644 1,513 1991 1,163 1,117 1,139 1,094 1,106 1,063 1992 1,300 1,220 1,164 1993 1,250 1,131 1,081 1994 1,250 1,110 1,056 1995 1,280 1,107 1,051 2000 1,610 1,173 1,097 2005 1,630 1,034 929 a/ Deflated by 0-5 Maufacturing Unit Valut (MUV) Index. b/ Ddated by 0-7 Comuer Price Index (CP). al Domtic producer prica. Prior to Jamuary 1991, high grad~/Prime Wetarn, incläde mal amomt of other grades; thereafter %peial mog grd~ d/ London Metal Ex~hang. For year, up to 1984, Good Ordiary Brand, or tandard grad for zin content of 98% or better; 1985-89, high grada, mm 99.95% z; from 1990 onward, apecial high grade, o~nmm 99.995% =. Sour: MMe Ma ~ Statisti Endr (actua); World Bank, International 162 Gold Swnmary * After having fallen for five years, gold prices seem to be near their cyclical bottom and real price increases are expected in the medium term. * Real price increases are expected to take place in a market where fabrication demand growth is around 1.5% p.a., the rate of expansion of supplies is projected to slow down, and Central Bank disposals are not expected to upset the market greatly. * Gold mine production growth has slowed since 1989. Production is expected to stagnate in 1992 and 1993 before rising at a much slower pace. For the largest producer, South Africa, mine output is expected to recover only as prices rise and political instability diminishes. * Gold fabrication demand, especially in jewelry, is projected to continue steadily during the forecast period, while investment demand is expected to increase markedly only in circumstances of political and economic disturbances which are largely unpredictable. Gold Prices a/, 1960-2005 C$/troy oz.) 1,000 800 Constant 1990 S b/ 600 £A 400 IfI I % I I I 200 Current $ 20 1960 1965 1970 1975 1980 1995 1990 1995 2000 2005 a/ 99.5%, fine, London, average daily final rate b/ Deflated by G-5 MUV Index 163 Price Outlook Gold price at cyclical bottom in mid-1992. There are four key components in the gold market that collectively set price levels and trends: total new supply, fabrication demand, private investment demand, and official sector transactions. These variables in turn respond mainly to expected changes in inflation rates, financial market conditions, and interest rates. Three of these four components have had a negative impact on gold prices in the late 1980s and early 1990s. First, total supply has been rising sharply, fueled by increasing mine production and higher exports from the FSU; second, there have been consistent net disposals by central banks; and finally, investors have had reduced enthusiasm for gold, given low inflation, high real interest rates, attractive alternative investments, and a perception that mine companies and the FSU were supplying too much metal to the market. The fourth factor, fabrication demand, generally rose throughout this period, a trend supportive of gold prices, but not sufficiently forceful to counter the negative price implications of the other three factors. Nominal prices consequently fell 32.5% from a peak of $497.10/oz in late 1987 to $335.30/oz in early 1992. There are reasons to believe that the price level of around $335/oz in mid-1992 is near a cyclical bottom, after five years of declining prices. For the projection period, the outlook is more balanced. Fabrication demand should grow moderately. In the four decades after 1950, the compound growth rate of gold fabrication demand was 4.5% p.a., although in the 20 years from 1970 to 1990, fabrication demand grew at a 2.6% p.a. rate, as prices rose sharply in a post-Bretton Woods adjustment period. The rate of growth projected for fabrication demand for the next 13 years are below historical growth trends, at around 1.5% p.a. Supply meanwhile has ceased increasing at the fast pace of the late 1980s. The growth rate of newly-refined gold supplies entering the market has slowed from a 5.4% p.a. average in 1988-89 to 2.6% p.a. in 1990 and 1991. A further reduction in the rate of expansion of gold supplies, including scrap recovery and exports from the FSU, is projected. Real prices to increase in medium term as demand grows and supply stalls. These two factors alone would be sufficient to push real prices up moderately over the next five to ten years. But in addition, investment demand can be expected to increase at times during the forecast period in the event of political changes and economic disturbances such as high inflation and financial and currency market instability caused by cumulative fiscal and trade imbalances. In order to stimulate sufficient supplies to meet projected fabrication and investment demand in the late 1990s, we expect real prices to increase. (T7hese real price increases are unlikely to be greatly affected by possible higher central bank sales.) Higher real gold prices should stimulate higher secondary recovery of gold from scrap as well as keeping mine production near recent high levels. Mine production in Russia and the other FSU republics is expected to recover from the declines in recent years and possibly reach new highs by the end of the century, especially with an infusion of capital and technology from the market economies. The projected rates of growth in mine production, total supply, and fabrication demand might appear low, but they are supported by a number of factors at work in the market. First, mine output growth is expected to be restrained by cutbacks and closures of existing mines, increased investor wariness towards major capital expenditures, and capital constraints for gold-mining ventures. Next, total 164 supply growth is likely to be restrained by the slow restructuring of mine production in the economies of South Africa, Eastern Europe, and the FSU and by the reduced profitability of recycling gold from redundant electronic equipment and other scrap. Finally, the rate of increase in fabrication demand is expected to be restrained by a combination of lower expectations of income growth in industrial countries, a shift away from heavy dependence on gold jewelry as a form of savings in some developing countries, and "portfolio saturation" among recent heavy buyers of gold jewelry in industrial countries. All of these factors have been present in the market during the early 1990s, and thus the projections here represent a continuation of trends currently unfolding, not a reversal. It should be noted that our projections for slower growth in mine production and total supply, are in contrast to some analysts' perceptions that supplies could decline over the forecast period-the information we have available does not support such a radical forecast. Supply Outlook The total supply of gold rose sharply in the past several years-from less than 50 million ounces per year in 1982-84 to 76.1 million ounces in 1991. Total supply consists of mine production (72% in 1991), secondary recovery from old scrap (15%), and net sales from the FSU, Eastern Europe, and centrally planned Asia (13%). Over the period 1971-91, total supply rose at a 3% p.a. rate, while mine production rose at a 1.3% p.a. rate. However, since 1977, the compounded growth rate of total supply was 3.3% p.a., while that of mine production was 4.3% p.a.-illustrating the tremendous increase in gold mining stimulated by the sharp increase in gold prices in 1979-80, and the fact that prices have remained above $300/oz for most of the subsequent 12 years. Over the forecast period, the rate of expansion of total gold supplies is projected to slow considerably. Between 1990 and 2005, total supply is projected to increase at a 1.2% p.a. rate, reaching 88.9 million ounces by the end of the forecast period. The growth in mine production is projected to slow even more dramatically, possibly to less than 1% p.a.. MINE PRODUCTION. Widespread restructuring causes fall in growth ofamine production. In the late 1980s, mine production rose at a very fast pace, in excess of 6% p.a. This illustrates the considerable lag in production in response to a price increase-a number of major projects conceived in the early 1980s in response to higher gold prices, were brought into production only in the late 1980s. In 1988, mine production rose 12.8% to 47.4 million ounces, in 1989 the rate of increase was 8.6%, in 1990 it slowed to 3.7%, and in 1991 to 2.5%. In 1992, the rate of expansion may be no more than 0.3%, and in 1993 total gold mine output may be roughly unchanged from 1992 levels. Beyond 1993, mine production is projected to begin rising again, although at a much slower pace than was experienced in the 1980s. South Africa remains the largest gold-producing country by a wide margin. In 1991, South African output totaled 19.3 million ounces. The United States has become the second largest producer, at 9.3 million ounces. While South African production has fallen in recent decades, it has stabilized in recent years and is much higher than market observers previously expected. It may decline in the middle of the 1990s as many of the mines in South Africa are experiencing severe cash-flow problems-more than one eighth of South Africa's present production capacity was operating at a loss as of the middle of 1992. Several mines have closed and cutbacks have occurred at others. Some mines 165 have redesigned their mining plan, raising their average head grade and thus increasing production in the short term. Other new projects, some designed as replacement capacity, have been postponed indefinitely, awaiting gold prices high enough to warrant the massive capital investments necessary, as well as of an improved political outlook for South Africa. Depending on the economic and political outcome of the current South African transition process, it is possible that South Africa could see a rejuvenation of its gold mining industry by the turn of this century, with total gold output rising again. Another country where gold production had been expected to decline was Australia. Australia's gold production rose extremely rapidly in the 1980s, however, from 867,000 ounces in 1982 to 7.9 million ounces in 1990. In 1991, the government altered the tax regime, imposing income taxes on gold mining profits. As a result of this, and the simultaneous decline in gold prices, the Australian gold mining industry has been restructured. Even so, production has held up reasonably well and is expected to stabilize at around 7.4 million ounces per year. Initially, Australia's gold production was expected to decline precipitously as a result of these developments. There have been a few closures in the past two years, but the major trend has been the consolidation of mining interests, rather than the closure of mines. Thus, there remains a large base of gold-mining companies in Australia, in possession of established, profitable gold-mining operations and resources. These mines should continue to operate for at least the next eight years. Some are expected to face diminishing ore reserves, but new mines are being developed or planned to replace the depleted operations. Given the extended period of relatively low gold prices, exploration and development efforts have fallen worldwide in recent years. One survey showed that there were 143 gold mine developments or expansion programs in progress in 1988, but by 1992 the total had fallen to 53 projects. While the pace of new developments has slowed, it nonetheless continues, indicating the continuation of incremental increases in gold production capacity for several years to come. New projects are in progress in many countries. In Ghana, once a major producer, the gold mining industry is being revived. Production has risen from 287,000 ounces in 1984 to 845,000 ounces in 1991, with further increases projected for the next several years. Papua New Guinea has been the site of several major new mining projects, boosting output from 564,000 ounces in 1982 to nearly 2 million ounces in 1991. Indonesia also has seen some major developments, with production rising from 72,000 ounces in 1982 to 828,000 ounces in 1991. Further significant increases in Indonesia's output are projected for the mid-1990s. Other countries showing significant increases in output include Chile while numerous countries have reported smaller increases in production. SECONDARY RECOVERY. Secondary recovery continues to be highly price-elastic. The recovery of gold from old electronic equipment, dental alloys, and, most important, jewelry and decorative items, increased sharply in the late 1970s, in line with the rise in gold prices. Scrap recovery totaled 4.5 million ounces in 1977, the first year for which such statistics are available. As gold prices doubled in 1979, so did secondary recovery, to 10.9 million ounces. Gold prices doubled again the following year, and secondary recovery once more doubled, to 21.2 million ounces. After that, gold prices and scrap recovery rates both subsided. Since 1981 secondary recovery of gold has ranged between 9 and 13 million ounces on an annual basis. Secondary recovery rates are expected to continue to be highly price-elastic so that when gold prices rise, as projected, secondary recovery of gold also would be expected to increase. FSU PRODUCTION. FSU production expected to return to previous peak. FSU production is estimated to have fallen from more than 10 million ounces per year in the middle of the 166 1980s to around 7.2 million ounces in 1991. Russia is estimated to have accounted for nearly two thirds of this output, with Kazakhstan and Uzbekistan producing much of the remainder. Even as mine output was declining, however, exports rose. Some of these exports were drawn from government reserves of refined bullion. Over the forecast period, gold mining in the FSU is expected to undergo a transformation to new levels of efficiency and output, aided by the introduction of capital and technology from the market economies. Already in 1992, several major joint-venture agreements have been announced, and many major mining companies are examining deposits and negotiating joint-venture programs. It is possible that by the end of the century mine production in the FSU will exceed the previous peak of 10 million ounces per year. Beyond the mid-1990s, increasing mine production from new projects can be expected to support an increase in total shipments. Demand Outlook FABRICATION DEMAND. Steady growth in demand for gold jewelry to continue. Fabricators are projected to use 72 million ounces of gold in 1992. This is 60% higher than the 44.7 million ounces of gold used annually as recently as 1986-87. Of this total, 89%, or 64.3 million ounces, is projected to be used in the fabrication of jewelry-the bulk of the increase over the past several years has been in the jewelry sector. Gold jewelry consumption has risen most strongly in the newly industrializing countries of Asia, although gold jewelry consumption in the industrial countries has also been growing strongly, partly because of favorable economic conditions during most of this period. Changes in labor force participation rates have contributed to the rise in jewelry demand, too, especially through their effect of heightening the role of women as purchasers in industrial countries. Strong growth in gold jewelry consumption has also been recorded in the Middle Eastern countries, Eastern Europe, and the FSU. During the next 13 years, fabrication demand for gold is expected to remain vibrant. Gold jewelry demand has not shown a high degree of price elasticity, o that demand would be expected to rise even in the event of moderately higher gold prices. Fabrication demand rose at a 2.6% p.a. rate from 1970 to 1991, although gold prices rose sharply during this time. From 1979 to 1990, as gold prices fell, fabrication demand rose at a compounded rate of 5.1%. The growth rate is projected to be 1.5% over the period 1990-2005. This would appear to be a conservative forecast. Market analysis does not point to a decline in consumer appetites for these products during the forecast period-market studies conducted by the World Gold Council and several other independent groups point to a continuation of strong, fashion-led demand growth for gold jewelry in the industrial countries. Additionally, demand growth in the newly industrializing countries is expected to continue apace, although economic and market studies in individual countries point to a slowing down in the growth rate from the fast rates seen in the late 1980s and early 1990s. OPPICIAL TRANSACTIONs. Central bank disposals not expected to upset gold prices markedly. The official sector probably represents both the greatest threat to higher gold prices in the 1990s and the market's greatest enigma. Central banks and intergovernmental agencies hold more than 1.1 billion ounces of gold, equivalent to 15 years of total demand at current and projected rates. For the most part, these holdings are vestiges of the various gold currency standards employed prior to 1971. A trend toward reducing these holdings has emerged since the mid-1980s, a trend which has gathered momentum since 1990. Since 1987, central banks have been net sellers of 8.8 million ounces, after accounting for the 9.7 million ounces purchased by the Bank of China (Taiwan) from 1987 through 1989 167 Gold Supply and Demand, 1970-2005 100 N 60 0 'II 1970 1975 1980 1985 1990 1995 2000 2005 otal SupplyFabrication Demand as part of a dollar sterilization program. In 1989 alone, net disposals totaled 6.2 million ounces. Should monetary authorities more aggressively seek to reduce their gold holdings, as some have indicated they wish to do, the impact on gold prices could be significant. Two aspects of the impact of potential official sales on gold prices need to be assessed. The first is the amount of supply which will be added to the market each year. Two or three million ounces of gold per year is not a large amount for the market to absprb, and sales at these levels would be expected to have only a slight effect on gold prices. However, should sales rise to around 10 million ounces per year, a significant decline in prices is likely to occur. The second aspect is the investment demand level. Annual investment demand levels are extremely volatile, fluctuating between 6 million ounces to as much as 26 million ounces. Should investment demand remain at the low levels of 1989-9 1 (below 10 million ounces per year), increased supplies from central banks could have a depressing effect on prices. But if investment demand were to rise to high levels and remain there during the forecast period, a major increase in net official disposals could readily be absorbed without substantial negative price consequences. Net official sales of 8.7 million ounces in 1977, 11.7 million ounces in 1978, and 17.5 million ounces in 1979 largely went unnoticed as the gold price rose from $127/oz in early 1977 to a peak of $825/oz in January 1980. In fact, gold prices would have risen sooner and higher were it not for these official sector gold sales, which at that time were the largest government gold disposals in history, and represented a 26% increase in new gold supplies during the three-year period 1977-79. For the projections made here, combined fabrication and investment demand is expected to be sufficient to accommodate central bank net disposals of 7 million ounces per year without significantly reducing gold prices. As outlined elsewhere in this chapter, even with a significant decline in the growth rate of jewelry demand for gold, fabrication demand is projected to rise from around 72 million ounces in 1992 to 78.2 million ounces in 2000. Investment demand levels of between 8.4 million ounces and 15 million ounces per annum are projected-levels that are within the lower range of investment demand levels in the past 25 years. Even at these low levels of demand, fabrication and 163 investment demand would absorb the higher amounts of gold disposed of by central banks and other monetary authorities. INVESTMENT DEMAND. Investment demand in 1990sfaces mixed pictum. Investors buy gold for many reasons, among the most prominent being: (a) as an inflation hedge; (b) as a portfolio diversifier; (c) as a currency hedge or surrogate; (d) as a safe haven in times of financial, economic, or political uncertainty; (e) as a commodity, when they perceive gold as having strong supply and demand fundamentals; and (f) as a form of savings, both in industrial countries and in areas where banks and other financial institutions either do not exist or do not have a reputation for security and stability. The interplay of these factors is crucial. In 1978-80, all of these factors stimulated investment demand for gold, which led to the massive rush on gold by investors worldwide. Conversely, for much of the 1980s these factors were not particularly stimulative of investor demand and prices fell. There were, however, periods during the 1980s when one or more of these factors stimulated some investor interest in gold. For example, in 1986 when the US dollar fell sharply against most major currencies, investors interested in gold as a currency hedge increased purchases thus stimulating prices. At the time, inflation was very low--consumer prices in OECD countries rose only 2.8% in 1986, while US consumer price inflation, at 1.9%, was at a post-war low. Thus, investors who viewed gold as an inflation hedge were not buying gold, which restrained prices. The net result was that the annual average gold price rose 15.9% in 1986. The nature of investment demand for gold in recent times has been that investors have been net buyers of 6.7-16.4 million ounces of gold annually in most years. Occasionally, investors have entered the market for greatly increased volumes of gold-in the range of 21 to 27 million ounces at an annual rate. This has occurred when investor interest in gold as a safe haven has been reinforced by expectations of attractive returns from gold relative to competing assets. Despite some market commentary to the contrary, there has been no evidence that investors have changed their views toward gold, abandoning gold for exchange-traded assets. In contrast, there have been indications, on both national and international levels, that investors continue to view gold as a viable investment alternative, given the appropriate economic circumstances. But over the past ten years, those factors that stimulate investor gold "rushes" have been largely absent. Inflation has remained under control in most industrial countries, at least compared with the experiences of the 1970s; the performance of stocks, bonds, and other assets that compete with gold have generally been good, again when compared with the experience of the previous decade; and most importantly, real interest rates have remained high for most of the time, while gold's fundamental supply and demand conditions have been relatively unsupportive of higher prices. In this environment, investors have found gold of reduced interest. Net investment demand has been around 9 million ounces per year in 1991 and 1992. The 1990s are likely to present a mixed picture for gold investors. Lower inflation is projected for the industrial countries during the 1990s compared with the 1980s, which would be a negative for gold investment demand. Interest rates are also projected to be significantly lower in the 169 1990s, which would be supportive of gold investment demand and higher prices. Our projection of lower real oil prices in the first part of the 1990s, followed by higher oil prices in the second half of the decade, suggests weaker gold prices to 1995, but higher prices later. However, the timing or occurrence of the kinds of economic events that lead to sharply increased investor demand for gold cannot be projected. 170 Tabbe Al: Iag-Tun GoM Supy mi D d (Mili Troy 0 ) Privata Iav~n Yefr prod~ie. iw f em a ey Tal Oioi yewely and Cos and 3T0~ia- inM ~dt Timein Reeovey Skpp Tre-m Indutuial Lk-ai--- Purcaae Ece---d Ec---il le ae (+) (+) erSSl(- or3Sals(-) 24.3 - N.A. 24.3 9.2 12.0 • 3.1 1950 33.5 5.7 N. 39.2 8.4 25.0 • 5.8 1960 40.9 -0.1 N.A. 40.8 7.6 41.1 3.2 -11.1 1970 39.7 L7 N.A. 41.4 -3.1 41.3 3.4 -0.2 1971 37.8 6.8 NÅ. 44.6 4.9 39.9 3.3 -3.5 1972 35.8 8.8 N. 44.6 -0.2 25.2 2.4 17.2 1973 32.8 7.1 N.A. 39.9 -0.6 14.3 9.0 17.2 1974 30.9 4.8 N.A. 35.7 -0.3 22.8 8.8 4.4 1975 31.2 13.2 N.A. 44.4 -1.9 37.2 7.5 1.6 1976 31.0 12.9 4,5 48.4 -8.7 43.9 6.3 6.9 1977 31.1 13.2 5.7 50.0 -11.7 45.2 10.8 5.7 1978 30.7 6.4 10.9 48.0 -17.5 39.7 10.4 15.4 1979 30.4 2.9 21.2 54.5 7.4 25.7 8.9 12.5 1980 31.1 9.0 10.5 50.6 4.5 33.4 8.2 4.5 1981 32.3 6.5 11.1 49.9 1.0 36.7 6.5 5.7 1982 34.1 2.7 11.7 48.5 -2.4 35.6 6.8 8.5 1983 35.6 3.9 10.4 49.9 -1.5 41.6 5.5 4.3 1984 37.8 7.6 9.0 54.4 1.5 44.1 5.2 3.6 1985 40.2 13.2 13.0 66.4 1.0 44.5 11.4 11.2 1986 42.0 10.1 12.8 64.9 -2.9 44.9 7.5 15.4 1987 47.4 10.1 11.1 68.6 5.3 56.6 5.2 1.5 1988 51.5 11.0 9.7 72.2 -6.2 63.0 5.7 9.7 1989 53.4 10.0 10.7 74.1 -2.0 68.6 4.6 2.9 1990 54.8 10.0 11.3 76.1 -3.0 70.1 4.0 5.0 1991 54.9 11.0 12.0 77.9 -6.0 72.0 3.5 8.4 1992 54.9 11.0 12.0 77.9 -4.0 72.7 9.2 1993 55.1 10.0 13.0 78.1 -6.0 73.0 11.1 1994 55.4 11.0 14.0 80.4 -8.0 73.4 15.0 1995 57.0 12.8 14.0 83.8 -7.0 78.2 12.6 2000 58.9 1.40 16.0 88.9 7.0 87.4 8.5 2005 Nots: N.A. - data on secondary supply is not av~ilabe for years prior to 1977. Demand ~atistics for years prior to 1977 r~presen usa of gold from minm production, trwaitaI ecofomy sal, and official sctor soue~s, and ex~lude use of gold from scrap, while demand satios beginning with 1977 sefleot total gold usm, including metal ecovered from acrp. *Coins and medallions prior to 1968 are included in dh~ jewelry snd industrial ct.m. Source: CPM Group. 171 Table A2: Major Gold Producem 1982-91 1982 1913 1984 1985 1986 1987 1988 1989 1990 1991e ('000 o ounces) South Africa 21,355 21,847 21,905 21,598 20,518 19,437 19.927 19,492 19,380 19,326 United States 1,466 1,956 2,059 2,475 3,739 4,947 6,460 8,537 9,330 9,300 Australia 867 1,035 1,257 1,833 2,508 3,533 5,048 6,752 7,877 7,555 Canada 2,081 2,223 2,491 2,815 3,308 3,724 4,334 5,128 5,284 5,624 Brazil 1,434 1,640 1,726 2,025 2,375 2,694 3,222 3.305 2,807 2,667 Philippines 830 812 773 810 1,139 1,048 1,044 964 940 800 Papua New Guines 564 584 835 1,067 1,157 1,190 1,286 1,085 1,093 1,967 Colombia 460 439 800 1,143 1,285 853 972 871 956 946 Chile 546 571 541 554 588 548 663 713 735 922 Zimbabwe 426 453 478 472 475 473 481 515 543 572 Ghana 330 304 287 299 287 324 390 428 487 845 Mexico 196 223 244 257 267 289 344 385 322 286 Dominican Republic 379 361 338 337 246 246 186 177 139 130 Peru 165 166 137 135 145 150 150 139 111 121 Indonesia 72 77 79 300 320 440 500 515 600 828 Others 1,161 1,443 1,661 1,701 1,871 2,142 2,393 2,464 2,758 2,872 Total 32,332 34,134 35,611 37,821 40,228 42,038 47,400 51,480 53,362 54,761 Notes: Totals may not equal the sums of the columns due to rounding. e = estimated. Sources: Chamber of Mines of South Africa U.S. Bureau of Mines; American Bureau of Mesis Statistica; Gold Insitutr; Staistics Canada; in.dusry aources; CPM Group. 172 Table A3: Gold - Prices, a/ 1963-91 (Actual) and 1992-2005 (Projected) (!tr:y oz.) Current $ - 1990 Constant $ G-5 MUV b/ G-7 CPI c/ Actual 1963 35 167 203 1964 35 163 197 1965 35 162 191 1966 35 157 185 1967 35 155 180 1968 39 173 193 1969 41 174 197 1970 36 143 163 1971 41 154 173 1972 58 202 223 1973 97 292 327 1974 159 392 483 1975 161 356 435 1976 125 272 326 1977 148 293 349 1978 193 334 373 1979 307 467 559 1980 608 845 991 1981 460 636 744 1982 376 528 616 1983 423 608 681 1984 360 530 583 1985 318 463 504 1986 368 455 483 1987 446 503 519 1988 437 459 472 1989 381 403 414 1990 384 384 384 1991 362 355 345 Prolete 1992 345 324 309 1993 350 317 303 1994 377 335 318 1995 393 340 323 2000 494 360 337 2005 567 360 323 a/ 99.5% fine, London, average of daily afternoon second fixing, or final rates. b/ Deflated by G-5 Manufacturing Unit Value (MUV) Index. c/ Deflated by G-7 Consumer Price Index (CPI). Sources: International Monetary Fund, International Pinancial Statistics to 1983; Shearson Lehman Brothers Metal Market Weel friFm1 to 1988; Metals Week from 1989 onwards (actual); World Bank, International Economics Department (projec. 173  ·，!,-,-,!-,；긔 P)Ice Outlook The silver market has experienced substantial changes in market fundamentals over the past 40 years. A turning point was reached in 1979 and 1980, when speculative buying pushed up silver prices from $5/oz to $50/oz. The surge in prices led to the reversal of many of the long-term trends. First, production and secondary supply rose and have remained high. Second, fabrication demand fell sharply in the early 1980s; it was not until the second half of the decade that silver prices fell to levels that allowed sharp increases in fabrication demand fbr silver, although this trend is being offset at present by weak economic conditions that are limiting demand for silver-bearing products. Third, and most importantly, investors abandoned the metal. Established mines have been faced with the possibility of closure, some development plans have been deferred, and other producers have stockpiled silver to wait for higher prices. Large inventorks and investor sales likely to eidendperiod of low prices. Fromalong- term perspective, silver production appears unsustainable at the current low price level. However, the large inventories and net investor sales are likely to extend the period of low silver prices for some time yet. The silver market historically has been well supplied; sharp increases in fabrication demand have been met easily with inventories, increased mine production, and secondary supplies. Any positive effect of increased silver use on silver prices has usually been minor. Typically, it has taken several years of high demand to tighten the silver market sufficiently to affect prices. For example, silver's supply/demand balance shifted into deficit in 1971, with the withdrawal of the US government as a major supplier of silver, yet it was not until 1979 that investor inventories had been drawn down sufficiently for the full impact of the shift in the market balance to be reflected in silver prices. The silver market in the early 1990s is looking much the same as it did in the 1950s and 1970s. Fabrication demand is rising strongly, while new supplies coming into the market are relatively stable. These two factors are supportive of higher prices, but the presence of large inventories of refined silver and the recent sales by investors from their massive holdings have kept prices down. Since 1990, the amount of newly-refined silver entering the market from all sources (total supply) has been less than fabricators have required, but the deficit has been filled by investor sales. Still, the silver market may have passed a turning point in 1990 and 1991. The surplus of total supplies over demand shrank consistently throughout the 1980s and turned to a deficit begi i in 1990 (see Figure 1). So silver prices may have stopped falling, reversing a trend that started in 1983; the low of $3.50/oz in early 1991 may well represent the cyclical bottom. Silver prices should strengthen in the years ahead, although the rise may be only mild and erratic. Real, inflation-adjusted silver prices are projected to rise from around $3.75/oz in 1992 (in 1990 dollars) to around $4.50/oz by the year 2000. Demand Outlook FABRicATioN DEmmD. Fabrication demand to rise steadily with technological innovation andprice competitiveness. Fabrication demand fell sharply in the early 1980s, following the high prices reached in 1979 and 1980. However, since silver prices have fallen to low levels in the latter half of the 1980s, fabrication demand has increased sharply (see the figures on industrial demand in Table Al). The weakening economic activity in the early 1990s seems to have slowed this growth. 175 Silver Supply and Demand, 1970-2005 Sao 700 600 400 - f I ' E 300 200 100 1970 1975 1980 1985 1990 1995 2000 2005 Supply Demand Fabrication demand is projected to rise at least 2.9% p.a. from 1990 through 2005-from 500.5 million ounces in 1990 to 772.3 million ounces in the year 2005. Technological innovations appear likely to support the increased use of silver in the 1990s. Equally important, silver's low price-nominally, inflation-adjusted, and relative to the prices of substitute materials-makes silver appealing to industrial users. Growth is expected to exceed 2% p.a. in all of silver's established markets, except electronics. Photographic use of silver, growing at 3.5% p.a., is projected to rise at the fastest rate. By 2005, photography is expected to account for 40% of total industrial use of silver-up from 37% in 1990. INVESTMENT DEMAND. Investor interest likely to remain low during forecast period. Investor buying of silver averaged 132 million ounces per year in the early 1980s, as buyers hoped for a repeat of the ten-fold increase in silver prices that occurred in 1979 and 1980. As the decade progressed and silver prices continually fell, investor disenchantment grew. By 1988, investors were buying less than 33 million ounces annually, while in 1989, investors sold an estimated 17 million ounces on a net basis. This was the first year of net investor disposals since 1978. The following year, 1990, net investor sales rose to an estimated 40.3 million ounces, and in both 1991 and in 1992, investor sales appear to be between 20 and 25 million ounces. Two factors readily explain the decline in investor demand for silver. First, silver was not an attractive investment during the late 1980s. Inflation was under control in most industrial countries, stocks and other financial instruments offered attractive returns, and silver's supply/demand conditions were not supportive of higher prices. The second factor was that the channels for distribution of silver investment products had largely deteriorated. In the past, most major brokerage houses and many large commercial banks offered silver investment products, but few offer these services today. 176 Over the forecast period, the second factor is expected to negatively affect investment demand, limiting investors' interest in silver. However, the first factor, the investment environment, is more critical in determining investment demand levels for silver, and this is expected to be mixed: at times some elements, i.e., inflation, silver's supply/demand outlook, and silver's overall ability to offer competitive returns vis-a-vis other investment opportunities, are likely to stimulate increases in investment demand. But such periods may be relatively limited in duration. Thus, over the coming 13 years, investors may prove relatively disinterested in silver. That does not preclude major, short-lived interventions of investors, such as were seen in 1973-74, 1979-80, 1982-83, and 1987. Supply Outlook TOTAL SUPPLY. Most supplies unresponsive to small changes in price. Newly-refined silver supplies are not likely to exceed silver demand for most of the forecast period. About one half of new silver supplies are in the form of by-product production from gold, copper, zinc, and lead mines. This source of supply is not very responsive to silver price changes. Few base metal miners would expand production because the price of their silver by-product had risen. Another quarter of supply derives from secondary recovery. Here, the production costs range from $1/oz to $3/oz, so that higher silver prices would have a relatively muted impact on the profitability of silver recovery efforts. In the 1970s and early 1980s, these sources were highly price- elastic, but in recent years, the price responsiveness has deteriorated. This new situation is likely to persist for much of the coming ten years. Indeed, a more compelling factor than price behind silver scrap recovery today and in the years ahead is likely to be environmental regulations related to protecting water supplies from silver-bearing discharges. The sources of secondary silver should be considered in two distinct groups. The first group is spent photographic products, used industrial catalysts, and redundant electrical and electronic equipment. The costs of recovering silver from these materials range from 075/oz to around $3/oz, in 1992 dollars. Following the sharp increase in silver prices in 1979-80, many manufactures implemented thorough recovery operations, and since silver prices have remained sufficiently high to make these recovery programs profitable in the ensuing 12 years, most have remained in effect (a move aided by the introduction of more stringent environmental controls and legislation regarding the disposal of these materials). As long as silver prices remain above these costs of recovery, changes in silver prices should have a minimal effect on the recovery of these materials. The second group of secondary silver sources is silver in old coins, jewelry, sterlingware, and decorative items. These items often have intrinsic value in excess of their silver content so that someone wishing to dispose of old jewelry, sterlingware, or decorative pieces would earn more by selling them into the antique market rather than into the silver scrap market. Similarly, old silver coins would usually fetch a better price in the numismatic coin market than in the silver scrap market. It is only when silver prices rise sharply that these materials appear in the silver secondary supply market. The one time in recent history when such materials constituted a significant flow of scrapped goods for refining into silver bullion was during the speculative binge of 1979-80, and then it was only when silver prices rose to levels above $20/oz (in nominal 1979 dollar terms). Given that silver prices are not projected to even approach these levels, such scrap flows are not anticipated. Most of the remaining quarter of total new silver supplies comes from mining operations in which silver is either the main economic product or one of the major economic products. Mine 177 production costs at these operations range from around $2/oz, at a few extremely high-grade Mexican operations, to $7/oz at some of the older, unionized underground mines of North America. In 1992, as in 1986, low silver prices are leading to closure of some of these mines. Perhaps 10 million ounces of production capacity has been taken out of operation since 1990 due to low prices. Not all of this would come back into production were prices to rise. Numerous silver-producing mines are being developed or planned at present. Most of these are nonferrous metals or gold-mining operations, with silver as a by-product. The McArthur River deposit in Australia is expected to be producing more than 2.2 million ounces of silver per annum after it commences in 1996. Several other major lead and zinc mines are being considered in Australia that also would have significant silver by-product, including the Century, Benambra, and Lady Loretta deposits. In the United States and Canada, several developments are underway, each of which should be producing 300,000 ounces or more per year of silver as by-product by 1996. The Eskay Creek deposit in Canada would have by-product output of 1 million ounces of silver per annum, if it is developed. The Montanore deposit that Noranda Mines is considering in Montana is said by the company to have the potential to produce up to 20 million ounces of silver per annum. The total silver supply is projected to rise at 2.2% p.a. between 1990 and 2005-from 505.1 million ounces in 1990 to 695.8 million ounces in 2005. The increase is expected to be equally divided between mine production and secondary recovery from scrap. Mine production is expected to rise at 1.4% p.a., especially in the form of by-product mine output. By-product silver output from major new mines is expected to peak in 1991-92 and then flatten out. Some of these new mines will experience a decline in by-product silver production by the mid-1990s. The 1990s also should see a continued decline in the recovery of silver from non-photographic scrap, although this trend will be more than offset by the projected increase in silver recovery from scrapped photographic products. INVENToREs. Inventory disposals expected to be high during forecast period. Market inventories are an important variable in determining silver prices. They may be grouped into two parts, reported and not reported. Reported stocks receive the most attention, since they are visible, but they are only a portion of the total silver stock available on relatively short notice. As of the middle of 1992, for example, reported market stocks were around 313 million ounces, while among those not publicly recorded were 200 million ounces or more held in Delaware and European inventories of 400 million ounces or more. It is estimated that total refined silver inventories in 1992 are of the order of 1.2 billion ounces. Silver inventories are held primarily by investors and bullion dealers, both of which are highly price sensitive. Disposals from these large inventories could accommodate a great deal of the projected growth in fabrication demand for silver over the forecast period. The outlook for the forecast period is that an extended pe7iod of deficits in annual supply flows are likely to be compensated for by large-scale disposals from inventories. The total worldwide inventories of refined silver bullion may fall by three quarters over the coming 13 years, with most of the inventory withdrawals projected to occur later in the forecast period. From 1992 through 1995, for example, the cumulative deficit is projected to be only 221 million ounces. An interesting historical corollary to this projection exists. From 1964 through 1970, investors, primarily in the United States, absorbed 620.5 million ounces of silver in the expectation of higher silver prices. Between 1971 and 1978 they sold two thirds of this, 415.8 million ounces, back into the market. 178 Table Al: Long-Term Silver Supply and Demmnd, 1950-2005 Supply Deand Year Mine Secondary Other TOal ld~sriul Co~iage Total Surphu or Out put De~ii (mil-ln troyounce) Actu8! 1950 169.5 N.A. NA. 169.5 157.4 44.1 201.5 -32.0 1960 201.8 52.0 90.0 343.8 235.3 103.9 339.2 4.6 1970 258.5 127.0 91.0 476.5 372.9 23.4 396.3 80.2 1971 247.3 127.0 7.0 381.3 386.6 27.8 414.4 -33.1 1972 248.9 112.0 12.0 372.9 427.4 38.1 465.5 -92.6 1973 254.0 122.0 46.0 422.0 516.5 28.5 545.0 -123.0 1974 236.6 192.0 21.0 449.6 466.3 31.6 497.9 -48.3 1975 239.0 177.0 18.0 434.0 404.5 33.4 437.9 -3.9 1976 242.9 235.0 18.0 495.9 481.0 30.0 511.0 -15.1 1977 263.3 169.0 13.0 445.3 451.0 34.5 485.5 -40.2 1978 266.8 152.0 16.0 434.8 444.4 39.5 483.9 -49.1 1979 272.0 216.0 17.0 505.0 426.0 31.0 457.0 48.0 1980 266.5 302.0 18.0 586.5 306.0 15.0 321.0 265.5 1981 283.2 184.0 12.0 479.2 326.2 9.5 335.7 143.5 1982 298.0 155.0 3.0 456.0 368.4 12.0 380.4 75.6 1983 313.6 197.5 20.0 531.1 330.3 10.2 340.5 190.6 1984 324.5 165.6 -14.0 476.1 368.3 13.7 382.0 94.1 1985 329.9 140.9 12.0 482.8 390.4 13.4 403.8 79.0 1986 320.3 129.3 -4.4 445.2 432.1 26.8 458.9 -13.7 1987 338.2 137.9 26.1 502.2 418.8 30.4 449.2 53.0 1988 344.0 143.9 14.2 502.1 453.0 25.3 478.3 23.8 1989 358.4 136.2 17.0 511.6 483.5 26.3 509.8 1.8 1990 370.1 118.0 17.0 505.1 500.5 29.8 530.3 -25.2 1991 366.6 111.2 19.8 497.6 503.9 29.9 533.8 -36.2 1992 369.0 111.0 16.6 496.6 516.0 30.0 546.0 -49.4 1993 375.0 116.0 15.0 506.0 529.5 35.0 564.5 -58.5 1994 380.0 126.0 17.0 523.0 541.9 37.0 578.9 -55.9 1995 382.0 140.0 17.0 539.0 559.2 37.0 596.2 -57.2 2000 419.0 190.0 8.0 617.0 658.0 30.0 688.0 -71.0 2005 454.5 237.3 4.0 695.8 772.3 28.0 800.3 -104.5 Notea: *Compounded growtc ate for tecondhry and other mupply it for 1960-90. N.A. = Not available. Sources: CPM Group; industry source. 179 Table A2: Silver Use, 1977-92 Photography Bsic X-Ray Graphics Total Ilectronics Jewey and Other Total silverware (million tray o ) 1977 N.A. N.A. N.A. 129.6 73.7 91.0 151.7 451.0 1978 N.A. N.A. N.A. 142.9 79.3 91.9 128.5 442.6 1979 N.A. N.A. N.A. 146.1 80.4 79.1 132.9 438.5 1980 N.A. N.A. N.A. 123.8 71.2 47.9 114.0 356.9 1981 N.A. N.A. N.A. 123.1 62.2 43.5 114.2 348.0 1982 N.A. N.A. N.A. 133.3 61.7 50.2 108.6 353.8 1983 N.A. N.A. N.A. 138.3 60.4 42.1 107.7 348.5 1984 N.A. N.A. N.A. 144.2 62.3 41.3 119.3 367.1 1985 N.A. N.A. N.A. 148.2 64.7 47.5 111.1 371.5 1986 N.A. N.A. N.A. 153.5 66.9 61.3 118.4 399.6 1987 N.A. N.A. N.A. 164.1 61.1 66.0 128.6 419.8 1988 N.A. N.A. N.A. 176.3 65.3 69.3 130.2 441.1 1989 87.0 50.0 40.1 177.1 70.6 75.8 142.5 466.0 1990 91.1 55.0 39.9 186.0 70.6 81.0 162.9 500.5 1991 84.2 61.7 41.1 187.0 65.2 85.4 166.3 503.9 1992 86.0 63.8 42.2 192.0 65.0 87.0 172.0 516.0 Source- CPM Group. June 29, 1992 180 Table A3: Silver -hlm., a/ 19501 (A~tual), 1992-2005 (Pr~jected (01:.:y oz.) Cueit $ 1990 Coa t - G-5 MUV bl G-7 CPI c/ 1950 74 454 598 1951 89 475 651 1952 85 430 592 1953 85 444 590 1954 85 455 581 1955 89 466 605 1956 91 459 604 1957 91 449 591 1958 89 433 568 1959 91 450 583 1960 91 442 574 1961 93 439 567 1962 109 505 647 1963 128 607 739 1964 129 603 728 1965 129 599 706 1966 129 579 683 1967 155 686 798 1968 215 958 1,074 1969 179 759 859 1970 177 706 801 1971 155 585 655 1972 169 585 647 1973 256 767 860 1974 471 1,158 1,428 1975 442 978 1,193 1976 435 950 1,138 1977 462 918 1,093 1978 540 932 1,042 1979 1,109 1,691 2,023 1980 2,064 2,867 3,365 1981 1,052 1,456 1,701 1982 795 1,117 1,302 1983 1,144 1,646 1,845 1984 814 1,196 1,316 1985 614 895 973 1986 547 676 718 1987 701 789 815 1988 654 686 706 1989 550 581 597 1990 482 482 482 1991 404 396 384 Pocted 1992 400 375 358 1993 390 353 337 1994 428 380 362 1995 450 389 370 2000 618 450 421 2005 709 450 404 a/ For 1950-1, umfnemd, producur price; begining 1962, Haady & Hrman, refmed, delivered New York. bl Deflated by -5 Mannrcwring Uni Va (MIV) Index. 0/ Deflatd by 0-7 Ca~sc Frice Index (CP). å~ =oca Auag~a Ifinaud osao FLd. ~5 n to 1im aud Me«ab g_ = 1980 o d aUa; WO (IO d). 181