The Location of Minerals Processing

(1)

NOT FOR QUOTATION WITHOUT PERMISSION OF THE AUTHOR

THE LOCATION OF

MINERALS

PROCESSING

Ruthann C. Moomy

April 1984 WJ?-84-31

Working Papers are interim reports on work of the International Institute for Applied Systems Analysis and have received only limited review. Views or opinions expressed herein do not necessarily represent those of the Institute or of its National Member Organizations.

INTERNATIONAL INSTITUTE FOR APPLIED SYSTEMS ANALYSIS 2381 Laxenburg, Austria

(2)

As part of its research, the Mineral Trade and Markets Project is exa- mining trends in the major metal industries since 1950. As now envisaged, this study will contain four substantive chapters. .The first, a draft of which is available as an IIASA working paper (WP-83-91), reviews changes in the location of metal mining and identifies the important determinants responsible for these changes. The second chapter focusses in a similar manner on geographic shifts in metal consumption.

while the t h i r d investigates trends in the location of processing. The fourth examines the role of transportation costs, political blocs, international ownership ties, and other factors t h a t influence t h e choice of trad- ing partners and introduce rigidities into international trade flows.

This working paper is a preliminary version of t h e third chapter concerned with the location of processing. It will be revised before publica- tion, and is being circulated a t this time t o elicit comments, criticisms, and suggestions for improvement.

John

E.

Tilton Research Leader

Mineral Trade and Markets Project

(3)

A popular perception has arisen that the mineral production of developing countries is too often exported in unprocessed form, thereby depriving the host country of the value-added and benefiting the developed country importer. The validity of this allegation and possible explanations for the location of minerals processing is the subject of this working paper. An examination of the patterns and trends in the location of processing for aluminum, copper, iron, nickel, tin and zinc, up to the refined metal stage, reveals that developing countries a r e processing a larger proportion of their mine output today than 30 years ago.

Despite significant progress, t h e developing countries as a group a r e processing less than 50 percent of their ore production, with the exception of tin and copper. When the major ore producers from all country groups are considered, t h e data reveal t h a t only two commodities, tin and alumina, exhibit a trend toward greater domestic processing. Thus, the pattern of exporting minerals in unprocessed form is not confined to developing countries, but is exhibited by a number of developed countries, particularly Canada and Australia.

To explain this pattern of behavior t h e chapter offers three plausible explanations. The first contends that economic factors, such a s comparative costs, a r e responsible for the location of processing. The second explanation attributes the location to primarily political factors, s u c h a s tariff barriers or government subsidies. The third explanation, though not entirely independent of the other two, emphasizes t h e dynamic nature of the location of processing by accounting for t h e effect of the historical legacy.

(4)

The analysis demonstrates t h a t comparative advantage a t one s t a g e of processing does not guarantee comparative advantage a t subsequent stages. Proximity t o ore production is not necessarily more advantageous than locations with an abundant supply of complementary inputs, such a s energy, or proximity to markets. Declining transportation costs have reduced t h e diseconomies of processing at g r e a t e r distances f r o m t h e mine. Other economic considerations, as well as the impact of public policy and t h e historical legacy differ by commodity a n d over time.

(5)

Introduction

Patterns and Trends in Minerals Processing Determinants of the Location of Processing

Economic Determinants Political Determinants Historical Legacy

Results

Conclusions References Appendix Tables

(6)

THE

LOCATION OF WNERAIS PROCESSING

Ruthann C. Moomy

INTRODUCTION

Minerals processing occurs in a variety of locations. ranging from those close to the mine site to those with proximity to commercial centers and including some third party countries t h a t a r e neither major producers nor major consumers but appear t o possess a comparative advantage of another type. A popular perception h a s arisen t h a t developing countries in particular do not process an optimal share of their ore production prior t o export. This situation has warranted international attention. The strategy of downstream processing of developing country exports of primary commodities has been t h e subject of such global reso- lutions a s t h e U.N. General Assembly's Program of Action on the Estab- lishment of a New International Economic Order, t h e Lima Declaration of UNIDOes Second General Conference, and t h e International Development

(7)

Strategy of UNCTAD. (Wall, 1979).

Downstream processing can provide significant benefits. The value added by processing contributes t o the country's export earnings a n d to government revenues. The processing industry provides employment opportunities. F u r t h e r domestic processing may enhance the country's control over t h e minerals sector of the economy and thereby p e r m i t t h e country t o r e a p g r e a t e r economic a n d social benefits from t h e minerals sector. In another light, minerals processing is not viewed as a n e n d in itself; rather, i t is desirable for its contribution to f u r t h e r economic development a n d diversification. This line of reasoning maintains t h a t local industries should arise t o supply the processing industry with required inputs or to use t h e output t o make finished products.

Economic nationalism provides still another justification for domestic processing and is reflected i n t h e idea t h a t mineral exporting developing countries should be more than t h e "hewers of wood and t h e drawers of water" for t h e developed capitalist economies (Balogh, 1978). This con- c e r n is not strictly limited t o developing countries but i s also expressed t o a degree by Canada, itself a major exporter of minerals in both pro- cessed and unprocessed form. The non-renewable n a t u r e of minerals contributes t o economic nationalism, engendering t h e attitude t h a t mineral resources m u s t be used t o foster domestic development o r con- served for f u t u r e generations, r a t h e r t h a n exploited for t h e beneFit of consumers in a foreign country.

Interest in processing is not limited t o ore producing countries.

Those countries whose processing industries a r e dependent on imported raw materials, seek a dependable supply of quality inputs a t a reasonable

(8)

price while maintaining some flexibility t o react to changing economic circumstances. The final consumer is primarily concerned with obtain- ing t h e product a t lowest possible costs, and indirectly with t h e impact of the location of processing on t h a t objective.

Finally, questions concerning the location of processing are important for the larger study, of which this chapter is a part, because it determines the stage of production a t which trade occurs. For our purposes, the scope is limited to t h e treatment of ores up to t h e refined metal stage.

The next section presents data on the portion of ores that a r e processed domestically. The data help t o answer the question: what progress have developing country mineral exporters made in acquiring downstream processing, over time and for which metals. Their progress is compared with t h a t of developed country mineral exporters, such as Canada and Australia. The chapter then examines t h e major factors influencing the location of processing. Three explanations for t h e observed behavior a r e described. The first contends t h a t economic factors, such as comparative costs, are primarily responsible for t h e location of processing. The second explanation attributes the location t o primarily political factors, such as tariff barriers or government subsidies.

The third explanation, though not entirely independent of t h e other two, emphasizes the dynamic n a t u r e of the location of processing. For example, historically a lag appears between development of mine output a n d smelting capacity. during which time ores are exported. Upon mine exhaustion, existing processing capacity is still usable and relies upon

(9)

imported ores from a new mining area. The final section of t h e chapter evaluates each explanation in light of t h e evidence and draws conclusions about the determinants of t h e location of processing.

PATERNS

AND

TRENDS

IN

MINEFWS PROCESSING

Table 1 presents t h e amount of ore processed by developing coun- t r i e s as a group, as a percent of the ore mined by developing countries.

The table shows t h a t developing countries are processing a larger proportion of their mine output today than 30 years ago, with the single exception of blister copper. Nevertheless, the developing countries still process a relatively low percent of t h e i r ore production, less than half, with t h e exception of tin and copper. Similar tables for t h e developed countries and socialist countries would show ratios consistently in excess of one for t h e developed countries, and almost as consistently equal to one for t h e socialist countries. Table 2 presents the percentage of ore processed in the country in which i t is mined, including all major producers from all country groups. Only two commodities, alumina and t i n , exhibit a trend toward g r e a t e r domestic processing. A comparison of Table 2 with Table 1 indicates t h e degree of success t h a t developing countries have experienced relative t o ore producing countries in general, in processing greater amounts of their mineral output. The p a t t e r n varies significantly by commodity, and for each commodity a look behind t h e aggregate d a t a merits attention. Appendix tables 1 through €3 provide production data by commodity, by country, for selected years from

1950 through 1980.

(10)

Table Lore Processed by Developing Countries as a Percent of Ore Mined by Developing Countries

Cqmmodity 1950 1960 1970 1980

Aluminaa Aluminuma Blister Copper, Refined Copper Pig Irog

Nickel Primary Tin Refined Zinc

Table 2.0re Processed by Major Ore Producing countriese as a Percent of Ore Mined

Commodity 1950 1960 1970 1980

Aluminaa Aluminupla Blister Coppet Refined Copper Pig Irog

Nickel Primary Tin Refined Zinc

Notes:

30

account for t h e non-metallurgical uses of bauxite, bauxite production was reduced by ten percent before the ratio of alumina production t o bauxite production was calculated. To account for t h e non- metallurgical uses of alumina, bauxite production was reduced by an additional five percent before the ratio of aluminum production to bauxite production was calculated.

'NA

indicates data a r e not available.

'includes

ferroalloys.

dlncludea primary nickel and nickel contained in ferronickel, nickel oxide sin ter and copper-nickel alloys smelted directly from ores.

' ~ a j o r o r e producing countries represent all countries with mine output equal to, or greater than, one percent of world production.

Source: Appendix Tables 1 through 8; Tilton (1983) Comparative Ad- vantage in

Mining.

Appendix Tables la-f. IIASA Working Paper No. WP-83

(11)

ALuminum. The transformation of bauxite into aluminum metal involves two stages. In the first stage bauxite is refined into aluminum oxide, commonly called alumina. In t h e second stage smelting of the alumina removes the oxygen and produces aluminum metal. Developing country bauxite producers have been much more successful in acquiring alumina processing capacity t h a n aluminum smelters. Data for alumina production in 1950 a r e not available, reflecting the lack of an established market for alumina a t t h a t time. In 1955, six percent of their bauxite was processed into alumina domestically; t h e corresponding number for 1900 is 39 percent. Between 1970 and 1900 almost all major bauxite producers, with t h e notable exception of Guinea, increased their share of alumina processing. Surinam now refines 70 percent of its bauxite and Jamaica 40 percent. Australia had refined 50 percent in 1970 and increased the share to 60 percent by 1900. For aluminum smelting, not only do developing countries process a much smaller share of their alumina into aluminum, but t h e aluminum smelting t h a t does occur does not take place in t h e countries t h a t a r e t h e largest bauxite producers. For example, Guinea a n d Jamaica produce no aluminum; and Suri- n a m processed only six percent of i t s bauxite into aluminum in 1900.

The developing countries with aluminum smelters are those with the larger domestic markets, such as Brazil and India. In addition, Venezuela and more recently Bahrain, have established aluminum smelting capacity, in spite of the fact t h a t they have no bauxite production and consume only a small percent of their aluminum output.

(12)

The pattern exhibited by the developing countries is also represen- tative of the developed countries. Australia, the world's largest bauxite producer, processes only five percent of its bauxite into aluminum.

Greece, the second largest bauxite producer among the developed countries, processes 20 percent of its bauxite into aluminum. Only 20 percent of the aluminum produced in 1980 was smelted in t h e same country in which the bauxite was mined, down from 35 percent in 1950. Aluminum smelting is more likely to occur in countries with large domestic markets, such as the U.S., Japan a n d the Federal Republic of Germany. Two significant exceptions a r e Canada and Norway who smelt most of their aluminum for export.

For the socialist countries data limitations permit only a few observations. Between 1950 and 1980 the socialist countries have developed the capacity to process quantities of alumina and aluminum in excess of their bauxite production. Over t h e past two decades Hungary has increased its share of alumina processing from 50 percent t o nearly 80 percent of its bauxite production. Despite rapid growth in both bauxite and alumina production, Hungary's aluminum production h a s remained fairly limited. This situation reflects the Hungarian-Soviet Alumina- Aluminum Treaty of 1962, whereby smelting capacity for Hungarian alumina was established in t h e U.S.S. R. to exploit surplus hydroelectric power there. As a result, the U.S.S.R. increased its smelter capacity a t a faster r a t e than i t s ore production.

Copper. The transformation of copper ore to refined copper metal requires t h e concentration of t h e ore (20-40 percent Cu), the smelting OF the concentrate t o produce blister copper (96-99 percent Cu), and then

(13)

the refining of the blister to produce copper more than 99.9 percent pure. An international market exists for each of the products: concen- t r a t e s , blister and refined copper. Developing countries have throughout the period smelted most of their concentrates domestically; however, since 1960 a downward trend has developed. In 1980 only 78 percent of copper ore produced in developing countries was smelted, compared to 90 percent in 1960. While it is true t h a t some traditional producers, specifically Chile and Zaire, are smelting a slightly smaller portion of t h e i r mine output, the primary cause for t h e downward trend lies with t h e relatively new mining areas opened u p in the Philippines, Papua New Guinea, and Mexico. The Philippines completed the construction of their first smelter-refinery complex in 1983.

The developing countries exhibit a general upward trend in percentage of ores processed t o the refined copper stage. Zambia now refines 100 percent of its ore production. Peru has made substantial gains in refining its ore domestically, from 16 percent in 1960 to 63 percent in 1980. Chile exhibits a similar, though less dramatic, trend from 42 per- c e n t in 1960 to 76 percent in 1980. Of t h e major developing country producers only Zaire processes a lower percentage today than formerly.

Zaire's refining capacity has remained almost unchanged since 1960.

while its mine output grew by over 50 percent. South Korea and Taiwan have recently developed processing industries. While their capacity is not large compared t o t h e developed countries, it is a n indication that developing countries without domestic ore production have acquired processing capability.

(14)

For developed countries the general trend has been to maintain smelting capacity in excess of the amount required t o process domestic ores. Japan and the Federal Republic of Germany maintain a large smelting capacity despite the lack of a substantial domestic copper mining industry. In 1950 and 1960. the U.S. smelted more ore than i t mined;

but for the two later years of the study, the U.S. smelted slightly less t h a n i t mined. Canadian smelter output a s a percentage of mine output fell significantly from 90 percent in 1960 to 66 percent in 1980, due primarily to the opening up of new mines in British Columbia, far from existing processing facilities.

In developed countries refining capacity is large relative. t o ore production, in part because they produce a substantial amount of refined copper from old scrap. In 1980, the

U.S.

produced 1,686 thousand tons of refined copper, of which 28 percent was produced From scrap. Neverthe- less, t h e U.S. refined less copper in 1980 than i n 1970. Belgium maintains the world's seventh largest refinery capacity, relying primarily on imports.

The socialist countries smelt all their copper ores, almost exclusively within the country of origin. Refined copper output exceeds smelter output in the

U.S.S.R.

by approximately 25%, which probably reflects the role of scrap and not the effect of imports. Poland, the second largest copper ore producer among t h e socialist countries, smelts and refines virtually all i t s ore domestically. China refined almost twice a s much copper as it smelted in 1980.

(15)

Iron. Data for pig iron production are used to identify where iron ore is processed. Unfortunately the data do not account for iron ore that is treated by the direct reduction method of ironmaking. While the magnitude of direct reduction is still not large, i t is relatively more important for developing countries. Hence, a slight underestimation of the quan- tity t h a t developing countries process results.

In 1950 the developing countries were not large iron ore producers, accounting for less than 10 percent of world output. In 1955 and 1960 they processed about 20% of their iron ore into pig iron, attributable largely to India and Brazil. Between 1960 and 1970 t h e i r iron ore production more than doubled, and t h e percentage smelted locally fell. By 1980, 31 percent of their ore was being processed in developing countries. Mexico had joined the ranks of significant developing country producers.

The early pattern of iron production consisted of the United States and t h e larger, developed countries of Europe a s t h e major producers, relying primarily on domestic ore and ore imports from Sweden. Mining, processing a n d consuming took place within one country or in close geographic proximity. By 1960 Canada and Venezuela had become important ore suppliers for t h e U.S. From the largest iron producer in 1960, the US. fell to t h i r d place by 1980 and actually produced less iron in 1980 than it had a decade earlier. By 1980 Japan h a d become the second largest iron producer, behind t h e U.S.S.R. Japan relies entirely on imports for its supply of iron ore. Australia is a major supplier for the Japanese industry. Australia processed only 13 percent of its iron ore in 1980. Canada, Sweden and South Africa also produce much more iron ore

(16)

than their domestic iron industry consumes. Less t h a n half t h e iron ore mined in 1980 was processed into pig iron domestically, and this t r e n d over t h e last 30 years h a s been downward.

The Soviet Union processes about 80 to 90 percent of i t s iron ore domestically, while Poland and Romania maintain an iron industry without significant domestic iron ore production. The result is a balance between ore production a n d pig iron output for t h e socialist countries as a whole.

Nickel. Nickel processing results in a variety of commercial products. For our purposes, in addition to primary refined nickel, t h e data include t h e nickel c o n t e n t of ferronickel, nickel oxide s i n t e r a n d copper-nickel alloys smelted directly from ores. These intermediate, processed products are used directly by t h e steel industry. Between 1950 and 1960 New Caledonia was t h e only developing country to process any nickel. From 1970 onwards Cuba processed half its mine output.

During t h e 1970s t h e Philippines and t h e Dominican Republic became o r e producers and simultaneously built processing capacity. As a whole, t h e developing countries processed 44 percent of their o r e in 1980.

While t h e developed country nickel ore producers do process a larger share of t h e i r o r e t h a n the developing countries. a significant amount is processed in developed countries without nickel mines. In fact, nickel processing is less concentrated t h a n nickel mining. For example, i n 1980 Australia processed only half of its output; a n d over t h e period Canada has processed between 60 and 80 percent. Among t h e ranks of t h e important processing countries t h a t lack nickel mines a r e t h e

US.,

t h e

U.K.,

a n d Norway in t h e early years. joined l a t e r by France,

(17)

and t h e n more recently joined and overtaken by Japan.

Over t h e l a t t e r half of t h e period covered by the study, t h e socialist countries began to process more t h a n they mined, due primarily t o t h e U.S.S.R. However, if Cuba is included with the socialist countries a balance between mine output and processed output results because the U.S.S.R processes t h e ore t h a t Cuba does not process domestically.

%. The role of developing countries in tin refining presents a dramatic contrast t o t h e o t h e r metals under study. Malaysia is t h e largest processor over the entire period. In addition t o refining all its domestic ore production, Malaysia also processed imported ore over t h e entire period. By 1970 Thailand and Nigeria were also processing 100 percent of their ore. Indonesia built a smelter in 1967 and by 1980 was processing almost all i t s ores domestically. Bolivia has embarked on a policy of more domestic processing and was able to smelt 50 p e r c e n t of its ore by 1980. The developing countries as a group smelted 94 percent of their ore output in 1980.

The role of developed countries in tin processing has been one of continual decline since 1950. The only developed country with signifi- c a n t o r e production, Australia, processes only half of its ore. The U.S., t h e U.K., t h e Netherlands and Belgium were the four largest processors in 1950 after Malaysia. From a position in 1950 of producing over 50 per- c e n t of t h e world's refined tin, these four developed countries' s h a r e of output fell to 6 percent by 1980.

Zinc. The developed countries dominate the zinc smelting industry just as they do t h e zinc mining industry. Nevertheless t h e developing

(18)

countries managed to process almost 50 percent of their ore in 1980.

Mexico's processing industry now t r e a t s 60 percent of t h e domestic ores.

While Peru's mine output has grown over t h e past decade, its smelter capacity h a s remained static. Although small relative t o world output, both Brazil and t h e Republic of Korea have smelting industries t h a t a r e more t h a n sufficient to t r e a t domestic ores.

Among the developed countries processing capacity is widespread.

Canada processes about half its output, a n d Australia processes a slightly higher percentage. The only developed country zinc ore producers without smelters a r e t h e Irish Republic and Sweden. Enough concen- t r a t e exports a r e available t o support processing in just about every o t h e r European country. Japan smelts about t h r e e times the amount t h a t it mines. The U.S. zinc smelting industry has been on the decline, following t h e t r e n d of ore production. In 1980 U.S. zinc metal output was only 40 percent of what it was in 1970.

The review of trends in processing raises a series of questions, s u c h as: why do even developed country ore producers, such as Canada and Australia, often export large a m o u n t s of m e t a l in concentrate form; what is responsible for t h e downward t r e n d in t h e proportion of copper smelted by developing countries; what is t h e cause for t h e decline in t h e position of t h e U.S. processing industry; is proximity t o markets more important for some metals than for others. Having established where processing takes place, it is now appropriate to investigate why. The next section examines t h e theories t h a t attempt t o explain this behavior.

(19)

DETEXMDJANTS

OF THE LOCATION OF PROCESSING

This section presents a brief theoretical discussion of the determinants of t h e location of processing. The determinants fall into three categories: those t h a t a r e economic in nature, those t h a t are political, and those t h a t a r e t h e result of the historical legacy of past activity. As each category is described in turn, some interrelationships will become apparent.

Economic Determinants

The economic objective behind t h e choice of location for processing is t o minimize costs. Factors influencing costs include input availability, economies of scale, transportation costs a n d marketing costs.

Input availability a t a competitive price is a necessary prerequisite for processing. This argument follows t h e line of reasoning of t h e factor endowment theory discussed in Chapter 2.' In addition t o t h e mineral raw material, inputs include labor, energy, technology and capital. As was also mentioned in Chapter 2. most inputs a r e mobile; however, mobility is rarely cost-free. The labor input required for most mineral processing operations is small. Energy on t h e other h a n d represents a large p a r t of t h e cost of most processing operations. While coal and petroleum a r e easily transported, hydroelectric power is much less mobile. Technology is frequently embodied in t h e capital input. Tech- 'Elton (1983) is the draft of the second chapter of the larger study of which this working pa- per is a part.

(20)

nology was once solely t h e possession of multinational mining companies; however, this is no longer true (Radetzki, 19828 and UNCTAD 1978). Technology can be purchased and the necessary amounts of capital have become so large t h a t frequently even the multinational mining companies engage in debt financing and joint venture.

The availability of capital affects the choice of location in yet another way. Larger absolute amounts of capital are required for a greenfield facility than for t h e expansion of an existing, or brownfield facility. Furthermore, t h e capital cost p e r unit of output is frequently lower for the brownfield facility. The costs of production from any incre- mental capacity must compare favorably with those from existing facilities, whose capital costs may be already partially or fully amortized.

Since "sunk costs are s u n k , existing facilities will continue in operation when price does not cover total costs so long as variable costs are covered. As a result, the capital investment involved prevents the processing location from changing quickly especially when overcapacity exists or little growth in demand is expected. In t h a t case, f u r t h e r investment in processing is not financially attractive.

Related t o input availability is the need to amass a sufficient quan- tity of inputs t o operate a facility of t h e minimum efficient scale (MES).

Otherwise, t h e firms m u s t choose between building a smaller, less efficient plant or operating a MES facility a t less than t h e optimal capacity.

Economies of scale determine how significant t h e cost disadvantage is for t h e small producer. Where a range of technological choice exists, t h e impact of a smaller scale operation may be very slight. In addition, a cost disadvantage due to smaller scale may be counteracted by a cost

(21)

savings arising from some o t h e r characteristic of the location.

Transportation costs encompass both the costs of assembling t h e inputs and the costs of delivering the output to market. The declining t r e n d in transportation costs, t h a t began long before t h e period covered by this study, continued to exert considerable influence over the m a r k e t scope for minerals during t h e 1950s and 1960s. Iron ore from South America can successfully compete in t h e mid-western region of t h e United States with domestically produced iron ore, basically because t h e higher grade of t h e foreign ores is more t h a n sufficient to compensate for t h e higher transport costs incurred. Over the past decade the rise in energy prices, specifically bunker fuel, has temporarily a r r e s t e d t h e downward t r e n d in shipping costs.

In minerals processing, significant weight reductions frequently take place, resulting in some transport cost savings. For example.

smelting a ton of copper concentrate with a 28 percent Cu content t o produce blister with a 98 percent Cu content results in a weight reduction of more than 70 percent. In contrast, t h e weight reduction from refining t h e blister copper t o a 99.99 p e r c e n t level of purity i s quite small. Over the period a relatively g r e a t e r decline in shipping costs has occurred for t h e bulk commodities. s u c h a s iron ore and bauxite. A result i s t h a t processing a t distances from t h e mine has become relatively more economical. The g r e a t e r decline in bulk shipping costs may be attributed to improved economies in loading a n d handling procedures.

a n d to the use of larger vessels. This t u r n of events f u r t h e r f r u s t r a t e s t h e developing countries i n their desire t o increase domestic processing a n d h a s resulted in allegations of monopolization i n t h e shipping

(22)

industry and discrimination against developing countries (UNCTAD, 1901A). Counter arguments include the rationale that shipping costs are higher for processed goods because processed goods a r e of higher value and shipping costs represent a smaller proportion of that value. Hence t h e price elasticity of demand for transport is likely to be lower for processed goods t h a n for unprocessed goods.2 While the general trend in shipping costs has aided the mineral exporting countries, the relatively greater decline in bulk shipping costs may discourage further processing n e a r the mine in favor of processing near consuming centers.

The final economic factor is t h e role t h a t marketing plays in t h e location of processing. Marketing skills could be viewed simply as another input, a specialized type of skilled labor. The reason for treating this separately here is t o focus attention on t h e structural characteris- tics of the market t h a t can affect t h e location of processing. For example, does an arm's-length market exist for the product a t t h e proposed stage of processing, or is this market foreclosed due t o vertical integration or long-term contracts? Some mineral industries exhibit a high degree of vertical integration. While this factor alone need not be a barrier to entry. i t may require t h a t a new producer integrate forward more t h a n one stage of t h e processing chain, thereby raising the capital costs a n d generally complicating t h e investment decision. Allegations of res- trictive business practices have arisen d u e t o a high degree of vertical integration. For example, the vertically integrated firm can s e t transfer prices with t h e objective of minimizing tax liability rather than

%or further, though still inconclusive, discussion the interested reader is referred to UNI- DO, 1981, UNCTAD l 9 8 l A and 18818.

(23)

representing t h e true market value. A nonintegrated firm would be unable to compete at the transfer price.

Long-term contracts have arisen as an alternative to vertical integration as a way of guaranteeing a steady supply of inputs, and in response to the need to raise very large amounts of capital from diverse sources. In both ways, long-term contracts seek to reduce or transfer some of t h e commercial risk. Long-term contracts need not bias t h e location of processing if they merely act to bring together buyers and sellers i n the market. Nevertheless, where long-term contracts a r e widespread, the remaining market may be so thin that its clearing price is unrepresentative of t h e market as a whole. In addition, the duration of long-term contracts limits t h e flexibility of a firm attempting to e n t e r a downstream processing activity either because their output is already committed or because potential customers are. Some flexibility remains, however, because long-term contracts do not customarily cover 100 per- c e n t of t h e firm's requirements.

Three further factors influence t h e marketing aspect of location:

domestic demand, byproducts, and growth in t h e industry. The existence of a domestic market facilitates domestic processing because of potential savings in transportation costs and a n ability t o respond more quickly t o changes in the market. A number of factors considered above, economies of scale, transportation costs, and a domestic market also apply t o byproducts. The revenue from sales of byproducts encourages a processing location close t o consuming centers if n o market exists a t alternative locations and if transportation costs are prohibitive. Finally.

when t h e industry experiences a period of growth all the problems

(24)

associated with marketing a r e diminished, whereas a period of stagna- tion exacerbates these problems.

Political Determinants

Government policies to encourage domestic industry a r e not uncommon, and in t h e minerals industries politics can sometimes play a more important role t h a n economics. Justification for government intervention ranges from t h e necessity to c o r r e c t a market imperfec- tion, to t h e argument t h a t economic efficiency is less important t h a n other objectives, such a s income redistribution, employment, or national pride. For example, tariffs can be used simply to protect domestic indus- t r y or t o compensate for unfair trade practices. In some countries tariff r a t e s on raw materials a r e zero or very low, and tariffs for processed materials escalate with t h e degree of processing. In this case, t h e effective r a t e of protection afforded t h e domestic processing industry is higher t h a n t h e nominal rate; and conversely, t h e barrier faced by the exporter is higher. Exporting countries can retaliate by enacting export duties on unprocessed materials. These duties seek t o neutralize t h e disadvantage of exporting processed goods. Other facets of t h e t a x sys- t e m i n addition t o tariffs a r e used t o encourage domestic processing.

Governments subsidize uneconomic capacity i n a variety of ways.

including low cost energy o r low-interest loans. Motivations include regional development objectives, national security or t h e need t o preserve jobs. Some countries, such as Japan, have adopted a series of policies to permit large processing industries to flourish while dependent

(25)

on imported materials.

Not all government policies favor domestic processing. Environ- m e n t a l regulations a r e especially onerous for the m i n e r a l s processing industry. An alternative to compliance with :domestic environmental regulations is t o move t h e activity abroad. In t h i s m a n n e r , developed countries c a n p u r s u e t h e option of exporting t h e i r most polluting mineral processing activities (Radetzki. 1982A).

Over t h e period of t h e study, many developing countries have become politically independent and have developed t h e political infras- t r u c t u r e n e c e s s a r y t o exercise a large amount of control over t h e i r domestic economy i n general a n d over foreign investment i n particular.

Along t h e way, some countries, in which t h e minerals s e c t o r played a major economic role, nationalized mineral operations owned by multinational mining companies. The nationalizations a n d t h e increasing sophistication of t h e developing countries i n pursuing t h e i r economic a n d political objectives have changed t h e rules of t h e g a m e for t h e traditional investor i n t h e minerals industry. The r e s u l t is g r e a t e r uncertainty a n d h i g h e r perceived risk t h a t may influence t h e choice of location for processing.

Historical Legacy

A review of t h e historical p a t t e r n of processing aids our understand- ing of t h e c u r r e n t situation a n d underscores t h e fact t h a t t h e situation is n o t s t a t i c . A t o n e time, mining, processing a n d consumption took place i n close geographic proximity. It was natural t o first explore close

(26)

to home. F u r t h e r m o r e , transportation costs encouraged processing and consumption n e a r the location of the natural resource. Over time the exhaustion of local deposits led to exploration f u r t h e r afield. While mining in unfamiliar locales was accepted as an undeniable necessity, the need to relocate processing facilities was not always apparent. The useful life of t h e established processing capacity frequently exceeded t h e mine life. In t h e event t h a t new capacity was required, locations close to home were favored for two reasons: the ability to exercise control more easily, and the desire to minimize political risk. The shifts in processing c a m e more slowly than those in mining.

Over time, a country's minerals industry is likely t o pass through several stages. Observations of this behavior in Europe led D.F. Hewett to hypothesize a general rule in 1929. Depicted in Figure 1, this model predicts five stages t h a t a nation experiences in regard to a mineral industry. In t h e first stage m i n e output is exported. As the n u m b e r of mines increase in stage two, a smelting industry t a k e s root (stage t h r e e ) and ore exports drop off. The fourth stage r e p r e s e n t s t h e p a t t e r n of metal production. As mine output declines due t o exhaustion, imported ore provides t h e feed for domestic smelters (stage five). Hewett warns t h a t t h e o r d e r of stages need not be invariable over a wide range of his- tory, and t h a t t h e successive relation of t h e peaks is m o r e important t h a n t h e relative heights of t h e curves. The model i s particularly appli- cable t o countries with domestic fuel supplies and t h e endurance of smelting capacity after m i n e exhaustion "depends upon t h e local fuel and power supplies a n d t h e maintenance of competitive technique"

(Hewett, 1929, p. 90).

(27)

u

I

Number o f P r o d u c t i o n o f

a Mines M e t a l

n I m p o r t s

t

i

/

t

P

. _--

T i m e

F i g u r e 1 . S t a g e s o f t h e M e t a l I n d u s t r i e s .

(From D.F. H e w e t t , " C y c l e s i n Metal P r o d u c t i o n , "

1929, p . 8 9 . )

A simplified interpretation of the Hewett model for o u r purposes would stress two stages. After mining begins, a period of ore exports is common due to a delay in building processing facilities. Upon exhaustion of t h e mines, a period of ore imports is common due t o the continued viability of t h e processing industry. The model is based primarily on behavior observed in Europe, and covers a time frame of an order of magnitude of 100 t o 200 years. Hewett provides some evidence from U.S.

behavior, reminding us t h a t ore from many early copper, zinc, man- ganese, and iron deposits traveled t o Europe for treatment. although

"the period of export of o u r crude minerals has been uniformly brier' (1929, p. 90). The model does not indicate how much time the cycle requires or what influences t h e amount of time. The model is perhaps more useful for explaining t h e colonial pattern of resource exploitation

(28)

than the more recent period of increasing importance of developing country mineral producers and their political and economic indepen- dence. In spite of its shortcomings, t h e dynamic element of the model provides a valuable contribution.

Before turning from the topic of historical legacy, a reiteration of the comment made above in regard to capital is appropriate. Since capital costs are frequently lower for a brownfield expansion than for a greenfield facility, expansions .at existing locations may be more advantageous than greenfield developments. Thus, the past location of minerals processing is important because of its ability t o exert a con- tinuing influence on the location of new capacity.

How relevant a r e these theories t o explaining t h e patterns a n d trends we have identified for t h e processing of aluminum, copper. iron.

nickel, tin, and zinc? I t is t o this question t h a t we now t u r n .

ALumina. We have seen a growing percentage of alumina production by developing countries and by major bauxite producers. The evidence indicates t h a t bauxite availability is becoming a m o r e important cri- terion for t h e location of alumina production. However, bauxite represents only ₃₀percent of t h e value-added for a l u m i n a 3 The availability of other inputs, particularly capital and energy, a r e also very important. The capital requirement is large due t o the scale of '~alculation based on data in UNCTAD, 1982 using 1980 prices.

(29)

production. Banks (1979) estimates that t h e smallest possible plant for producing alumina should have a capacity of 300,000 tpy and t h a t capital costs represent about 40 percent of total production costs. Transporta- tion cost savings represent another incentive for processing near t h e mine. Conversion t o alumina reduces the weight by approximately half, and the ease of handling and shipping of alumina is comparable t o bauxite. Alcan built t h e first refinery in t h e Caribbean because its refineries in Canada were a t a greater distance from the mines and thus a t a disadvantage, relative t o U.S. producers. Over the period, a number of new producers have entered the market, frequently via joint ventures, thereby reducing the degree of vertical integration in the industry.

Long-term contracts are common a n d help to provide supply security for the non-integrated firm. The International Bauxite Association reports t h a t over 80 percent of internationally-traded bauxite and alumina involves inter-affiliate transfers or long t e r m contracts (Rodrik, 1980).

Political influence i n t h e alumina industry has taken many forms.

Early refining capacity in t h e U.S. was encouraged for security reasons because during World War I1 shipments of bauxite were vulnerable t o enemy a t t a c k The cold war mentality engendered the attitude that i t 4 was more desirable t o ship t h e low-value bauxite than t o risk losing t h e

higher-value alumina (Girvan. 1978 and Davis. 1982). As a result. both Reynolds and Kaiser expanded their alumina production capacity in t h e

U.S.

in the 1950s despite evidence t h a t i t would have been more econorni- cal t o build new plants next t o their mines in t h e Caribbean (Girvan, 'During 1842 and early 1943, 28 ships transporting bauxite were sunk out of a total 815 sail- ings between Sllrinam and the U.9.A. (Daviu, 1982).

(30)

1976). Of more r e c e n t significance is t h e increase in the bauxite levy from $2.50 to $11.50 per ton, enacted by Jamaica in 1974, and adopted by a number of other producers. Auty's (1983) analysis For the late 1970s concludes t h a t t h e bauxite. levy rendered a greenfield refinery in Jamaica so high-cost t h a t even without Further taxation of profits such a plant would be 15 percent more expensive to operate than a greenfield u n i t in Western ~ u s t r a l i a . ' Nationalization in t h e bauxite industry per- mitted developing country governments to exercise more control over t h e building of alumina facilities. The risk of nationalization and t h e increased uncertainty about taxation have undoubtedly influenced t h e investment decisions of the private aluminum companies; however, t h e magnitude of t h e impact remains unquantifiable and subject to specula- tion.

The aluminum industry has exhibited higher r a t e s of growth t h a n any other metal under consideration. This factor has contributed t o t h e development of new alumina capacity and a shift toward more competitive locations. While bauxite availability is an important factor for alumina processing locations, t h e availability of capital and energy have also been shown t o be influential. Political factors have played an impor- t a n t role, especially as they contribute to t h e perceived political risk of competing locations. From a position of virtual insignificance in 1960, Australia has risen to t h e world's largest alumina producer, producing over 20 percent of t h e world's output in 1980. Australia combines t h e '.Jamaica seems to have recognized this result and announced the objective of restoring Jamaica as a "priority area in the aluminum companies"' investment plans, during negotia- tions for the extension of the agreement effective January 1904. See Mat41 Wcllrtin, De- cember SO, 1883, p. 9.

(31)

a t t r i b u t e s of bauxite availability, large domestic energy reserves, a n d a well-developed capital m a r k e t , while maintaining a political s y s t e m t h a t

appears to be acceptable t o m a n y multinational mining companies.

Aluminum. In c o n t r a s t t o a l u m i n a , aluminum exhibits a decline i n t h e a m o u n t produced within developing c o u n t r i e s or by m a j o r bauxite producers. In fact, t h e ratios for a l u m i n u m production a r e t h e lowest of t h e eight commodities under review. Clearly, bauxite availability is less influential t h a n o t h e r factors a n d t h i s i s n o t surprising, given t h a t baux- i t e r e p r e s e n t s only 6 p e r c e n t of t h e value-added for a l u m i n u m (UNCTAD, 1982). Energy is a considerably l a r g e r portion of t h e value-added. P e c k (1983) e s t i m a t e s energy c o s t s a t

U.S.

s m e l t e r s i n 1982 ranged from 3.2 t o 26.8 c e n t s per pound of a l u m i n u m , with a weighted average of 16.8 cents.

F u r t h e r m o r e , h e found t h a t t h e cost of electric power is a critical d e t e r - m i n a n t of t h e location decision b e c a u s e i t s price varies m o r e by location t h a n t h e costs of t h e o t h e r inputs. Hence, t h e energy cost savings a t t h e m o s t advantageous location m o r e t h a n compensates for t h e additional c o s t s i n c u r r e d by a less t h a n optimal location for t h e o t h e r inputs. One r e a s o n for large energy c o s t differentials is t h a t electric power c a n not b e t r a n s m i t t e d long distances economically. Energy availability, m o r e s o t h a n economies of scale, is largely responsible for t h e u l t i m a t e size of a s m e l t e r (Peck, 1983). While t h e weight reduction in the conversion of a l u m i n a into aluminum i s approximately half, t h e t r a n s p o r t cost savings i s l e s s t h a n t h a t because a l u m i n u m ingots a r e more expensive t o t r a n - s p o r t on a per ton basis t h a n alumina. Some evidence exists t h a t m a r k e t a c c e s s influenced t h e processing location. In t h e 1960s, t h e m a j o r North American producers began t o invest in a l u m i n u m s m e l t e r s in Western

(32)

Europe because demand there was growing rapidly. At the same time.

European producers began to build smelters in the U.S. (World Bank, 1983). Peck (1983) distinguishes between two types of smelters built abroad by U.S. firms. The first were those designed to serve the U.S.

market and these locations offered low-cost, hydroelectric power. The second type were built to serve regional markets; for this second type, energy costs appear to be a secondary consideration. Since t h e North American producers were able to develop smelters abroad and export aluminum into t h e U.S. market, it appears reasonable t h a t European companies could have pursued the same option. Therefore, t h e desire of European companies t o establish capacity within t h e U.S. r a t h e r than simply export t o t h e U.S. would indicate t h a t m a r k e t proximity was considered important. The location of aluminum production in 1980 is more closely aligned with consumption than with t h e production of bauxite or alumina.

Due to i t s defense industry applications aluminum is a strategic mineral and thus subject to a large degree of political attention. For example, during World War I1 the U.S. government built nine smelters and sold t h e m below cost to private firms after t h e war. The Canadian industry also received support during t h e war from various governmen- tal sources. In 1950, the U.S. a n d Canada together accounted for more than two-thirds of world aluminum production. More recently, govern- m e n t subsidy has taken the form of low-cost energy. For example, From t h e mid-1950s to t h e mid-1970s electricity r a t e s paid by aluminum smelters t o the BonneGlle Power Administration were about one-fifth of t h e rate available to other large consumers (World Bank, 1983). While a n

(33)

incentive certainly exists for aluminum smelters to locate n e a r low-cost power, the presence of t h e low-cost energy may reflect e i t h e r a t r u e savings o r a political 'decision to subsidize the industry. In t h e event of t h e l a t t e r , t h e buyer m u s t be convinced that t h e long term c o m r n i t m e ~ t is a s e c u r e one before the large amount of capital will be committed. Thus t h e search for low-cost energy is influenced by political factors, and therefore by political risk. Another political factor is t h e use of tariffs.

The EEC a n d Japan have ad valorem tariff rates on aluminum ingot of six a n d nine percent, respectively. While lower or zero r a t e s a r e available t o some developing countries u n d e r t h e generalized system of preferences, these tariffs can still p r e s e n t an obstacle to developing country efforts to export aluminum.

The impact of t h e historical legacy on the c u r r e n t location of alumi- n u m processing is mitigated by t h e fact that, of all t h e metals under study, aluminum has had t h e shortest commercial existence and t h e highest growth rate. The high growth r a t e has led t o significant expansions in capacity, permitting t h e industry to adopt more competitive locations. The application of t h e Hewett model is hindered slightly by t h e youthfulness of t h e industry: no country, with the possible exception of France, has yet progressed through the five cycles. Nevertheless, t h e evidence t h u s f a r suggests t h a t only those bauxite producers t h a t a r e also well-endowed with energy a r e likely to develop a smelting industry, t h e r e b y progressing to stage 3 of t h e model.

BZister Copper. In c o n t r a s t t o t h e aluminum industry, developing countries and major ore producers have historically accounted for a large proportion of blister copper production; however, over t h e last

(34)

decade a downward t r e n d has appeared. This indicates that mine production is a fairly important determinant of smelter location, although i t s importance may be diminishing. A further contrast with the aluminum industry is the high portion of vaiue-added that accrues a t t h e mining stage, estimated a t over 75 percent based on 1977 values ( ~ i r t h i s i n g h a , 1902). The majority of t h e copper that is exported before smelting comes from t h e Pacific rim, including Canada, Australia, the Philippines, a n d Papua New Guinea. Much of this mine production is of a fairly r e c e n t vin- tage and located convenient t o port facilities. For example, the Bougain- ville mine of Papua New Guinea opened in 1973 and ships concentrates t o Japan a n d Europe under long t e r m contracts. These contracts constitute part of t h e financial package t h a t was required to develop the project. In the past much of t h e concentrate production of the Philippines has been smelted i n Japan; however. t h e Philippines brought a smelter-refinery complex on s t r e a m in late-1903. Much of t h e concentrates previously exported t o Japan will be processed domestically and Japan will import refined copper instead. The Philippines have found it necessary to offer a slight discount in order to a t t r a c t buyers, presumably due t o i t s relatively unknown reputation as a supplier.

Smelting of t h e majority of Canadian ores takes place domestically;

however, a downward t r e n d has developed since t h e late 1960s a s m i n e output in British Columbia h a s grown. Output from the initial mines was too small to warrant building a smelter. The lack of a local m a r k e t for the byproduct sulfuric acid f u r t h e r hindered the development of a local smelting industry. Transportation costs t o smelters in Manitoba or eastern Canada proved t o be high relative to ocean transport t o s m e l t e r s

(35)

in Japan. Furthermore, Japanese smelters offered very competitive t e r m s under long term contracts, such t h a t when t h e mine output in British Columbia grew to a size sufficient to support a local smelting industry, t h e mining companies continued to find i t advantageous to export concentrates to Japan r a t h e r than invest in smelting capacity (Wittur, 1974). In fact for m u c h of t h e 1970s t h e "sellers" m a r k e t t h a t existed for concentrates discouraged most new investment in smelting capacity in Canada (Wittur 1974, Balogh, 1978).

Transportation costs have also influenced t h e smelting location for African and South American concentrates. Zaire and Zambia have always smelted their concentrates before exporting. The mines in both Zaire and Zambia a r e located far from ports and overland transportation facilities a r e limited. The economies of transporting bulk concentrates by rail, relative to metal ingot, are less substantial t h a n in ocean shipping (Wittur. 1974). The saving in t r a n s p o r t costs due to t h e weight reduction from smelting compensates for the fact t h a t much of t h e materials for the smelting process m u s t be brought into the remote areas. Mine production in Chile and Peru is invariably within access of port facilities. As a result i t is economical to export concentrates, however, in 1980 both countries were smelting almost 90 percent of their concentrates.

The only countries t h a t have major smelting industries without significant m i n e output a r e Japan and, on a m u c h smaller scale, t h e Federal Republic of Germany. Japanese blister copper production ranks as the world's fourth largest. Reasons for Japan's position appear t o be influenced by political objectives, and import restrictions on blister have subsidized domestic smelters by supporting a domestic price in excess of

(36)

t h e wdrld price (Balogh, 1978 a n d Whitney, 1976). In addition, both coun- t r i e s have m a r k e t s for t h e byproduct sulfur.

Another concern of t h e political system, environmental quality, affects t h e copper i n d u s t r y m o s t seriously a t the smelting stage. This is

i

d u e t o t h e s e p a r a t i o n of t h e copper from sulfur and o t h e r impurities t h a t t a k e s place during smelting. Environmental regulations i n m a n y developed c o u n t r i e s call for reduction of sulfur dioxide emissions, often n e c e s s i t a t i n g a large capital investment. In 1980 a n d 1970 t h e U.S.

s m e l t e d less c o n c e n t r a t e s t h a n it produced, reversing t h e p a t t e r n exhibited in 1960 a n d 1950. F u r t h e r m o r e , a smaller absolute a m o u n t of copper was s m e l t e d in t h e

U.S.

in 1980 t h a n in 1970. Could this situation b e t h e r e s u l t of environmental regulations? Whitney's (1976) analysis covering t h e period u p t o 1972 concluded t h a t environmental problems w e r e n o t a d e t e r m i n a n t of s m e l t e r locations; however, t h e i m p a c t of

e n v i r o n m e n t a l problems and t h e a t t e n t i o n t h e y have received have mul- tiplied since 1972. Environmental c o n c e r n provides a possible i n t e r p r e - t a t i o n of t h e Japanese contribution t o t h e r i s e of t h e Philippine smelting industry. J a p a n e s e firms provided both technical expertise a n d financing assistance. Japan h a s also c o n t r a c t e d t o buy considerable a m o u n t s of t h e o u t p u t , a t t h e s a m e t i m e t h a t t h e Japanese smelting industry is operating well below capacity. This p a t t e r n of behavior would be con- s i s t e n t with t h e goal of exporting polluting activities t o protect t h e domestic environment.

The behavior of t h e copper smelting industry provides some support t o t h e Hewett model. New mining a r e a s , s u c h as those in British Colum- bia, Australia, P a p u a New Guinea, a n d t h e Philippines, exhibit a p a t t e r n

(37)

of export-orientation prior to developing a local processing industry. The period will be prolonged, as it was for mines in British Columbia, if opportunities to sell concentrate a r e more economically attractive t h a n investment in new capacity.

In summary, most concentrates a r e smelted n e a r t h e mine to economize on transportation costs. Where smelting occurs a t long distances from the mines, primarily in Japan and t h e F.R.G., these smelters tend to be m h k e t - o r i e n t e d and revenues from t h e byproduct sulfur help t o offset t h e cost disadvantage incurred in transport. Import restrictions further aided t h e viability of t h e Japanese smelting industry.

Aefined Copper. The location of copper refining has historically been somewhat more market-oriented t h a n smelting, although t h e share processed by developing countries has increased over the period. The share of copper refined by the major ore producers has remained fairly steady.

In addition to Japan and t h e F.R.G., who have smelting industries, Bel- gium and t h e United Kingdom constitute the countries with substantial refined copper production without substantial domestic mine production. Their comparative advantage appears t o be their location near large industrial markets. The availability of copper s c r a p may also be advantageous, though this can not be considered a major determinant.

The

weight reduction between smelting and refining is not large enough t o generate transportation cost savings, nor does t h e transformation from blister to refined copper r e s u l t in significant differences i n handling or loading. Historically refineries have often been located a t t h e point where t h e mode of shipment changes, to economize on handling costs, such a s packaging and terminal charges (Whitney, 1976).

(38)

Economies of scale t e n d to be larger for refineries than for smelters, partly due t o t h e economies of by-product recovery. Blister copper frequently contains significant a m o u n t s of gold, silver, or lesser byproducts. Refining s e p a r a t e s these metals from the copper a n d some a m o u n t of additional processing is required for e a c h byproduct t o be in a marketable form. Economies of scale in t h e byproduct recovery circuits can influence t h e economically-efficient size of t h e overall refinery. The increasing complexity from multi-product recovery requires g r e a t e r technical expertise a n d skilled labor. Since t h e r e a r e fewer refineries t h a n smelters, a refinery frequently depends upon blister copper from more t h a n one s o u r c e (Whitney, 1976). Hence, locations within a well- developed transportation network a r e advantageous. The output of the refinery involves a variety of products, including t h e byproduct metals plus t h e copper i n several s h a p e s such a s cathodes, wirebars, billets a n d alloys, t h a t m a y require special handling a n d packaging. A savings in handling a n d packaging costs can arise if t h e refinery is n e a r semi- fabricating plants. While e a c h factor alone m a y appear inconsequential, t h e combined effect is a slight advantage t o locating refineries n e a r markets.

Fron. The downward t r e n d in t h e a m o u n t of iron ore processed domestically, which was slightly less t h a n 50 p e r c e n t i n 1980, is evidence of t h e declining pull exerted by t h e location of inputs. In the nineteenth century, when u p t o eight t o n s of coking coal were required to produce one t o n of pig iron, t h e iron making i n d u s t r y located most efficiently in a r e a s where iron o r e and coking coal were found in close proximity (Manners, 1971). Industries dependent on iron and s t e e l tended to locate

(39)

close to t h e mills (Manners, 1971). Improved iron making techniques have reduced both the iron ore a n d coking coal requirements significantly, and in conjunction with transportation economies, have enabled the industry to become more market oriented. Changes in s t e e l making technology have also led to a more important role for scrap, as both a substitute and a complement for pig iron. Therefore, proximity to consuming centers is advantageous for facilitating scrap collection.

Over the period of the study, significant reductions in transport costs for bulk commodities, which includes both iron ore and coal, have occurred. The 10-20,000 ton ore carriers of t h e 1950s have been dis- placed by carriers as large a s 150,000 tons (Santos, 1976). In 1950 the cost per ton of ore for transportation from Brazil to Japan was about $20;

by 1965 the rate was as low as $4 ( c u r r e n t dollars) (Santos, 1976). While one example serves to document the magnitude of the cost reductions, i t should be kept in mind t h a t shipping rates vary considerably by specific route and Length of journey, frequency, port charges, a n d a myriad of o t h e r factors. The n e t effect of transport costs on the processing location m u s t also account for t h e transport of other inputs, s u c h as coal.

a n d t h e transport of t h e output. While stressing t h e complexity of t h e countervailing forces, Manners (1971) concludes t h a t t h e n e t affect favors a market orientation. The result is due in part t o t h e fact t h a t t h e output is not a bulk commodity a n d its distribution involves a larger n u m b e r of buyers. Most iron o r e a n d coal exporters a r e dependent upon one or two major importers (UNCTAD, 1981A), while t h e trade i n finished steel mill products is more diffused.

The Location of Minerals Processing

MINERALS

E.

Results

THE

AND

IN

30

'NA

'includes

Mining.

U.S.

U.S.S.R.

US.,

U.K.,

OF THE LOCATION OF PROCESSING

I

t

/

. --

U.S.

U.S.

U.S.

The

. _--