Report No. EC -161 FILE COPY Second Edition J This report may not be published nor may it be quoted as representing the view of the Bank and its affiliated organizations. They do not accept responsibility for its accuracy or completeness. INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT \\ INTERNATIONAL DEVELOIJMENT ASSOCIATlbN MANUFACTURE OF HEA VY ELECTRICAL EQUIPMENT IN DEVELOPING COUNTRIES February 10, 1969 Economics Department Prepared by: Ayhan Cilingirog1u TABLE OF CONTENTS PREFACE i I INTRODUCTION 1 • II CHARACTERISTICS OF THE INDUSTRY 5 Definition of Heavy Electrical Equipment 5 The Products 6 The Market 8 Raw Materials and Semi-finished Components 9 Technology 11 Finance 12 III THE INDUSTRY IN THE WORLD CONTEXT 14 and Trade P~~~.. duc tion 14 Dominance by Large International Concerns 16 Capacity, Orders, Deliveries 19 Mergers 23 International Manufacturing Arrangements 25 IV THE INDUSTRY IN DEVELOPING COUNTRIES 29 Growth of the Industry 29 Country Experiences 30 Capabilities 41 V PRICES FOR HEAVY ELECTRICAL EQUIPMENT 45 World Market Prices 45 Prices in Developing Countrie5 51 VI COSTS AND COMPETITIVENESS 57 Concepts and Methodology 57 Materials 60 Manpower 66 • Charges Against Capital 69 Combined Impact of Cost Differences 71 Measures of Competitiveness · 72 Foreign Exchange Savings 81 • Review of Findings 85 VII THE DYNAMICS OF COMPETITIVENESS 67 Market 87 Learning Process 91 Institutional Framework 93 11 VIII SUMMARY AND CONCLUSIONS 96 ANNEX A ANALYSIS OF PRICE TRENDS 105 ANNEX B EXCHANGE RATE ADJUSTMENT IN ARGENTINA 112 ANNEX C PRODUCTIVITY AND THE LEARNING PROCESS .... - SOME ILLUSTRATIONS 114 ANNEX D TABLES 116 TABLES 1. Maximum Specifications of Units Manufactur.ed or in Process of Manufacture in European Countrlea, 1951, 1958 and 1967 8 2. Consumption of Heavy Electrical Equipment in 1965 14 3. Distribution of Exports of Nine Major Exporters, 1964 16 4. Major Producers of Heavy Electrical Equipment, 1964 18 5. Orders and Deliveries: 1964-66 21 6. The Maximum Size of Equipment being built by Domestic Manufacturers, 1967 42 7. Price Premia for Electrical Equipment in Developing Countries Compared with World Market Prices, 1964 52 8. Prices of Principal Materials, 1966 62 9. Comparison of Wages and Salaries, including all Supplementary Benefits, 1965' - 66 10. Employment and Manpower Cost against Sales, Heavy Electrical Equipment Industry, Seven Countries 69 • 11. Cost Structure in the Manufacture of Heavy Electrical Equipment 71 12. Measure of Competitiveness - Example 1 73 • 13. Measurement of Competitiveness - Example 2 75 14. Measurement of Competitiveness - Example 3 77 15. Protection and Domestic Resource Cost per Unit of Foreign Exchange Saved for Various Products 78 iii 16. Measuring the Foreign Exchange Savings of the Industry, Six Countries, 1964 ' 82 FIc;URES I I 1. Evolution of Size of Vertical Generators Produced in Spain 43 2. Development of Price Level for Hydroelectric Generators 1956-1965 49 3. Mexican Transformer Prices compared with the Lowest International Bid 54 4. ~er,Capita Spending on Heavy Electrical Equipment in Relation to GNP per Capita, 1964 88 4a. United States: Price Indices for Selected Products 107 5. United States: Average Transformer Prices by Size Groups, 1958 and 1963 108 6. Price Indices of Distribution T~ansformers in Selected Countries 109 7. Price Indices' of Rotating Machines in Selected Countries 110 8. Price of Distribution Transformers, Pakistan, Mexico, USA, 1965-66 . 111 ANNEX TABLES 1. Heavy Electrical Equipment Manufacturers: Structure of Assets and Liabilities, 1964 116 2. Export and Import of Heavy Electrical Equipment, 1965 117 3. Degrees of Self-Sufficiency in the Production of Electric Power Machinery and Switchgear, 1965 119 • 4. Maximum Specification,s of Units Manufactured or in the Process of Manufacture, by Countries, to January 1, 1967 120 5. Importance of Different Materials ,in Cost Structure for Power Transformers and Heavy Motors, Mexico, 1966 121 6. Cost Structure of a Generator, Argentina, 1965 122 7. Breakdown of Material Costs, Firms in De'/'e1oping Countries, 1965 123 iv 8. Inputs as percent of Total Material Cost, Mexican Firm, 1964 and 11~65 and a PaRis tan Firm, 1965/66 124 9. Comparison of Direct Labor Requirements of Electric.:tl Equipment, 1966 125 10. Manpower Costs in the Manufacture of Heavy Electrical Equipment, Related to Sales and Value Added~ Industrial and Developing Countries, 1964 126 11. ?~ice Assumptions for the Analysis in Chapter VI 127 12. Examples of the Calculation of Price Levels 128 \ 13. Compos i tiot:l of Gross Profits, Profit-to-Sales and Sales to Assets Ratios for Representative European Manufacturers of Heavy Electrical Equipment 129 14. Production of Transformers by Manufacturers, India, 1962 130 ~ 15. Price Indices for Mexican Large Motors Compared to Imports - 1966 131 I 16. Price Comparison of Transformers in Mexico 132 17. International Bids - Comparison between Hexican Price and Lowest Foreign Price c.i.f. of Trans- formers, 1957-64 133 18. Comparison of Prices of Motors in Mexico, India and Pakistan, 1965/66. 134 PREPACE The study on the heavy electrical,equip.ent industry is one of three case studies prepared as part of a br~ader research project dealing with the experience of the developing countries with the establishment ~ . I. . I of capital goods industries, which was under the directio't\ of Barend A. de Vries. The case studies of industries, dealing with the automotive, heavy electrical and mechanical equipment industries, were directed by Bertil Walstedt. Herman van der Tak reviewed the studies in their later pliase. The industries s~lected for case studies play a strategic role , in the more advanced phases of industrial development. Theil.' establishment takes up a substantial share of investment and is, therefore, often ac- companied by stt~ains on real and financial resources and poseb difficult problems of domestic and external economic policies. In addition, these industries produce many items used in the construction of the kind of projects financed by the Bank. . The industry studies were designed to consider the follOWing questions: What has been the experience thus far of the industry in countries with relatively small markets? What are the economics of the industry in more advanced indus- trial countries and what are the conditions of international markets for the products concerned? What aspects are relevant for the growth of the industries in the developing countries? What are the cost and benefits of establishing these indus- tries in developing countries' In particular, what is the cost of the saving of foreign exchange made possible by the industry? What can be said about the efficiency of the industry, at present and over time, and bow i8 cost efficiency affected by such industrial factors as economies pf scale, ~vailab~lity of supplier capabilities, skilled manpower, product design, access to new technology? What haa been the i.,act upon the industry's cost efficiency of governaent policies - protection, excbanae and t.port regulation, requir..enta as to domestic content of production, - i1 - credit', etc. - &8 well as of the structure and the extent of aonopc;ly of dOlle8tic industry? What general indications, if any, can be given OD polici.. of developing countries, as vell as of industrialized countries, to foster the healthy growth of the industry? What contributions might re..onably be expected fr~. large multinational firms in promoting manufacturing in developing countries? What is the future outlook for the- industries in an inter- national context and what might b,l the lI4Ilufact\lring role of developing countries? B.ca~e of the nature of the problems to be analyzed, heavy reliance had to be placed on direct interviews with, and information provided by, ma~ufacturing concerns in developing countries and a8sociated or parent firms in advanced industrial countries. The case studies are, therefore, largely based on findings from field visits to a number of developing countries which have experience with the capital goods industries, including Argentina, Brazil, India, ~exico, Pak1st~n, Spain and Yugoslavia. In addition, visits I were made to firms in Europe and the United States. Though the field visits were limited only to the developing coun- tries indicated, and varied somewhat in their indUstry coverage, the inter- views with multinational companies covered their operations in much of the developing world. We are grateful to the various company representatives for their generous and invaluable assistance. They have shown great interest in the studie8, have discussed the problems of the industry with frankness and made available information without which the 8tudy could not have been undertaken. They have supplied comment8 and criticis.. on an earlier draft, permitting us to te8t the validity of the analysi8 and the accuracy of the factual infor.ation used. The present study was undertaken by Mr. Ayhan Cilingiroglu in close collaboration with Mr. Bertil Walstedt. Mr. Cilingiroglu .a8 a180 aided by Mr. Jose Datas-Panero and Mr. Sanjaya Lall. Miss H~l.n May and ·. iii - ~ ~ss E~iana Barahap contributed much of the statistical work. There were also the helpful criticisms and suggestions by electrical equipment manu- facturers and by other experts both inside and outside the World Bank Group. The author alone, of course, is responsible for th~ facts and opinions presented. The views expressed in this paper are in DO way to be taken as necessarily representing the views of the World Bank. Andrew M. Kamarck Director Economics Department • I INTRODUCTION 1. , This study analyzes factors in the growth and competitiveness o:f the heavy electrical equipment industry in developing countries. For the purpose of the study the Jleavy electrical equipment industry i8 defined as the manufacture of equipment for power generation, transmission, and sometimes industry (e.g. large motors and electric furnaces). It does 1 not cover cOlllDlunication equipment and consumer dUI'ables (automotive or domestic) of any kind. Thus electric locomotives and cables are excluded. 2. The study is based on findings from field trips to Argentina, Brazil, Mexico, . . Pakistan and Spain, on documentation available in the Bank on the Indian industry, and on interviews with sixteen major international manufacturers in the United States and Western Europe. The field trips were limited to developing countries with advanced development in electrical equipment manufacturing, but an attempt was made to cover the rest of the deve1op1ng world in interviews with international companies. The main focus of the study is on the comparison of prices and costs in developing countries with those in the international market. The first three chapters review the distinguishing characteristics of the industry while the following three chapters discuss factors affecting its competitiveness - international prices, factor costs and production efficiency in the developing countries, and factors which may affect comparative costs'over time. The final chapter summarizes the study and indicates main findings. 3. . While Chap~er 11 deals with some of the. general features of ' the industry, Chapter 111 describes the international structure of its production and trade. Because of heavy concentration in the international . market (about 20 firma cover 90 percent of the world trade) and the fact that all manufacturers in developing countries are linked with international manufacturers - either through capital participation or licenaing agreement. or both - a knowledge of the international structure is essential for under- standing the problema which face the developing world. 4. Reviews of some individual country experiences are siven in Chapter IV to provide a comparative picture of the industry. We discuss - 2 - there briefly the historical l~ackground and current status of the industry in terms of physical capacities, technical capabilities and major pt'oblems. 5. Any intemat:f:,onal comparison of values is affected by the rates , I • I of exchange {or ,rates 6f conversion) used. There is an obvi()u8 difficulty in arriving at a reasonable cost figure for production of a heavy electrical I { product with a long lead time (over a year or aore) during periods of rapid inflation'. Besides, successive devaluations make intemational comparisons I extremely complicated. For instance, in Brazil, there were three devaluations within two years (1964-1965); thus the costs of producing a generator over several months cannot simply be fldded up it! cruzeiros. In tl:1e majority of countries studied, overvaluation of the currency haa been compensated b~ high tariffs. Quantitative restrictions, and sometiaea prohibition of particular ~mports, introduce other dimensions in determining the domestic prices of imported goods. Complex exchange systems, e. g. multiple exchange rates, compulsory deposits, and surcharges, are among the eleaents which cause difficulti~s. The problem is ,particularly 4ifficult in Argentina and Brazil, and we have derived exchange rates for these two countries. (See Annex B). The basic Indian data for this study was collected before devaluation. Thus the old rupee rate (Re. 4.75 - US$I) haa been used. 6. Quality is another issue to be considered when comparing costs and prices for products as well as inputs. Since manufacturers in devel- oping countries are linked with international manufacturers, the products they offer under international competition are generally based on the latest designs. Despite this there are great differences in production technology and in quality between products with a high domestic content and those sold in the international market. Efforts to maximize domestic content impose a" strain both on manufacturing facilities and on the product. Their impact on operational difficulties in developing countries (additional maintenance requirements, frequencies of sbut-doWDa, etc.) can 'only be evaluated after several years. ~e industry is young in most develo~ing countries; therefore quality differences have been described rather than measured. 7. Chapter V reviews world . .rket prices in 80ae detail and compar•• them with prices in developing countries. All information on prices i. - 3 - based on ,actual transactions. It is important to bear in mind that the prices of:t heavy electrical equipment are irregular and are Dot like pt:iices of items such as steel, cement, or fertilizers for which continuous price I . info~t{on is quoted in trad~ journals. Any price comparison is true only for !l!!. product at lh!. .ll!!!. of purchase in £b!. country under the .!.!l. i . ' of conditions (financial arrangeaents, political aotivations, etc.) prevail- ing. Deriving a price pattern or reference level, as we attempt to do in this study, requireJ, a very large sample because of the cllstoa-derived nature of product and prices alike. 8. Because of the scattered price information and the Breat variety of products, the price co.parisons in Chapter V are incomplete.. The cost comparisons of major raw materials in Chapter VI .re more complete because, in contrast to the prices of final products, the prices of material inputs are relatively easy to obtain. About 90 pnrcent of the total aaterial cost is composed of six or eight materials (copper, silicon steel .heets, ordinary steel, paper, porcelain, castings). The basic material mix for a given product may vary under different manufacturing conditions. Pirst, there may be a difference in quality of raw materials (e.g. non-uniform thickness of steel sheets); secondly, the working scrap ~atio in a less advanced manufacturing environment is higher; and thirdly, larger pieces of material of h~gher grade may be used for less demanding jobs because the right ones are not immediately available - for instance, thick carbon steel plates may be used for transformer tanks where thinner plates would do. 9. Chapters VI and VII analyze the conditions for economic production in terms of four major determinants - the sum of factor costs (materials, labor, capital), the si~e of the market, the learning process, and finally, what, for want of a better word, is termed the "institutional framework" (mainly government policies). In a 8ense, the first three factors represent a country's intrinsic comparative advantage. The fourth factor determine8 whether a latent potential can be translated into actual competitiveness. The results of the evaluation of actual costs and foreign exchange savings in Chapter VI should be looked upon Dot as an indication of the-differing - 4- success of developing countries, but as a comparison of a developing cou~­ try's industry with the international market at a particular time. Some dynamic factors are considered in Chapter Vl~. , 10. Chapter VIII provides a su.ary of the atudy and ita findlngs. 11. In the atudy we di.tingu1a~ factors inherent in the industry (type a,nd nature of product, cost .~,ructure) from factors controlled to '1'1 some le~!tent by corporate 1Il8na,sement (cost, price) and from other. factors not 80 controlled (size of market, institutional framewOrk, etc~). We have not made specific policy recommendationa for the partie. concerned - manufacturers, developing countrie., development agencies. In general, the s.tudy leads to the concluaion that corporation;) would profi t by learning to look upon the world not only as one 1Il8rket but also as one manufacturing baae; developing countries would benefit by following carefully what happens i i in the international market; and, consequently, development agencies st~uld encourage sucb efforts. - 5 - 11 CHARACTERISTICS OF THE INDUSTRY Definition of Heavy Electrical Equipment 12. In our study the term "heavy electrical equipment" includes equip- ment for the generation and transmi.sion of electric power and for certain major industrial uses. The principal products are main line generQtors, power transformers, switchgear, rectifiers, large motors, and electric furnaces. Light motors and distribution transformers have been excluded, as have components such as wire and cable. Though some electrical equip- ment manufacturers manufacture wire and cable, much output a180 comes from specialized producers in the non-ferrous metals and rubber tnduatries.· On the other hand, steam turbines have been included in some of our trade and production figures. Though by technology an item of aechanical equip- ment, steam turbines have traditionally been supplied by electrical equip- ment manufacturers as a major component in complete steam power lenerating plant. Nevertheless, since steam turbines are not generally produced in developing countries 1/, no comparative analysis has been made for this line. 13. Heavy equipment is normally made to individual design. It re- quires considerable investment in special testing facilities and in heavy handling and machinery facilities. The production cycle is long, varying from 18 to 24 months for very large equipment, and from 8 to 12 aGnths for smaller ones. The concept of heavy as just defined forms the basis for specialization by some major firma and, more commonly, for the separa- tion of heavy equipment divisions from other. within large international concerns. These divisions, of cour,e, may not coincide exactly with our definition. Thus, small lenerators, traneformers, and .witchaear may be produced ••parately from the larger units. Where the market 1. larle, I !I With exceptions such as India and Spain. - 6 - small_r units may be produced in series, and different manufacturing tech- nique~ may be adopted. 14. Within the heavy eqUipment category, there is a significant differ- ence between equipment at the heavy end of .the range and equi~ent at the light end. To avoid confwaion the terms "largen and .....~1 .. will be used .to indicate such differences in size or the range of powe'r or voltage will I be givenJ In d~veloping countries with low domestic and industrial consump- tion,individual generators and transformers are likely to be relatively small. Again, since individual regions are often unconnected with one i another or with large power sources, at least in the early atases, trans- mission lines usually ~ave low voltages. In this respect, there could theoretically be a certain hanao~y between the development o1f the power network and the development of the technical capability of the local in- dustry. An indlastry could gaiD experience OD equipment of aediuaa .izes before tackling very heavy and complex ones. 15. Our study is focussed on the conditions for developing countries to enter the manufacture of heavy electrical equipment on an economic buis. It includes ther"fore countries which have takeu only parti81. st('fJpS in that direction (Mexico), or may be OD the verge of entry (Pakistan), as well .a those which are more' advanced (e.g. Spain and Brazil). This has led 08, in some casea, to investigate how well countries such as Mexico and Pakistan ate doing with products repre.entative of their present level of technical capability; these products may, at leaat in part, fall below the range we heve defined as IIheavy equipment". This has led us to dis- tinguish three categories of electrical equipment - large, interaediate, and small. They are' rough divisions, and their border lines may vary ac- cording to the country, the age of the industry and other factors • .'J:he Products 16. Though there is a tendency to apply mass production aethods to I the .anufacture of distributiontranaformers and .mal1-size Bteam turbines, the ...in characteristic of the industry is the custom-designed nature of its products, which are in general very large unit.. Thi. calla for ability to design according to the customer'a specification, and forhishly 'killed - 1 - workmanship at the bench level to interpret drawings and to carry the respon- sibility ~or ind~pendent jobs on large pieces with very ema1l tolerances. Manufacturing techniques vary greatly with the type and .ize of product. 17. Economies of scale in power generation and tr,r.nami.sion (or some- times the geographical remoteness of major power .ourc.. ) have forced the industry to strive for higher voltases and to concentrate anor.ou. pover in single units, i.e. on a single axis (Table 1). Before the 1950'. the maximum commercial A.C. transuda8ion voltage was 220 kv; in the laat decade this has been raised to 400-500 kv. At present the effort is to obtain 750 kv or more. Similarly, not more than t~n years ago the largest turbine built in Europe was' at the moat 200 MW, whereas in 1966 the European firms were manufacturing sizes above 600 Hi and the capacities are greater atill now. Higher voltages place new strains upon the aupp1ier industries (espe- cially for insulaticn material), nec.ssitate special technique. and equip- ment for manufacture and testing, and also take up more space. Larger equipment a180 requires higher quality components (forgings and castings especially) and larger, more powerful production equipment and faciliti.s. 18. The size of the equipment ordered today is 80 large and construc- tion periods are 80 long that at a given moment one unit or one order may be taking up 20-25 percent of the production capacity of a plant. The unit may be on the factory floor for a year - generally even longer. The programming of overall produ~tion is rendered extremely difficult by the irregularity of orders and the size of each individual order. Large order. and long lead time. demand large amounts of working capital. The need for fixed capital has increased because economies of scale in selling cost. and in research and development, as well as the marketing advantages of offering a full line of equipment, have prompted most of the international firma to manufacture practically every type of heavy electrical equipment. - 8 - . ' TDLE 1: Maximum Specifications'of Units Manufactured or in'Process of Manufacture in European Countries, 1951, 1958 and 1967 Unit 1951 1958 1967 Steam turbines Rating MW 110 250 660 2 250 Maximum pressure kg/em 89 145 I Maximum temnet.ature ~, I °c 520 565 565 ~ermal generators MW 125 300 660 Hydraulic generators MW 160 180 242 Transformers MVA 200a 900b 1,000 a Three 8ep~rate phases. b Three separate phases of 300 MVA each. Source: Organization for Economic Co-operation and DevelopmeDt, Twentieth Survey of Electric .Power Equipment (Paris, 1967). 19. Manufacturing techniques vary with the product group. The manu- ·facture of transformers, which are atatic equipment, differs fro. making switchgear and rotating ..chines, and among rotating ..chines, ateam equip- ment is very different from a hydro-electric senerator. In the largest international concerns" plants are often specialized in a given type of equipment. This variety of heavy electrical equipaent add. another diffi- culty for developing countries. The Market 20. The customers for heavy electrical .quipaent are lenerally lars_ cone ems : power companies, railroad., and larg_ ind,... trial fins. The n_ber of customers is liaited, so_times to a sill8le cueto.er. Thus, in Prance, Electricite de Prance accounts for .o•• thing like 40 percent of the market, the Prench State Railways (SHer) for another 30 percent. 21. Electrical equipment is alway. subject to regulations and mini- mum specificationa for aafety. Besides, each customer haa certain stand- ards and rules of his own, and evaluatea each tender on the basis of a - 9 - "tationale" dev~loped over the years. Customer pr.ferences are based not onlY,:'Dn price and quality but also on the suitability of the equipllent to the existing system, the expetience gai~ed by manufacture::a on similar work and the relation between lIat,lufacturera; and purchasers. ~rhe price of heavy electrical equipment, in fact, is part of a package wbich includes many DOn-price elements such as speed of delivery, credit terms, ease of ordering, and quality of aft~r-sale services. These non-price factors explain some of the preferenee enjoyed by local manufacturers in their national markets. 22. If manufacturers are tied essentially to a local market, the regularity of orde~ing by the main local purchasers becoaes highly signi- ficant. A manufacturer of heavy electrical equipaent baa a high invest- ment in fixed assets and in a highly specialized staff and labor force. Unless be can ensure reasonably stable utilization for these resources, operations will be neither profitable nor economical. Raw Materials and Semi-finished Components 23. Material costs dominate factory costs; they are 77 percent of ex-factory cost for a transformer in Mexico, for instance (aee Annex Table 5). The most important raw materials for the beavy electrical equipment industry are copper wire or bars and silicon ateel sheets. Por ex..ple, in a transformer manufactured in Brazil, copper accounts for 23 percent and silicon steel for 40 percent of total material costa. Ordinary steel sheets are also important, as are bearings and insulation materials such as paper, porcelain bushings and transformer oil. Theae materials are ~ported by moat countries and are normally freely available on the world market at established prices. Though supply difficulties in copper in 1964-1965 led to a disorganized price atructure, it is difficult to idantify any important long-run cost advantasas for any country or fi~ with reapect to supplies of primary copper. Nor is there much advantage in firma manu- facturing their own silicon sheets. On the other hand, a competitive dis- advantage may arise for 8011e developing countriea where protaetlon haa raised prices for semi-manufactures of copper or aluminum. - 10 - 24. Steel castings and forgings as well as n~n-ferrous caatinl;s are used quite extellsively by the manufacturers of hea'~ electrical eq\rl:pment, as well as by other branches of engineering, industries. In indu.trial- ized countries there are specialized foundries and forging plants fro. which heavy electrical equipment manufacturers purchase their requifeaent. of all semi-finished parts while they themselves concentrate on design, machining, elec~rical engineering and assembly. In developing countries, however, the heavy electrical equip_ent industry is often forced to eatab- lish its own facilities for casting, forging and the like. Electrical , I equipment manufacturers in Pakistan are even forced to make 'cheit' own nuts and bolts. Even where outside suppliers are used, electrical equipment ; manufacturers must often help these with financing and technical assistance. 25. The establishment by electrical equipment manufacturers of special facilities for the production of' parts, coupled with the protection offered to local suppliers of cable, wire, ~U8hings, etc., have raised production costs in several of the countries studied. Though electrical equipaent manufacturers are ~arge buyers, they have had to accept high costa (even where these are caUsed by monopolistic pricing practices) as a price paid for the desire of the local government to industrialize and to aave foreign exchange'. There has been resistance (not always succea.ful) when the quality of the local product is inferior; power companies, for obvious reasons, are concerned about the performance of their equipment. 26. Heavy electrical equipment, simply because of ita size and nature, offers substantial savings in freight and handling when partly built or assembled in the country. Some international firma have developed special designs suitable to local manufacturing capabiliti•• , e.g. part. normally produced by cast steel are replaced by welded piece., etc. Thus, a German company developed a design for the Eatreito Power Plant in Brazil by which they arrived at • 33 percent domestic contribution (- ••vinsa in for~i8n exchange baaed upon c.i.f. value) and 8 60 percent reduction of the weight to be .hipped. - 11 - Technology 27. Though the basic principle8 in the construction of beavy elec·' trical equipment bave r ..ained. the ._e since the beginning of the indu8try, there ha. been a steady 8tre.. of technological improveaents and .one major breakthroughs.. Patented invention" now cover almoet every cOllponent and manufacturing process. These have been directed, in part, towards the reduction of electrical losses and also towards the ••nufacture of equipaent which i8 lighter and more compact. Y But, from the beaianing, the aain thrust of technological progress haa been in the capability of designing and producing ~he ever larger equipaent demanded by customers. In this f respect the industry may now be close to a ceiling .ince the enormous size of some modem equipment causes many difficulti... On the other hand, this situation a180 pre8ents a challenge tb introduce revolutionary changes in the generation, tranaai8sion and application of power. 28. The introduction of nuclear energy for power generation has been a major revolution for the power supply industry, and haa had important repercussions on the electrieal equipment industry. All, . . jor electrical equipment manufacturers have become involved not only in the con.truction of steam power plants baaed upon nuclear energy ,095 105,938 (r~1VA ) 1965 99,996 100,330 223,563 2.24 1966 d 103,307 92,246 217,982 2.11 1967 119,905 206,921 1.13 Nuclear PO'\vcr 1965 2,01~ 750 Reactors O'IVle) 1966 1,312 2,156 4,527 3.45 1967 1,293 5,396 4.1'1 USA - Generators for Steam 196~. 12,394 21,112 Turbines (H'~~) a 1965 13,744 26,975 42,496 3.09 1966 15,492 47,023 55,444 3.58 1967d 24,364 87,159 3.58 Generators for Water 1964 2,397 5,477 Turbines (M-l) C 1965 1,621 2,504 9,347 '5.77 1966 3,071 1,002 10,230 3.33 1967d 3,414 8,161 2.39 POl·rer Tran::;f.ormers 1964 68,967 100,613 (~WA)b 1965 80,2h8 107,838 121,066 1.51 1966 102,lh6 153,103 1h$,864 l.h3 196,(d 116,031 190,825 1.64 Nucle.ar POti0r Reactors (H\~e ) 1966 1967 926 2,420 .. •• •• 26,4h1 10.9 Ja:ean Genera tc)rs for Steam 1964 2,318 2,137 Turbines (NH)a 1965 2,954 3,457 4,510 1.53 1966d 2,930 3,688 4,714 1.61 1967 2,812 5,765 , . ~ 2.05 -. 21. - TABLE 5: E..~loym8nt of Hanufacturin~ Capacity, 196h-67 Orders in Hande Orders in Hanel Deliveries NelV Orders January 1 /deliveries (I) ''"12) T3l (3~J ~ (1) Generators for Hater 196L~ 1,755 749 Turbines (N\'J) a 1965 1,078 2,460 1,365 1.27 1966 827 3,883 3,222 3.90 1967d 2,119 6,351 3.00 P01'1er Transformers 1964 17,452 24,471 (MVA)b 1965 22,268 26,895 24,189 1.09 1966d 20,792 22,800 28,969 1.39 1967 25,497 29,124 1.14 a 10 MIl] uni ts up~Tards. b 10 l{vA Units upwards. c 4 :H\I] units upl·lards. d Orders booked at January 1/67 for del:i..very in 1967. Japa'1.: orders booked at April 1/67 for delivery in 1967. e Japan: orders booked at April 1 of the year in question. Sources: Organization for Economic Cooperation a...'1d Development: Nineteenth ~nd Twentieth Surveys of Electric P01-1er Equipment, (Paris, 1966 and 1967). Note: New orders (+) and deliveries (-) do not entirely account for the differenc.es in orders in hand between one year and the next. The balance is evidently explained by cancellations. II • - 22 - 48. To produce a steam turbine of 50 ~ega~atts , capacity requires in Europe 60,000 man-hour8, whereas a 100 m~8awatt unit requires only 80,000 .In-hour8. Production of one 1100 MW unit requires the holdinl of .ore Baterial. and stock at one ti~ and deaands more powerful manufacturing capacities (larger buildings, longer lathes, heavier crane8, higher as- sembly .hops, etc.). The manufacture of such unite, if never undertaken before, requires very large research and develop.ent expenditure. When purcha8ers buy two 600 MW turbines to meet a given power demand instead of buying six 200 MW unit., there is work for only two firma, aince pro- duction of steam turbines can hardly be broken down to emall fractions, and purc~asers like to have only one major equipment supplier for a project. Therefore less work is shared/by the industry. Yet an individual fi~'s R&D expenditures and additional investments, etc. will be much hiaher • This favors the trend towards concentration. 49. Another example can be given froll the transformer market in the United Kingdom in 1967: 1/ "In consequence (of ever l~rger transformer sizes) the average price of transforaers remained at about 11 (then US$2.80) per kva from 1930 to 1960 and has dropped slightly since then. With the smaller distribution transformers ••• prices now barely cover variable costs. With the big transformers ••• last year (1966) prices came down 20 percent. EYen then manufacturers of the aize of A.E.I., Bruce Peebles and C.A. Parsons got no new transformer orders from the Central Electricity Generating Board. This year manufacturera were bracing themselves for another round of cuts of UF to 30 percent. At this point the C.E.G.B. ssw trouble ahead. Once prices get below even variable costs a manufacturer can increase hie profits by st.,ly clo.ing his factory ••• Prices for this year's orders will be 'frozen', in the .ense that the e.E.G.B. will not let them go down any further. It will be ordering at last year's price.. There 1. enough capacity in Britain to make about 146 million of grid transformers (200 kv and above) a year. In 1964, when the C.E.G.B.'s orders were over E35 .illion, the industry was only 80 percent loaded. Ita break-even point is around 70 percent. But the current ordering rate is only around 118 million (i.e. less than 40 percent capacity uti1ization) ••• kDocking the big eight manufacturers down to four would not be so difficult." 1J 11 ~ Econoai.t May 20, 1967. pp. 820-823. There ate now only two or three big ..nufacturere. - 23 - 50. Excess capacity, large capital requirement. for reaearch and develo,.eDt, high overhead costs. lumpy or fluctuating demand, and the fear of competition becauae of regional tra~e .greements, are the factor. which explain the pressure on prices and the pres.ure for aergera and/or "arrangemente" among producers. In co.peti~ive condltiona thi. pressure r also gives a atrong incentive to producers ,to seek .eparate markets iD order to minimize price competition. Thus, excess capacity al.o explains the I , j~aloU8 protection of nationa+ markets and the attempts to extend these protected markets abroad through various financial and other arraDgeaenta. 51. Finally, such pressures explain both the very .harp price competi- tion on the free world market and attempts to get around thi~ competition by licensing agreements with firms working in protected markets or direct investments in such firma. An international cartel known as I.N.C.A. (Inter- national Notification and Compensation Agreement) was in operation in the early 1930's. ~ In the late 1950's and early 1960's the producers in the USA made an agreement on price fixing, which was stopped by court ac- tion. !!I If attempts to reform an international cartel have been _de sub- sequently, they do not .ppear to have been successful. 2..1 Mergers 52. Leading manufacturers have tended to produce a full line of heavy electrical equipment. This has made them less vulnerable to fluctuations ,. in the demand for individual items and has had certain advantages in com- peting for jobs for complete power plants and in relations with suppliers (because of the large total volume of orders). The speed of technological progress today i8 such, however, that enormous investaent. would be needed to keep in the forefront of technology over a wide front. In Prance, this See: Report of the Federal Trade eo..i•• ion'on Interoational Electrical Equipment Cartel., U.S. Govara.ent Printing Office, 1948. The Wall Street Journal, Marchl6, 1964, aDd Ausust 4, 1964. The New York Tim•• (and other daily papers) on or about July 1, 1964. There i. instead a trade aeeoclation. The International Ilectrical Aaaoc:lat:lon, London. of which all major Europ. . uuufacturars are _berfJ. - 24 - problem has been solved partially by:Electricite de France bearing a con- I siderable portion of total research lInd development costa. Electricite de France haa become more courageous in its order policy by encouraging new techniquEs, e.g. for large units. In Germany, on the other hand, the several power companies provide little .uppbrt to the research effort of the Gel'lDaD. heavy electrical equipaent inawitry. Hance, in apite of t the size of individual companies, thia industry ha. lagged behind other countries, e.g. in ste.. turbines. It is only recently that German power companies have taken a step towards large UQits; they have ordered two nuclear reactors of 600 MW and related equipment. 5j. The movement towards mergers has changed the structure of the industry, especially in Europe. The main manufacturers of heavy electrical eqUipment in the United Kingdom were AEI and English Electric, which vere formed through .ergers of several companies, and C.A. Parsons, which bought the business of General Electric Company (UK) in rotating machines, and has now joined with Reyrolle. In September 1967, Johnson and Phillips Ltdo were taken over by English Electric. Then G~C, more successful profit- wise than its rivals, bid first for Ail and then for English Electric, which merged with GEe in late 1968. In 1965, two major French producera, Alstham and CGE, decided to aerge some of their operations to fora ODe new company, Del1e-Alsthom, for the production of high and ••dium volt- age switchgear and another company, Alstho~Savoi.ienne, for large trana- formers. CGE holds the majority of the shares in the first company vhi1e Alatho. is the leading shareholder in the second one. CGE and Aistho. have a180 joined hands, through UDelec, in the production of aotora, a.a1l tranaformers and low voltage equipment. In all theae merger8 the pr•• - , sure in Prance from Electricite de Prance, and in Britain fro. the Industrial Reorganisation Corporation, va. the significant element. Por inataDce, Electricite de Prance refused to give any more ordera till the . .nufacturera grouped together. In Italy, An8a1do San Giorgio, a atate-owned manufacturer . " of heavy electrical equipment, merged ita operations with ,CGE (an Italian 8ubsidiary of General Electric, USA) under the new naae of ASGIN. In Austria, " there have been numerous bids by foreign competitors to take over ELIN Union, which is itself a merger of ELIN A.G. and ABG-Union. III Switzerland - 25 - (1967), the Brown Boveri group took over the Ma9ch~nen r.brik Oer1ikon i I with aales of about $40 1Ii11ion and employment of 4,500. ,Pinal1y, 'lven in Germany, Siemena and AEG-Telefunkflin have announced that J a. from April 1, 1969, they will develop, produce and eell jointly their h~avy electr.ical equipment in two jointly-owned subsiaiaries (though both will continue to develop their own nuclear reactor .odels derive,l fro. different Merican sys tems). !!J i International Manufacturing Arl~angeaents 54. National markets reaerved ,for domestic producers. The internal markets of exporting countries, in moat cases, are he.vily protected nearly I to the point of complete exclusion of imports. First, there i. tariff protection ranging from 10 to 20 percent, or higher in some ca.ea. 11 Secondly, I • etlch industrial country has ita own standards with respect to operational conditions and safety, equipment producers accustOiled to working with a given standard and with existing purchasing habits'are thus favored. Finally, and perhaps most important, power companies tend to support their domestic industries and 1lSually do not even invite foreign producers to quote. Export and import figures given in Annex Table 2 and consumption figures in Table 1 a1so support this point. Prices quoted by American manufacturers 1n 1 the USA (e.g. T.V.A.biddings) have been substantially higher than inter- national prices. This is made possible in part through the Buy American Act. 55. Attempts ~ extend the domestic markets internationally. Three circumatances have facilitated the extension of protecteci domestic mar- kets beyond national borders: !I The Economist, November 2, 1968, Dar Spiegel, No. 45 of 1968. ,,11 Even in the cue where the tariff is insignificant, -the import ia apparently restricted by the purchaser's preference. Por instance, in Switzerland ~port duties for large hydroelectric genera tore amount at present to 2.7-2.8 percent and for larse tr.n8fo~ers to about 2.5 percent ad valor... laports originating in EFTA countri.e are exempt from duty. Yet heavy electrical imports into Switzerland are very minor (less than 10 percent') though .tandard. are identical to those in Continental Europe. This phenomenon cannot be explained only by the price competitiveness of the dOilestic aanufacturers; there may be ea.e el.-ent. of ,purchaser's preference. . - 26 - I (a) Special relationship with a developing country arifling from previous political association, such as language, educa- tion and experience of administrators and engineers, exi.t- ing equip.ent and trade channels, etc. Against thia, it is the policy of some governments to 1008en economic ties with former I8Other-countriea. In 8pite of ~hia, and aore importantly, severe competitic,n frOID continental European and Japanese companiea, the British industry is still by far the largeat supplier in the Commonwealth countrie., though its ahare ma:1 have decreased. (b) The role of consultants hired by the developing country, who are more f_iliar with the equipment of their tlOIle country than with competing equipment from other countrie•• (c) Tied aid and bilateral agreelllents. Pric.aa under .\'Jcb arrangements tend to be ai.1lar to boae aarket prices in the exporting country, i.e. considerably hig~er than price8 in the free market. Some companies (.ai~ly from Ge~any and France) disagree that tied aid has an effect on the price. They argue first that inter- national c01lpetition exist:; between ~upplying cf)untries for both equipment and credits and that beneficiary countries can select both the equipment and the credit terms that suit them best ..ong a generally wide range of proposals. Secondly, they say there will be a number of firms from the donor country participating in tendering, thus assuring competition. There is reason to doubt whether this does assure competition, however. General experience is that prices quoted in tenders restricted by a "tied aid 'competition" are much higher than tho•• in world wide open competition. In one particular case, even though badly in need of foreign exchange, the purchaser switched from tied aid procurement to own financing because the price v.. ex- cessive. It i. also argued that the low prices .ometiaea obtained under competitive bidding are to .o.e extent made pos8ible by higher prices obtained through tied-aid pro- curement, and in .ome ca.ea, through high prices in ho.e markets. Heavy equipment producers a~e natural .upporter. of bilateral aid programs. It i. sugg.sted that without tied aid, co.petitive price. would tend to riae. But this argument aSBu.es that all exporters are equally efficient and none are currently sheltered by tied aid. Faced with increasing ca.petitlon in third aarkets, US manufacturer. have campenaated by concentrating on .ale8 to developing countries under tied aid arrange..nts. 56. LicenainR aRreements and direct investaent.. Licensing agree- ments in the heavy electrical equipaent industry are so widespread that every . .jor firm is a licensor and a licensee at the 8ame tiae. Even among the larKe fi~., only very few have a positive b.~ance of licen.e payaents. - 27 - 57. In standard light equipment (e.g. motors and distribution trans- form~rs), the basic technology and manufacturing procedures are well known and could be copied. Nevertheless, leading manufacturers have experience of great value to newcOlllers, ,and introduce many refinements l~oth in products and pro~uction techniques wh:tch can form the basis for liceDfJe agreements. Atypical fee in this area would be about two to three percent of the value of the output. 58., Heavy equipment requires skil::ed design worlt and t~Lme-conuUID- ing engineering. In this area newcomers cannot progress without a license and with.,ut a continuing working relationship with an established produc,er. Hence, license fees are higher. From an average of perhaps four to five percent, they may reach ten to twelve percent on large steam turbines. Higb license fees are, of course, the reward for auccesaful reaearch and development which has placed a company, at least t1emporarily,; in a unique position in the field. 59. Many international companies would prefer to aell licenses with- out getting involved in direct investment and production abroad. This, for many years, was the policy of We3tinghouse Electric of the USA with respect to the European market. But policies such as theae were disrupted by the firm resolve of several countries to build their own electrical equipment industries. Some companies had established particularly close contacts with certain developing countries, and were induced to help in initiating manufacturing operations in those countries both by their inter- est in preserving a market and by the pressure of their distributors on manufacturing licenaees for light equipment. As manufacturing of heavy equipment was extended to industrializing countries, a need waa felt for assistance and supervision of a degree which can only be achieved by parti- cipafing in management and in equity. At the same time 80me developing countries insisted on equity participations by the foreign lic~n80r as a condition for permission to manufacture. 60.' Partly as a result of these trends, there are substantial .hare- holdings by foreign companies, General Electric (United Statea), Alsthom and G.e.E. (France), Brown-Boveri (Switzerland), ASEA (Sweden), Siemens and AEG (Germany) in the Spanish electrical equipaent industry. General - 28 - Electric and Brown-Boveri are the major producers in Brazil but ASEA, Siemens, and ABC are also manufacturing the~e. In Argentina, the principal pro- ducers are CEGELEC (financial participation by eCE France) and SIAM Electro- Mecanica (Westinghouse 40 percent). Mexico's major producer is lEM, in which Westinghouse has a ieading participation. G'aneral Elet:tric, I'lexico, is primarily in the light end of the bU8i~ess; there are DO other large producers of heavy electrical equipment. In India, three foreign firms, including AEI, (United Kingdom) are among the major manufacturers of tran8- formers. It has been the Indian Government's policy for the state to enter the production of heavy generating equipment and heavy traction and indus- trial equipment; even so, All was employed as a consultant. In Pakistan, which is just starting to produce heavy equipment, two out of four domin- ant electrical equipment manufacturers are Pakistan-owned with license agreements; the other two have foreign majority participations -- Siemens (Germany), and Johnson & Phillips (UK). - 29 - IV THE INDUSTRY IN DEVELOPING COUNTRIES Growth of the Industry 61. The heavy electrical equipment industry in the developing coun- tries is the 'Dutcome of two sets of fac·tors. The ,first set consist. of Goveraaent policies favoring the creation of capit;:al goods industries. The second is the interest of international firma in keepinl a foothold in every important aarket as long as pos.ible, which is explained in part by excess plant capacity in the home country and high technical overhead costs. 62. Governments have shown their vital concern in the creation of thia industry by giving it a prominent place in their development plans (e.g. India's second and Pakistan's third plans) and by encouraging private investment through protection, preference in public purchases, or even financial assistance. Because of the know-how required to build cuetom- designed products of gL'eat technological sophistication, the dl.aveloping countries could not start the industry without the assistance of interna- tional concerns. Even in thEl one case where a' wholly-owned govemaent plant was built (the Bhopal plant in India), the purchase of know-how and even managerial assis tance f rom foreign companies was nece.sary. Govern- menta desired quick forceful action in the hopes of achieving industrial adulthood and of radically cutting t.port bills. Being unf.ailiar with the industry, 80me of them tended to preas too hard for a telescoped increase in capacity with too many producers. 63. The international companie. were taken by .urprise, and their first IIOves were es.entially defensive. In the extreme case, tbey 10okt.'4 upon manufacturing activity as a facade behind which exports could go on in the traditional way. At best, they were skeptical about the return from'manufacturing iuveataents; their profits would come inatead from li- cense fees and from s.les of such parts and components as could not yet be made in the developiD.R country. These profit~. of course, would be - 30 - increased if it were possible to capture the markets of those competitors who were unwilling to come in, or were too late in establishi'ng a manu- facturing base. Almost without exception, the international firms question the economics of heavy electrical equipment manufa~ture in developing coun- tries. Theoretically, if they had realized the determination of the devel- oping countries to build these industries, they might have devised some scheme fdr sharing the markets rationally and setting up plants at a few s~lected locations in the developing world. Such a solution, however, was ruled out equally by their own rivalries and by the ambitions of each developing country to possess its own industry. 64. tHe disorganized bsckground of the industry and the scramble by the many previous suppliers to defend or capture protected markets explal11 the present characteristics of the hea~, electrical equipment industry in developing countries - small size plants, high cost operations and doubt- ful growth prospects. With a longer history of prntection, the Spa~ish industt'Y' in some ways provides an exception to this rather discouraging pattern. Country Experi.,ences 65. Spain. The Spanish heavy electrical equipment industry achieved large output during World War II and was given a rapid boost by intensi- fied power development after the war. More recently, railway electrifica- tion and industrial expansion have helped to increase sales. International companies, including General Electric; Westinghouse, Alethom C.G.E., and after the. ,.'ar, Brown-Boveri, Siemens and ASEA entered either alone or in joint ventures with domestic firms. As.a result, .tandards improved steadily, and there was no significant technological lag. 66. At the present moment, the main producers are General Electric. Espanola (General Electric, United States, has now acquired a dominant interest), whi.ch probably accounts for about one-half of the total dom•• tie production, and Ceneae.a (a Westinghouse licensee) which .ay account for another quartet". La Maquinista, a major Spanish eq1.lipaent producer, . .k•• generators and motors for die.el electric locomotive. and i. preparing to make turbo-alternators for the~l power stations under a licen.e from - 31 - 1 Bl,:own-Boveri, Switzerland, an iaportant sbateholde±'. Other producers aake mainly transformers and motors. 67. The Governaent has facilitated the industry's growth by iaport cont~ols and pr~tective duties. Duti~. are generally high on light equip- ment (50 percent, plus 10 percent on induction motors) but .00000hat lower (20 to 25 percent, plus five percent) on beavy equipaeDt. ~til recently, tmport licensing was quite restrictive and was as important ua the tariff in prote,cting the domestic industry. Nevertheless, there has been a gradual i and efficient growth in the capability of the indu-try and in domestic content. 1/ 68. The Spanish industry does not appear to have suffered froa those drastic changes in equipment purchases by the local power indu8try which have been a disruptive influence in several of the other countries studied. Prices for electric power have been aet so as to facilitate aelf-finan- cing by power companies, and protection has reserved the aarket for the domestic industries. In recent years, nevertheless, there haa been a sharp upturn in imports of electric paver machinery and switchgear - frOID about $13 million equivalen~ in 1961 to $32 million equivalent in 1965. This reflects both the increasing importance of thermal power equipment (with a large import cOlDPonent) and the full order books of the Spanish indu.try. 69. The Spanish industry has taken full advantage of existing tariff protection, and is not competitive with world market prices. Its main problem is adjusting to rapid changes in ttl~h1iology and size of equipaent. This is accentuated by the world market price structure (relatively high prices on new equipment and specialties against low prices on equipaent with well-documented technology and manufacturing techniques) and by the increasing shift in Spain from hydropower to large thermal power atationa. 70. Though domestic demand is sufficient to pe~it full capacity utilization in most heavy electrical equipaent plants, there is an increas- ing interest in exports, particularly to Latin Aaerica. Spanish electrical !I Present capabilities and dOlDe.ti~. content for all the countri•• studied are su.aarized ~n paras. 97-100 below. - 32 - ~ equipment manufacturers , have joined with United States and Canadian manu- facturers for the supply of diese1-electric locomotives to Argentina and Brazil. These orders have been faci1itat~d by spe~la1 trade agrea.ents i between Spain and these Latin American companies abd by United States and Canadian financing. The international companies do not 8eem averse to assigning Spain an export role within their international c01~orate frame- work on items or components which can be made cheaply in that country. ) 71. Brazi1. Brazil is similar to Spain in having a very large power market; the annual growth in electric generating capacity exceeds one mil- lion kw. AS in Spain, most of the present demand can be met through hydro- power, but whereas in Spain thermal power seems to be gaining ground in I new orde~s, in Brazil reliance on hydropower will be possible for many years. Another difference is that in Spain the industry grew ateadily, while th~ growth in Brazil was explosive; the share of imports in the total consumption of heavy electrical equipment fall from 94 percent in 1956 to 55 percent ih 1964 and to about 33 percent in 1965. 72. The provenance of the industry dates from 1949, when a c~is- sion was formed to study ways of meeting the growing demand for heavy elec- trical equipment. As a result, several international firms were invited to start manufacturing heavy electrical equipment and were given facil- ities for duty-free imports of the necessary plant, as well as strong assur- ance. of protection for the finished products. 73. The first major international producer to eatablish large-scale manufacture of heavy electrical equipment in Brazil was Brown-Bover!, Switzer- land. Ita Brazilian subsidiary, almoat wholly owned, achieved a rapid buildup in capacity during 1958-1962, and in the latter year was equipped to design and manufacture prac,tically every type of rotating equipment" The only other company with a similar broad capability is General Electric S.A., a subsidiary of General Electric, United Statee. This co.pany, which had been manufacturing 11gb~ and medium electrical equipaent for yeara, started operatioDs in ita new heavy eqUipment plant in 1962. Other major foreign firms, like ASIA, SieaeDs and AEG, have also initiated aanufactur- ing operations in Brazil, a. yet primarily for power transformers. !I 1:.1 8i..en8 ha. alao delivered so. . large hydroelectric generator. produced in their local factory. jl - 33 - ! 74. The Brazilian industry is capable of producing pOWf!r equipment up ~to voltages of 500,kv, although its preaent experience is li.ited to 70 mva for generatora and 70 ava for power transfor-era ae a three phaae unit. Heavy- switchgear and heavy duty moto~s ! are • not yet produced. The import component aay range fro. less than 10 percent for a 15 ava gener- ator or 700 hp synchronous aotor to 30 percent for a 33 mva transforaer or a 50 mva generator. ~ported inputs include copper, grain-oriented I silicon steel, .ica, bushings over 60 kv, and aOlle types of insulatin:g ! paper. Autosaat:Lc tap changers for transformers and regulators for gener- ators are also imported as are certain castings and forgings outside the ! t capacity of the Brazilian industry. 75. The most important problem faced by the industry was the recruit- - I I ! Dlent of staff at all levels. Although the industrial enviro1.1l1lent of Sao Paulo was highly favorable, the creation of the electrical equipment industry I ! . coincided with unparalled growth and expansion of the aechanical and auto- motive industries and the steel industry. Craft..en capable of operat- ing production machinery were scarce_ as were engineers and araftsaen. There were also some supply problems. For instance, several firms attempted to manufacture unaolded mica 'sh&ets for commutation segments (needed for uniform properties and high heat stability), but the product turned out to be very expensive (up to four times the cost of the imported material, including duties) and, worse, of poor quality. The newly established plants had t.o operate in an inflationary environment and had difficulties in securing a steady flow of work. Local procurement of heavy equipaent has been so.e- ~bat errati~, reflecting both econo.ic fluctuationa and variations in foreign finanCing for power loans. 76. Brazil's heavy electrical equipment industry has not yet reached even the com,etitive level of the Spanish industry. There have been diffi- culties in meeting delivery schedules and in spite of relatively higb pric•• , the profitability to date has been low. Yet, the outlook is good, given the capability attained in a short time and the l~rge domestic market; unlike Spain, Brazil will have a market for hydroelectric equipment for many years. - 34 - 71. Argentina. In Argentina, the existence of a light electric equip- ; : aent ind,.try helped to give rise to a heavy industry. Earnings fro. exist- ing enterprises, which were mostly foreign-owned, were Dot transferable and had to be invested in the country; severe co.petition in the light seeto,r also drove firms into the new field of heavy equipaent. The industry was given substantial goven.ent encouragement. until 1958 b)! quantitative import restrictions and. more recently, by high duties and import 8urcharges. 78. Between 1958 and 1964, domestic production of heavy electric equipment grew rapidly from 1.5 billion pe89s, to nearly 3 billion pesos ( at constant 1960 prices. Three finas accouklt for the bulk of the output of heavy electric equipment. Cegelec, a subsidiary of the French fi~n i of e.G.E., is the oldest manufacturer of heavy transformers in Argentina; it produces mainly power and distribution transformers and also some switch- it~g equipment. Electrica Mechanica J~gentina was founded in 1952 to make control panels; it has diversified production to include switching equipment. The third major producer, Siam di Tella Electrica Mechanica, was created in 1959, with a participation by Westinghouse of the United States, to manufacture generators, transformers, motors, capacitors, electric trac- tion equipment, etc. 79. MOst firms were originally established, under 80me government pressure, to manufacture mainly one product (say, transformers or oil field pumps) for vhich the public sector was the major buyer. When it became clear that public sector orders were very irregular and payments often slow, all companies were forced to diversify their operations in order to achieve a more continuous flow of work. This has led to a high capital- ~,tput ratio and also to an excessive number of producers for each major product. 80. The industry produces transformers up to 0'0 mva and Beneratore or up to 8 mva as well as electric traction equipment and medium switchgear. In addition to beavy equipment, the .ain companies also produce capacitore, control panels, light 8witchgear, and several other items. 81. Excessive diversification, unused capacity, large inventories because of import controls, and difficulties in obtaining outside finance explain the high price level for Argentine heavy electrical equipaent. - 35 - Domestic aupplier industries are generally heavily protected and pricea are high for copper, a_i-manufactures, and for in,sulatien materials. On balance conditions in Argentina in recent years have been ao abnormal that ohe cannot use the past as the basis for future projections of growth and development. Nevertheless, governaent policies appear to have been wanting in two respects - lack of planning in procureaent and excesaive protec- tion both of the electrical equipment industry and of ita aupplier indus- tries. 82. Mexico. The light electrical equipm*nt industry was launched in the middle 1940's when the Mexican Goveraaent began a conacious policy of industrialization. Substantial protection 'nd tax incentives were granted and some important foreign producers were attracted. The induatry ia atill producing mainly li:ght equipment, but one fina, 1EMS A, in which Westing- house Electric, USA, has an important participation, haa emerged as a domin- ant firm in the manufacture of heavy transformere, motors and switchgear. Generators above 30 ava are not produced locally. 8l. There has been a gradual and efficient build-up of tbe indus- try's capacity and, as we shall see later, today's prices for tranaformers in the 15 to 20 mva range .are competitive by international atandards. On the other hand, heavy motors are about 40 percent above the world market price level and light motors (which fall outaide this study) are very expen- sive. In Mexico, prices for major inputs such aa silicon steel and copper are still comparatively high. Tranaformergrade aheets, which are imported, are generally expensive. Furthermore, high protection has encouraged the ~~tabliahment of industries aupplying secondary raw materials of lower t\~an average quality and very high prices by international .tandard•• Originally, the Mexican industry va. faced with a ahortage of engineers, draftsmen and ekilled laborers, which called for very exten- 1 sive in-plant training programs for skilled labor and apecial training of the professional staff abroad. Today, the .ajor problems faced by the industry are said to be exceaaive fluctuations in government procurement (due to insufficient planning, shortage of funds, etc.) and foreign co.peti- tion. - 36 - 85. India. Unlike the other countries analyzed in this chapter India • was not visited for the purposes of this study: the data here are from reports of other visits and ftom published sourcea. However, it wa. felt that the Indian experience was too important to leave out, despite the quality of data on which it is based. The analysia of India's industry has thus to be read with caution. The first transforaer factory in India was the Government Electric Factory at Bangalote, built in 1936, which had an output of 6,000 kva before WOrld War 11. Only two firma were engaged in the m*nufacture of electric motors before the war, P.S.G. and SODS of Coimbatore and K1rloskar Bros. Ltd., of Kirloskarvadi. Total output was around 1,000 hp. Due to shortages of imported equipaent during the war and especially because of the industrialization policies of the Govern- ment after independence, the industry expanded its capacity 24 percent per year between 1950 and 1963. However, production has lagged. very much behind the growth in production capacity. ,The Indian experience is prob- ably a unique illustration of the difficulties which aay arise from a very fast expansion of output and size of product in the electrical equipaent industry0 86. The number of manufacturers of electric motors increased from 31 in 1958 to 108 in 1962 and 127 in 1963. The private sector of the induatry consists of a large number of firms. Over 20 firas produce transformers. Nevertheless, in both transformer and electric aotor production four or five firms account for over 60 percent of total output and a auch higher proportion of output of heavy equipment. (For a list of the important private manufacturers see Annex Table 14.) These firma, several of which have foreign participations but no foreign majority control, are very small by international standards; the 1962 8ales volumes of the five largest manufacturers of transformers (above 1,000 kva) , ranged between 25,000 kva and 165,000 kva. A representative plant would have about 1,500 e,m- p10yees and a sales volume of Rs 25 to 30 million (US$5 to 6 million equiv- alent). The most representative type of ownership is a public limited company wholly owned by Indians having a license agreement with a foreign . company of international repute, or jointly owned by Indians and foreign companiea. There are wholly private and wholly state-owned companies. - 37 - Sometimes public bddies such as state governments are partners with foreign cOmpanies. Several of the international firms, and the Soviet and Czecho- slovakian Governments, have collaboration arrangements in manufacturing in India. The distinctive charai~teriatica'of this expansion of plant can be explained by the .ize of the country, the government's plana for .xpansion of power generation, regional differences, and high profits in .pite of low plant utilization. 87. When imports of electrical equipment into India were curtailed, I new operations were started by Indian and foreign companies, attracted I by the Government policy towards industrialization. and by what appeared to be a promising new market. Continental European and US products had played little part in the Indian market prior to independence and the oppor- tunity suddenly opened must have appeared difficult to resist. The aize • I of the p'''pulatil)n indicated a potentially good market but infomation about income distribution, supplier capabilities, availability of .killed man- power and managerial capability was not well documented. The result was that many operations were hurriedly entered into, poorly planned and wrongly equipped. Supplier capability has remained a problem, and, more impor- tant, there have been continuous shortages of raw materials, and inadequate standards in materials and component parts. Many difficulties have now been overcome, and a survey by the Tariff Co.-is.ion of India indicates that even before the recent devaluation, in many cases ex-factory co.ts were lower than c.i.f. prices of competing imports. For instance, a 250 kva Indian transformer was over 10 percent cheaper and a 1,250 kva trans- former was over 15 percent cheaper; prices were about equal or alightly higher for a 5,000 kva transformer. Profitability is good (after tax it . averaged about 14 percent of net worth in 1964-65) in spite of exce.. capacity and shortage of imported raw materials. 88. Low price. for motors and tranaformer. do not Dece •• arily demon- strate competitivene•• , however. For instance, in procur.eDt of tran.- foraers t a prime consideration would be low energy 10..... If all desigD characteristics are con.idered, it ..y be judged that the Indian electric motor and transformer industries have not yet attained full international competi- tiveness. Improved design would not only assure the conlumer better value - 38 - but would also permit Bubstantial foreign exchange .avings ou iaported .aterials. In 1965, only about 15 percent of the aotors produced in India had die-cast rotor.:. If theae rotors were used univeraaily, copper iaporta I for the production of electric motors could be cut by 20 percent. Similarly, the use of cold-rolled, grain-oriented aheets in tranafor.era, apart fro. a·lO percent reduction in 10s.es.(8 tremendous gain) would al.o yield 10- . ~. ~ 2S percent savings in the requir,sents of uteel and copper w'bich are the ... jor raw ..,.terials. The Tariff Co.m.ssion of the Government of Iudia suggests that "the long and expensive tiae lag between the establi.hed , use of improved design and materials in overaeas countries and their ac- ceptance in India is due to t.he inertia of indigenous producer. who enjoy , a sheltered market and higher price level for their product. 1I -3/ But this "inertia" is itself related to the fact that, in 'L rigidly controlled econoway, competition is thwarted by raw material allocation and the producer has little flexibility for introducing new products or Dew aethods of produc- tion. 89. The public sector in this industry has tackled a far 1IIOre diffi- cult task. The Indian Government has considered for many years the setting up of a plant to manufacture heavier items than the private sector produced, e.g. generators, electric locomotives, etc. Heavy Electricals Ltd. India was formed in 1958, and construction vaa begun' at Bhopal of a large plant to manufacture virtually every type of heavy electrical equipaent. 4/ There were difficulties in utilizing the huge production potential. Thouah the Government had anticipated losses for a long initial period, the actual ~ Report of the Tariff Commia.ion on Protection of Electric Motora, 1963. !/ This was contrary to the advice of con.ultanta (1953) who had concluded that the production of heavy electrical equipment in India would be an uneconomic proposition. If the Governaent felt that tbe national interest dictated the construction of .uch a plant, independently of economies, they aUlsested the following alternativ.. to be adopted singly or together: (1) to develop the .ingle line of manufacture which appeared to offer the greateat chance of beiDI an ecol~ic venture, namely traction equipment and/or (2) to take over an exi.ting factory and expand ita operation as and when it becaae econa-ically practicable. - 39 - losses were much larger than expected and foreign exchange savings almost negligible. The total investment in Bhopal up to March 31, 1964 wal about Ra 600 million '(then US$125.4 million) al~at equal to the ~otal capital : . employed of Ra 604 million in the private eector in 1964-65. 1/ Though the value of annua~ output ia planned to reach: Ra 310 million (then US$64.8 million) eventually, it was only Ra 45 million (US$9.5 million) in 1963- 64. The Bhopal plant hu had difficulties in starting up but these diffi- culties may be overcome in time; it will be far more difficult to offset the effects of excessive diversification and high overheads. ~ ~ 90. Despite this unpromising beginni!tlg, in 1963 and 19'64 conetruc- tion was begun on three new heavy electrical equipment plants, and Bharat ! , Heavy Elect'ricals Ltd. was incorporated as a separate public manufacturing enterprise to administer these three units. Total final investment is estimated at Rs 1,310 m1lliol~ (then US$276 million) with projected eales of Is 760 million (US$160 million). Partial production has begun, and capacity production in all three plants is scheduied for 1970/71. 91. According to the Bureau of Public Enterprises in India, !! the High Pressure Boiler Plant, Tiruchirapalli, which has Czech technical and financial assistance, is designed for the manufacture of high pressure boilers equivalent to 750 MW. The generating capacity will eaploy 6,000 people when full production is reached. The Heavy Power Equipment Plant, Hyderabad, is also being developed with Czech assistance. The plant is designed to manufacture ateam turbinea, turbo-alternators and pumps equiv- alent to an output of 800 MW per annum in sizes up to 100 MW each. A separate switchgear unit is a180 being undertaken in collaboration with ASIA of Sweden. The employment potential is 4,350. Heavy Electrical Equip- ment Plant Hardware has Soviet asaistance to manufacture ste.. turbines and turbo alternators in ranges of 50 to 200 MW hydro-turbines and gener- ators of unit .izes up to 100 MW, and large A.C. and D.C. motors. 21 Theae figure. are for all the operations and output of the 21 largest private firma (including their light eqUipment manufacture). All dollar valuee are at the pre-l966 parity rate of $1 • .. 4.7619. 6/ Annual Report on the Working of Industrial and eo..ercial Undertaking. of the Central Government, 1964 - 1965, Bur..u of Public Enterprise, Goveraaent of India. - 40 - , 92. Pakist'an. There was practically no manufacture of electrical equipment in Pakistan prior to partition. Independence led to .ome aanu- f~cture but growth during the 1950~. was relatively .low and . .inly concen- ttated on light equipaent (fana, low-tenaion .witchgear, light aotors and distribution tran~foraers). In the .arly 1960's Pakistan entered a new phase as a result of deliberate government policy. New protective tariffs for motors and transforaers were published in 1961,-62, and the privllte development bank, PICIC, financed investments to the extent of about: $2 aillion equivalent in investment credits. 93. There are nine producers of electrical equipaent in Pakistan but most of their production is in items falliftg outside our definition of heavy electrical equipment (distribution transformers, light motors, low-tension switchboards, etc.). With one or two exceptions, the .ajor units have been started by foreign capital or with the aid of foreign partners (Siemens, Johnson & Phillips, AEG, English,E1ectric). 94. The increase in output of major items was quite dramatic, as may be seen, from the following summary: Value of output. $ million equivalent. Current prices 1956/57 1962/63 1964/65 Transformers 0.28 1.2 4.0 Motors 0.06 0.3 2.3 Switchgear 0.10 1.3 2.8 Total 0.44 2.8 9.1 95. In 1965, domestic production covered about 60 percent and 50 percent respectively of the deaand for aotora and transformers. Output wa. a. yet on a very •.all .ca1e; the largest manufacturer of transformers had an output of about $1 .illion equivalent, while the largeat aanufacturer of ,ator. produced le•• than 2,000 motora, worth only about $250,000. In te~~ of capabilities, the industry produced a range of 11/0.4 kv tran.- formers up to 1,500 kva, motor. up to 50 hp, awitchge.r of 11 kv, with a ..x~um rating of 350 ava. Even though Pakistan'. total production of . ' heavy electrical equipaent vas not far below $10 million equivalent in 1965, imports of power machinery ros. from $15 .il1ion equivalent in 1961 - 41 - to $21 million equivalent in 1965. Aided by the government's export in~entive 1 ; progr.. , the Pakistan industrY has received an important export contract for switchgear to Kuwait and • smaller order for switchgear to Australia. i 96. The Pakistan industry is not, as yet, nearly competitive. Prices for motors, tra,sformers, and switchgears are generally 25 to 35 percent above the equivulent c .i. f. price of imports. M ve .hall .ee presently, raw and semi-finished material costs (except for iron castings) are 40 tJ 60 pe~cent or more above Westent Europe levels. The scale of produc- tion is small for the types of iteaa presently produced, capacity utiliza- tion is low (generally one shift only), and inventories are extremely high, reflecting fears of a cut off from imported supplies. The industry suffers from a shortage of technical and management skills, and there are illstances of friction with overaeas partners. Pakistan ia a relatively amall market; the installed capacity in 1965 was only 1,150 MW in five systems which were not interconnected, and which used a variety of generating equip- ment (hydro, steam, and gas turbines). The desirable lines of future devel- opment of the industry therefore are far from clear, and would merit a special study. Capabilities 97. The industries in the six countries studied, except Spain, were established during the last decade. In all six countries, they have achieved the capability of aupplying the bulk of the distribution equipment. Capa- cities for heavy power generation and transformation are summarized below: - 42 - TABLE 6: The Maxi• • Size of Equipment beiJlg built by Domestic Manufacturers, 1967 Generators t Country Power Hydro Therma:. Transformers (3 phase) Spain 200 255 81 Brazil I 70 70 Argentina 60 8 Mexico 100 India 75 37 Pakistan 2 Portugal 150 Sc',urces: Portugal, 20th Report, 1967, op.cit; Sp.'in, General Electrica Espanola, see Figure 1; others, interviews with manufacturers, 1966. 98. All the countries listed except Pakistan can supply most of the power transformers needed; Brazil and Spain have the capability of pro- ducing transformers up to 400 kv. In generating equipment, where the learn- ing period is the longest, Spain has produced hydroelectric generators of 255 mva. The Alcankara Storage Plant is to be equipped with four 200 MW generators, the largest hydro units in Europe. Brazil has produced 70 mva hydroelectric generators, and could build 150-200 .,'a units. India bas produced 40 mva units. in contrast, Argentina has built only small units and Mexico none. Only Spain has built thermal generating equipment; the industry there is presently preparing to produce units above 100 mva. 99. The pace of growth of capability is as im,ortant as the limits of capability actually reached. It took the Spanish industry about 8e."en years to move from hydraulic generators of 2.5 mva (1952) to 30 .va (1959) and another seven years to reach 255 mva (1966). (See Figure 1, p. '3). Very similar progress can be seen in Brazil. In 1958, the upper limit of generators built was 4.3 mva; it rose to 50 mva by 1965. Today, the capability is about 150 to 200 mva, though the industry is not as coapet:i.- tive for these large generators as it is for the ...ller types. 100. In discussing capabilities, the do.estic component should a180 be considered. In transformers, the import content is dete~ined .ore by raw . .terial availabilities than by the technical capability of the PIG U R!i, 1 ,.. .,t'dH\1 IlllIHI(\ I "".\'fJl \ EVOLt1l'ION OF SIZE OF VERTICAL GENERATORS PRODUCED IN SPAIN (Spanish Experience) 1952 1953 1956 1959 1961 1t62-63 1965-66 ~ UJ 6 ~ ~~ -A. _ _ _.., ~::A. -~ ,~~~ -'"' -:;1;''' ,- ~~. 4tl~r~j;t, ._ _ • •UCII -- c. t. I. t lif£lll _ _ 11111"* _.fWlIJI.11ICf1 _1I1 • ...u.I.. ~1 I. _mIClUHIIU ~t·-......c &.t Relative ....llness of world market. Compared to total sales of heavy electrical equipment, the market subject to competition is relatively amall. The French and the German markets are caapletely closed to outsiders; the United Kingdom .arket is virtually closed, even to partners in EnA. Historically, there have been few' imports of electrical equipment to the United States, due, in part, to domestic preferencea. 11 1/ The above statements may seem contradicto~ to the analysis in Chapter III showing that many industrial countri.. are both exporter. and taporters of heavy electrical equipment. Nevertheless, judging from our conversationa with European manufacturer., import. are largely (though not exclusively) of it... not .ade in the purch.. ing country. Some Ganan firms would not agree with the stat_eat that the,ir home market ia closed and adduced global statistic. to aupport their arg_ant. But if the trade in specialized patented i t _ is excluded, the import of power equipment by power utilities CaD be shown to be wery 8mall indeed. (See Annex Table 2.) - 46 - (b) Importance of tied aid. For equipment moving in international trade, there are two pf,ice level~. One, applying t:o "tied-aid" contracts, is aligned un domestic.. price levels in lthe: export- ing countries. On the. free world market, on the other hand, p~ices quoted are generally below home market prices in mAjor exporting countries. (c) Impact of excess capacity. Reference bas already been made \,to the overcapacity in the industry. But the prclblem of reasonable capacity utilization is, to some extent, present in the industry even in normal times. Because of the long production cycle and the large .hare of fixed ex- penses in the total cost, each international firm is anxious at times ito capture a particular order which would fill its plant cap',acity nicely and help keep its key staff and skilled labor occupied. On such occasions, a firm will quote a price desi~ed to secure the order and will forget about the normal profit margin. (d) It cannot be delaied that domestic prices are substantially above world market prices. This situati'on peraiat,) becauae on its home ground each national industry is protected from the full force of international competition. 104. ' When discussing world market prices we shall reserve this term for prices quoted in international competitive bidding. Since at any .am- I ent a developing country is notionally free (subject to foreign exchange availability) to procure equip~nt on the world market, this is the s~andard against which it mUst judge the economics of domestic production. It has been suggested that free world market prices are not a relevant standard because so much-heavy electrical equipment is bought by developing countries under tied aid at considerably higber prices. But first, tied aid could sometimes be switched to some other commodity, and .econdly, even 1f it could not, the higher cost of tied aid is best considered separately, as a price paid for financial resources which would not be fort~coming without it. lOS. There may be many reasons why a company quotes • certain price. 'The price quoted in the free competition would reflect the long-term busine•• interests of the company related to the work load of the plants and its anxiety to capture a particular job at a particular moment. Firms will often decide their prices on the basis of current coste only, and some may even accept losses in order to break into a new market. - 47 - There may also be some international or inter-company motives other than economic. Nevertheiess, an attempt to analyze prices is ~ade"here. 105a. Each large generator or transformer is a unique piece of equipment. A serious study of international prices therefore requires some method or formula by which price~ for different sizes of generators lIay ~1e reduced to a common reference basis. This can be achieved approximately by expressing all prices in $ per kva and .. .. .. 1---- ... 'j" .~ '" '·.. ·-1I..·_·'-- f f- - .. - .. f.- ......._+_.• ,..,. .• ,_.- ." .,. ..~ •• ",~, . ,._- .... "., . t' -I- .6 ..., . II' l ,1 ".~ LOGARITHMIC SCALES .4 L.....-_ _,...L....._-1-_-l..-_--L.._ L,,._~ ~-_'-. _ " , , _ , _ , . , _, ~ ••'~_'__ . ,. ".1. j [ LL ,L .2 .3.4.5.6.8 1.0 2.0 3.0 4.0 6.0 8.0 10.0 SIZE, MVA PER POLE KEY: 1 - PRICE LEVEL OF, FURNAS-BRAZIL, 1959 (# 10) ® BIDS PRIOR TO FURNAS (1959) 2 - PRICE LEVEL OF, CHAVANTES-BRAZIL, 1963 (#24) X BIDS FROM FURNAS UP TO CHAVANT£S(l963) 3-PRICE LEVEL OF, RIO NARE-COLOMBIA, 1965 (#40) [!] BIDS FROM CHAVANTES UP TO RIO NARE(1965) • BIDS FROM RIO NARE ONWARD NOTE: The underlying data are shown in Annex B (Figure B- 6) SOURCE: Courtesy of an international manufacturer. (3R) IBRD- 3455 -49- - 50 - Legend Successful Bids - Data Underlying Figure 2 MVALPole .vJS~lKVA I 1. Almus (Siemens) 1959 10.66 10.75 2. Sefid Roud (Jeumont) , 0.98 7.25 3. Caroni 5/6 (AEG)-60 C.p.s. 1.11 9.92 4. Matahina (E.E.) , 1.21 9.92 5. Rio Macho {Shibaura) 60 C.p.s. 1.17 9.25 6. Mareta (E.E.) 1.40 8.12 7. Yanhee (AEC) 1.82 7.25 8. Salto II (AEG) C.p.s. 2.25 7.50 9. Calima (Mitsui) 60 C.p.s. 2.40 4.75 10. Furnas (Siemens) 60 C.p .,S. 1959 3.45 5.00 11. Poatina (Siemens) 5.42 6.39 12. Murray (ASEA) 8.60 2.24 13. Cundinamarca (Oerlikon) 60 C.p.s. 0.50 15.00 14. Brokopondo (Siemens) 60 C.p.s. 1.10 8.50 15. Volta River (IGE) 2.75 5.66 16. Rape1 (Hitachi) 2.35 6.25 17. Konya II (AEG) 5.80 4.60 18. Angat River (ASEA) 60 C.p.s. 2.25 8.65 19. Furnas (Exp.) Siemens 60 C.p.s. 3.35 6.25 20. Mang1a Dam (Hitachi) 4.00 4.20 21. E1 Co1egio (ACEC) 60 C.p.s. 5.23 4.15 22. E1 Co1egio (ASEA) horz. 60 C.p.s. 5.30 4.75 23. Maria Christina (Toshiba) 60 C.p.s. 2.60 6.50 24. Chavantes (Toshiba - ASEA) 60 C.p.s. 2.05 4.62 25. Paulo Alfonso (ASEA) 60 C.p.s. 2.55 4.14 26. Cachi (Toshiba) 60 C.p.s. 3.35 4.56 27. Nam Pong (AEG) 0.345 20.00 28. Murray III (ASEA) 8.00 3.05 29. Guri (AEG) 60 C.p.s. 3.80 4.55 30. Guri (Hitachi) 60 C.p.s. 4.40 3.75 31 .. GU1"i.,\(Westinghouse) 60 C.p",s. 3.70 3.65 32. Awash' II and III (Elin) 33. Manapouri Siemens 4.30 3.75 34. Corani (Siemens) 1.65 7.45 35. Cumbaya (AEG) 60 C.p.s. 0.91 6.75 36. Kovada II (AEG) 2.60 4.12 37. B10wering (AEG) 2.35 4.10 38. Yanhee III - IV (AEG) 2.22 4.40 39. Yanhee V (AEG) 2.22 5.01 40. !!!! Nare (AEG) 1965 6.50 5.50 · 41. Aviemere (Siemens) 1.10 5.50 - 51 - Prices in Developing Co~tries 110. There are technical difficulties in comparing prices for heavy eiectrical equipment in different countries. First, the equipment is not identical: varying technological envirODllent, indu.trial capabilit:les, domestic content rules, and purchaser's.preferences .ay cauee substantial differen~es in design .nd in ,~erformance (e.g. 1n reliability, .aintenance requirements, and life expectancies). Secondly, the delivery time and, perhaps most important, the financial arrangements will have a bearing upon the pr1.ce. Ill. Power companies try to compare bids in d:i.f ferent ways. Thus , payment plans may be made comparable by reducing tile total prices to pres- ent values. Differences in power losses for transformers may be capitalized in a sillilar manner. Nevertheless, come element of guesswork or j udpent is inevitable for certain estimates, e.g. the future loading pattern, dura- tion of aervice, discount rate, etc.). Our findings on comparative prices can only be crude approximations, though it is our hope that through crosa- checks with manufacturers participating in the study, we have avoided gross errors of fact or judgment. 112. We a180 encountered two major analytical difficulties: (a) In some countries, the exchange rates in effect at the time of our comparison were misleading because import duties and other import restrictions were used not only for infant industry protection but also as • aubstitute for devaluation. Though this aspect was present to some extent in several of the countries studied, it was a major factor in Argentina and required the adjustment described in Annex B. 11 Some adjustments also bad to be . .de in the case of Brazil. (b) It vas difficult to establish what point of the "learning curve" each country had reached. This is diacuased in Chapter VII. The implications of, say, a 20 percent price difference are very different for an industry cl.arly in an infant stage, and for one which has had time to acquire the necessary production experience. Differencea 1:.1 Our adjustment is only a working hypothesill subject to critique and refinement. Yet ve found it the moat meaningful method available for ca.paring Arlentin. costs and prices with those of other countries. - 52 - in dOme8tic content and in the sizes and capacities of equipment should be considered when comparisons among developing countries are made. 113. The following table 8UDIDarizes the available price information, which is given in more detail in Annex A. It includes prices for some light equipment where these are more representative of the present capability and production experience of the count~J: TABLE 7: Price Premia for El~ctrical Equipment in Developing Countries Compared '~ith WOrld Market Prices, 1964 (2ercent) Spain Brazil Mexic~ Argentina India Pakistan Heavy Equi21lent Generators 24 30-60 50 Trans formers 40 MVA anel above 50 40 (80) 40 75 Motors, 3-phase, 100-200 hp 65 High-voltage accessories 70 Transformers, 5 MVA 10 (40) I Light Equi2ment Transfomers, 200- 1,500 KVA 25 45 36 44 Motors, 2-20 hp 95 78 77 Switchgear, up to 35 kv 15 LA Figures in parentheses indicate the approximate price differential in Mexico four years earlier, in 1960. Source: See Annex A. 114. The Mexican experience in transformers, described in Annex A. is particularly interesting. Annex Figure 1 ahow8 that Mexican prices for 5 to 15 mva transformers fell from 1.4 to 1.6 time. the lowest interna- tional bid in 1960 to only about 1.0 to 1.2 times that price in 1964. (In some tenders, Mexican bids were below the international price.) Increased competitiveness may have been due, in part, to a lowering of the Mexican profit margin (present prices do not fully cover costs), but the m~jor factor was probably increased production experience. Brazil show8 a similar - 53 ..;. trend. In bids for 20 mva and 112 mva transformers for the Jupia power plant (1962), Brazilian prices were 100 and 120 percent, respectively, I . above the lowest foreign bid. In the Ilost recent tenders, on the other hand, Brazilian prices have been only )0 to 40 percent t above the comparable • c.i.f. price. The Brazilian prices might be profiLtable if profit vere based upon marginal costs; on an average basis~ however, to earn a reasonable I return on the investment, Brazilian transformer manufacturers today would probably need prices 50 percent above the world market level. 31 115. Generator prices are very much influenced by the size of the domestic component. For two 50 mva generators for the Chav~ntes power plant in 1963, Brazilian offers based upon,' a 15 percent domeistic content were only about nine percent above the international price. Yet, based upon a 90 percent domestic content, Brazilian offer. were more than twice the lowest foreign bid. Again, with a 33 percent domestic content, Brazil- ian offers for two 50 mva generators far the Peixoto power plant (.:arly 1965) were virtually identical with the lowest United States tied-aid bid (say, 2~ to 30 ~ercen~ above·the world market price). In a more recent i tender, 'Brazilian prices for one 50 mva generator were 56 percent above the lowest foreign bid; the price increase apparently refle~ting an effort to raise the domestic content to 40 percent or higher. The Brazilian industry . has not yet reached the competitiveness of the Spanish industry. . With a 45 to 50 percent domestic component, Spanish prices for 100 .va generators, according to two different sources, are only about 19 to 24 percent higher than domestic prices in France or Austria. !I 11 The reader is warned that the price differentials qUOted above are not to be taken as indicationa of a country's comparative advantage in pro- ducing heavy el.ctrical equipment. In particular, they may in some c.... ,be inflated by high prices for major material. like silicon sh.et. or ...i-.anufactures of copper. The country may have a well e.tabli.hed and efficiently run heavy electrical equipment induatry but its s.i- . .nufacture indU8~ry (like steel) may )lave so.e deficiencies. Thus doaestic steel prices, which are a major input of heavy electrical equipaent, ..y be auch higher than international price•• !I The cOlipariaon would be sOIlewhat lea8 favorable if . .de with the world market price c.i.f. Spain, aince the difference between French dc:.e.tic pric.. and the lower French price. for export i. greater than the coat of .hipping generators fro. France to S••in. MEXICAN TRANSFORMER PRICES COMPARED WITH THE LOWEST INTERNATIONAL BID, 1957 -1964 (U.S. $ PER KVA) 9.0 I I I -y- I 8.0 9 1962 : 7.0 · · o MEXICAN PRICE X LOWEST INTERNATIONAL BID ! 6.0 · · i 0 1960 I ~ 1957 5.0 1961 I 0 i l957 ~ I ii I 4.0 ,1964 · · · I :. ; I ~ a: i I It I LLJ 0- '964 I I ::!! en a: 3.0 i : ; .. 0 ; 1957 li l960 ; C)1960 G'l C ::u ITI *1 L64 * • I I Ij~i . . 2.0 196411 :1964 * ;i ~ I ~ * 0 1962 1.5 11 1.0 o SEMI-LOGARITHMIC SCALE I 5 I 10 I 15 --~---~---~--------- 20 25 -- 30 -~------- ~____ J 35 40 45 I X 50 MVA l2R) 18RD-3478 - 55 - l 116. There is aD intermediate range of equipment for which the competi- tive disadvantage of manufacturers in the developing countries, at least for aOlle items, may be less than that for light . .ss-produced articles I or very heavy equipment. These items are produced in relatively small numbers (often to individual specifications) and do not require very powerful manufacturing facilities or very advanced techniques. This intermediate range the borders of which are not easily defined, includes medium-sized transformers (s.y, 5 to 20 mva) , and switchgear. Thus, Mexican prices for 10 to 12.5 ava power transformers are very close t~ import prices, excluding duty. Argentine prices for switchgear, I(other tha~ ~'ery heavy switchgear), are only about 15 percent higher than French domestic prices, while Spanish and French domestic prices are virtually identical. 117. On the other hand, the prices of Mexiean-made three-phase 75 to 200 hp motoT.s range between 23 and 55 perc£Dt above those of iaports from the United States, the average being perhaps on the order of 40 percent. (See Annex Table 16.) As a rough estimate, the premium ~ver imports from Europe might be of the order of 65 percent. A similar situation exists in high-voltage accessories where Mexican prices range from 45 tOr 90 per- cent above the comparable import price. Though neither of the.e it... are ~ss-produced in Western Europe, manufacturers there probably achieve aavings by making small production runs for inventory. 118. Very tentatively, we may summarize the price situation for heavy electrical equipment in developing countrie8 as follows: (a) The fir8t generator or transformer manufactured of a size beyond the previous production experience will be expensive: probably' the cost will be 50 percent or more above imports from Western Europe or Japan • .,. (b) As more experience is gained, this premium may be re- duced to, say, 20 to 25 percent - the case of Spain. At this stage, competitiveness will depend. very much upon the size of the domeatic content (beyond 50 percent, for instance, prices for generators arelikely to riae rapidly) and the increase in prices for major materials and components due to protection. 11 '2} In international tender8, electrical equipment manufacturers often submit two or more bids corresponding to different.as8umptions about the domestic content. - t 56 - (c) Developing countries will be most competitive on equipment of intermediate size made to special design when the same equipment i8 not produced in series in the industrialized countries. f 119. Though beyond the 8:0pe of the pre~ent report, the queetion is pertinent whether, comparatively spe~king, develop~Dg countries are in a better position to produce light electrical equi:pment (e.g. light motors ox distribution trapsformers) than heavy equipment. Table 7 sugge.ts that this would not no~lly be the case (see again Annex,A). The main reason probably lies in the limited size of the market, particularly where a small '! I market i8 shared by several manufacturers. Where the industry is heavily protected , and the domestic content high, manuf.cture of light equipment, like motors or distribution transformers, is 11kely to become very uneco- nomical. - 57 - VI COSTS AND COMPETITIVENESS ConceRts and Methodolosy 120. In this and the following chapters we shall compare the costs of producing heavy electrical equipment in developing countries with those in industrial countriese Ideally, we would like to attribute observed cost differences to four major variables: - factor costs (for materials, manpower, and capital) - market size - ti~e and cost of learning process - institutional framework. 121. The first three factors taken together determin-: & country's intrinsic comparative advantage. The fourth factor, the institutional fram'!work, is used as a shorthand phrase for all those condj:tions, includ- ing government policies, which determine whether actual costs exceed costs determined by the country's comparative advantage. For example, govern- ment policies might be too protectionist, management-labor relations may be poor, the devel~pment of labor and p'rof~ssional skill~ may have been neglected. 122. In practice, it is extremely difficult to separate the i~act of these different factors. Factor costs depend not only upon the prices and productivity of different facto~s but ~lso up~n their .ubatitutability , following changes in technology and the organization of production. The cost disadvantage associated with a amall market for, say, hydroelectric generators, might be reduced if the same manpower and the same facilities could be used for production of other items. The length of the learning process will vary with the regularity of orders, which is closely linked with the size of the market but also determined by the institutional frame- work e.g. procurement policies. 123. Faced with these difficulties, we chose the following procedure: - 58 - (a) In this chapter we compare actual costs, 1.1 and their majclr components (materials, manpower, eapttal); (b) In the next chapter, we discuss the extent to which actual costs have been influenced by market size, the ~earning process and institutional factors. 124. Even the comparison of actual costs is not a simple matter. For analytical purposes, we would like to break down each major cost element (materials, labor, capital) into.quantitie~ used per unit of , production I and unit prices. But we are faced with thu dilemma that different plants have a different product-mix and, for the same final product, do not perform exactly the same work - some are more fully integrated and make certain components which others purchase from the outside. These difficulties we have sought to overcome as follows: (a) In principle, we have taken as our frame of reference the cost structure of firms in Western ~Iurope that are representative of the heavy electrical equipment industry in terms of an individual product or of a firm's total output of heavy electrical equipment. 11 We have then estimated how much the value of each major cost element would deviate in the developing countries from the West- ern European standard. (b) This comparison is relatively simple for material costs where differences s.re determined mainly (though not exclusively) by price variations for major materials. On manpower, considerable information is available on relative wages and salaries. Reasonable assumptions regarding the returns on capital in industrial and de- veloping countries can also be made. (c) The real difficulty arises in estimating labor and capital inputs per unit of output. Though we have some 11 . . As distinct from shadow prices. But we make allowances for inflation (see Annex B). Costs of inputs are taken 8S the manufacturer pays them, and the question of the indirect tax content is not discussed. In a few cases costs will therefore also include import duty on components. 11 Firms in developing countries seem to be closer to European firms than to American companies, in terms of size, market, and more impor- tantly, product diversification. - 59 - information on individual products for the countries studied, their comparability is suspect and their interpretation exce.r!dingly difficult. We have there- fore relied mainly on a different approach. Published profit and loss statements for firms engaged in similar produdtion in Western Europe and in the ~eveloping coun- tries have been used to calculate comparative manpower and capital costs per US $ million of sales valued at constant (Wet.tem Europe) prices. Tl1is ~ethod givE~s significant results to the extent that the follow- ing conditions are fulfilled: (i) the product composition of total sales is 'reasonably similar; (ii) the degree of vertical integration is reasonably similar; (iii) the conversion of , the prices of a given developing country into Western Europe prices is reasonably correct. (iv) capital costs are properly reported. 125. It is not necessary that all the above conditions be perfectly fulfilled. For instance, when the product composition or the degree of integration differ, it may be possible to make a rough adjustment for such differences. Moreover, our final estimates have been checked for consis- tency with some cos't comparisons made by international manufacturers between their home and overseas plants. Finally, we have asked the manufacturers participating in our study to check our estimates against their own experi- ences and such internal cost comparisons as they may have. Even so, in t~e end, our findings cannot be precise but only broadly suggestive of existing costs differences. And one leading United Kingdom manufacturer has in fact found it impossible to make meaningful comparisons between different plants in the United Kingdom, let alone comparisons with overseas subsidiaries: throughputs would not be compar'able, production methods would never be the same, and the proportion oe materials and components bought outside would always differ. These are telling points; nevertheless, other manufacturers have tried to make comparisons. - 60 - 126 .I~ A special study grol.lp, International Committee for Economic Compari- sons in the Electrical Industry 3/ in the European Common Market, has come - I across similar difficulties while trying to arrive at cost cCII'parisons within the common market countries. The differences in accounting systems . I seemed to be the major factor affecting the comparisons even among industrial countries. The Committee therefore, made a special effort to put these on a similar basis - see "Guidelines for Calculation of. P~ociaction Cost in the European Electrical Industry". Students of American industry report the same kind of difficulties and would agree that the coot data aVI,ilable hardly allow a meaningful comparison. Our approach is different in nature from .letailed product comparifJons between individual plants. We start out with a measurable difference between domestic prices and world market I prices, and then ask ourselves three questi()ns: are material costs higher, r are labor costs higher, are charges against capital higher? . Materials 127. Materials, including processed and semi-finished components, weigh heavily in the cost structure of heavy electrical equipment. In industrialized countries, they generally account for 50 percent or more of the total selling price. The main materials in motors and transformers are copper semi-manufactures and silicon st~el ,sheets; in generators large steel castings. are also important. Other important items are insulation materials, porcelain bushings and transformer oil. (See Annex Tables 4 to 7.) 128. All these ma~erials (except. perhaps castings)' are normally freely available on the world market. Domestic production therefore is not neces- sary. In fact, heavily protected domestic production of many of these inputs has sometimes been a major commercial disadvantage to manufacturers in developing countries. They are not always in a position to charge prices for their end products which (a) cover the higher cost of domestic materials and/or import duties and (b) provide some protection for their processing margin in the learning stage. It 3/ International~r Arbeitskreis fur Betriebswirtschaftliche Vergleiche in der Elektrotechniscb£n Industrie, Betriebavergleiche von Kosten und Ka1kulationen, ZVEI, Frankfurt/Main, 1963. - 61 - 129. A comparison of raw material prices for heavy electrical equip- . , ment in developing and industrial countries is given in Tabl.: 8 (p. 62). Wire in various forms (blank, enamelled, iIisulated, etc.) is the dominant copper item for the electrical industry, but other copper products are also used extensively in heavy electrical equipment. The countries studied generally import wire b~rs which are thf~n turned into semi-manufactures. At present, most copper and brass semi-manufactures cost about 20 to 40 perce,ltt thore in the three Lat'ln American countries studied than the lowest f.o.b. price in the international market. In Pakistan and India, this difference rises to 75 and 100 percent, respectively. , The pl':ice for drawiug copper into enamel wire represents a mark-up of 20 to 30 percent on the cost of bars in Western Europe but of no less than 100 percet~t in Latin America and Spain. 130. Due to sharp fluctuations in copper prices and a long-tenn widen- ing of the price differential between copper and aluminum, the latter metal has been increasingly substituted for copper in industrialized countries, e.g. in small motors. Such substitution has been lagging in developing countries. 131. The situation for the other main raw material, steel, is interest- ing. The Brazilian steel industry is almost competitive in ordinary steel sheets. In Mexico, Spain, and Argentina, on the other hand,' prices for ordinary sheets are about one-third above the world market level and roughly reflect the present needs for infant industry protection of the domestic steel industries. Spain hopes to be close to ~nternational competitiveness in steel when present expansion and modernization plan.s have been completed by about 1970. The Indian price level xeflect~ the black market prices which are paid for marginal supplies. At the present moment and perhaps for quite some time to come, steel can be imported to anyone of the six countries studied at prices, including oce~n freight, very similar to those paid in Western Europe and considerably below the prices in tbe United States. This is because, under present wO,rld steel surplus conditions, major steel producers charge considerably lower prices on export sales than on.domestic sales. 'J TABLE 8: Prices of Principal Materials, 1966 (lmiest :r .o.b. price in international trade = loot .- Prices in the • Majority of. Advanced Coun- tries a Argentina Brazil Mexico Spain India Paldstan b Copper and copper products 104 123 b 119 135 c 128 252 156d - Wire bars Round coppere 102 185 136 193 143 188 173 p:lat copper 102 n.a. 137 140 133 232 174 f Enamelled copper let> 216 159 120 127 194 139 Brass 102 184 n.a. 159 123 168 n.a. I Q'\ rv steel and steel products Silicon'stee1 (dynamograde - 3.6 w/kg. - 0.5 mm) liS n.a. 199 196 130 187 167 (Trans'former grade grain oriented g h i j 0.6 W/kg. 0.35 mm) 108 152 154 . 112 105 n.a. 123 Steel sheets 106 n.a. ~ 130 136 1:22. 150 Hardware 100 n.a.. 140 n.a. 135 125 300 Cast iron 106 88 150 94 n.a. 69 kl 181M n.a. Insulation material 112 110 ' 178 190 200 Varnish 105 n.a. n.a. 270 n.a. 168 n.a. n Transformer oil 114 n.a. 98 n.a. 165 19U- 209 Footnotes on next page. - 63 - Note: The international price (-100) represents the iowest f.o.b. price in the international market, (information made available by courtesy of inter- national manufacturers.) a. Even industrial countries do not always buy materials at the lowest prices in the international market. Their home industries produce at. higher p~ices o~ even ~f they import from the lowest price sources they have to pay ocean freight, handling and import charges. b. Price comparisons for copper, as far as possible, relate to the situation after the copper crisis. Due to great fluctuations in copper prices, particularly in 1965, the time of purchase has a bearing on the price paid. The price movements were especially harmful to the electrical equip- ment industries in developing countries. Being in very early stages of production and not large consumers, they had no links with producers and had to purchase copper at the high London }oreta1 Exchange price. In Argentina. the unit value of imported unwrought copper (electrolytic copper ingots and slabs) ran fairly close to the LME price -- the 1964 unit value was the equivalent of 40 US cents per pound. In additionl the unit value (If copper wire imported into Argentina exceeded the unit value of imported unwrought copper by a substantial margin. In November 1965, while industry in ad- vanced countries was paying U8$851 a ton for wire bars, Argentine industry had to pay U8$l,430 a ton f.o.b. (US$l,576 landed), or 185 percent of the inter- national price. Brazil tmported from Chile at the LME price. The Government and manufacturers tried to establish agreements based on the producers' price but they were not successful. The Indian unit import value of unwrought copper ran fairly close to the producers' price in Europe from January to August 1964, but exceeded it by a widening margin thereafter -- by the eqpivalent o~ about 5 US cents per pound from September to NovL~ber 1964, 11.5 US ce~ts per pound in December 1964 and 12 to 17 US cents per pound in 1965. Even at its highest, this was well below the LME price, but the price of domestic copper was very high, so the average cost to the industry was considerable. The Spanish wholesale price of unwrought copper, like the Argentine unit import value, ran with the LME price, exceeding it by the equivalent of some 10 to 12 US cents per pound. At present (1969), since the Zambian Government levies taxation on copper at'the LHE price, the producers have been constrained to retreat from the double pricing system. c. Copper is produced In Mexico. d. High prices for copper in Pii~kistan were due mainly to small lot size purchases and tied aid. e. This t'efers to insulated blank soft copper wire of 8mall diameter. The thickness of wires and also its treatment considerably affect price differences. f. If allowance is made for the low quality of Indian enamelled wire, the price difference would be substantially greater. - 64 - ! g_ . The price difference isimainly due to charges levied on imports. The composition of price. is as follows (January 1965) -- fre1.ght 3.6 percent, import charges , 49 percent. h. Silicon steel has been imported at international prices. The price of silicon steel imported from the USA was $535 per ton whereas from Europe it may be tmported at $409 per ton. lmport chirgea on silicon steel total around 55 percent. i. 1 The Mexican price includes ocean freight and handling (5.7 percent of f.o.b8 value) plus charges and expenses related to import (5.8 percent). j. Based upon a comparison with the German industry for 1963 and 1964. k. In many developing countries the procurement of insulation material ptesdnts'serious difficulties mainly because of poor quality of domestic production and high protection. As an example of the protection granted to domestic suppliers a summary of import expenses for Brazil is given below (percent of f.a.s. value): License and other expenses 6.7% Freight 10.7% Insurance 0.9% Banking expenses 6.1% Financial expenses 3.7% Customs duty 70.4% Importation expenses 9.4% Total 207.9% 1. One of the most important items causing serious difficulties in developing countries is porcelain bushings. The quality of locally manu- . factured porcelain for voltages higher than 33 and 69 kv has been poor in practically all countries. The irregular business of the equipment industry does not allow a continuous flow of porcelain bushing imports; therefore there is an important ttme element involved in the procurement of porcelain bushings for higher voltages. Because of differences in the type of por- celain bushings and their prices, it has been extremely difficult to make an international comparison. One example, however, may serve as an illus- tration: Porcelain bushings (SPM - 3241 - 04200 A DIW 42531) Germany 100 Brazil 235 The Brazilian price is made up as follows: (cost of import f.a.s. - 100) .... 65 - License charges and others 15.0 Freight 5.6 Insul:ance 1.1 Banking expenses 6.7 Customs duty 97.1 Importation expenses 8.9 Total 235.3 m. Kraft paper in thickness from .0012" is locally available. Elec- trical (neutral) grades are made by several paper mills up to laO" ill width or slightly more. There are no quality problems. Nevertheless, these papers are not made for inventory stock and it is relatively difficult to obtain quantiti_s of less than 10 metric tons per order. Some simple glass fabrics are locafly made using imparted glass yarn. The maximum width is 6". Cost is competitive with imported equivalents. Glass cloth larger thaI) 6" is imported. n. The reason why transformer oil is excessively high in Spain may be monopolistic pricing. - 66 - 132. More important for the industry are the silicon steel sheets of dynamo grade. They are produced in Brazll,'Spain, Mexico and India. The price handicap in Spain is about the same as that for ordinary steel . sheets. In contrast, prices in Brazil, Mexico and India are almost twice the world market level, , reflecting a high level , of protection. Trans former- grade, grain-oriented sheets are not produc!ed in the countries studied and with few exceptions their prices are close to the world market level. The high' price :In Pakistan may be partly the result of tied-aid procurement. The high 'Brazilian price is explained by a tariff duty, apparently imposed in the hope of inducing domestic production. 133. Above we have dealt with prices only. We must also consider the quantities of major material~ used. The raw material recluirements per unit of output in developing countries are larger because of (a) higher I • rejectio,~and wCJrking scrap ratios and (b) obsolete design. Manpower 134. Manpower costs will depend upon (a) the cost of the average man- year and· (b) the manpower requirements per unit of output. Wages and salaries .. re, of course, much lower in developing countries than in industrial coun- tries, as shown by the following figures for heavy electrical equipment plants: TABLE 9: Compatison of Wages and Salaries, including including all Supplementary Benefits, 1965~. ---- Absolute Figures Ratio Wages Salaries Indices ~ Salaries to Country US$/hour US$/month Wages Salaries Wages Ic (Averages) Germany 1.25 305 100 100 1.39 Spain 0.51 213 41 70 2.38 Brazil 0.40 189 32 62 2.70 Japan 0.39 125 31 41 1.84 Ind:!.a 0.25 94 20 31 2.15 Pakistan 0.20 116 16 38 3.31 L~ Also see Annex Table 10 for 1965 data on some countries. ~ The figures are averages for a representative group of jobs, both at the staff and at the workman level. Hence, they are not - 67 - infl~~nced by differences in the compositibn of the workforce. Fedetal Geluany has been chosen to represent industrial countries. ~ Assuming an average 175 hours per month for workmen in all the countries. Sourc~: Data submitted by international manufacturer. 135. Ac~ording to the table, wages in West Germany are five to six , I times those in India or Pakistan and about !~hree times those in Japan or Brazil. Differences in salaries are considerably less. German salaries ~ a~e 2.5 to 3.0 times those in India, Pakistan, or Japan, but only 1.5 to 1.6 times salaries in Spain or Brazil. The relatively low ratios in developing i countries reflect primarily the relative 9\1Iil~plus of workers. . To a lesser extent, they reflect the scarcity of highly skilled staff and of high level , : I acministrators who can find employme~t anywhere and have an international market price. Where foreign personnel are employed, these would normally receive higher pay in developing countries than in their homE: countries. 136. , Even allowing , for the smaller gap in salaries, the sizeable dif- ference in pay scales would seem to give developing countries a competitive advantage in the labor component of cost. In industrial countries, this component accounts for up to 40 percent of the total selling price. Large equipment is virtually handmade, and after the initial learning period operatin~ times for a given production machine are said to be the same in a country like Brazil as in, say, Sweden. Nevertheless, the extent to which.this potential advantage can become a real advantage will depend upon the organization of production, the scale of output, the utilization of plant,capacity, etc. 137. Comparisons of manpower productivity are extremely difficulto The average composition of the output is not the same. The proportion of work done outside the heavy electrical equipment industry varies. Fi- nally, different manufacturing steps and processes will call for different skills and different· investments per worker. ~I !I Labor productivity also influences material and overhead costs. Low productivity may express itself in more rejects,' repairs, or other supplementary work. One manufacturer in Brazil recalled that none of the transformers built in the first three years of operation passed the work tests. Many additional manhours were required to improve the equipment already built. - 68 - 138. Neverthe~ess, there is enough evidence to suggest that differences in productivity arE: very large: (1) Direct labo~. Annex Table 9 indicates that direct matt-hour requirements for a range of heavy items were two to three times higher in Mexico and Brazil than in industri~;; coun- tries. For small motors (light equipment), the di:fference was even greater; because of the larger scale of o'utput in industrial countries, production can be more mechanized and automated. (2) Indirect labor. The proportion of indirect labor tends to be high in developing countries. f Th1.s category includes, in particular, production of jige, and fixtures, plant main- tenance, etc" (3) Staff. ~i:'he staff-to-worker ratie/s in the SIAM plant in Argentina, the General Electric plant in Br.;lzil or the , IEM plant in Mexico are not too different from, say, those at the ASEA works at Ludvika or the Alsthom works at Belfort - about 0.55. The ratio of 0.62 for the Brown- . Boveri plant in Brazil is somewhat high~r; more design work is probably done in this plant. Tbe similarity of the ratios suggests that relatively low productivity is characteristic of staff as well as of workers. One ex- plana:tion is' the need for close supervision in developing countries. 2/ 139. The following figures for manpower costs per'US $ million of sales at constant prices clearly suggest that low wage and salary scales in developing countries were more than offset by lower productivities. ~I ~I For light equipment involving a high proportion 6f standard items, the ratio of staff to workers can be drastically reduced, particularly if no special design capacity is needed and the sales work is routine. For this type of equipment ratios of 0.25 or lower are quite typical. One c~pany ~.n Pakistan ,roducing small motors and transformers in 1958-1960 had a ratio of staff to workers of about 0.08 which was re- duced to about 0.05 by 1965. This company lives on purchased or copied technology. When a company develops its own design potential, the ratio is bound to increase. fl./ The exception is again Pakistan. As indicated earlier, the Pakistan industry cannot be compared to the heavy industry of Western Europe in the same way as can that of, say, Brazil or Spain. Pakistan pro- duces mainly light and medium transformers and motors with a typically small labor component and a high bought-out component. In this type of production, relatively little supervisory labor is need_d. - 69 - Figures are also shown for em~loyment per $ million of sales. 11 Since i the p.roduct composition and t:,le degree of out-of-plant work vary, these are not exact measures of productivity; nevertheless we believe that the general picture is fairly conveyed. TABLE 10: Employment and Manpower Cost against Sales. Heavy Electrical Equipment Industry, Seven Countries Employment per Ra tio of Manpower $ million of sales cost to 'sales At actual At W. Europe At Actual At W" Europe Index prices prices LA prices prices W. Europe (000 manhours) - 100 United States 52 .323 France 92 92 .268 .268 100 Spain 159 176 .347 .396 148 Brazil 144 170 .286 .337 126 Mexico 179 228 .302 .438 163 Argentina 342 440 . 503 .645 240 Pakistan 179 211 .170 .200 90 Source: Calcula'ted on the basis of information available in the Annual Reports of various companies. 1i i.e'. the manhours required per .$million if pt'~ces had been, not at inflated home price levels, but at levels in Che Common Market (re- presented by France). The price indices used were as follows: Spain (114), Brazil (118), Pakistan (118), Argentina (129), Mexico (133) • (See Annex Table 11.) Prices in the Commo.t'& Market were chosen rather than world prices in order to give a fairer comparison. Charges Against Capital 140. We shall define charges against capital (or the gross return on gross assets) 11 as the sum needed to meet deprectation, interest, income and property taxes, and profits. The incidence of these costs will depend upon the ratio of assets to output and (b) upon the minimum acceptable gross return on mssets. As a preliminary working hypothesis, we have 11 I Th~ figures on which these comparisons were based are shown in Annex Table 10 where they are presented in a somewhat different form. !I Defined 8S gross fixed assets ,plus gross current assets. - 70 - assumed that a 15 percent gross return on assets would be acceptable in industrial countries while a 20 percent return would be needed in developing countries. 141. We have not been able to collect meaningful comparative data o~ capital costs for specific items of heavy electrical equipment. The difficulties are readily understood when one considers that most. heavy items are made to order, and that the average size and type of e,quipment demanded by the Brazilian electrical power industry differs greatly both from the equipment used in Germany and that used in Mexico. 142. We made considerable efforts to compare capital costs in two firms in Spain and Brazil with those in more fully developed countries. We found. that the ratios of gross assets to sales for the Spanish and the Brazilian firms were about 1.40, whereas those in northern Europe varied from 1.40 (an Austrian firm) to 0.80 (a German and a French firm). However, the difficulties of valuing assets, and incomparability between firms, made this an unreliable measure. Even without relying on a difference in the assets to sales ratios, however, it can be said that ~he cost of capital for producing a given item of equipment in a developing country is higher a~ because the price of the equipment is higher, and so a similar rate of return will indicate a higher capital cost, b) because the rate of return expected in developing countries is also higher. ·In Table 11 below, we have assumed the real cost is double -- though the comparison in that table does not depend on this assumption (see' note 9). 1430 Two important operating factors also serve to make capital charges higher in developing countries. The first is heavy initial operating losses which add greatly to the cost of the investment. The second is a low rate of turnover of ~apital in developing countries compared with, the speed with which really efficient producers in industrial countries push a large volume of output through their plants. This also covers the large inventory carrying cost which mainly arises because of fear of the discontinuity of supply of imported parts and materials. This latter factor is related both to the skills of production management and to the characteristics of the market and the quality of the sales effort which determine whether an even and suitable flow of orders can be secured. . ..'. - 71 - t Combined Impact of Cost D:lfferen,;es 144. Having considered the three main elements of cost and their variation between industrial and developing countries, we must look into the total 1 " impact of observed cost differences. Ideally, ~e should make this analysis separately for ~he main classes of equipment: generators, transformers, s~itchgear, Since this is impossible for reasons already large motors, etc. f ' indicated, we shall start with the eost stlucture in an induutrial country for its entire actual output of heavy electrical equipment and ask ourselves I . , what this cost structure would look like in a developing country based upon our previous analysis of differences in factor costs. If that analysis is at all correct, there shou'id be at least a rough order of cOIl:parability between the index of comparative cost arrived at in this manner and the co~parative prices reviewed in Chapter V. 145. This comparison is shown below: TABLE 11: Cost Structure in the Manufacture of Heavy Electrical Equipment Index of Comparative Cost Costs LA Developing Countries Cost JA WeEurop~ Spain Brazil Spain Brazil Material,!b 51.5 128 133 65.9 68.7 Manpower-- 35.5 148 126 52.5 44.6 Charges ~ins t capita d. 13.0 200 200 26.0 26.0 100.0 144.4 139.3 Processing Margi~ 48.5 78.5 70.6 LA Cost to international manufacturers (in Germany) • 100. /b Derived from Table 8. ~ See Table-10, last column. :L! Assumed to be double the cost in W. Europe in real terma. ,h! Margin between selling price and materials coate. I • 146. According to the above table, the present productiQn cost disad- vantage of Spain and Brazil is of the order of 40 to 45 per~ent. This difference. is greater for Brazil than the difference shown in'Annex Table 11. ~part from er~rs involved in our estimating procedures, one-likely explanation - 72 - is that neither the Brazilian nor the Spanish firm earned anything like the 20 percent gross. return on their assets which we have postulated. The paucity of information in offici~l financiai statements precludes confirmation of this hypothesis. ~ I i ll,7. The above discusfiion has been concerned excl'llsively with comparative average costs, making no allowance for the fact that world market prices I I are based upon marginal costs. According to Annex Tablell~ world market prices c.i.f. Brazil or Spain would probably be about 10 'percent below domestic prices' in Western Europe (France). It follows' that a pric.~ allowing a normal profit margin to producers in Brazil or Spain would be nearly 50 percent higher than the equivalent import price. Measures of Competitiven!ss 148. In the following paragraphs the degree of protection implied in price differences is analyzed by a uniform method under different con- cepts. In order to avoid lengthy explanations the mechanics of the calcula- tion are explained on the basis ~f three d~fferent products used as ex- amples (Tables 12, 13 and 14). All costs and prices are from the actual operations of the equipment manufacturers. The concepts under which we present results can.be summarized as follows: Simple Price Difference -- the ratio of domestic price of the product to the c.i.f. value of a comparable import. This measure reflects gross protection in the price of final product. Efffective Protection" -- the percentage excess of domestic value added OveT value added in a free trade situation, including freight etc. ~ 11 A simple arithmetical example will illustrate the point, however. If actual gross profits in relation to assets were 10 percent only instead of 20 percent, this would reduce the price disadvantage from. say, 40 to 25 to 30 percent. ~ See Table 12 for definition of value added in this context. Protection I is defined as anything which makes it possible for the domestic pro- ducer to charge a price higher than the c.i.f. price of an eguivalent import. - 13 t t. TABLE 12: ~"1easurement. of Competitiveness, Example 1; Costs and Prices of Manufacture of a Transforr~r in Brazil against International Prices, 1965 ( (million cruzeiros $Gr 2220 = $US 1) Domestic International MarJ:ufacture ' Manufacture l ...., d i ~ ~ p'. Price 2 168.8 140.0 B. Direct~ Imported Material 1) 1 - f.o.b. price 44.6 2) B2 - c.i.f. price 48.3 3) B3 - duty & charges 9.9 4) B4 - clearing expenses 3.3 5) BS - 1anded costs 61 •.5 38.1 C. Directly ~orted Services 3 8.4 ••• D. Domestic Materia1~4 11.9 11.6 1) D1'''' import content 2.46 2) D2 - net of import 15.48 content E. Other Domestic Supplies 4.2 1) E1 - import cont'3nt 2) E2 - not of import content F. Depreciation on Imported Equipment 3.9 G. Foreign Exchange Expenditure 64,.0 , (B 2 + C + D1 ' + El +F) H. Materials & Services net of import content 20~2 (D2 +.E 2 ) I. Import Duty on Material (B3) 9.9 J. Value Added (Y-B-C-D-E-F) , ' 11.4 83.1 K. Foreign Exchange Savings <9i- G) 76.0 .. L• Domestic Resource Cost . CJ) z 8 LOGARITHMIC SCALES .1 I I--LJ.-tl-t-IJ-I_ _ _.._~_. __ "_ _ _ _ 30 40 50 60 80 100 200 300 500 1,000 2,000 5,000 GNP PER CAPITA (2R) IBRD -3356 -88- - 89 - the other one on self-sufficiency aald industrialization, irrespective of comparative advantage. Finally, all three European countries (Sweden, Austria and Switzerland) produce hydro-ele~trical equipment, mainly needed in their domestic markets. 160. Recognizing the apparent existence of economies of scale in the manufacture of heavy electrical equipment, the next problem is to identify them. Only some tentative guesses are ventured in this area. ~he most obvious advantages of scale ~re in research and development. If a company delivers, say, five 500,000 kva gen~rators per year, rather than one such 1 generator every other year, its chances of writing off development expense are immensely increased. It is not entirely clear how much effort it takes far a subsidiary of an international concern, say in Brazil, to absorb experience gained by its parent in Switzerland or in the United States in making a new.-size generator. If the transfer of such experience is very efficient, there will be only moderate scale economies on this score. 161. Compared to those of generators or switchgear, the technology of transformer manufacture is simple and the number of units purchased is lArger (particularly for intermediate "systems" transformers and dis- tribution transformers). Even so, a large industrial country such as the United Kingdom needs only about 500 systems transformers per year. Though flow line techniques have been developed for assembling such transformers, the massive operations of winding and test~ng generally have to be carried out separately. 11 On balance, therefore, economies of scale in transformer manufacture are probably modest, and would not normally preclude their manufacture in fair-sized industrializing countries, as long as a certain m~nimum ~emand can be foreseen over a period of y~ars, for a given type and size clas8. 162. In manuf~cturing generators, it would seem that t~e prinCipal factor in realizing scale economies i8 enough ~ spread uniformly 80 8S to keep existing producers busy. Roughly speaking, this requirement is likely to be met by one order for a large generator every 18 months. 11 See "Transforming Transformera",The Economist, May 20, 1967, p. 822. - 90 - Costs could be reduced if major power supply companies established long- I term plans, giving th~ estimated compositian df their future equipment demand. In the United Kingdom, the Central Electricity Generating Board makes a ten-year projection 6f its switchgear Irequirements which has been followed up, at regular intervals, with "bulk allocation" of work between the (then) fOU1~ manufacturers. The firs t step penni ts the manufacturers to make long-run investment in production facilities; the second ste~ enables them to go ahead with productior'. planning :and supplier contracts. !I 11 163. Such allocation of contracts should not, however, conflict ~ith ~ompetitive bidding. Any departures from this principle, such as direct negotiation or restricted bidding, should be only such as will push domestic producers towards competitive international standards, or force a streamliningg of the industry structure if there are too many producers. 164. As already indicated, the whole European heavy electrical equip- ment industry is undergoing concentration. It seems highly likely that there are also too many producers of hydraulic generators in Brazil (2), or of medium switchgear in Argentina (3), or, taking an illustration from light equipmen~, of distribution transformers and electric motors in India (17 and 25 respectively of some importance, plus a large number of very '1:.1 See G.B. Richardson, The Pricing of Heavy Electrical Equipment: Competition or Agreement, Oxford University Bulletin of Economics and Statistics, May 1966, p. 86. The problem of feeding orders to as 'many as four suppliers was one of the factors which led the authorities to support mergers between manufacturers. In most sectors of heavy electrical industry, there are now only two competitors. The importance of careful market study for economic production can l' hardly be overstated~ Such studies. 'according to expert8, should cover at least a lS-year span, and should project the breakdown be- tween hydro and thermal capacity, the length .nd voltage of trans- mission lines, the size of generators and transforuaers, the demand for heavy electrical equipment by manufacturing industries and rail- road8~ etc, Some of the plants visited embarked upon manufacture without clearly identifying the future structure of demand and later were forced to diversify Qperations in order to fill unuaed capacity, sometimes duplicating established producers, even beyond the elec- trical equipment line. - 91 - small manufacturers).!1 Governments have sometimes justified their en- cduragem.nt or toleration of an excessive number o'f producers by alleging I 'the benefits of competition. But protected competiti.on in a small pond may lead to an excessive number of weak fish. Within the LAFTA area, econ~ic production of heavy electrical equipmen~ might be possible but probably only with a limited number of producers, orderly procurement by powt!r companies, free trade within the area and, at least in the beginning, moderate protec- tion against imports from the outside. 51 - , If equ1.pment manufacturers have merged their operations in Western Europe (though not yet over nationsl boundaries), corporate policies may conceivably not be an absolute obstacle to mergers overseas. Learning Process 165. In arriving at competitive manuf~cture of heavy electrical equip- ment in developing countries, it is convenient perhaps to distinguish three steps of the learning process: (a) initial learning (b) gradual buildup of capability (c) independent research and development potential. 166. In the first step, skilled labor has to be recruited and trained, supply capabilities developed, expatriate managerial p~rsonnel adjusted to decision-making in an entirely new environment, and a flow of orders secured in the face of buyer skepticism regarding product quality and !I For details, see Annex Table 14. In South Africa, there are five manufactuers of motors up to 250 hp. We aTe told that, ,,,ith standardization, one would be sufficient. ~I . Atthe third sectoral meeting of Latin American manufacturers of generators, transmission and distribution equipment (May 21 to 31, 1967), the delegationsrepreaented agreed to present a joint recolllllendation to the international financial organizations suggesting that the bene- fits (preferences) which are accorded to national manufacturers should be e~tended to the manufacturers in other member countries. For their part tbe delegations from Brazil and 'Mexico concluded a project~or a "complementation agreement", .ccording to which all duties and restric- tions would be removed in the trade between two countries in a number of products. They agreed to aak their respective governments f'or offi- cial formulation of the agreement. - 92 - on-time delivery._ In one case, a well-known heavy electrical firm invested as much money in' tTaining programs and early opeTat.lng l(v\sesin its overseas subsidiary as its whole fixed plant investment. TheTe is a dilemma here. I . If one wants to ,compress the peTiod fOT achieving optimum plant opeTation, one has to recruit on a large scale for multiple-shift operation, and the 1 • investmen,t in manpower goes up_ If, on the other hand, one follows a more gradual course, plant utilization stays low and long delivery times may make ~he firm less competl~iv~. If on top of this, the infant industry fails to $ecure an adequate volume of orders (because of irregular procure- ment, poor markE':t research, exchange rate overialuution, etc.) it will be in deep trouble. Annex C gives some actual productivity profiles for the learning period but these are too fragm~nt~ry and the difficulties in year-to-yeaT eompaTisons are too great to permit safe generalizations. Under favorable circumstances, the learning petiod for heavy electrical equipment production with well established technology in an industrial though still developing environment can probably be reduced to four to five years. In actual fact, it is perhaps likely to reach seven years (more, if the environment is unsuitable or fif there are unforeseen diffi- culties). 1~7. The problem of growth in capabilities was discussed previously in Chapte~ IV. Because of the trend towards large units, any new manufacturer of heavy electrical equipment has a strong incent1.ve to strive towards new frontiers of production experience. It is doubtful whether this can be achieved economically through license agreements. It is certainly achieved more easily by a local affiliate of an international concern. In the latter case, the ,main cost is in the initial leaming period. Growth into new capabilities becomes a natural process, less burdensome to the developing country than the similar phase some years earlier in thE indus- trial country. 168. Even major technical advances are readily introduced where the domestic producer haa strong corporate ties with a large international concern. Where this knowledge has to be acquired by an independent pro- ducer through license or know-how agreements, on the other hand, there is a danger that he may stay too long with the originally acquired tech- nology. Often be would lack an independent technical capability permitting - 93 - him to adjust promptly to new techniques or changes in the relative pri:es of major inputs (say, aluminum versus copper). To establish such a capa- b:J.lity in an early stage of industrialization is probably not economical. First, the adva~tages of scale in research and dev~lopment are one major reason for recent and proposed mergers in Western Europe. Secondly, de- veloping I countries have no comparative advantage in this area: salaries of research pereonnel are high, and the research environment is more diffi- . cult • I 169. If our obger~ations on the learning process are correct, they would suggest the following pattern of infant industry protection: (a) An initial period of protection could be provided each time a new major line of ulanufacture is taken up This would last, 0 say, seven years; (b) The actual rate of protection could be based upon a realiatif. appraisal of the average cost difference between domestic and imported equipment during the infant period. The exact level of protection might be a matter for negotiation between the government and the interested producers. The main end sought by the government is prestDlla6ly competi- tiveness at the end of the learning period. Within this framework, a production license might be given to that manufacturer or manufacturing group which, under otherwise equal conditions, would be satisfied with the smallest protection. Institutional Framework 170. When comparing costs and prices, we defined the institutional framework as those conditions, ~uding .!!!.. particular government pol1:.- c~es, whJsh ~uld make. actual cos ts exceed cos ts determined .l!I !h! £2!!!!- try's ~mparativ~ advantage. Though appropriate to the discussion of compara- tive costs, this definition should be elabora~ed for more general analysis. It views the institutional framework as a negative factor. It may be said with equal justice, however, that the institutional frlllDework "encompasses those instruments which enable 1.\ country to realize its full industrial potential:'. In ~rief, the institutional framework may 8upport or thwart healthy growth. 171. The emphasis is on the word "healthy". Historically, the manu.... facture of heavy ~lectrical equipment in the countries studied was started - 94 - in response to the twin desirl!s of governments to industrialize and of the international manufacturers of heavy electrical equipment to preserve and expand markets. To this end, governments were normally prepared to provide heavy protection and sometimes additional :I.ncentives •. Protection §.l was clearly the indispensable condition for the establishu,ent of the industry since~ in all cases, heavy electrical equipment could be imported from abroad at prices lower than those of the prospective domestic producers. 172. Yet, even if we accept protection as a necessaI)4 condition, cer- tain policies adopted under this umbrella will have a major bearing upon product~on costs. These polidies may be considered under four headings: Shape of protection. Protection should be time-limited and held as low as possible, recognizing that excessive protection breeds inefficiency and is inextricably interwoven with, and tends to perpetuate, econo~ic controls and poor industrial structure. Protection should not become a compensation fQr inefficient I management or for bad planning. It is only justified for indus- tries which are ~conomically viable in the country concerned and efficiently managed. In this case the/industry will presumably become internationally competitive as soon. as the difficulties of the learning stage are passed, Number of producers. An excessive number of producers is likely to result from excessive initial protections This type of industrial structure should be discouraged. Mergers across national frontiers in the same geographic area should be promoted. Economic controls. Economic controls to a large extent, are a by-p1:'oduct of foreign exchange stringencies. We are not con- cerned here, however, with possible justification but only with visible effects. S~ID~ manufacturers of heavy electrical equip- ment were not granted sufficient foreign exchange to complete their plants; delays resulted. Many plants encountered a mare acute difficulty: after only a few months of operation, they suffered import resCrictions on needed materials and components. This ,~aused bigh costs because of unused capacity and idle trained p~"1:'~onnel. Yet the worst damage is probably an environ- mental one:,;' management is diverted away fTom productive tasks or left idle. Procurement. Procurement should be based upon a long-range plan, taking into .ccqunt the need for competitiveness and for .ound utilization of ~xisting cap~city. il We define protection as any meaaure taken by the government which .akes it pos8ible for the domestic producer to charge a price higher than the c.i.f. price of equivalent tmpor~s. - 95 - 173. Besides protection, the major factor affecting the development, of heavy electrical equipment production has been the climate for foreign investment. In developing countries it has been virtually impossible to mahufact~re heavy electrical equipment economically without the support of interllational manufacturerl3; it should normally be advantageous for government policy to smooth the path for such participation. Major general ! policies promoting international investment (e.g. political !;tability, guaranteed transfer of earnings, reasonable protection against expropriation, satisfactory labor relations) are well known. 174. Something should be said, however, about economic nationalism. This word is used here in 8 purely descriptive sense, concerned only with proximate economic effects. (No attempt is made here to evaluat~ or to judge poiitical necessities or psych~logical or emotional factors.) Eco- nomic nationalism has many fa~ets, e.g. insistence upon: (8) minimum domestic shareholdings or even the acceptance of specified domestic partners; (b) limitations upon the employment of foreign technicians; or (c) limitations upon license fees. Some policy moves in these directions have in practice often turned out to be self-defeating. Many foreign companies are anxious to find domestic partners and/or sell shares locally. An unsuitable domestic partner forced , upon a sound international producer reduces the attractiveness of the in- vestment. Similarly, many foreign companies in this industry are anxious to replace their own nationals With local personnel; France, Germany, Sweden, Switzerland are all short of key staff, 8'fid very great cllpability i8 :required to succeed in a developing country. Limitations on license fees, in a competitive situation, may me~ely impose similar needless obstacles. - 96 - VlII SUMMARY AND CONCJ~USIONS Characteristics of the Industry 175~ The heavy electrical equipment industry, in the definition we have adopted, covers the production of po,",er generating and transmissio'n ( equip~ent together with certain industrial items such as heavy traction equipment and electric furnaces. 176. Such equipment is generally custom-designed and much of it is ! large and expensive~ It therefore calls for design capability and sub- stantial skill and care at the manufacturing end. Large investments are required in testing facilities and, increasingly, in heavy handling and m~chining facilities. 177. To keep plant and skilled manpower occupied becomes a prime ob- jective for management. This is made difficult by the fact that orders are typically very large, few and infrequent. One order or one piece of equipment may absorb 10 to 20 percent of the annual production capacity of even a relatively large heavy electrical equipment firm and the trend is towards even larger units. Hence, if there are ~hree manufacturers of turbines or generators for hydr{Jpower applicati~ps, only one may recei",e an order in a giveh year; even if there is olnly one manufac turer he may have no orders for this type of equipment in years when expansi9n is focus- ed on thermal generating capacity. This situation, one suspects, is one reason why most major companies have tried to produce a full line of heavy electrical \1 i)'equipm'£')nt ,,)' \-, which, in tUTn~ forces them to keep in touch with a largeriilmber of technologies and manufacturing techniques. 178. The \:!.ndustry is very dependent upon outside financing •. Fixed investments aTe perhaps not unusually high but much capital 1s tied up • i in inventories and ~ork-~n-progress. At the same time, at least in de- veloping countries, the customers want credit; some of the equipment (e.g. hydro) they purchase will last 30 to 50 years and upwards, and their re- ceiptsfrom the output of this equipment will come in only slowly- - 91' - International Structure lt9. The tc.tal annual demand for heavy electrical equipment (outside Comecon countries) is estimated at US$16 billion, of which 80 percent comes from ten industrialized countries. Trade is important, even among the maJor producing countries, but is largely resrtricted to items not produced in the importing country. World exports are valued at $2.2 billion equival- ent; of these $0.9 billion go to industrialized countries and $1.1 billion to developing countries. Some medium-sized industrialized countries like Sweden, Belgium and Aus tria are vi,rtually self-su~ ficient on a net basis; yet their imports of heavy el~ctrical equipment rlnge from 40 to 60 per- cent of their total consumption. The developing countries are large im- porters. As a group, they account for the major portion of world imports of hydraulic generating equipment, transfol"mers and switchgear. 180. The in.dus try is dominated by a 1imi ted number of large companies, practically all of which are engaged in international operations. Many . . ' of the international firms produce both he,lVY and light electrical equ:Lp- ment and the largest ones have branehlad out substantially into related fields. Nevertheless, sev~ral major producers have heavy electrical equip- ment as their main production line. Major firms like General Electric, Siemens and Brown-Boveri have manufacturing interests' in a number af indus- trialized cquntries as well as in several of the developing countries studied. \\ ,\ 1.81. Be'cause of intensive competition in conventional items of ,equip- ment, the bulk of the profi,ts are earned on i terns where a particular f1 rm has pioneered new developments. Investments in research and development are enormou~;. 182. A pronounced trend towards very large individual units of generators and transformers has created excess capacity since it takes far fewer man- hourB and machine time tt') produce one large generator than to produce two small generators of the same combined rating, Excess capacity combined " with the large capital requirements ,for research and development (and in- creasing competition within the Common Market) have led to a wave of mergers in Western Europe. - 98 - 183. Excess capacity also explains the jealous pt:otection of nutional markets ~nd the attempts to extend thes~ markets abroad through various ,j' ; financia.L and other arrangeme!:lts, particularly tied aid. Finally, it lies behind the very sharp price competition on the free world market and has encouraged attempts to get around this by direet manuf~cturirig investments in protected markets abroad. The Industry in Developing Countries 184. We have analyzed the experience of six countries - Spain, ,Brazil, Argentina, Mexico, India and Pakistan. In Spain, the manufacture of heavy electrical equipment on a substantial scale dates from World War II. There has been gradual and efficient growt'n during the post-war period. Spain leads the other developing countries in capabilities, has the lowest prices, and has developed some exports. Nevertheless, the industry is not as yet price-competitive in the woxld market. The future shift in Spain from hydro-power to large thermal (ma;,"\ly nuclear) power stations is probably the major difficulty and the major challenge faced by the Spanish industry today. 185. The most impressive growth has been in Brazil. Since 1958, it has been moving towards self~reliance with a very rapid buildup of capa- bilities. Enormous efforts have been made, particularly by two large in- ternational companies, in creating modern facilities to produce very large equipment, in recruiting and training maQPower, in developing supplier potentials, and in gaining production expe;!ence. Though ther~ have been great strides in competitiveness, prices are still rather high, profitability poor, and capacity in excess of needs. 186. Until now, the production of heavy electrical equipment in Mexico and Argentina has been limited mainly to tran.sformers and motors, with some small generators being built in Argentina. In Mexico, one firm dominates the industry and bas apparently achieved good results in transformer pro- duction. In contrast, prices for motors are high - .the ~arket may not permit reasonable economies of scale. In Argentina, there are two major j producers of power transformers and switchgear. Tbe industry,has been troubled by excessive product diversification and unused capacity. - 99 - 187. In India, the "privE~te sector man\lfacturing transf01:mers and motors grew rapidly in 1950-1963. TheTe are many firms, all small by international standards. Prices are (tuite, low tJ but the industry bas lagged behind in technology. This· may bf!~ due, in part, to rigidi ti~s ~aused by econcjmic controls. The Indian Q')vernment entered into the producti0t.l of generators, turbines ~ and heavy swi'tchgea!' through public enterprises, and construction was oegun at Bhopal, in, 1958,. of a large plant to manufacture most of these items. Subsequently, t:bree more specialized plants have beet;! built at other locations. The total investment in Bhopal a~ the end f)f March 1965 was about $125 million equivalent, includin~ accumulated losees. Sales were only about 15 per;cent of the planned output. The reces9ionof 1966 and 1967 did not advattce the industry's profitability. Clearly, Bhqpal experienced enormous difficulties, one of which was perhaps trying to do too many things too f'as t • 188. Pakistan's. production of electrical equipment for power system and i,ndustria1 uses ;does not, as yet, include much. heavy equipment. In the early 1960's mB:1l1ufacture of transfonners, motors and switchgear was strongly promoted by the government, and, in 1965, domestic producers (all private) supplied .;::ore than one-half of the total demand fo~ electric motors and transformers. Prices are still high by international standards •. Price Comparisons 189. World market prices for heavy electrical equipment have falle.n sharply in recent years, and there is no indication that they will improve in the near future. The large unused capacity in the industry exerts pres- sure on prices. Firms which are aggressive in marketing probably earn a profit. Mergers and specialization should reduce future costs. Hence, no developing country should plan the manufacture of heavy electrical equip- ment on the assumpt:l~,on \\ that the present world market prices are abnormally low and therefore unlikely to endure. 190. If world market prices are accepted as a proper standard for comparisons, the next question is how prices in the countries s~udied com- pare with world market prices. Such information has been deriy~d in this study from analysis of prices quoted in international tenders supplemented .• 100 /:- by interviews with manufacturers. This comparison is valid only for cate- gories of equipment which are both produced in the country and imported. The categories compared differ between countries, and the conclusions do not necessarily apply to equipment which is;' not presently imllorted. With these and other reservations, we found typical price d:f~.fferences of 25 to 60 percent for a ~ange of items (1964 data). The difference tended to fall substantially and quite rapidly as mor~ production experien~e was gained. Mexican transformer production provid~s a goad illuiltration of this. Price differences were also very' much influenced by the size of the do- mesti:c component, particularly in generators. In Spain, ~here the industry has reached considerable matJrity, present domestic prices are probably no more than 20 to 25 percent above the international level. The price comparison was most favorable for medium-sized transformers and switchgear - these are items produced in relatively small numbers to individual design and they do not require very powerful manufacturing facilities. 191. tn our discussions, we often encountered the argument that, even accepting that world market prices are going to remain low for a long period to come, it is incorrect to compare prices in developing countries with prices below average production costs in industrialized count~ies. Under a rational allocation of res~urces, there is no r~ason for protection other than infant industry protection; balance of payments equilibrium .ay be restored essentially through the exchange ~ate adjustment.· 11 'The fact that equipment is offered on the world market at prices below the ave~age cost of production or even at subsidized prices, is a bargain for purchasers. tn this sense., it is no different from, say, a discovery of important new cOl'per resources which would reduce the price of copper. The world market for copper has to be taken as given; there 'is no reason for exploiting uneconomic low-grade domestic deposits. 11 ttis obvious that in the case where one industry has a pronounced ,co~parativeadvantage in relation to any other, such a& oil in Libya or Veneauela, mere reliance on the exchange rate may kill off the other industr.es. - 101 - Present Cost Comparisons 192. We compare actual costs in Western Europe and in the countries studied, and then analyze actual costs in relation to theoretically attain·- able cos ts • (: )1 • . I! Ii 193. Major raw materials for heavy electrical equipment are consider~ ably more expensive in the countries studied. than in industrj,al countrias. In a few cases, this reflect~ high duties' CJ:n imported materiuls (e.g. trans- former grade steel sheets in Brazil) or tied-aid procurement. In other cases, importers taking advantage of restricted supplies may'charge high prices. ~fost often perhaps, material c~.sts are high because of high prices charged by protected domestic industrieu (ordinary steel sheets, dynamo- grade steel sheets, semi-manufactures of copper). Often these, high prices primarily reflect the high costs associated with infant industries; in man" cases they also reflect various inefficiencies and monopoly profits. 194. Wages for comparable skills in 1965 were onlY'about40 percent of the German levels in Spain, ~about ,- one-third of that , level' in Brazil, • and about one-fifth in Indi~:./ Differences in salaries were considerably less: in Brazil, wages-'f~r comparable skills were 32 percent of the Ger- man level while salaries were 62 percent of the corresponding level. On balance, the average cost per man-year will be considerably lower in devel- oping countries. This could give them a competitive edge, since manpower costs in industrialized countries account for up to 40 per.cent of the se~jling price for heavy electrical equipment. Up to now this advantage has been offset by lower productivity. Our best evidence suggests that two to three times as many direct man-hours have been us~d in making heavy electrical equipment in Brazil or Mexico as in Western Europe. Since the proportion of indirect labor is high in developing countries, and ratios of staff . . to labor very similar to those in Western Europe plants. the lower product- tivity of direct labor also holds true for total manpower. 195. The above productivity figures should be interpreted cautiously. Their comparability is limited by differences in the composition of the labor force and different degrees of vertical integration. High labor productivi.ty is not necessarily synonymous with economic efficiency; this would also depend upon the quantities of other factors used such 8S mate- ~ials and capital. - 102 - 196. In order to consider all cost factors, ~~ use the following ap- proach. First, we select manufacturers, both in Western Europe and in developing countries (specifically Spain and Brazil) who have specialized in the production of heavy electrical equipment. ,Secondly, we divide all I costs into material costs, manpower costs, and charges against capital:. and express each category as a ratio to sales. Tl~irdly, we ·':deflate: sales of the firms in the developing countries to Westel~ European prices. This made it possible to expr~ss each one of the three cost elements, both for' Western Europe and for the developing countries, as a ratio to sales valued at Western Europe prices and to compare these cost elements both singly and in terms of their total impact. 197. Our comparison s1.:iggests that manJ~ower costs' per dollar of sales at constant prices were 25 to 60 percent higher in Brazil and Mexico than in Western Europe, corresponding to a cost difference of about 10 cents for each dollar of sales. Charges against capital in developing countries were about twice the Western Europe level which makes a difference of about 13 cents for each dollar of sales. Under capital charges we include depre- ciation, interest, profits, and income taxes. 198. The final balance sheet of present cost comparisons indicated Spanish arid Brazilian costs 40 to 45 percent above the W. European level. About 16 percent could be explained by higher prices for materials and components, 10 to 15 percent by higher manpower costs and 13 percent by higher charges against capital. 199. ~t the time our comparison was made and under the ~ditions in which the industry was then operating in Brazil, it would consequently have needed the following average protection to earn a 20 percent return on the total assets employed: - 103 - Reguit'ed Protection Industrial Country Based on Based on Home prices Expc)rt Brazil Home pric(~s Export prices Raw materials and purchasti.d compon- ents ("inputs") 51.5 51.5 67.3 30% 30% I ' Conversion margin 48.5 32.5 74.5 Final selling price f.o.b. 100.0 84 141.8 42% 68% c.i.f. developing country 92 141.8 54% Future Outlook 200. Only to a minor extent do the above cost. differences reflect differences in productivity In a narrow sense. A~ter the initial learning period, the operating times on the same productiQ~ machines are virtually identical in Brazil and in Western Europe. 201. Nor would it be true to say that a large home market is a decisive condition for comparative advantage. Most of the countries studied have home markets as large as some very competitive Western European manufacturers. True economies of scale in the production of heavy electrical eq1.d:,.:ment I are apparently quite modest. In other words, costs would not differ greatly between a plant with a capacity for making two very large generators over a given period, say 18 months, and a plant with a capacity for only one such item in the same period. What matters is that the plant should be well adjusted to the size and needs Qf -the market, and that it should re- ceive an even flow.of orders. 202. The key to success in the manufacture of heavy electrical equipment would seem to lie in some combination of the following factors: (a) A gradual and efficfl,ent build-up of capacity, avoiding' excessive diversification; (b) Government policies aiming at an even flow of orders for the local industry and preventing the multiplication of producers; (c) Efficient management, skillfully coo·rdinating every aspect of production and sales; - 104 - (d) Resea~ch and development, permitting good profits on advanced types of equipment where world market prices are relatively high. 203. In the latter area, deve],oping countries have a cOplparative disad- vantage which will not be overcome in the short run. The other three condi- tions, in our view, will be generally fulfilled to the extent that the I technological and ~anagement kno~-how of internationallconcerns are combined with a proper institutional framework in the developing c~untry. I The Role of Protection 204. In most of the countries studied, the industry is already emerg- , ing from the infant stage, and is heavily protected. Progress has been reasonable but has fallen short of what was actually attainable in several of the countries studied. In retrospect, efficient growth might have been promoted by the following type of arrangement with an international manu- facturer covering a specified infant industry period: (a) A detailed plan for the development of the industry during this period, including arrangements with power companies regarding domestic procurement. (b) Support by government during the infancy of the industry, but not thereafter. 205. In most of the countries studied there has probably been enough time for learning with respect to the types of eqUipment presently made. In practice, learning is a function of jobs done rather than time elapsed; an excessive number of producers and an uneven flow of wo'rk has lengthened this 'Period in some countries .. In order tp encourage the gr.owthof exist- ing industries, some extension of protection during a period of adjustment might be defended. The corporations, for their part, should encourage manufacture in developing countries with the ultimate aim of achieving the optimum location of plants around the world. - 105 - ANNEX A ANALYSIS OF PRICE TRENDS 1. The prices for heavy e~ectrical equipment quoted iq the last five years show a significant dec!line in hc1ine as' well as int.~rnational markets. This is illustrated by the first four charts which follow. Fig- ute 4. shows price changes for various products in 'the USA; the market de- ciine since 1958/59, especially for transformers, representative of inter- national price trends. Figur·e 5 traces the price fall in the USA between 1958 and 1963 by size of transformer. It shows first that the east per J kva is less for the larger transformers, and secondly, that the price fall has been greatest in the intermediate sizes. Figures 6 and 7 show changes in price levels in some of the developing countries and the USA. 2. These figures indicate that no one price level curve is valid for all products, but that there is a price pattern generally applicable to the majority of products. 30' Prices for transformers have shown a decline in developing coun- tries as well as in industrial countries (Figure 6)1\ In the ease of motors and special machinery, e.g. in Mexico, prices have not follo,ved the same pattern of decline (Figure 7). This is mainly because production of large motors is relatively new in these countries. and also because pricing poli- cies differ. In fixing prices a producer has to deCide whether he wants to compete internatio~ally and therefore fix tpem low, or seek protection and char~e higher prices. More detailed information on Mexican motor prices is given in Annex Table 15. It is claimed that the manufacturers in Mexico calculate with a higher profit margin on motors (sold to private industry at protected prices) in order to cover any drop in the margin on tranS- formers (sold to utilities; which may be deemed to have more "international competition"). 4. ' Price dif~erenees change by size and other specifications, and may be best expressed by ranges of price premium. From interviews with Mexican firm$, we found that in 11 cases of pricing for high voltage accessories, the price of the domestic product was: ANNEX A Page 106 45 to 56 percent higher in 1 case; 56 to 67 percent higher, in 3 cases; 67 to 78 percent higher in 5 cases; 78! to 89 percent higher in 2 c:ases. 5. The price decline per kva demonstrated itl Figure 2 over the years 1959-65 presents an important challenge to the producers in devel- oping countries. From time to time the 1n~ustry i~ called upon to adjust to a new and lower price level set by a determined competitor. Annex Tables 16, 17 and 18 show the. effect of this pattern on the experience of the Mexi.can I' industry. In 1957-61 local producers were in competition ! with the p!tfoducers from Central Europe; later they had to face Japanese . 'I competition (1962 onwards). The effect of the learning process and efforts towards competitiveness of Mexican transformer manufacturers can be seen in Anne]!: Table 17.' By 1964 the price ratio of Mexican to foreign bids had markedly improved. For medium-size transformers Mexican producers are now b1dding lower than international pr1.ces. 6. Figure 5 and Annex Table 18 show the competitive position for disttibution transformers as it was believed to be in 1965/66. It shows the Mexican price curve per kva tending to intersect the us curve at the medium-size range. Most of Figure 5. however, covers light equipment. The estimated price level in Pakistan is given for comparison. ~ . UNITED STATES : PRICE INDICES FOR SELECTED PRODUCTS (INDEX; 1958 =tOO) 120 , - - j J t f ! ---T-"--T I ,- -' ----: ,- - ,- I I I ~-l-' -'-.-~--: 120 '~~ I ---if-, i i t i 110 I i I I I -.,~-tp-- t ,110 I r---l:~~~~~·-~-rf,~···~·, ~.~rl, -'~~~r~~~~--~----~---L--~ . __ MOTORS ,I, ' " I I r • . ; 100'; .1~ • . . ',~·· I. . . . . . . . .L 1 .... 1 100 ~,~ I I ' 90 I ' I i i (! l .i,f, t I I - I' I ! , "I t---- 1!, '..~~..! I 'I' '" .~TURBO-GENERA-TO-R-S--+t---~t90 .It· , ......J . ., Gi c :,1 -..,j. i ' If p ! .. (' :fI ' " I' I .' I I '-..'ir·········r.····J,.......... i 'I i •••••• ••••••••••••••• ;:0 m -'--~ '-r- .$:I- o 80 1.. t ----t~--+---L--~ . . . . 1 1 I f ! t -.. -L-,i lao " I ! ,I ,. , i . 1 ' ! ' I' I ,- I ' 70' I . _._{ i TR~NSFORMERS~I" , ! ~j ",;o+-, ','" ' I ._..-. .... , .... ' -li.,. i i '" " I f - . - r 'I I I . I I 1 170 ! ~ ~ ; 60~ i • I I - I , I - t - - + - -L t.----._--l I , - - 60 ~ ----'- ____1.__ -.- ____. 1 t I 50;<--- . 1956 1957 1958 ! I I 1959 1960 - 1961 1962 1963 1964 1965 J 50 SOURCE Handy Whitman mdex (R) IBRD -3359 UNITED STATES: AVERAGE TRANSFORMER PRICES BY SIZE F GROUPS. 1958 AND 1963 16 ' i i I I -------- . I 14 ! • i I 1 12 --------- , 10 I ~ ~1958 I a::: w a.. I : " G) C ~I (I) ;0 i a::: 1963 J ! 1 4 - 2 I - 01,---- ----I. -L ---L_----i i 15 50 168 500 5.000 10,000 SIZE GROUPS - KVA (SEMI-LOGARITHMIC SCALE) SOURCE U S. Deportment of Commerce J 963 Census of Manufacturers (R)IBRD -3360 PRICE INDICES OF DISTRIBUTION TRANSFORMERS IN SELECTED COUNTRIES (INDEX, 1958 130 ~~-~~-----'--I =100) T--------- I r--~-~ 130 I _~_ _ YEARLY 120'~'-, .--------r-/,.,."\ ._ . _____ (MID-YEA? I 120 ,..-,•..-' \\) ~ SPAIN .1' I, ~ -\" : . / . . ~ .- I lOr--' • ",,------, • , --.._---,-- , -~- ... \ . -~""' .• -,'='<-"'''" f "" .-..... "'---~--- "'''''''-~'---~~'-'''''''-~-'-"-"".~",---~--,~-~---,-. ; I t : .: : I 10 ," L-" . / tI' / ,tI' J..,...................... .....,. , ' \. •••••• IIc••••••• -. • ARGENTINA t ~ : .: 100 ..oWl' . . . . . . , '... \ ••••• ~ ! 100 2! . - J.... _ ""'" _.* " '. , , .."....... __ ~ •••• ',.... 1 I l :' G1 C o U> • ~., \ .\ -, " e.. , ...... •• /:' MEXICO I . ~.- ::0 rTJ m ••. ~ ~--- a.. ... _____ ...ae__ . 90 -.--.----~-------.---~- • '~----t--"..--, --'."-\-'-~- ~-----'---'-----'-- .. -~.~-t-----.---- ••~---.--' ---.- 190 \. ·· · 1 . .l e..- .. ! .i •••• ..f. I 80 -.--~-----~,-'-.~---- ~---t----:·~.~~--¥,...-·'------·-~-------l8 0 70~!;--- - - - - '----~1-- ----t '-::-::UNIT~~TATES -----1 I 70 60 ' ~ 1958 _----1 _. _ _ _ ~4~ i ____ , _ 1.• • I _.L., , ,'-_ _ _ _ , _ _L -_ _ _ _ _ j . .60 1959 1960 1961 1962 1963 1964 1965 1966 SOURCE: Interviews with manufacturers. (R) IBRD - 3379 v PRICE INDICES OF ROTATING MACHINES IN SELECTED COUNTRIES (INDEX, 1958 = 100) 140 1 I I ,140 1301 ~! 1130 t 120 I I 1120 .... .. .. · IIO} 1 • ••• . •• ~-----..-'-,' : 110 I I .. •• ••- I ~.. I·. -. _-~...... ; ! zl!u.S I I ." • o t 100 •- - .•• "W ----- ...... ..".;,.---. ~.. ,• • i --. ...... --. f .j... " ! ' " I . I I ~ I • I I I 100 G) C ::u ITI / ~ ',! I I ~ ... . -. " ,I ~-----~ ~ i . __ f i -__._- + __ -f ; .J ' ""'" 90~ ~.. ~.e. ~ 4·'--· .-~+'------t---- i I I I I --190 ~ J . . .: ,i j •• ••• ····1 ! r -. '.. I '" ······t··· j ••• f 80 J j SPAIN ! ......... 1 ••••• I I ~. 80 '-r.~. ~ I 701 I . J I ~70 60 ! L._q"~ __ H _____ ~J.. _ _ _ _----->._ .___ .,._~L __ , _. ___ ,_.. ___ ~ l-_ -' - - - ."J 60 1958 1959 1960 1961 1962 1963 1964 1965 SOURCE: Interviews with Mexican and Spanish firms (R)IBRD -3358 FIGURE 8 PRICE OF OIISTRIBUTION TRANSFORMERS-1965/1966 300~------~----~--~--,--.-.-o.-.---------.----.---.---r-r -+1- "Ti" LOGARITHMIC SCALES I 200 f---~~+---+---+--t---+--t--t--t--t--~---. I I I 'II ; 1001---- I --+--I---'I-t·-+-... ·~..·----+----I---t----+---j 80 t--------t---+---l-'-i- ---+--~__+--+---+~--t---+- 60 --+------+--,.+-+---+. 501--- .,,+,- ! t·-,~ I t'-' ~ 30 I 1 ~ ~ w a.. 20 (J) 0:: 4.24 48,250 2.68 3 25MVA 11,/13.2KV 1957 1.17 121,860 4.87 104,450 4.18 4 IB.5/24.6MVA 110/44113:.2KV 1951 1.57 128,594 5.23 81,800 3.33 5 33.3144.bMVA 1958 1.14 128,900 2.90 113,100 2.55 220/85113.2KV 9 2$MVA 115113.2KV 1960 , 1.63 80,800 3.23 49,500 1.98 7 15/20MVA 134/33KV 1960 1.63 116,000 5.80 71,000 3.55 8 3O/40MVA 15V13.8ICV 1960 1.92 109,500 2.74 57,000 1.43 9 30/4oMVA 1Sb/69KV 1960 1.42 123,500 3.09 86,700 2.17 10 8.3/10.4MVA 1961 1.39 51,500 4.95 37,000 3.56 106.6/6.6KV 11 3O/40MVA 110/13.8IV 1962 1.54 110,100 2.75 71,:00 1.79 12 15/20MVA 138/13.8/1).8kV 1962 2.21 82,800 4.14 37,400 1.87 13 35146MVA 1)8/69/13.8KV 1962 1.47 80,800 1.76 54,800 1.19 14 SMVA 138/13.8KV 1962 1.43 )8,300 • 7.66 26,700 5.34 1S 16 17 lS/20MVA 11S/34.51l3.8rVl964 0.91 40,160 2.00 44,000 2.20 . 18 15120MVA 138/13.8IV 1964 1.23 40,100 2.00 32,610 1.63 19 10/12MVA 110/13.81V 1964 0.84 27,230 2.27 ,32,400 2.70 20 7.SMVA 11S/13. 8/13. Bit' 1964 1.04 25,150 ,3.35 24,160 ,3.22 21 2.S1fVA 34.5/1).8I(V 1964 1.25 9,.590 3.84 7,670 3.07 2~ SlIVA 115/34.srv .1964 1.09 26,160 4.03 18,470 3.69 Hote.- In cases of transformers with a range in capacitY', i.e. 15/20MVA" .))/40MVA, etc., - the price per KVA has been calculated on the highest capacitY'. Source: Information subndtted by Mexican firms. I - 134 - ANNEX TABLE 18: Comparison of Prices of Motors in Mexico, India and Pakistan, 1965/66 (international Erice a 100) v 8 International Mexico India Pakistan Motors Priceb • 133 131 2 hp 100 153 5hp 100 1.59 139 133 10 hp 100 15" 149 145 20 hp 100 15'3 ~ 147 ~ Unweighted 100 1".5 142 .142.5 Adjusted figure c 100 195 177 178 a All motors are squirrel cage induction JOOtors, 4 pole, up to 5 hp totally enclosed; ,-20 hp, sc~een-projected, tri~-proof. \ b International price is the price at which imports to the particular country could have been realized. These are c. i. f. prices, and do not include handling charges which would add another 1 to2 percent price. c "Ex-factory costs" (the figures which happen to be available) do.not include selling and administrative overheads, nor do they include interest (except, in some cases 6 percent imputed interest), profits, or income taxes. Based upon available info~tion on costs and prices for individual products, and on the detailed profit and loss stateJOOnts for the companies studied, we have added a 20 percent margin for sales and administrative overhead, interest, profits and income taxes in the case of transfomers and a 30 pe:~cent m.argin :In the case of electric motors. This' seoms to be close to the margins actually practiced; tlJ,ez--e is some doubt whether these margins permit a reasonable return on investment at this stage of the industr,y's development. Since the ex-factory cost already includes a 6 percent imputed interest, the net addition to the ex-factory cost becomes 15 percent for transformers and • 2, percent for electric motors. ' Source: Based upon infcnnation supplied by individual firms at the end of 196.5 and the beginning ot 1966. Domestic prices are ex-factory costs ~ including 6 percent calculated interest.