ov- F‘”‘ .. . v H ' -"v- fl.“ V .n ,. \rvi“".“ I... “..v"‘ '0'” n ’. ’ " Q_‘.. ....I ...~ ”I ‘0 nvbvtv L4 0 ~ arhr;::p ~v- - . 00‘. 5-nv-~—¢g~. .tt q -:IOQ-~~ “n F" :"""‘D a4“ 1‘. I 9 L) In \L) : ‘an- "t s A o t-1 V‘f‘“ f' o‘.:‘, .aut: “ 4 "“th a“. - ..._._' DJZS“* .‘ 5‘ b .a b :A :“1 ‘. “ “7‘3 tra'j ..1 “In I ‘L . M. —.:_‘l" V“ . t yrl‘fiary b tun .; .2, 's .3. , «in (em ,. "' a}, ‘ M “St eyza r: t ‘. ‘ “L. H .‘ '3 S ”mere .I “f. c ‘ O 'laln ;._ I: "910“. 'vlna C, SE"\." ~ .'« JG .1 ABSTRACT THE COMPARATIVE ADVANTAGE OF DEVELOPING COUNTRIES IN THE MANUFACTURING INDUSTRIES AND THE EFFECTS OF GENERALIZED TARIFF PREFERENCES BY Zubair Iqbal Accelerated economic development requires a substan- tial increase in the imports of industrial materials, capital goods and technical services, besides an expansion in domestic savings and investment. However, the developing countries are unable to generate sufficient exports to pay for the im— ports required to attain a high growth rate. This has created a large "trade gap" which has to be filled either by aid, or import substitution or export expansion in the developing countries. Import substitution cannot be carried too far due to various structural limitations. The developed countries may not be able to increase their economic aid sufficiently to fill the trade gap. There remains only the expansion of exports of the developing countries. There is little chance for the developing countries to generate a dramatic expansion in their primary goods exports because of low income elasticity of demand (among other reasons) in the developed countries. They must expand their manufactured exports to the developed countries where the demand for manufactures is the greatest. One of the main obstacles in the way of manufactured exports of developing countries to the developed countries is the ef- fective protective tariff barriers in the latter countries _. . ’43 “'C‘DY’B' "(3 our. ~-y-»§usa~vv . . . ":lar'rrv rwv'O-1 ""9'“": v-Acny4 . in... b.“ :55! COH' IV”. ‘35 A::,, c ~ -E C. .. any 4 '1'1 -,. ‘. I . S‘an ~' . "u .S a r91 A . -36-ua 5 i: p_ . u ‘ “4": “ t O +. ya 5._ r .T ‘ inacturej 6?, it __ . ..-Nc* . ~~~~s WitnO‘ t 6;; c ~.»r~.nce SI "*r v; 35:» {Vt/EC COJF“ t “m w. 3.:ectiv‘3 t ""241 Prefere' We establis (‘e r: I;h~ “its * f 5 Hg 0 “5 l . 1 1-,. \— x. v “a V Zubair Iqbal against manufactured imports from the former. The effective rate of protection is the real measure of restrictiveness imposed by the structure of tariff rates. Tariff preferences in developed countries for the manufactured imports from the developing countries is one way of increasing the export earnings of developing countries. The preferences will give competitive advantage to the de- veloping countries over the producers in the "third," devel- oped, countries equal to the effective protective duties in the preference-giving country. While it is possible to con- ceive of many preference schemes, the generalized preference system is a relatively easier method of granting preferential treatment to the manufactured exports of the developing coun- tries. Such a system would require all developed countries to grant a duty free access to the manufactured and semi- manufactured exports of all developing countries to their markets without limitation on volume. Under the general preference system, the higher the effective protection in the developed countries, the higher the preference for the devel- Oping countries. Therefore, based upon the existing structure of effective tariff protection in the developed countries, the general preference system would give the greater incentives to the establishment in the developing countries of those industries that are subject to higher effective protective duties in the developed countries. The question is: Are these the industries in which less developed countries have (potential or actual) comparative advantage and developed M‘“'Y‘HD ‘Q. - - so"~" CSEQS tr: ~ I av ' .10" ’ ‘ .....oqn+‘~rhq "I .- . H ‘ _‘.‘-:\'v‘-vu -: . :Vu/‘Ai 0' \v‘al‘vv L 'M-v-zy-JJ-‘vva a;- 5‘ '1-¥“ub-0v § w ~ ._'A" ‘ f‘f‘v C.‘U.F‘.‘ ' " r‘ " w J‘— {-9 ‘ a. ‘ “2:."rflpnn ‘ . .V‘ U¢‘J.~D ' '- t -~-‘ p. ‘l( 'a VF ‘ . ~~n .‘y .N“:‘. dv‘ b. r»... v: 'V s,"[ . g. S 1: O ‘ w. u I k u": "6 In 1 D )4 Zubair Iqbal countries comparative disadvantage? This is what the present study seeks to answer. To do this we compare the effective protection rates in the advanced countries with the compara- tive advantage schedules of developing as well as the devel- oped countries in the production of manufactured and semi- manufactured goods. (”Since there is no single explanation of a country's comparative advantage, various theoretical models were con- sidered to develop indexes in order to rank industries by the degree of their comparative advantage. We used the factor proportions, human skills, R & D, and economies of scale models besides Balassa's "revealed" comparative advantage to rank the industries. The input-output technique was used to obtain total requirements of capital, labor and human skills in order to estimate the input ratios for ranking the indus- \\ x. ‘\ tries. } -J/’The effective protection rates in the developed coun- tries were obtained from the estimates prepared by Professor Balassa. These rates were "normalized" by obtaining a weighted average of the effective rates of EEC, Japan, U. K. and U. S. A., with the rate of each developed region weighted by its total trade in the product to which the rate applies. Linear regression and rank correlation tests were used to determine the nature and significance of correlation between the structure of effective protective rates in the developed countries and the schedules of manufactured commo- dities in the developing and developed countries ranked by ,c {-n4:§r~:"t .. 3 5-:~--‘“““ w‘v-as, 51;..1 gua‘ovob‘- .34 cannd‘ s’ rd Via.» -h— L: general, the vfi-vc "V: + ‘0‘. 51“, w ' 4 Zubair Iqbal the various ranking indexes developed in this study. There is a significant and positive correlation for the developing countries, significant and negative correlation for the de- veloped countries. Our investigation, therefore, appears to show that, in general, the generalized tariff preference system will give greatest incentive to the establishment of industries in the developing countries in which they have greater compar- ative advantage. Furthermore, preferences will encourage the developed countries to import those commodities from the developing countries in which the former have least compara- tive advantage. Hence trade diversion will be minimized. 'PY' lfi‘ !’"" H Pr|.vH—1“ cu.“ VJ “though . purr- ‘\oo-v- [HP ‘j! v- ‘ H p- 3.“ .hu‘us a. . m in Part THE COMPARATIVE ADVANTAGE OF DEVELOPING COUNTRIES IN THE MANUFACTURING INDUSTRIES AND THE EFFECTS OF GENERALIZED TARIFF PREFERENCES BY Zubair Iqbal A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of SOCTOR OF PHILOSOPHY Department of Economics 1971 Airiser and C"‘-‘ .L‘Jo I...“ fl"; {No gV‘" ..:‘:a' au 'vva .4...“ " C 5‘»; : 2333210!) 0:. t”-.. ‘ Q "‘§‘ ‘~\n ,‘ A I ."~‘Luke to a 1". 33.3 to the '1' fcr .i. apcifl s :25. pass 1b is . The Si ACKNOWLEDGMENTS I wish to thank Professor M. Kreinin, my Academic Adviser and Chairman, Thesis Committee for his unfailing help, advice and encouragement at every stage in the pre- paration of this thesis. I would also like to thank Profes- sors V. Smith and W. P. Strassmann, members of my Thesis Committee, for their suggestions that helped me to improve the final draft. There is no way to fully express my gratitude to a large number of fellow graduate students and members of the Faculty in Marshall Hall whose intellectually stimulating company helped me to sharpen my ideas. My thanks to the Yale/Pakistan Project and the Ford Foundation for financial support without which this study could not have been possible. The sins of omission and commission are of course my own . ii TF;\' 0 , ...Juv y. & -.,.‘ $3.216! _‘ -‘.q..-;.H.vm .- . r t. ... .... _l t. .. I .l a 1 I w. .q H... Mi swucrrrxeteesxon pw. ”A. AJ .1 nu Pu PU . 7' Cu Wu LL ‘0 C; CH n.5,; .9.“ wsdufi¢ *ucww.rw.shnwmv »~ A: H.“ .1 r0 0 Fri o H .0 Mu. ?“ “V fin. Y.‘ y I. J NU PU 71‘ E T P. 2;; LBaPIEE..L, a. .I 1“” w... Ni he Ti. TABLE OF CONTENTS Chapter Page INTRODUCTION . . . . . . . . . . . . . . . . . . . . l I. TARIFF PREFERENCES AND THE PROBLEM OF ECONOMIC DEVELOPMENT . . . . . . . . . . . . . . . . . 3 Introduction . . . . . . . . . . . . . . . 3 Export/Import and Saving Gaps Projections and the Aid Requirements . . . . . . . . . S Policies to Fill the Residual Trade Gap . . 11 Import Substitution Policies . . . . . . . 11 Export Acceleration . . . . . . . . . . . . 12 Exports of Developing Countries . . . . . 12 Manufactured Exports and the Process of Development . . . . . . . . . . . . . . . . 15 Obstacles to the Manufactured Exports of Developing Countries . . . . . . . . . . . 17 Internal Obstacles . . . . . . . . . . . . 17 External Obstacles . . ... . . . . . . . . 18 Trade Preferences for Manufactured Exports of Developing Countries . . . . . . . . . . 20 General Preferences . . . . . . . . . . . . 22 II. CRITERIA OF COMPARATIVE ADVANTAGE . . . . . . 30 Introduction . . . . . . . . . . . . . . . 30 Criteria for Resource Allocation and Com- parative Advantage . . . . . . . . . . . 31 The Classical Criterion of Comparative Ad— vantage . . . . . . . . . . . . . . . . . . 31 Foreign Exchange Costs Minimization Criterion 33 Programming Criterion and Shadow Prices . . 35 World Price Criterion . . . . . . . . . . . 37 Domestic Price Criterion . . . . . . . . . 38 Growth Criteria . . . . . . . . . . . . . . 39 Factor Proportions Model . . . . . . . . . 42 Human Skills Model . . . . . . . . . . . . 49 Research and Development, Technology and the Scale Economy Effect . . . . . . . . . 53 "Revealed" Comparative Advantage . . . . . 54 Choice of Models . . . . . . . . . . . . . 55 iii n! ‘ I’:"Or 5.5“.“ V- - F... HAMTA\' \v- " \ bun-4.0V. -‘l. C'T o... Introdu. A ‘ ‘ DEIBCtl electi p \I \ “HGQQV‘ P- P b ‘. ‘kL'i 23-1 “6-. K ”v tro- 6. 75 A. Mumi . .3 C N ‘ find 7:” a; v.“ 2. Yu P we a... mi PF. nhv ‘mHA II. a C UH” Q» nu was: :nun t..b rt Chapter III. IV. VI. SELECTION OF COUNTRIES AND COMMODITIES . . Introduction . . . . . . . . . . . . . . . Selection of Countries . . . . . . . . . . Selection of Commodities . . . . . . . . . DERIVATION OF THE VARIABLES USED IN THE ANALYSIS . . . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . Effective Tariff Protection . . . . . . . . Estimation of Comparative Advantage Indexes Estimation of Capital/Labor and Human Skills Ratios . . . . . . . . . . . . . . . Capital/Labor Ratios . . . . . . . . . . Human Skills Ratios . . . . . . . . . . . Research and Development . . . . . . . . Aggregation of Effective Rates . . . . . . Conclusions . . . . . . . . . . . . . . . . EFFECTS OF GENERAL PREFERENCES ON DEVELOPING COUNTRIES WHEN COMPARATIVE ADVANTAGE IS ESTI- MATED BY THE FACTOR PROPORTIONS AND HUMAN SKILLS MODELS . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . The Factor Proportions Model and the Devel- oping Countries . . . . . . . . . . . . . . Statistical Tests and Results . . . . . . . Human Skills Model . . . . . . . . . . . . Statistical Tests and Results . . . . . . . Combining The Factor Proportions and Human Skills Models . . . . . . . . . . . . . . . Productivity Model . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . EFFECTS OF GENERAL PREFERENCES ON DEVELOPED COUNTRIES WHEN COMPARATIVE ADVANTAGE IS ES- TIMATED BY THE FACTOR PROPORTIONS AND HUMAN SKILLS MODELS . . . . . . . . . . . . . . . . The Factor Proportions Model . . . . . . . Regression Tests . . . . . . . . . . . . . Rank Correlation Tests . . . . . . . . . . Human Skills Model . . . . . . . . . . . . Regression Tests . . . . . . . . . . . . . Rank Correlation Tests . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . iv Page 61 61 62 65 69 69 69 74 78 78 85 94 94 95 99 99 99 104 106 106 109 111 112 115 116 117 118 119 119 121 121 ”z'ra bub-I vv D gvvv I... I IT‘ECU J‘ba‘IV‘Il VflVI'JC (V: vo-ouv VA on m own: .\ “fin“"fl'fi‘filflu ~".“-1" ‘I r...» “bi“. \y. ‘lv . :Ft1"“ ~3 h,_ COPC1“C Unob-‘d v.9 cvv‘.|“-.' ‘ u... ‘ . . “—4 v v~._“_. ‘ r “"-~ .‘ ""‘ 51' n. -.-“U-“ . I c MI I ‘4 ‘7! “ :-"‘-. §.~.‘ka II o u.‘ d Y'f‘ (~— c- , ,.~ .d~.J'-c:4r‘rvhv." 1“ Chapter VII. RESEARCH AND DEVELOPMENT, TECHNOLOGY, ECO- NOMIES OF SCALE AND THE GENERALIZED TARIFF PREFERENCES O O O O O O O O O C O O O O O O 0 Introduction . . . . . . . R & D and the Structure 0 Trade . . . . . Neo—Factor-Proportions Approach . . . . . . Market Scale Approach . . . . . . . . . . . Product Cycle Approach . . . . . . . . . Economies of Scale, Comparative Advantage and Generalized Preferences . . . . . . . . M048 1 O O C O I D O O O I O O O O O O O O 0 Estimates of the Economies of Scale . . . . Conclusion . . . . . . . . . . . . . . . VIII. SUMMARY AND CONCLUSIONS . . . . . . . . . . . APPENDIX I. "REVEALED" COMPARATIVE ADVANTAGE AND APPENDIX II. BIBLIOGRAPHY THE GENERALIZED PREFERENCES . . . . . Introduction . . . . . . . . . . . . Estimation of "Revealed" Comparative Advantage . . . . . . . . . . . . . The Export Performance Index . . . . Empirical Results . . . . . . . . . Conclusion . . . . . . . . . . . . . P RIMARY DATA 0 O O O C O O O O O C O O Page 124 124 124 125 127 128 130 132 136 141 146 153 153 154 155 158 160 163 203 1-3 III-l III-2 V?’ c.1-3 1"? _ 1 IV- 2 f'h Put 4' No. III-l III-2 III-3 IV-1 IV-7 LIST OF TABLES Projections of Trade Gap of Developing Countries by Regions, 1975 . . . . . . . . Projections of the Trade Gap of Developing Countries (Total) . . . . . . . . . . . . Exports of Developing Countries by Commo- dity Groups and Projected Growth Rates of Exports . . . . . . . . . . . . . . . . . Critical Values Assigned to Economic Indicators 0 O O I O I O O O O O O O O O 0 Exports of Manufactures and Semi—Manufac- tures of Selected Developing Countries, 1966 O O O O O O O I O O O O O O O I O O 0 Selected Sample of Manufactures . . . . . Weighted Effective Rates of Selected Commo- dities . . . . . . . . . . . . . . . . . . Total Capital and Labor Requirements of Manufacturing Industries in Mexico (1960) Total Capital and Labor Requirements of Pakistan and India . . . . . . . . . . . . Taiwan's Total Labor and Capital Require- ments 0 O C O O O O O O O O O O O O O O 0 Total Requirements of Capital, Labor and Human Skills in the United Kingdom (1960) Total Capital and Labor Requirements of Various Industries Per $1 Million of Out- put (1963) in the U. S. A. . . . . . . . . Total Requirements of Human Skills of Manufacturing Industries of Mexico (1960) vi Page 10 14 63 64 66 75 80 81 82 83 84 88 50. cf (1‘ Prod- Ifidgs No. IV-8 IV-9 IV-10 IV-ll VI-l VI-2 VII-1 A-I-l A-II-l A-II-Z A-II-3 A-II-4 A-II-S A-II-6 A-II-7 Page Labor Skills Requirements of Taiwan . . . 89 Labor Skills Requirements of Pakistan and India . . . . . . . . . . . . . . . . 90 Total Skill Requirements Per Million Dollars of Output in U. S. Industries (1963) . . . 91 Productivity Indexes of Manufacturing Industries I O O O O O O O I I O O O O O O 93 Results of Rank Correlation Test--The Factor Proportions Model . . . . . . . . . . . . 107 Results of Rank Correlation Test--Human Skills Model . . . . . . . . . . . . . . . 110 Results of Rank Correlation Test--The Factor PrOportions Model . . . . . . . . . 118 Results of Rank Correlation Test--A1ternative Human Skills Indexes . . . . . . . . . . . 122 Results of Correlation Tests on Effective Protection Rates and the R & D Index . . . 127 Results of Country-by-Country Tests for Correlation Between Effective Tariff Rates in Developed Countries and Export Per- formance Indexes of Developing Countries . 159 Projections of Savings and Export/Import Gaps of Developing Countries . . . . . . . 164 Per Capita GDP at Factor Costs, 1967 . . . 167 Size of the Manufacturing Sector in the Developing Countries . . . . . . . . . . . 168 Percentage Share of Manufacturing Exports to DMEC in Total Exports . . . . . . . . . 169 Value of Manufactured Exports (Total - A) of Developing Countries to DMEC . . . . . 170 Annual Trend Rate of Growth of Manufactured Exports of Developing Countries to DMEC . 171 Imports of Selected Manufactured Products By Developed Market Economy Countries . . 173 vii IIIIII I . v 0. . 4 I v » c It r“ t a. CV "L t t K 11 To a . .E < .1 y .- a .1 C. 4.. 3 PJ 1. .1 Cu L .L I n.» .. :1 m. rw. r. L. r e C. r :13 .1. Y; .1. Au . «nm m. at +. .u a «d ”u 0. up“ QC 0 a pH; 7.1” «U 0 . nu . v 1 3. ab AU v. u C no nu. t C C. 7.... ‘IE :u f a u Yb 7.. L“ V1. Q» N “LL C 9 1| 6 . .. 1.1 J1 7. .l. l l n n "I“ . . . . 1.1 7... - ~ .. . YM. o ,. vi. I H “In In. I.“ ‘3 1.. o o c . a '- ’hu ‘- 7““ No. A-II-8 A-II-9 A-II-lO A-II-ll A-II-12 A-II-l3 A-II‘14 A-II-lS A-II-16 Page Nominal and Effective Tariffs in the Major Industrial Countries on Selected Commodities Before and After Kennedy Round (KR) o o o o o o o o o o o o o o o o 179 Japan's Capital and Labor Coefficients, 1954 .185 Capital Coefficients (K/O) Per $1 Million Output and the Labor Coefficients (L/O) per $1 Million Output (In Man Years) in the U. S. A. . . . . . . . . . . . . . . . 186 Capital, Labor, and Skills Coefficients for the United Kingdom (1960) . . . . . . 189 Japan's Skill Coefficients (Man Years) 1950 . . . . . . . . . . . . . . . . . . . 191 Skilled and Unskilled Labor Coefficients for U. S. Industries per One Million Dollar Output (1960) . . . . . . . . . . . 193 U. S. Occupational Skill Requirements By Industry in 1960 for Manufacturing . . . . 195 Hufbauer's Scale Economies Indices for Manufacturing Industries According to SITC Classification System . . . . . . . . 198 "Revealed" Comparative Advantage Index Estimates for Selected Developing Countries (1960-61--1967-68) . . . . . . . 201 viii 1". JV! VII-l VII-2 LIST OF FIGURES Factor Intensity Hypothesis . Factor Intensity Reversal Hypothesis The "Required" and "Justified" Functions The Scale Effect of Preferences ix Page 47 47 134 134 o . ”'8. A Hw‘ F ‘ : u.£:,aqhyh-“ ‘3 "h“nd- . _. out-v»: VU.‘.'§- - V ‘ ”an N. I’d: 2b odyvu H1 ”‘4‘. ' ~ .‘ ‘ ‘ ‘ F'B'Y“ “LG V‘" *9“. -.4I.u., “I 1’... . 4 :‘:.“~+ ‘uv‘n ‘ an ‘oal: V. &‘ ‘V V b ‘ A Y A ' W“ H 1.:5 15 ; «C u 'A . ‘4 fip"n-~." a» VUAID‘CI“JF.( o 3-4:“ +. A,“ ' ‘ ”1‘ a . 'V . kn .I‘ES 1’: u c. Y H s‘ t“?'~ ‘ ._ H Nf‘ ‘ ‘ Q . ‘Q '»,': § s‘.‘ ‘. ‘ v.‘ . ‘ v. ‘ a- -. . ‘ “I s.: h ‘ ' “rfiv‘; ' “I‘ ‘ ~A ‘~‘%v .~.A‘:.~ a1 “rt: ~~ ‘ ‘. “— . . ifi‘ v.1“.Lr INTRODUCTION The manufactured exports of developing countries to developed countries are subject to various restrictions like quotas, administrative controls and the tariff duties in the latter countries. Since the nominal tariff rates are esca- lated by the stage of fabrication, the effective protection provided by the tariff structure in the developed countries against imports of simple manufactures from developing coun- tries is much higher than the nominal rates would suggest. The consequence of this type of tariff structure is to ad- versely affect the capability of developing countries as potential or actual exporters of manufactures that are of special interest to them. One way to alleviate the position of developing countries is to remove the trade barriers against their manu- factured exports in advanced countries while maintaining them against the imports from other developed countries, i.e., grant preferential treatment to their exports. A general preference system eliminating tariffs in the developed countries on manufactured imports from developing countries, without limitation on volume, will grant the greatest pref- erences to the commodities that are subject to the highest effective protection in the developed countries. This will sin-'9 :reatest 1: ing car-.tries c highest effect; tries. Then; if. only if ti; $73.1 19$ 11 (3'36: 'C.‘?;v‘v . _ . “wast-El! 10‘“ 0...,‘1,’ h .. jfln ."U“' 11-3..er . v 11’“ “a . .m, 0395 i . ::'3‘fiv .h“"-.v\, I Q 2‘; ‘ u; 4. u 5:18 dc“. \w'fe 1"... N36 V 7° alga-a?“ gyi‘Fe greatest incentive to the establishment in the develop- jJ1g; countries of those industries that are subject to the highest effective protective duties in the developed coun- tries. The generalized preferences will, therefore, be use- fhil. only if these are the industries in which developing countries have relatively greater, and developed countries relxatively lower, comparative advantage. If this condition does not hold, that is, if the industries subject to rela- tivwely higher effective protection in the developed countries are: the ones in which developing countries do not have com- paJLative advantage, then general preferences will not promote thee industrial development of the developing countries. They magr actually worsen the resource allocation and retard the PIK>cess of economic growth in the developing countries. The present study seeks to find out whether general- izena preferences will promote the establishment and growth of jJldustries in developing countries in which they have greater ‘anparative advantage and developed countries relatively 1orts of developing countries would tend to diverge, on the huissis of past experience, if these countries sought to achieve tales overall growth rates indicated. The trade gap thus pro- 'Vidfles an indication of the magnitude of policy adjustments required if the growth objectives of developing countries are to be realized. KPORT/IMPORT AND SAVINGS GAPS PROJEC- EQNS AND THE AID REQUIREMENTS The most comprehensive projections of savings and ‘Export/import gaps of developing countries have been made by the United Nations Conference on Trade and Development (UNCTAD) in 1968.5 A country-by-country study was undertaken for thirty-seven countries of Latin America, Africa and Asia to Project gaps for 1975 at two different target growth rates. n... v A v' ”*0 exrc" 1.: I V ‘ gfiov- Qrar- H ‘F v‘v- ~U G. ‘C «A u (I) “R n . UQCCD. . L ‘n l.’ 0' A A 'J-K—ESS r“ u 'A 1". ~. . n :‘ut 1a; a . .:¥na+ .‘_ '3" 3:7, .3 . ( I t n; (1‘ (f :1": ‘ fr‘ «P. . m: u'v.‘ A I “Jr C b ‘ V. ‘3" d3”e “yo U D :c \I.e r0 r‘ I to U .‘ I. u” ‘.F “sac?“ ably- k b“‘1 . . . ‘& CE (I ‘1 'I 113 E‘flw ., P‘ m . 9“; \U H 5:2. “-~’&' r. This investigation is reproduced in Table 1, Appendix II. The export/import gap and the savings gap as defined here are comparablein terms of national income accounting concepts. These estimates of gap projections are carried out for a low and a high target growth rate. The low rate, con- sidered to be substantially below the absorptive capacity of developing countries, represents something between the high and low rates in the period 1950-1965, averaging out at 5.2 Per cent for all developing countries. The "high" rate is con- sidered somewhat higher than the absorptive capacity, avera- ging out to about 6.1 per cent for all the developing coun- tI‘ies. Table 1, Appendix II, shows that twelve countries have an excess of exports over imports, i.e., a "negative" export/ i1“port gap at the low target growth rate, but for the high taarget growth rate, there are only six such countries. Out of the twelve countries showing an excess of exports over im- POrts-at the low rate [in 1975] , only Brazil and the Philip- Pines have more than 15 per cent of their exports in manufac- tures. The rest of the countries either export only primary gOods, or non—ferrous metals or other raw materials deirved from domestically available natural resources. On the high rate no country with more than 15 per cent of exports in manufactures shows an excess of exports over imports. This Will be discussed further in Chapter III of this study. As for savings gap projections, fourteen countries exhibit surpluses under the "low," and four under the "high" assumption- I ‘nfi?’" rtolu‘ 'OI z N '""an'sg QY' io'v‘vvv 4.: 4 J l a a. ‘ In ~~A‘rn r5 u.ui-u-l .... 4 A“. A‘ t}; V+ V; 3"" 4“ . . vu~ O I v (n o; I - (I) ,u a ‘1' v. r F“V¢n'-o be.--._u “V . V F + u. ‘i; 1.“ Q E :"::‘ 1 1 “‘-L '- .2. ~ , . :+Vs “n\' 3": b“ 5;, “4.1 . n c... a .3 ~~ :. '»~_‘ “ ‘ 9“ AP‘Q; '1' A "‘31:- - | i"? 4‘ ~,. “ .1135? R u . h . “A ‘.‘.‘\.-Q (:6.- .. 1‘5: . ‘. '«ZC . . ~ a-A .. “" .-v~‘ ., .‘u fl.‘ ‘ § . ~~ S‘ts" ‘§ r.” V. .. I t t. \- P” v“ I ‘ v‘f‘" “.J . A. *‘L‘f "b N“ wt: Q \‘N. Another important result that can be drawn from fradale 1, Appendix II, is that there is a tendency for the projected export/import gap to dominate the savings gap. Ex- caltmding the cases where both gaps are "negative," twenty-one orrt. of thirty countries show an export/import gap larger than tale: savings gap under the "low" projection and nineteen out «of ‘thirty-five under the "high" assumption. The dominance of the export/import gap would become HKDIKB significant if payment of interest, profits and dividends féfllling due on past loans, etc., were added to export and im- Pert.of goods and services. This would give us the trade gap 98 timates; these are usually more important as a policy guide tlléan the export/import gap. But no long run or even medium teirmestimates on the trade gap and debt requirements are axredlable for most of the countries. Even for the existing SILIucture of debt and repayment, there is very little informa- tidon on an individual country basis. The problem is compli- cfitted further because factor income payments in 1975 will CKDnsist of service payments not only on initial debt but also cu) new debt incurred during the entire period up to 1975 and lNance depend upon the assumptions made regarding the term, £3ize and the composition of new inflows needed to fill the ‘gap during the entire projection period. However, on the basis of some rough assumptions regarding the terms and compo- sition of new inflows, the UNCTAD Secretariat has prepared Projections on the trade gap on regional levels. In order to determine factor income payments, it was assumed that only =""=*=s or a vb!” i bv P 1 g I? or“ A“ r v. “A. tra‘: ' . H ‘ ”9 rm 5. ..J.. ‘V VI: 1:5”. V’AA *f“ .. .1...& LC... 5“ . 3 ‘ ‘ Lei‘Of‘. as .‘n. ‘ i751" 1‘ tn 4;. iv?“ I...“ ‘ . “ ‘ .5 v‘ ‘ 7‘: “I ‘1' 9 § ‘~ . “:‘.A‘ ’5 “7“:l'ad-CC ‘ I. J 4r " ICE 1‘1“. ‘1 I ~ ‘ l “"‘v3 1 1,.1‘ U ; ‘f‘. ““V“‘ I "Rx. ‘ I‘I '- fi‘ 2‘ I ~\ ‘ ‘ 'u '0‘ p, " than- -A“: \ .J' ‘ . \|I_‘S ‘ C Q.- ‘ ~ "~\s-. K H“: -“\\‘u one-half of the projected trade gap will be filled by public capital inflows if all aid-giving countries contribute one per cent of their gross national product.7 The trade gap estimates on a regional basis are given in Table I-l. For the developing countries as a whole the projection of the trade gap for 1975 amounts to $15.4 billion on the low growth rate assumption and $24.2 billion on the high assump- tion. For the low growth rate assumption the trade gap is estimated to be composed of an export/import gap of $3.4 billion and factor income payments of $12.0 billion; while for the high growth rate assumption, the export/import gap is ex- pected to be $10.0 billion, and net factor income payments amount to $14.2 billion. This is shown in Table I-2. The trade gap figures indicate that there is substantial increase in the trade gap by (base year) level even on the low growth 9ap>will, however, be much larger if the fail.to contribute one per cent of their assistance . going to be a very 1975 over the 1963 rate assumption. The aid-giving countries income to foreign If all developed countries contribute one per cent of thfiir respective GNP in public capital flows and about $4 billion in private capital movements totaling about $17.5 bi-11ion, on the high target rate of growth assumption, there Will.be a residual trade gap of over $7.5 billion to be covered by means other than those giving rise to factor income pay- ments if the target rate of growth necessary for early self— Sustaining growth is to be realized. ul'l‘l‘I‘II 1|. I llll‘l <.u-mu.~z< Z—._.<.~ 0P9H .uluznuunuv.: .‘l’lu‘ . I‘lll‘ I1: 1 I‘ll‘ II .I|I n l. I 'l "fll \f $.- <_Tu< ll‘.""|.|\l Av...anvu.h.~ naafiud‘ Ufiw ”4.”; - v-.—.z-n »- v “#2 ~ .~Av.—h.~>h.~a— Turk away .H CA~ p-AV C... - ...~.-‘<.~. .n/Nuv “gas/\vrT LC :2: .. a a my: O.~.F V MHZA VIN nN§7,Va\- \dfi H-a~ .mv .mcasouuon an omaaam ma mam woman man mo wom umcu coaumssmmm map so muaoammo waammx omuomnoam map Scum meamaam ammo 30: wsu mm aamB mm Damp awauaca map mcaoa>amm mo mmumEHuwm moosaocH .m .xuow 3mz .z .D .mmma .umaumumaomm Q¢BOZD =.mmauucsoo mcamo Iam>mo mo mommz amuammu paw muommmonm momma: ca mm magma Scam omummom ma HIH magma mamm mmmm mum omom ommm mmmm «mam mmom amm mum mamas .m ammv mmmm mmma mmmm mmmm momm meow mama oom mmucmsmmm OEOOCH HODDMh .w mmmm mmma oam- moma mm mam- mmam mmm ama mam uu0m5a\uuomxm .m mmmm mmmm amma mmmm mmmm mama mmmm maom omma mmanama>ca an ommma mmmma mmvm mmmmm mmomm ommma momma mmmaa mmmm mmamamoeeoo am meH>Hmm 0G0 mmomm mmaom mmmaa omomm mmomm momma mmama mmmma mmom mmooo mo muuocsa .N mmmm mmmm moam meme moms aoom mmmm mmmm moma mmanama>ca an mmmma ammma «mama Nmmom mmmmm «mesa mmama Namaa mmmm mmaaamoEEOUAm mmaom mmmma mmmma mmmmm ammam mmoma mmoma mmmma mmmm mmoa>umm mam moooo mo manomxm .H sown 30a mmamI 30a seem. 30a mmma mmma mmma mmma mmma mmma moammzm 2aama mamm moamm< mnma AmmOflHm ommH um mHmHHOQ MO mCOwHHHZV .mZOHOmm Mm mmHmBZDOU OZHmOAm>mQ m0 mdw madmfi ho mZOHfiumhomm «H IH mqmdfi -.1. 99" L ‘ .. 4". EX:orts of a) C.—~,..:; --- «A..- F'-‘ o D)Serv;ceg 1, Y _ 4' iT'Cl’ts of \ m CczxodL‘ 1‘ w ‘ “link: .: 10 TABLE I-2* PROJECTIONS OF THE TRADE GAP OF DEVELOPING COUNTRIES (TOT AL) (Billions of Dollars at 1960 Prices) 1963 1975 Low High 1. Exports of Goods and Services 37.6 67.4 73.5 a) Commodities 31.8 56.6 61.9 b) Services (Invisibles) 5.8 10.8 11.6 2. Imports of Goods and Services 37.5 70.9 83.5 a) Commodities 31.9 59.7 70.0 b) Invisibles 5.5 11.2 13.5 3. Export/Import Gap -0.1 3.4 10.0 4. Factor Income Paymentsa 4.9 12.0 14.2 5. Trade Gap 4.8 15.4 24.2 * Table I-2 is derived from Table 22, Trade Prospects and Capital Needs of Developing Countries, UNCTAD Secretariat, New YOfk: U. N., 1968, p. 43. a-The method used in estimating the factor income payments is given in Annex III of Trade Prospects and Capital Needs of Developing Countries, p. 68. DC A. . A n, . 1r . M I; F~L Hr“ Y: o n V v ,. CI d .rw‘” O 5.. 5.. \ F a tr v' ‘0- .. .4.» a. ‘V- s; PPIT‘V ‘Vv..b‘ ll POLICIES TO FILL THE RESIDUAL TRADE GAP The residual export/import and associated trade gap can be met by two groups of policies: (a) Policies to increase import substitution, and (b) Policies to accelerate exports. To these policies may be added the acceleration of private foreign investments. IMPORT SUBSTITUTION POLICIES Import substitution, essentially internal to the de- veloping countries, has been carried on for quite some time in the more advanced of the developing countries. Import substituting industrialization is usually the first step in the developmental effort of the less developed countries be- cause the existence of "imports still provide the safest, most incontrovertible proof that the market is there."9 However, as a development strategy it has certain severe limitations. It is limited by the size of the domestic market because a narrow domestic market may not permit the scale of operations in manufacturing that is compatible with the technological requirements for efficiency. Even an extension of the market due to economic integration among the developing countries may not allow import substitution to become a long run policy. Furthermore, import substitution tends to work towards autarkic tendencies in the development process, eliminating international competition and perpetuating inefficiency so that domestic infant industry tends to stay infant forever. In some cases inefficiency has reached such proportions that r-v ' 0 Ab L) .1 . ’ . p'..r‘|c,x l Erlld‘ v 5.1". mustu ‘ “5"."- 2' "Yf‘ 0 v3v’ l“ t,‘ 4 ‘ u . ire r - r4 "“vs Vaua . 1 ”ma" ' 'l—' -‘ 'Iflr‘. AI mun" 4-v..- Fu L4... ' Q va. 9: I "C‘svfi‘ " \L‘\y_.- ‘ 9": "Q‘\ ~ “‘1 12 the net subsidy received through tariff and non-tariff pro- tection exceeds the total value added, and the operation of such industries subtracts from, instead of adding to, the national product.10 Import substitution, therefore, cannot be expected to fill the dominant trade gap for most of the developing countries.11 EXPORT ACCELERATION Export acceleration appears to be the only means avail- able to bridge the dominant residual trade gap. Export earnings depend upon the volume as well as the price of exports. There- fore, it is necessary to either increase volume, or price, or both. Unlike import substitution, acceleration of exports of developing countries involves the policies of both developed as well as the developing countries. EXPORTS OF DEVELOPING COUNTRIES The export trade of developing countries is largely dominated by primary products and until recently, their manu- factured exports to developed countries played an almost negligible role. The demand for foodstuffs and staple con- sumer goods is generally much less responsive to income growth than the demand for industrial products and services. And technological progress leads to the increasing substitution of synthetics for natural materials. These factors underly the United Nations' evaluation of prospects of individual markets suggesting that the prices of traditional primary products exported by developing countries may well be lower V I AIJV‘H {Cr tr‘. blank“ p . .. _ r . .r“ f. w . 4 t . . . v fid h to 1‘ . 1 . . vb » . . . . a; Fv 5 C it e h.V w 3 F A F as I mt ”a P . r .4 .J .U. .l a -5 L. m .0 .0 .I e E + . .1 S f m0 P... v. E C E . Au .0 A G P. .. t S t Ru 6 N .. r; .b C C. 1 l 1-. ab 2. . +u Nd: \ Nu m a O a . VJ Flu “U r‘ Q u run x» . n... 1‘. C. .Fd Du Pu 111 C . Ax. a -7. fa a .p . Wu. S Cw Cw v . A: J . u as . vi . a *»H u .01. 6 I4 .W it .\~ . k A»; 5r” fin. «1.. w.. pm . NJ 7. r. a; t ,5. an“ ml «run. NJ L 4 Mr“. M 34.1 o a 4 “AM “4* Wm . UAW pk”. ”fun. mm M n5 o M 5. My u! i O ‘ Hi.“ u§.i ... 13 in 1975 than in 1960.12 Based upon the income elasticity of demand for the exports of developing countries in the im- porting developed countries, the projections of values and gorwth rates of different kinds of exports are given in Table I-3. As the table shows, the growth rate of traditional pri- mary goods exports is expected to be far below the overall annual compound rate of growth of exports. A slight reduction in the growth rate of developed market economies will bring about a further decline in the projected growth rate of ex— ports of developing countries, especially for the primary ex- ports as indicated in the UNCTAD study.13 From these projec- tions it appears that for the developing countries as a whole it would be unrealistic to look for any improvement from addi— tional exports of primary commodities unless there were reasons to expect a major shift in the import policies of developed countries with respect to primary products.14 The tariff cuts on primary products negotiated in the Kennedy Round were modest and it is known that in certain commodi- ties the policy intentions of some developed countries will increase rather than diminish the tendency toward self- sufficiency--notab1y in the case of sugar.15 On the whole, the developing countries cannot pin any hopes on the expansion of their primary goods exports. In fact the U. N. projections of the performance of primary exports of developing countries are too optimistic. On the other hand, the exports of manufactured and semi-manufactured goods, though very small as a percentage of nfi'fifijm" “4,13: -) “I he: a), -‘9 ~v.:‘U... - sC I 5;...C “\4"“~ I ‘ z1-= ‘f‘N‘o "‘31.: 14 * TABLE I-3 EXPORTS OF DEVELOPING COUNTRIES BY COMMODITY GROUPS AND PROJECTED GROWTH RATES OF EXPORTS Value of Exports Annual Compound (Millions of Dollars Growth Rates of at 1960 Prices) Exports (Percen- tages) 1960 1975 1960—1975 Item Low High Low High 1. Foodstuffs 9262 14160 15644 2.8 3.5 2. Agricultural raw materials 5064 6389 7144 1.6 2.3 3. Non-fuel minerals and metals 3176 5976 6413 4.3 4.8 TOTAL (1, 2, 3) 17502 26525 29201 2.8 3.5 4. Manufactured goods 2416 7916 9086 8.3 9.3 5. Fuels 7422 22220 23620 7.5 8.0 TOTAL (1—5) 27340 56661 61907 5.0 5.6 * Table I-3 was derived from Tables 3, 4 in Trade Prospects andlCapital Rquirements of Developing_Countries, Vol. I, UNCTAD Secretariat, New York: U. N., 1968. .J nflfi“ b o. "00 n! F" ~p'u-1‘ n'f‘v F ' . a..uu.¢a “a- b . .C :13"?! .d‘.._- ”'w 4-- i.-."r-:~ ~ -F .“v‘flr‘va, ." .1", F ”a ‘ 's‘... -r~ "v n’.~ . a .‘s...:.v‘v ' .- a” \ F '- U F ‘~~QV+ ‘g § . Fun: ”U 15 total exports of developing countries, have shown a very signi~ ficant acceleration over the last decade and the trend is likely to continue as shown by the projections for 1975. While the annual world trade in manufactures has been growing at the rate of around 8 per cent, the manufactured exports of developing countries have been growing at an average annual rate of 9 per cent. 80 the relative share of developing countries might be expected to increase. This is so even when most of the exports of manufactures from developing countries are subject to quantitative and/or tariff restrictions in the developed countries. If these restrictions are eased, then the actual performance might be much better than the projected rate of growth of manufactured exports. Where public and private foreign capital flows are un- able to fill the dominant trade gap, where further import sub— stitution would be acutely harmful to the economy, and where primary exports are not expected to grow significantly, the support for accelerating the exports of manufactures from devel- oping countries tends to become very obvious and natural. This has become "conventional wisdom," as Professor Raymond Vernon calls it, to stress the role of rapid expansion of manufactured exports in speeding up the process of economic growth.16 Manufactured exports and the process of development. The development of manufactured exports in particular and all ex— ports in general has a dual function; one of which is related to the structure while the other is related to the general level of industrial development-~both also are, of course, closely 1 +‘Av- ~.vuu\.(,“vxu A bit“ Q "Vhflv A rwv RV V. _C r». .5. 3. M "\L “H PM FA .4 C\ ‘Id pp» f\ C 3 aid 7 MW .5 .... L C u." C .l E .C d u .. r C .v v LY“ s... H“ v rd mid Cu .u . u q I ~ ¢ ‘[ s Q.» \n .u .HL r C» Q.» .1 n 7 «D J} as. a: an a ”an «d t Cw a L n tn A: DJ 1‘ . n u t H.“ P . .x. .3 av AC .3 - Fm 9» .$ ‘Hq ‘h-.. C a h» I» QC .. . a: e +5 .3 a. .. . Z c. 2 v. 71 L . t... J- C. 2.... .. . n. .3 .. 1 Y. .u in -f. P. we. any we a.“ s. C\ Q . «L .4 1L .. . . . 4 . 1. .1. T. u}. 2.. s . by .x n “u .3 C N m. u u s . .5 2 . s u .2 L . .1 A: 5 . 2. .U «4‘ n8. \< A u . .r... 9:; r:. 46% 16 related to each other. Industrialization requires the establishment of a complex and highly capital intensive structure of production that will permit a diversification of production and eXports. The types of goods demanded are usually not the same for which production capacity exists. There is a structural imbalance making the developing countries greatly dependent upon imports. This structural dependence on imports may be temporary and for a very large developing country, import-substituting industrial- ization may help in diversifying industrial structure. But this system has well known limitations, reference to which has already been made. The dependence on imports may require con- tinuously growing export earnings since reliance on foreign aid can be considered a temporary solution at best. Secondly, exports accelerate the process of industri- alization both by raising its level and influencing its struc- ture. While the first function could even be fulfilled by the expansion of primary goods exports, the second function can be realized only if the structure of exports is diversified and the share of manufactured exports grows rapidly. This would help in providing for a growth of industrial structure much wider in scope and depth than the one based exclusively on the domestic market. Dependence on a small domestic mar- ket for the growth of industrial development does not permit efficient production for technological reasons. Raul Prebisch has therefore stated that a continuous dependence on a small domestic market has created «ma-‘ac‘~w . gooa facturec because I I I LVNF F; u.V.l Ma. ~ 1 wcu.c e: sible t custrla con-‘1’: ‘r— ‘V‘éd AA . u . 1.n« 1 M ‘ RyA'. ‘h ‘cJ. L. :5.” '- "Nw’ bVoaJi..‘C + k .A —F6 --».. . n5 v- E) gnu P ‘4 A; A 'st’h‘o ‘ ‘9b' ”Y7 A Hy 4 q Fv. bv'uh‘Y-‘a “ *4. ‘Cs s. *3 H‘ lx'ycr v1 7" run ‘4 Vhtq ‘ - “‘z‘lr h vJ . ‘ rE=»P g . :4 J: 1”.» 841*“ 1 "NE l7 . . . a real vicious circle as regards exports of manu- factured goods. The exports encounter great difficulties because internal costs are high, and internal costs are high because, among other reasons, the exports which would enlarge the markets are lacking. Had it been pos- sible to develop industrial exports, the process of in- dustrialization would have been more economical, for it would have made possible the international division of labor in manufacturing. It is this vicious circle which limits the ability of devel- oping countries to carry out meaningful industrialization and growth. OBSTACLES TO THE MANUFACTURED EXPORTS OF DEVELOPING COUNTRIES There are two main obstacles confronting the develop- ing countries in the expansion of their exports of manufac- tured goods--internal and external. (a) Internal Obstacles. The most obvious domestic economic trends and policies which appear to work to the detri- ment of export performance of the developing countries in manufactured goods are inflation and overvalued currencies. Inflation causes balance of payments difficulties and ineffi- ciency by effectively keeping the price levels in developing countries above the world market level. The same is the effect of overvalued currencies coupled with unwillingness to devalue. There is a "structuralist" View which suggests that inflation and overvalued currencies are not the basic obstacles to exports of manufactures, but rather the outcome of a severe structural disequilibrium that exists in these countries which is responsible for limiting the growth of their manufactured 19 . . exports. The factors of production are not mobile, factor .' '- V‘ 2: '" a a~ . (I) a'fi'g'ti C b'di' P‘ . n ”'9“? c b‘VOVOOV (I) o- :1 Fr"- v..A‘ bx“ ‘ . 'th‘xn. Vouuud» 'v' . T. -“"“v\ 4 ‘Ovvv 4.. “A. b.‘:“_ 5‘ . 18 prices are out of line with their relative endowments,the economic system is monopolistic and the institutions are inef- ficient and hard to change. These factors create serious supply inelasticities. Therefore, even if a country has poten- tial comparative advantage in the production of a manufactured product, it may not be realized due to structural rigidities. The structuralist View has led some economists to discount the absolute disparity in prices due to inflation and overvalued currencies as the basic reason for slow growth of manufactured exports of developing countries.20 They be— lieve that devaluation alone will not be able to eliminate the obstacles in the way of their manufactured exports because structural flexibility is lacking.21 According to this asser- tion, inflationary policies cannot be discontinued, at least for the time being, because stagnation without export promo- tion will be the only alternative to inflation. A continuing, inward-looking import substituting industrialization without any export horizon is another reason for slow growth of manufactured exports of developing coun- tries. It has also been emphasized that the underdeveloped countries suffer because their entrepreneurs have no informa- tion and knowledge of the existence of a potential export market abroad, or that the level of acceptable risk on the demand side in exports may be too big for them.22 (b) External Obstacles. Perhaps the most talked about external obstacle is the protective barriers imposed by devel- oped countries against the import of manufactures. These $:y\‘ {in ta'A v“.... \ruu l I . hnc‘r1 are nc..»53~1 ‘ . .n A ray ‘ 5-313, {.mt. .4 n1, v~ . , Lost: auu VA~ 1’" ,‘ irri- . .e O; :J.'v “ ”r; ...,._' ‘lv “ . -nac‘ - - "h&~v‘ .- ovV-i'e rat- 5‘ \— WAR;- b..'L\ c“::: I Q Inn-A, 9 ~ .. Ara a.._‘ ~(‘V n, H "vu b‘ «a. . < - -nt, "" .35: t???“ Hg . . .1 CF A-“":l"’c‘"- ». ‘ .‘ ‘1. g h“ . w ‘-A. -._= .p. ,. «C:~QS. ‘ .. h F» a R .‘::‘Q:‘ I I‘VL l? O 9.‘ L .,"‘v~ \‘l LA. “- r “I A P" ‘J . 1 ‘ \"Ffi wvr‘; 1'“- ‘ch4V Vt . :hn‘ ‘\ l9 tariffs tend to be escalated by the stage of fabrication, and are noticeably higher in some categories of products than in others, particularly in labor intensive products like tex- tiles and clothing. The escalation means that the effective rate of protection on the final product is far above the nominal rate assuring the import competing producers in the developed countries a much higher effective protection than the nominal rates would suggest.23 Therefore, it is the ef- fective rate that should be considered as the measure of res- trictiveness imposed by the structure of tariff rates in the developed countries on manufactured imports from the develop- ing countries. This tends to keep the potential manufactured exports of developing countries out of the markets of the developed countries where the demand for such products is the highest.24 Apart from this, there is a tendency for the ef- fective tariff rates to rise with the degree of labor- intensiveness as observed by Professor Hal B. Lary.25 Although substantial tariff cuts were successfully ne- gotiated in the Kennedy Round over a wide range of products, the deepest concessions made were concentrated in the areas of greatest interest to the industrial nations (such as sophis- ticated chemicals, machinery, and transport equipment) while tariff reductions in the labor-intensive products were rela- tively small. This left the structure of tariffs after the Kennedy Round in the major develOped economies biased against labor-intensive goods. Similarly, the Kennedy Round left the StruCture of effective tariff rates on goods of export interest AC .Hv «flu Au 1" :‘r ..-n-' b- p‘.»" F' v‘a -"‘". VH- 1|. 2. v. .-- :.:Py--' { ~ I“; “Q.~4 a A P““—y ' U ' ‘ "ot..“ ‘ a... \& ‘U - c...‘ 20 to developing countries more or less unchanged. Besides tariffs, quantitative restrictions and ad- ministrative controls constitute important trade barriers in some areas. With respect to a number of cotton textile and clothing items in particular, it may well be considered that quotas in the framework of the International Long Term Cotton Textile Arrangement, rather than tariffs, provide the real protection for the domestic cotton industry of the de- 27 veloped market economies. TRADE PREFERENCES FOR MANUFACTURED EXPORTS OF DEVELOPING COUNTRIES One way to alleviate the position of developing coun— tries is to remove the trade barriers against their manufac- tured exports in advanced countries. Besides increased export earnings, the removal of trade barriers facing developing countries would help to promote an export-oriented method of industrialization. Additionally, freer access to developed markets would encourage a larger proportion of future invest- ments in processing industries to be made in developing coun- tries close to the source of raw materials with a view to exporting the manufactured goods, rather than raw materials, to developed countries. However, the removal of trade barriers will be much less important for the developing countries if they are non- discriminatory. It is claimed by the developing countries in the UNCTAD that the acceleration of their export earnings requires the establishment of preferential access for their . . 1 QI". fl..- “Alsfvavl-I a v‘doo I '3..- v . bung“ '- ...-... a : pnn'. '0‘- § - .s 4 ‘4 4 4 a .m A». .~« .4 ‘ an» C .L 3 C r. . _ . . 1 2 J r r . . a 3 3.. Lu #- v. a A ., .2” NW C. mm :L ..~1 r. C «L c 1 .C. :1 e To .C we fly a" n. .. e a. :1. at a. u .. Q» t 3 s F. p . p . r . .3 C . .. _ a. .t A: :N v; v . s . .l A-.. .1. u a C . A . : . u .. LN ya fly L . -. F. .1 .u. ... t. -7. A. 2. n. .u .1‘ s u? “a ox 21 manufactured and semi—manufactured exports into the markets of developed countries in violation of the non—discrimination- in-trade clause of the GATT. Furthermore, it can be argued that while developed countries are unwilling to increase their foreign aid programs substantially, they may find it convenient to assist the developing countries by providing them a subsidy in the form of preferential treatment to their manufactured exports.28 A tariff preference system might work in two ways, or combinations of these: It can increase the returns to an exporter of commodities already being sold abroad, and it can reduce the buyer's price sufficiently to permit new manufactured and semi-manufactured products to be sold in foreign markets. The competitive advantage that preferences in industrial products might give less developed producers must be assessed in terms of the resulting reduction of the effective protection enjoyed by the producers in the develOped countries against less developed producers and of the margin of effective preferences given less developed over the producers in the "third," developed, countries will continue to face the full restrictive effect of effective protection in the preference-giving countries, the developing countries will no more be so constrained in their manufactured exports to the developed countries. The gains from tariff preferences for the manufac- tured exports of developing countries can be assessed in terms of trade-creation and trade-diversion effects of preferences. C I ‘ a tzaze-c; 9 hot“! Val at t}. ' 'W.e'u.e ‘. .‘fi U05 I (IA in» AF «3 Cu 3'. a. 22 On the consumption side, the magnitude of trade-creating and trade-diverting effects depends on the effects of preferences on the relative prices of commodities from different sources, and, therefore, on the level of existing nominal tariff rates on which the preferences are based. On the production side, however, the magnitude of these effects depends respectively on how the preferences alter the effective rate of protection of producers in the preference-giving against producers in the preference-receiving country and the margin of preference it establishes for producers in the preference-receiving coun- tries over their competitors in non-preferred, "third" coun- tries.29 According to Professor Johnson, "the trade- creating incentive might be substantially greater than the calculated effective protection rates would suggest."30 The dynamic effects of preferences are similar to those of the customs unions such as internal economies, scale effects due to the extension of the market, and external economies 31 such as learning through experience and growth of skills. GENERAL PREFERENCES A preference system is determined by factors such as definition of manufactured goods, the tariff quotas, es- cape clause, the definition of countries as preference-givers and preference-receivers, and the duration of preferences. They can be combined in different ways to give rise to all conceivable types of preference schemes. A general preference system would require all developed countries to grant a duty-free access to the manufactured and semi-manufactured +1 5 1 Mite» N‘""“’: rfir+ uh..- " yd“. LL -E 1 ‘32-: FF‘IV‘ 9V‘ V»: “a -4». - '1”. ‘v r ‘ f“ u ‘a‘ 6" h.‘ E A u 9» .9. 23 exports of all developing countries to their markets without limitation on volume. This might be the best method of pro- viding preferences for the manufactured exports of developing countries because it would be the easiest to administer and leave the specialization and industrial exports of developing countries to their relative competitiveness in production and trading.32 Under the general preference system, the higher the effective protection in the developed countries, the higher the preference for the developing countries. Therefore, based upon the existing structure of effective tariff protection in the developed countries, the general preference system would give the maximum incentives to the establishment in the devel- oping countries of those industries that are subject to the highest effective protective duties in the developed countries. Are these the industries in which less developed countries have (potential) comparative advantage and developed countries comparative disadvantage? This is the question that the present study seeks to answer. To do this we compare the effective protection rates in the advanced countries with the comparative advantage schedules of developing as well as of developed countries in the production of manufactured and semi-manufactured goods. If a high degree of positive corre- lation between the comparative advantage schedule of developing countries and the effective protective rates of the developed countries exists, then a generalized preference system will promote the expansion of industries in the developing countries . I ‘ ,, vvr‘ a; tr *ll.\f“ ‘° 50‘ UF‘” a .u-V ."”;‘i r?” ...-.E as ‘2' nA-v-r fi'V'f ¥.-~~:-ab.os . ‘ u . was: 1 W o‘vvd ‘04 U. average 5p‘ v.\- onD -U. N'OUA a; r- T‘ ' ...., 5-3“ .,..',... "-35. "‘L. L.‘ :‘t’ . 5‘ Ch‘ “ I V. ‘5‘ "C: \— ‘:-- ‘. n‘ ._,fi 5‘,“ v- 24 in which they have a comparative advantage. This will pro- mote a better allocation of resources in developing countries. Similarly, a significant and negative correlation between the comparative advantage schedules and effective protective rates in the developed countries will mean that general pref— erences for manufactured exports will involve minimum trade diversion and hence, minimum losses for the developed coun- tries. This requires (a) ranking of industries by compara- tive advantage for various less developed and developed countries, and (b) the estimates of effective tariff rates levied on manufactured products in the developed countries. Since the comparative advantage of a country is a function of many factors, the ranking of industries must be approximated by the use of various models explaining comparative advantage. This will be the subject matter of Chapters IV—VII. The effective tariff rates in the developed countries are esti- mated and analyzed in Chapter IV. c C C. .C . . . - MHH u;- s ‘. .Pu “L‘HVJ .“ S H..- ' C rm \4 w n "xv“ C u‘.‘ to fil\» “I. Y§ INIQ PM. 0 y “V “It 9 v. nu a ... \\ I 2* it 5 Wu To .11 3.1L . O e .t 3 vi .. l D; an t .1. C. . u-.. a: .11.). , L. e. d M... 3 :49. r u .r.. V. 4.. E 6 .1 .3 t t a 2 “x“ in «I. .x. u... .3 Pw an». .. u an. 4.1 My. M3. MW...£~. Mi. raw-W NOTES TO CHAPTER I 1Hollis B. Chenery and Alan M. Strout, "Foreign As- sistance and Economic Development," American Economic Review, September, 1966, pp. 689-733; "Comment" by G. Ranis and J. C. H. Fei, American Economic Review, September, 1968; and Chenery and Strout,fiReply,“'American Economic Review, September, 1968. For further detail and alternative approaches to the theory of "dual gaps," see the following: John C. H. Fei, and Douglas Paauw, "Foreign Assistance and Self Help: A Re- appraisal of Development Finance," Review of Economics and Statistics; I. M. D. Little and J. M. Clifford, International Ai§’(Chicago: Aldine Publishing Company, 1965); R. F. Mike- sell, Economics of Foreign Aid (Chicago: Aldine Publishing Company, 1968). 2See Trade Prospects and Capital Needs of Developing Countries, UNCTAD Secretariat, TD/347Rev. 1., New York: Uniteleations, 1968. 3The logic underlying this relationship between the savings gap and the trade gap can be inferred from G. Ranis, and John Fei, "Comment," and H. B. Chenery, and A. M. Strout, "Reply," American Economic Review, September, 1968, pp. 897- 916. 4Bela A. Balassa, "The Capital Needs of the Devel- oping Countries," Kyklos, Vol. 2, 1965; Trade Prospects for Developing Countries, Economic Growth Center, Yale Univer- sity. Homewood, Ill.: Richard D. Irwin, 1964; United Nations, United Nations Wprld Economic Survey, Vol. 1, 1963; Trade and Development: Trends, Needs and Policies, (New York: U. N., 1964); P. N. Rosenstein-Radan;F"International Aid for Underdeveloped Countries," Review of Economics and Statistics, 1961. 5United Nations, Trade Prospects and Capital Needs for Developing Countries, Chapters I, II, III. Robin Morris, 1'Can We Measure the Need for Development Assistance?" The Economic Journal, September, 1970, pp. 650-68. 6See The Outlook for Debt Service, U. N., TD/97, Vol. IV, New York, 1968. 7Trade Prospects apd Capital Needs of Develppipg Countries, U. N., Annex III. 25 v . a. 1C? ‘3' 8. h .‘v Q Pa V!» n A "Q. ‘4‘" ‘ I 26 8 Ibid., Annex II, pp. 66-67. 9Albert Hirschman, The Strategy of Economic Develpp- ment, New Haven, 1958, p. 212; "Political Economy of Import Substituting Industrialization," Quarterly Journal of Econo- mics, September, 1968. 10The concept of negative value-added is very contro- versial. Some economists like Ellsworth think that it is absurd to even imagine that the value-added can be negative. However, strong evidence to the contrary is available. The concept of negative value-added does not imply that the pro- duction actually falls, but rather that the cost of inputs in foreign exchange is greater than the foreign exchange value of the commodity produced. This need not be true in domestic prices. For elaboration see the following: Ronald Saligo and J. J. Stern, "Tariff Protection, Import Substitution and Investment Efficiency," Pakistan Development Review, Vol. 5, 1965, pp. 249-70; Maizels, Industrial Growth and World Trade (Cambridge: Cambridge University Press, 1963), pp. 70-75; Stephen R. Lewis, Jr. and Stephen E. Guisinger, "Measuring Protection in a Developing Country: The Case of Pakistan," Journal of Political Economy, 1968, pp. 1170- 97; Stephen E. Guisinger, "Effective Protection and the Concept of Negative Value Added, " Quarterlnyournal of Economics, November, 1969. llDavid Felix, "The Dilemma of Import Substitution-- Argentina," Development Poligy, Theory and Practice (Gustaf F. Papanek, ed),(Boston: Harvard UniverSity Press, 1968). If it were possible to reduce the ratio of import to gross domestic production via import substitution, then each percentage point reduction would mean a drop of about four billion dollars in the import gap by 1975. Referring to Trade Prospects and Capital Needs of Developing Countries, U. N., Table 14, the imports in 1975 would be $70 billion and ’G.D.P. = $432.3 billion, therefore, 70 x 100 R = (Imports/GDP ratio) = 432.3 = 16.2%. If now R has to be reduced by one percentage point, i.e., to 15.2%, through import substitution, then the import requirements would fall to: 15.2 x 70.0 16.2 = $65.8 billion This indicates that for one percentage point fall in R. the import gap will fall by 70.0 - 65.8 = $4.2 billion. 12See Trade PrOSpects and Capital Needs of Developing Countries, U. N., pp. 19-20, 47. 13 Ibid.,pp. 18-19, Ch. 2. SW A ‘.v~-vl d ( yfhh Fn—CI‘ Uswn‘.nnc UV! HLJ" \ q . sons 9 +"rs M..- D we 1... Cs a u. a: . . . . c .. n D. C H :11 7. t .3. a; E a C A O 13 5.. “ii; .7. r v1 HA .3 .. s: n... 5. ”43.1 F»; u n: .~.¢ h... .n o ..u ?s “us An» .N‘ V}; u‘ 59 .n . x.‘ AN» ~ . its U]. Pu H..- .~..\ t .1 .m t. a... C ..\ t (x . . u... .~ T. .5 ..: r? a. a. 4.1 g . C O x. f S C a. 27 14 . . . Harry G. Johnson, Economic Polic1es Toward Less Developed Countries,(New York: Praeger, 1967), Chapter V; Boris Swerling, Current Issues in Commodity Poligy, Essays in International Finance, No. 38, (Princeton University Press, 1962). 15Proceedings of the United Nations Conference on Trade and Development, Second Session, Volume II, Commodity Problems and Policies, (New York: U. N., 1968); 16For detailed analysis on this point, see: Raymond Vernon, "Problems and Prospects in the Exports of Manufac- tured Goods from the Less Developed Countries," Proceedings of UNCTAD, Volume IV, Trade in Manufactures, (New York: U.N., 1964). pp. 200-10. l7Raul Prebisch, Toward A New Trade Policy for Devel- Opment, Proceedings of UNCTAD,(New York: U. N., 1964) Vol. II, p. 40. 18 Harry G. Johnson, op. cit. 19The "Structuralist" view has been propounded and popularized by Raul Prebisch and his disciples at the U. N. Economic Commission for Latin America. Theoretical formula- tion of this concept has been done by Hirschman, Hagen, Felix and Diaz-Alijandro. This view is, however, distinct from the demand-pull and cost-push types of inflation. It involves supply bottlenecks and price-price Spiral as the basis for price disparities and lack of price competitiveness of devel- oping countries in the exports of manufactured goods. For a very good summary of the structuralists--as opposed to mone- tarist--view, see: E. E. Hagen, Economics of Development (Irvin, 1968), Chapter 11; see also, Hagen,"AnEconomic Justification for Protectionism," gparterly Journal of Eco- nomics, Vol. 72, November, 1968. 20See R. Vernon, op. cit. Professor Vernon considers that price competitiveness is not the only thing in the deter- mination of manufactured exports from developing countries because of structural disequilibria. According to him, if exports were only a function of prices, and if questions of price were the principal obstacles to increased manufactures exports, one suspects that the past record of underdeveloped countries in this field would have been much less dismal than in fact it has been. Lack of information on export opportuni- ties and domestic distortions may completely neutralize the price incentives provided by devaluation. A similar view has been presented by Raul Prebisch in Toward a New Trade Poligy Afor Development (New York: U. N., 19675. '0 -9 ‘cv! ‘ t" "C “‘0‘ 75...- 2" hot a: n: i Q fifi' IR ‘4‘-“ * r- 'AI‘WV‘L V‘v‘lng. a a.) Cu . v1 4' 28 21See Carlos F. Diaz-Alejandro, Exchange Rate De- valuation in alSemi-Industrial Country (Cambridge, Mass.: MIT Press, 1965). Besides considering the structural dif- ficulties, he believes that income effect of devaluation is more significant than its substitution (i.e. price) effects, and that the income effect is always detrimental to the de- valuing economy. 22 R. Vernon, op. cit., pp. 207-11. 23cf. Bela A. Balassa, "The Structure of Protection in the Industrial Countries and its Effects on Processed Goods from Developing Nations," Programs for the Liberaliza- tion and Expansion of Trade in Manufactures and Semi- manufactures of Interest to DevelopingCountries, Proceedings of UNCTAD, TD/B/C, 2/36, (New York: U. N., 1967), p. 1; See also the following: W. M. Cordon, "The Structure of a Tariff System and the Effective Protective Rate," Journal of Pplitipal Economy, June, 1966, pp. 221-37; Harry G. Johnson, "The Theory of Tariff Structure with Special Reference to World Trade and Development," Trade and Development (Geneva: Librarie Droz, 1965); Economic Policies Toward Less Developed Countries, pp. cit. 24 See A Study of Industrial Growth (New York: U. N., 1963. 25Hal B. Lary, Imports of Manufactures from Less Developed Countries (New York, National Bureau of Economic Research: Columbia University Press, 1968). 26B. A. Balassa, "The Kennedy Round," Estimated Ef- fects on Tariff Barriers, Part II, UNCTAD, TD/6/Rev. 1, (New York: U. N., 1968). 27H. G. Johnson, op. cit., pp. 75-76. 28H. G. Johnson, "Trade Preferences and Developing Countries," Lloyds Bank Review, Vol. LXXX, April, 1966, pp. 1-18. 29H. G. Johnson, loc. cit., Chapter VI, pp. 163-211. 30H. G. Johnson, op. cit., p. 16. 31M. Kreinin, "On the Dynamic Effects of Customs Union," Journal of Political Economy, April, 1964. 32In reality, it is not possible to have such a gen- eralized preference system due to various administrative, political and economic reasons. The escape clause and posi- tive tariff quotas, etc., will have to be incorporated into the system. For exhaustive sutdies of this subject, see the 29 following: Sidney Weintraub, Trade Preferences for Less- Developed Countries, An Analysis of U. S. Policy, Praeger Special Studies in International Economics and Development, (New York: Praeger, 1966), Chapters 3-5; UNCTAD, Problems and Policies of Trade in Manufactures and Semi-Manufactures, Vol. III, (New York: U. N., 1968). Y‘"§"P 7-4-»: , . .‘.-C\ A ,\ «d L. 3. ~\u IDA C‘w‘ a; \H l ‘Vck L, “2" m '5‘ §ey ‘9 ~. I.“ ”A .— bun, Vt II'I‘H ‘8VA. CHAPTER II CRITERIA OF COMPARATIVE ADVANTAGE INTRODUCTION This chapter develops the models which will be used to determine the comparative advantage by which manufactured commodities will be ranked. LThe theory of comparative advan- tage states that a country will maximize its gains if it specializes in the production and export of goods in which it has greater relative efficiency, i.e., it can_produce these goods relatively cheaply vis-a-vis the rest of the world, and import those products in which it is relatively less ef- ficient, i.e., which cost more at home than abroad. In a perfectly competitive system, market prices will reflect the comparative advantage of a country. But price distortions, especially in the less developed countries, are such that the price mechanism is not an indication of comparative advantage, neither potential nor actual. Hence there is a need to draw upon certain theoretical models to determine the comparative advantage of developed and developing countries, and to rank their industries by the degree of comparative advantage. There are several alternative explanations of a country's comparative advantage, each subject to its own set of limitations. This is perhaps due to the fact that 30 31 production and trade patterns are a function of many factors, thus it is not possible to incorporate all of them into one \_. sophisticated theorem of general validity. CRITERIA FOR RESOURCE ALLOCATION AND COMPARATIVE ADVANTAGE Planners in developing countries follow various al- location criteria to choose the export and import competing industries that would make the most efficient use of the available resources. Since the principle of comparative advantage also aims at the best allocation of resources, these allocation criteria may be used to rank the industries of a country in a hierarchy that will be consistent with its comparative advantage; these criteria are considered below. The Classical Criterion of Comparative Advantage. The classical criterion of comparative advantage, though widely accepted as it is, is also the one which has been greatly dis- regarded by actual policy formulation. If product and factor markets are perfectly compe- titive and there is universal free trade and no transport costs, the domestic market price of products will be the same as their world prices. Then the net social benefit (NSB) from starting or expanding production of any good will be given by the difference between the world price of the good in question, say Pi' and the factor cost involved in produc- ing it valued at market prices, say Ci' so that (NSB) = pi - ci GIRI 7. {m ”a 7 C V. . . a . pm. .3 3 3 E S V l .J C S : . z 2. . I E a. a. P .3 S C .C .3 + .. I E E C. C .. E i t .l t r. 3. Lu .. . S C .. a. r. a. C .u 4‘ ,T\ A a z. n. t. : . Lu :. .3 v. r. r. i. E a: . . .1 C. .. a v 1. v. .3 .. A. — . nu Va my. .ws .pu ... .~ ..\ H 3. .7. ~x~ ass 32 or an index of comparative advantage can be attained as: for this particular product, i. A ranking of industries by their comparative advan- tage index will indicate that industries at the top of the schedule should be developed as exporting industries.2 This is an index of the net benefits generated by using one unit of domestic resources in producing this particular good. This criterion was developed by Deepak Lal.3 However, because of certain unrealistic assumptions on which this theory Crests, it has not been used anywhere. The criterion is based on the assumptions of perfect product and factor markets and free trade, hardly satisfied in the developing countries because of structural disequili- brium in the economy. The domestic prices of goods and ser- vices and factors do not represent the real social costs and benefits of producing or using these goods and factors. This is particularly true of labor under situations of disguised unemployment. Moreover, the degree of distortion is not the same in all industries. Imperfections are also caused by government inter- vention in the form of many protective devices, taxes and sub— sidies. There is a need, therefore, to find some alternative criteria that are more operational and serve as proxies for e cla Cr. 3,. 1 V ‘ ,r .--— G ._ ‘x—n.‘ I!“ An‘ mama‘- v.1... up- ~-~.' SC UA¢- v Q. «Q nfiu .2 ‘ C y... 5“ v :r «5 CL 2» add a: ’ PM; P. A‘s C. AF» 0. C. 33 the classical concept of comparative advantage. This would require that some way be developed by which the index of comparative cost, including potential changes in the future, can be made ope ational by finding a method of valuing inputs and outputs at prices which reflect real social costs and benefits. Once this can be done the social value of setting up a particular industry in line with a country's comparative advantage can be determined, and industries in which the country has the highest comparative advantage will then be potential export industries. Foreign Exchange Costs Minimization Criterion. While analyzing the dual gaps approach to the foreign exchange re- quirements in the first chapter we found that most of the developing countries face chronic foreign exchange scarcity. The foreign exchange criterion, therefore, assumes that the only social costs incurred by such an'economy are foreign exchange costs, and hence those industries should be chosen and developed for which foreign exchange costs per unit of foreign exchange saved or earned by producing the commodity are the lowest. By producing the commodity the economy may save foreign exchange equal to the c.i.f. price of the good, say P, or if it is earning foreign exchange (if it is an exportable) equal to the f.o.b. price of the good, say P‘. Further, if the social costs in both cases are the direct and indirect foreign exchange costs of importing the raw materials and intermediate goods which go into producing the good, and the sum of these costs is, say F, then on the I ~ I“ U- -. A. ~‘-\.. .. .. or~ b-‘. .. ‘V- : g ‘ ’ ‘4 “Q- A. ‘ 34 foreign exchange costs minimization criterion, industries would be ranked according to the index: P-F P'-F F F Ranking by this index indicates comparative advantage. This index was used to select export and import competing indus- tries in Pakistan.4 Bruno, Krueger and Balassa take excep- tion to the use of foreign exchange cost per unit of foreign exchange earned or saved. They would rather introduce the domestic resource cost per unit of foreign exchange as the basic criterion. Therefore, for a given commodity, the cost of a unit of foreign exchange (earned or saved) is taken to equal the direct and indirect domestic resource costs incurred in supplying it domestically, divided by the difference between the foreign price of the product and the foreign exchange cost of direct and indirect imported in- puts.5 It is extended into a programming criterion analyzed in the next subsection. This criterion is a special case of the scarce- factor intensity criterion which indicates that a country should specialize in producing goods drawing more sparingly on its relatively scarce factor, capital or labor as the case may be. The foreign exchange cost minimization criterion suffers from the same shortcomings as any other factor inten- sity criterion because of common underlying assumptions. It is very misleading to assume that only foreign exchange costs represent real social costs to the economy, unless the economy u Inn!- ay- ._l ‘r‘ "'U‘|\‘.- ‘ 7‘ ‘u rr ht“ L s.,." I --Vv ‘ :.~u9 ~Q‘..-‘ ‘s 4V‘spf‘ K I} 35 is facing a "strict" foreign exchange bottleneck implying that necessary importables cannot be obtained from domestic production or imports, i.e., both the foreign and domestic rates of transformation are zero. The validity of this cri- terion thus depends upon the existence of a strict foreign exchange bottleneck. Programming Criterion and Shadow Prices. The mathe- matical programming approach, especially linear,6 has come to be used extensively in the developing countries because it provides a convenient link to the principle of compara- tive advantage as the optimal pattern of trade is determined simultaneously with the optimum allocation of investment. Bruno has used this approach to select the potential export industries in Israel.7 The social welfare function—-growth target-~is maximized subject to the constraints of available primary inputs like labor, capital and foreign exchange. In the process of solution to such a program, accounting or "shadow" prices of the primary inputs are obtained as a by- product of the solution. Such shadow prices are a reasonably accurate measure of relative factor scarcities, and real costs and benefits of different lines of investment providing for the most optimal allocation of resources. The shadow prices of primary factors of production are the social marginal value products of the factors. For the factors in fixed supply, the shadow prices also represent the social opportunity cost of using the factor. Because foreign exchange supply is not ‘fixed and can be increased, the shadow price of foreign ex- r) 1‘ {)J t C) (I) .«a . unpfl I! C ‘ .. ..n.~-' .. A ['1 .u‘ativ U - a ‘. ‘v‘- Q 36 change is determined by both the costs of increasing the supply of foreign exchange in terms of domestic resources, and the benefits from having one more unit of foreign ex- change given by the marginal productivity of foreign exchange in domestic factories. Since in equilibrium the marginal costs must be equal to the marginal benefits, the exports should be expanded to the point at which the resource cost of earning another unit of foreign exchange is equal to the contribution which foreign exchange makes on the margin to the domestic social product.8 Now, given the shadow prices of primary inputs, one can follow Bruno9 and use the shadow prices to calculate the domestic resources cost per unit of foreign exchange saved (or earned), say e Given the shadow exchange rate es, j. industries can be ranked according to the index: Such a ranking is consistent with comparative advantage. This criterion, however, has its limitations. The linearity assumption on which the approach is based implies constant returns to scale, and hence disregards the oppor- tunities for economies of scale, which may be of great im— Portance. With non-linearity the accounting prices will no longer reflect the marginal social productivity of the input. Therefore it is necessary to modify the development program Particularly in the case of exports where the price elasticity , A. p", ‘v- ~I 4 VJ‘.‘ g- 37 of demand is also rather low.10 Moreover, the input-output relationship on which such a model is necessarily based may not be known and may not be stable as is assumed in the programming approach. This method comes close to the concept of comparative advantage and may be used in ranking the com- modities for comparison with effective protection rates for countries for which information is available. The World Price Criterion. Mr. Littlell has developed a fairly simple and operational social cost-benefit criterion in choosing potential eXport industries. Assuming for the moment that a country has perfect trading opportunities, that all inputs and outputs are tradeable, and that there are no trade controls, then clearly the shadow prices of inputs and outputs are their world prices. The rule is then to value all inputs and outputs at their world prices, and the ranking of industries by the ratio of net benefits to costs will indicate ranking according to comparative advan- tage. To make this criterion operational two of the assump- tions must be relaxed: (1) that there is perfectly free trade, and (2) that all inputs and outputs are tradeable-- where in fact land and unskilled labor, as well as many goods, are either non-tradeable or not traded. If the world price criterion is to be operational the first modification would require that instead of taking world prices reflecting average costs and revenues of the in- puts and outputs, world prices relfecting the marginal costs r S to . .s .. Jud «H. a u by. L . a . l ,9-“ t FF“.- I‘vl‘ PH". and. I' Jar ..-. l 1 x I fin e 4 x .3- 3... .1r. .5“ O ~ AH. 38 and revenues should be taken. While this involves judgement, it is unavoidable in any estimates which try to peer into the future. .The second adjustment requires the pricing of non-traded goods and inputs at world prices. The non-traded inputs can in turn be broken down into tradeable inputs, and the latter can then be valued at world prices. Little and Mirrlees12 have given various methods of converting non- traded goods into tradeable values. But their methods are exceedingly arbitrary and ad Egg, making the technique of questionable value. Domestic Price Criterion. In analyzing the compara- tive costs,factor proportions and industrial efficiency in Pakistan's manufacturing industries, Nurul Islam ranks the commodities by the ratio of ex-factory prices of specific domestic products to c.i.f. prices of closely competing imports.13 If the ex-factory price of the i-th commodity is, say, Fir and the c.i.f. price of the competing import is Pi' then the following index can be used to rank the commodities where Pi' is the domestic price of competing imports adjusted for overvaluation of domestic currency. This criterion involves serious data problems. The ex-factory prices have to be based on a direct estimate of the costs of individual manufactured goods. No such information H b on V‘v.‘ 54 ‘nop 9'" ‘ {waldo “A... vv ‘0. R1 ,1 o. :y 2. :‘ ‘a~ «am av flu.- .w‘ ~HH Av . ...§ a ”I“ “v. .a c w oh" 1 n - .~.~ Vs .6. s Vs Nu —W ‘.n.o a N. ... a. I. y .h.r~ — . ~ . 39 is available for developing countries other than Pakistan. Growth Criteria. The problem of defining compara- tive advantage in the context of economic growth is very com- plex. It is not only a matter of determining the most effi— cient method of allocating existing resources, but also of mobilizing new resources. There are possibilities of en- countering increasing and decreasing returns to scale; the external demand for exports may be highly inelastic, not to speak of structural disequilibrium and uncertainties about the future. A more serious limitation is caused by the non- quantitative interdependence among sectors such as is assumed by growth theorists like Hirschman and Lebenstein. For example, Hirschman, by supposing that one growth sequence (backward or forward linkage) is more effective than another because it economizes on decision making ability or provides a greater incentive to political action, implies a set of criteria having little to do with the criteria already con- sidered. Furthermore, growth is considered a cumulative process which feeds on itself by generating new and more in- puts such as increases in labor productivity and managerial ability as‘well as technological changes which are not only the cause, but also the consequence of the growth process. One way to take these dynamic factors into account 14 1" is to follow Nurkse and define "incremental" or "marginal comparative advantage, which may be different from the exis- ting or "established" comparative advantage. One way of defining "incremental" comparative advantage would be to t V. . a I ‘ d 11 \ Abs «C [In ‘uh .hu q I .3 . Q. o. u... S .r. r c. Ir. a. v z .(a S F r .w . .1 W a e . 3.. C n“ .M n c I C .5. L E r a. f. by a. .G .L . . .3 p. i .J .3 .3 u-.. .1 . v; . J LC. 2 a . L“ r . .,; : . a» r... n . : . r. .1. r. «a n : . a 4O consider the technology and production functions as given, and then state that it is economical for a country to move towards the production of those commodities that draw more heavily upon the factors with a relatively greater growth rate. But technology will not remain unchanged. It will make our task of determining comparative advantage more dif- ficult. One could assume that all inputs are growing at a constant, proportional rate and then look at the direction of technological change to define a partial or neo-incremental comparative advantage.15 This formulation, however, fails to take into account factors such as external demand conditions and differential rates of technological change at home and abroad. Moreover, incremental comparative advantage cannot be drawn upon to rank industries. Baer and Kerstenetzky, while writing on import- substituting industrialization in Brazil, make use of a} "repercussions effect" index that quantifies the backward and forward linkage effects to rank the industries according to the degree of comparative advantage.16 Using the input- output table they computed the indexes of dispersion (back- ward linkage) and sensitivity to dispersion (forward linkage), using the following formulas: U. = (j 1 m __._.. z. mzzj i=1 II |-‘ N B l’LC 41 and __1_ z m i Ui= (1:1, 2,....m), 1 m m2 El 21 i=1 and (2U, + U ) 3 . U= ' 2 where Uj = index of the power of dispersion, Ui = index of the sensitivity to dispersion, Zj = sum of the row elements of the transposed inverse matrix, Zi = sum of the column elements of the transposed inverse matrix, m = number of in- dustries. The index Uj indicating the extent of the expansion induced by industry j in the economy as a whole corresponds to an estimate of what Hirschman calls the backward linkage effect. Ui indicates the extent to which industry i is affected by an expansion of the economy at large and is an estimate of the forward linkage effect. The backward and for- ward linkages were combined in a single index (U) for ranking, giving double weight to the backward linkage because, according to Hirschman, backward linkages are more important than for— ward linkages. The repercussions effect index, however, fails to incorporate factor endowments, rates of factor accumulation and technological changes as the determinants of comparative advantage. All the allocation criteria considered above to rank V' _ M""A."‘ +7 » bunny/U. 5‘ - 2.; ' ‘ 3...: pro:. to inccrgi ii? "i up) tive explj are analyt grcblems. .L'TEPJJATI‘J N IIMDD hr. "'V¢J.. ‘yu . C. “\- arra,' ‘ L. N. vl.’ 166 the . C.) P1 1 42 commodities according to comparative advantage have serious data problems and analytical shortcomings. Most of them fail to incorporate various important influences on comparative advantage. It is, therefore, necessary to draw upon alterna- tive explanations of comparative advantage and trade that are analytically sound and which do not involve serious data problems. The following models meet both of the conditions. \ r x ALTERNATIVELMODELS, (PACTOR PROPORTIONS MODEL K). The factor proportions model states that a country Specializes in the production and export of a commodity that draws more heavily on the factor that the country is rela- tively well endowed with, i.e., a capital abundant country will export capital intensive products and a labor abundant country, labor-intensive products. Therefore a ranking of industries by their relative capital intensity will be con- sistent with the principle of comparative advantage>3 Fol- lowing Professor Kreinin's suggestion]:7 if all the industries are arrayed according to simple ratio of capital per worker, then, leaving aside some industries falling in the middle of the schedule, the pattern of trade between any two countries can be explained adequately by the factor proportions model. The model is based upon the following assumptions: (a) Perfect competition exists in factor and product markets. (b) Fixed quantities of the two homogenous factors of s A . A . Q N 43 production, capital and labor, are fully employed in each country. (c) For each commodity there is a common production func- tion everywhere, and these production functions are mathematically homogenous of the first order and have the same elasticity of substitution everywhere. (d) The production functions for the commodities are tech- nologically distinct, i.e., they could be distingui- shed by factor intensity. This implies that at any ratio of wage rate to capital cost, the optimal ratio of capital to labor in a given industry, i, is always greater or always less than in any other industry, j, in a two-country-two-commodity framework. This is the so-called strong factor-intensity hypothesis. (e) The quality (but not the quantity) of each factor in each country is identical, as are the production functions. (f) There is completely free trade and transport costs are zero. However, there is perfect immobility of factors of production between countries. Assumptions regarding relative factor abundance, iden- ‘tical production functions and strong factor intensity hypo- thesis are crucial to the use of the factor proportions model 'to approximate the principle of comparative advantage. Relative Factor Abundance. There are two alternative :formulations of relative factor abundance. Ohlin defines this concept on the basis of the pre-trade ratio of factor prices in :_ ‘s 44 the two-country, two-commodity framework. Denoting countries by the subscripts A and B, and capital and labor by K and L respectively, country A is relatively capital abundant in the Ohlin sense if before the trade begins. By definition capital is rela- tively cheaper in the capital rich country before trade, and labor in the labor-rich country. Jones, however, defines factor abundance in terms of physical factor endowments.18 A country is relatively capital abundant if, and only if, it has a higher proportion of capital to labor than the other country. Thus country A is relatively capital abundant if E R _ > _. L L A I where K and 3 represent the total capital and labor stocks respectively. If there are two commodities, X and Y, with identical production functions for each in both countries respectively, i.e., ox f(KX'Lx) QY = g(Ky,LY), and if (:99 E). then country A will have comparative advantage in the production (If X and B in the production of Y. This implies that, in general, proportic' country w I‘ '3) 333% l '-. .C. .44V" cther co; PIOp-orti : functions 45 general, "if the output of the two commodities is in the same proportion in both countries, the relatively capital abundant country will be able to eXpand its production of the capital- intensive commodity at a lower Opportunity cost than the other country."19 This is the definition of relative factor abundance that underlies our analysis. Identical Production Functions. For the factor- proportions model, the assumption of identical production functions for a product in different countries is crucial. Having subsumed all supply phenomena under production func- tions, given identical tastes, we are left only with factor proportions to explain differences in prices as the basis of trade. As Romney Robinson has suggested, admission of dif- ferent production functions will be fatal to the factor pro- portions analysis of trade because "comparative advantage theory, to be of the slightest analytic value, would then re— quire an explanation of when and how production functions come to differ. The problem is to stop the theory from de- generating into a surface eXplanation, explaining anything Exfppgp and nothing gxfappg."20 Hence the need for assuming an identical production function in our analysis. Strong Factor-Intensity Hypothesis. The assumption of strong factor-intensity specifies that whatever the ratio of wage rate to capital cost, the optimal ratio of capital to labor in any given industry, i, is always greater or less than in any other industry, j. If this assumption holds, then a unique ranking of industries by capital/labor ratios can ry- ~-l u . i 46 be obtained. The geometric interpretation of this hypothesis is presented below with the help of diagrams II-l and II-2.21 In diagram II-l, Illland 1212 are the unit isoquants for the commodities l and 2. It is possible to find a factor price ratio such that the resulting isocost line, PKPL, is tangential to both the isoquants. If the factor endowment rays of the countries a and b, under consideration, E and Eb’ a lie within the diversification cone, Z OZ then there will 1 2' be incomplete specialization, international factor-price equalization, and a ranking of commodities by their respective capital/labor ratios will be consistent with their ranking in order of comparative advantage. This is the strong factor- intensity hypothesis. However, the violation of this hypothesis will make it impossible to uniquely rank commodities by capital/labor ratios that will be consistent with a ranking by the compara- tive advantage principle for all commodities and countries. Diagram II-2 shows what happens when this assumption is violated. Diagram II-2 drops the assumption of a one-to-one correspondence between factor prices and commodity prices and constant elasticity of substitution between labor and capital. This gives rise to a factor intensity reversal. If the isoquants intersect more than once, there will be more than one common factor price ratio tangent to isoquants, giving rise to more than one diversification cone with no :0 0...... .356 Jleu—QIKU Pun 47 FACTOR INTENS I TY HYPOTHE SI S 2‘ E 11 as 6 ,2. PK ’/ E 6 \ ’/’/22 v 1 ’ v/ I / 12 0 PL LABOR FIGURE II-l FACTOR INTENSITY REVERSAL HYPOTHESIS g‘ / Z; E / 5,’ 2H ’ t: o. 6 II /’ Z 1 P x’ K, // I I ’I \ I” x” 22 I I, ll ’1’ \ I; ”” ,./’ o , I It 1 0 PL PLz LABOR FIGURE II-Z points in double in factor er. ferent di in; the < 33-: a un lie in t two inte tion co: hast ass or in Z! basis 0: Sim. g j 4:, . kv’ersm mltiple IaCtOI-i, seminal M 48 points in common (save in the degenerate case in which a double intersection is in fact a tangency). Now if the two factor endowment rays in the two-country model fall in dif- ferent diversification cones, there is no unique way of rank- ing the commodities consistent with comparative advantage. For a unique ranking it is necessary that the endowment rays lie in the same diversification cone. Diagram II-2 shows a two intersections of isoquants and two possible diversifica- tion cones ZiOZé and ZiOZ". Under these circumstances we must assume that all the endowment rays either fall in ZiOZé or in ziozg to obtain unique ranking of commodities on the basis of capital/labor ratios. In our analysis, therefore, we have to include a strong factor intensity assumption, or a much less strong assumption of limiting the factor endowment rays in a single diversification cone when factor intensity reversals and multiple solutions are possible. Empirical evidence regarding the relevance of a strong factor-intensity assumption is not conclusive. Minhas' seminal work22 seems to have shown that factor intensities were reversible as the C.E.S. production function fitted by him to international data showed elasticities of substitution both significantly different from unity and zero and also from 23 has pointed out, such one another. However, as Leontief reversals were practically insignificant (18 to 20 out of 210 possibilities) within relevant and observable ranges of factor 24 endowments. Moreover, Fuchs and Merle Yahr have also found little evidt‘ investigati: also tended ir ' $u.0 Cork 3 49 little evidence of factor intensity reversals. Similarly, investigations by Hal Lary, Moroney, and Hufbauer25 have also tended to support the rejection of the reversibility hypothesis. There are alternative methods of ranking industries according to capital/labor ratios. Hufbauer ranks the fac- tor intensity of the final industry only, and does not take into consideration the decomposition of intermediate goods into labor and capital. Lary26 uses the non-wage value added by manufacture per employee (roughly, value of output minus value of materials used and wages divided by employment) as a guide to interindustry differences in capital intensity. The higher the non-wage value added per employee, the more capital intensive the industry. Both Hufbauer and Lary con- sider only the direct input requirements and not the total-- direct plus indirect--requirements in the production process. The Leontief27 criterion of ranking industries according to the ratio of "total" capital and labor requirements is a more comprehensive and analytically better technique. \. \| HUMAN SKILLS MODEL #1:) \\~ The factor proportions model of trade has been sub- jected to severe criticism and controversy following the demonstration by Leontief that U. S. trade does not behave in conformity with this model.28 Does this mean that the factor prOportions model is oversimplified and fails to in— corporate other and perhaps less systematic factors which have pervading influence on the structure of trade? Corden " .‘A ‘I-LA—.W t: r: Lw . a v s Q.» .s .- >1 i s 50 tries to answer this by suggesting that realism demands the consistent introduction of nontraded goods, of inter- mediate goods, of economies of scale, of product differ- entiation, of technical change as a determinant of the trade pattern, of transport costs, and of the size and nature of the home market--all of which would probably9 alter the model so much as to make it unrecognizable. This is an impressive array of causal variables not explicitly incorporated in the simple factor proportions explanation of trade patterns. Is it possible that the factor proportions model stressed the wrong type of "generic" factors as Profes- sor Harrod suggests?3O Recent empirical investigations by Keesing, Yahr, Waehrer, Kenen, and Bhagwati and Bharadwaj con- clude that physical capital and labor are not the most effec- tive generic factors whose relative availability determine trade patterns.31 Moreover, evidence shows that differ- ences in supplies of human skills afford a better factor proportions explanation of trade and location in manufacturing industries than do endowments of physical capital and labor within the framework of the factor proportions model. When human skills are treated as generic factors, the factor pro- portions theory seems to show results that accord with the intellectual appeal of the factor proportions model. The Human Skills Model postulates that the availa- bility of labor skills or human capital determines the pat- tern of international location and trade for a broad group of manufactured goods, those not closely tied to natural re- sources and those that are produced by the "footloose" indus- tries. A country with relative abundance of highly trained bP-z—AA.O~" v__T—q¥ fh 5" «Au “’2? ~v-4 $¢nh “v- . ~A . v-P. o..‘ L. r a .a A s 51 personnel and skilled labor will have a comparative advantage in the production and export of skill intensive products while a country with a relative scarcity of skilled labor will have a comparative advantage in relatively less skill-intensive manufactures)» A ranking of industries by their relative skill intensity will, therefore, represent a ranking according to the principle of comparative advantage. Assumptions of the Skills Model. The assumptions underlying the human skills approach are similar to those used in the factor proportions model. We assume identical, homogeneous production functions for a commodity in all countries; perfect skills mobility internally but no mobility internationally; perfect competition and full utilization of resources; and the absence of "skill-intensity reversals"; in order to ensure a unique ranking of commodities according to their skill content. It is necessary to make the "strong" skill intensity assumption so that skill ratios or coef— ficients of one country could be used to rank industries that will be consistent with the principle of comparative advantage for other countries as well. Based on this assumption we can use the skill content of say, Japanese industries, to repre- sent the same in developing countries.32 There is strong logical basis for regarding differ- ences in labor skills availability as the basic eXplanation underlying location, specialization and trade in manufactured goods. First of all, in most industrial activities, labor is the most important factor of production as reflected by factor N—n—nn—I A‘.L-;IA~ _ 'M m-a‘w i a ll IL‘ ‘y. .l a» L» z a p.- nw A UA.« u LL a. ti .... nu .. n C g .. taine 52 income shares. Moreover, although human skills can be augmen- ted and are somewhat mobile internationally, in many respects the human resources of a country are subject to slower change and less international mobility than man-made resources such as physical or financial capital. The slower international mobility of human skills ensures that its initial availability in one place compared to another will have strong and persis- tent influence on industrial location. Furthermore, it is not possible to carry out rapid transformation of skills of a labor force. Some occupational skills can be acquired only through a long process of professional training. Broad classes of skills in any population can only be altered slowly. This is enough to maintain skill differences suffi- cient to produce persistent patterns of trade among nations. Another reason for taking skills into account in trade theory is that they appear to play an important role in explaining economic growth,33 and we know that growth and trade are interrelated. Moreover, various other factors like historical dif- ferences in skill supplies propagated down to the present moment by a need for skilled workers to train skilled workers, cultural differences among nations, unequal incomes, selec- tive migration and the arbitrary division.of labor that is sus- tained by trade among developed and developing economies, have created very definite differences in the endowment of skills among developed and developing economies. These factors are sufficient to produce persistent patterns of trade among O L Y‘ 4‘ 7 IV. LP» r“ a A» L. K» n.-. b C a: n44 IR. f\ \u n Ch .u. fi Vi a.» .p a t L; r a C . a. a .c .3 \Fu Q) Q H:v4 . . qls ~ t C» . ‘1‘ O Q . I... .s t ‘ QM h v‘ \Lt 71‘ C t s 53 nations. There are alternative methods of ranking industries according to the skills model. Kravis and Waehrer have used the differences in wage rates as proxy for differences in skill intensity of commodities because higher wages and greater skill intensity go together. Similar approximations were employed 34 by Roskamp, Yudin and Kenen. Recently Baldwin applied r... J.- nin. an q Schulze's concept of human capital in the form of average years and cost of education of labor to rank commodities and Wt! h determine the structural basis of U. S. trade.35 Keesing has used the U. S. skill input coefficients to rank industries according to the ratio of "direct" requirements of skilled labor to "direct" requirements of unskilled labor in man- years.36 All these criteria of human skills are closely re- lated to each other. One expects a positive one-to-one re- lationship between the level of skills and wage rates. Simi- larly,average years of education and cost of education are positively related to the level of skills, i.e., the greater the level of skills, the greater the cost in time and resources to acquire it and vice-versa. It is, however, desirable to use the "total" skill requirements to estimate the ratio of skilled labor to unskilled labor in order to rank the indus- tries. RESEARCH AND DEVELOPMENT, TECHNOLOGY AND THE SCALE ECONOMY EFFECT Keesing and certain other writers have come out with yet another explanation of the pattern of trade in manufactures. r‘ ivi r egar flat tlss= ‘stx 561““ E 1!“. “film: 4C: 7?: p § - s 1"“ ”P! 4 & '1 ea. I ‘Iatlflr 54 It is suggested that a country with greater research and de- velopment (R & D) expenditures will have a comparative advan- tage in the production of newer and more sophisticated products. This is so because on the one hand R & D results in technical progress, lowers costs, attracts resources into industries with greater R & D expenditures and confers com- parative advantage on them; on the other hand, it results in the development of new products not available elsewhere which leads to the creation of a comparative advantage for the country in the new commodities. One can, therefore, rank industries by the index of R & D that will be consistent with the comparative advantage. One index of R & D is the percentage of engineers and scientists employed in R & D activities in the total employment of the industry. The impact of scale economies in production and dis— tribution on the determination of comparative advantage is very important but little quantitative knowledge is available regarding many relevant dimensions of economies of scale so that no really good indicator is available. One commonly used indicator is the size of the home market; this suggests that a large home market is conducive to the export of goods produced under increasing returnsto scale, while a small home market is conducive to the export of goods produced under . 38 constant or decreaSing returns to scale. "REVEALED" COMPARATIVE ADVANTAGE Professor Balassa has developed a measure for approx- imating the comparative advantage to assess the effects of C l ”t u 1.... -oAn on own unnv LIIC C R \t n ad ad «MU .u 5 Va .c 450. 55 trade liberalization among the developed countries of Western Europe, Japan and North America.39 According to Balassa, cost considerations are not sufficient to explain the com- parative advantage as it would leave out of consideration the non-price variables if their influence on the determination of comparative advantage is to be assessed. However, it may not be possible to do so in practice. In light of this limitation of the factor-proportions model, it seems desirable to provide information on "revealed" comparative advantage in order to consider the effects of generalized tariff prefer- ences on the exports of developing countries. The "revealed" comparative advantage can be indicated by the trade performance of individual countries with regard to manufactured products, in the sense that the commodity pattern of trade reflects relative costs as well as differ- ences in non-price factors. The revealed comparative advan- tage can be estimated either by the export performance index or the export/import ratio index. The export performance index is more relevant to rank industries for analyzing the usefulness of a generalized preference system for the devel- oping countries. (See Appendix I for detailed analysis.) CHOICE OF MODELS The theoretical models considered above tend to make the pattern of trade a function of one variable (or one set of variables) or another, while in fact the explanation of trade patterns is not so simple. It is therefore necessary to follow an acelectic approach to the estimation of cat: car: :ist .Ll] tars .n u 3» av eere ‘. ‘1!- 56 comparative advantage. Instead of considering just one model, several alternative models will be used to rank manufactured commodities in developed as well as developing countries con- sistent with their comparative advantage. This approach will help us to incorporate in our analysis most of the fac- tors that explain the structure of production, specialization and trade. If the ranking of industries comes out about the same on all models, this approach will at least lend some credibility to the results. The factor proportions, human skills, R & D and scale economies models and the "revealed" comparative advan- tage are analytically sound and enough data are available to estimate their indexes to rank commodities. We will, therefore, use these models to test whether generalized tariff preferences will grant the greatest incentive to industries in which developing countries have relatively greater, and developed countries relatively lower, comparative advantage. The ranking indexes are developed in Chapter IV. 11:6 1.“. U ca .‘Js 9. C .0 . .r» r... m. are ~.. Y. A} at 2. «f» .u a; r. 0 CL. .\ n 5 v .5 I. RV ‘ .J .Lu P“ ll- v.» nu. Y.“ Pb in” .9 -U .5 :4 ~1. Tu a In my .1. C l 5-. is... miwm‘. -14 .w» LL ~». 4.. L.» . vL h... To PJ .2 A1» a L.. a. A h I u : A . . u .. u. l. N y ‘4» .14 n s. v a \nd Had Eu“ N. A xiv... \ s h. 1.: he AJ I; .‘ NOTES TO CHAPTER II 1 See: J. Bhagwati, "The Pure Theory of International Trade: A Survey," Economic Journal, Vol. 74, 1964; reprinted in Surveys of Egonomic Theory, Vol. 2 (Macmillan, 1965); J. S._Chipman, "A Survey of the Theory of International Trade, Parts I and II," Econometrica, Vol. 33, 1965. On the Heckscher-Ohlin model; Bertil Ohlin, Inter-regional and Inter- national Trade, Harvard University Economic Studies, Vol. 39 (CamBridge, Mass: Harvard University Press, 1933); Ronald Jones, "Factor-Proportions and the Heckscher-Ohlin Theorem," Review of Economic Studies, Vol. 24, 1956-57, pp. 1-10; I. Kravis,FfAvailability and Other Influences on the Commodity Composition of Trade," Journal of Political Economy, Vol. 64, April, 1956; S. B. Linder, An Essay on Trade and Transforma- tion (New York: John Wiley & Sons,l961). 2Economic Survey of Asia and the Far East, "Appendix to Chapter 1: Factors in the Asian Export Performance," (Bangkok: EDAFE, 1967), pp. 32-37. 3Deepak Lal, Foreign Exchange Bottlenecks, Balance of Payments, Investment Criteria and the Optimum Pattern of Trade and Production, Oxford University, January, 1967, mimeographed. 4Economic Survey of Asia and the Far East, Govern— ment of Pakistan, First Five Year Plan, December, 1957, Chapter 21. For further elaboration on this point see: Michael Bruno, Interdependence, Resource Use and Structural Change in Trade (Jerusalem: Bank of Israel, 1963); Anne 0. Krueger, “Some Economic Costs of Exchange Control: The Turkish Case," Journal of Political Economy, October, 1966, pp. 466-80. For a critical evaluation of the Bruno-Krueger approach, see: Bela A. Balassa and Daniel M. Schdlowsky,"Effective Tariffs, Domestic Cost of Foreign Exchange and the Equilibrium Exchange Rate," Journal of Political Economy, May, 1968, pp. 348-60. They suggest that it is better to use their effective tariff measure rather than cost of foreign exchange for the purpose of indicating the desirability of individual industries. 6Hollis B. Chenery, "Comparative Advantage and Devel- opment Policy," Survey of Economic Theory, Vol. II (St. Martin's Press, 1965), pp. 125-55. 5:7 nu ‘3‘ It'lv he 2. n\.u rte All s1. IIIL ,u‘ ~-~v.t .A. I .3. l 12¢ :04 buy v-7! u\v 58 7M. Bruno, op. cit. 8Bruce Glassburner, "Aspects of the Problems of Foreign Exchange Pricing in Pakistan," Economic Development and Cultural Change, June, 1968. 9M. Bruno, "The Optimal Selection of Export Promo- ting and Import-Substituting Projects," Planning the External Sector, Technigpes, Problems and Policies (New York: U.N., 1967). 10Hollis B. Chenery, op. cit., pp. 145-46. 111. M. D. Little and J. A. Mirrless, Manual of Industrial Project Analysis in Develpping Countries, Vol. II, (Paris: OECD, 1969). 12 Ibid., Chapters XII—XIV. fink-a. 0 3.. l3 . . N. Islam, "Comparative Costs, Factor Proportions, and Industrial Efficiency in Pakistan," Pakistan Develppment Review, Summer, 1967, pp. 213-46. 4See Ragnar Nurkse, Patterns of Trade and Develop- ment (New York: Galaxy Books, 1967), PP. 162-226. 15A brief survey on this issue is presented in M. O. Clement, Richard L. Pfister, and Kenneth J. Rothwell, Theore- tical Issues in International Economics (Boston: Houghton Mifflin Company, 1967), pp. 104-12. 16Werner Baer and Isaac Kerstenetzky, "Import Subs- titution and Industrialization in Brazil," Supplement, American Economic Review, May, 1964. 17Mordechai Kreinin, "Comment" on Hufbauer's Paper, (Raymond Vernon, ed.) The Technology Factor in International Trade, Universities--National Bureau of Economic Research, Conference Series No. 22 (New York: Columbia University Press, 1970), pp. 296-98. 18Ronald Jones, "Factor Proportions and the Heckscher- Ohlin Theorem," Review of Economic Studies, Vol. 24, 1956-57, pp. 1—10. 19Ibid., p. 7. 20Romney Robinson, "Factor Proportions and Compara- tive Advantage, Part 1," Quarterly Journal of Economics, Vol. 60, 1956, pp. 173-74. 1 . . The exposition of strong factor-inten51ty hypothe- sis and factor-intensity reversals is based upon the . C. l‘ I ~ .- . ‘ .1 +t : . Pd rd A . u n1. PL alt. , Um: .I; 59 following two studies: J. Chipman, "A Survey of the Theory of International Trade, Part III," Econometrica, Vol. 34, 1966; Murray C. Kemp, The Pure Theory of International Trade and Investment (Prentice-Hall, 1969), pp. 77-82. 228. S. Minhas,"The Homohypallagic Production Func- tion, Factor-Intensity Reversals and the Heckscher-Ohlin Theorem," Journal of Political Economy, Vol. 72, 1962, pp. 138-56. 23 . . . W. W. Leontief, "An International Comparison of Factor Costs and Factor Use," American Economic Review, Vol. 54, June, 1964. 24V. Fuchs, "Capital-Labor Substitution: A Note," Review of Economics and Statistics, Vol. 45, November, 1963; M. Yahr, Estimating the ElastICity of Substitution from International Manufacturing Census Data, Ph.D. Thesis (un- published), Columbia University, 1967. 25Hal B. Lary, Imports of Manufactures from Less Developed Countries, National Bureau of Economic Research, (New York: Columbia University Press, 1968), pp. 51-80; G. C. Hufbauer, Synthetic Materials and the Theory of International Trade, Appendix B (London: Duckworth), pp. 115-20. um”. . ffiflm .L . 26 Hal B. Lary, op. cit. 7For details see Chapter IV of this study. Also see, W. W. Leontief, "Domestic Production and Foreign Trade: The American Capital Position Re-examined," Proceedings of the American Philosophical Society, Vol. 97, 1953. 28 W. W. Leontief, op. cit. 29 , W. M. Corden, Recent Develgpments in the Theory of International Trade, Special Papers in International Economics, No. 7, International Finance Section, (Princeton University, 1965). P. 31. 30R. F. Harrod, "Factor-Price Relations Under Free Trade," Economic Journal, June, 1958. 31Helen Waehrer, "Wage Rates, Labor Skills, and United States Foreign Trade," The Open Economy: Essays on Inter- national Trade and Finance (P. B. Kenen and R. Lawrence, eds.) Columbia Studies in Economics, Vol. I (New York: Columbia University Press, 1968, pp. 19-40; P. B. Kenen, "Nature, Capital, and Trade," Journal of Political Economy, October, 1965; Merle I. Yahr, "Human Capital and Factor Substitution in C.E.S. Production Function," The Open Economy, pp. 91-97; Donald B. Keesing, "Labor Skills and International Trade: -.v—v< L') I h / 60 Measuring Many Trade Flows With a Single Measuring Device," Review of Economics and Statistics, August, 1965; R. Bharadwag and J. Bhagwati,fifiHuman Capital and the Pattern of Foreign Trade: The Indian Case," Indian Economic Review, October, 1967. 32Donald B. Keesing, op. cit., p. 289. 33Donald B. Keesing, "Labor Skills and the Structure of Trade in Manufactures," The Open Economy, p. 4. 34Irving Kravis, "Wage and Foreign Trade,‘ Review of Economics and Statistics, February, 1959; H. Waehrer,—"Wage Rates, Labor Skills, and United States Foreign Trade," The Open Econpmy, pp. 19-40; Karl W. Roskamp and Gordon C. McMeekin, Factor Proportions, Human Capital and Foreign Trade: The Case of West Germany Reconsidered," Quarterly Journal of Economics, February, 1968; P. B. Kenen and E. B. Yudin, Skills, Human Capital and the U. 8. Foreign Trade, International Economics Workshop Paper (New York: Columbia University, 1965), (mimeographed). 35R. E. Baldwin, Determinants of the Commodity Struc- ture of U. S. Trade, Social Systems Research Institute, University of Wisconsin, 1969, (mimeographed). 36D. B. Keesing, The Open Economy, pp. 1-18. 37 , D. B. KeeSing, "The Impact of Research and Devel- opment on United States Trade," Journal of Political Economy, Vol. 75, February, 1967 (reprinted in The Open Economy, pp. 175-89); W. Gruber, D. Mehta, and R. Vernon, "The R & D Factor in International Investment of United States Indus- tries," Journal of Political Economy, Vol. 75, February, 1967. 38G. C. Hufbauer, Synthetic Materials and the Theory of International Trade (London: Gerald Duckworth, 1966). 39Bela A. Balassa, "Trade Liberalization and 'Revealed' Comparative Advantage," The Manchester School Economic Papers, May, 1965. CHAPTER III SELECTION OF COUNTRIES AND COMMODITIES INTRODUCTION Our aim is to find out whether or not generalized tariff preferences would grant relatively greater incentives to the establishment of industries in developing countries in which they have relatively greater comparative advantage and developed countries relatively lesser comparative advantage. This requires the selection of a set of countries and manufac- tured commodities. Only a small number of developing coun- tries have a large manufacturing sector and a great number of manufactured exports. Therefore most of the gains from a preference system will accrue to the more advanced of the developing countries. The selection of developing countries for our investigation will, therefore, be made from these countries. Similarly, not all manufactured commodities can be of immediate interest as potential exports for the devel- oping countries. The selection of countries and commodities will, therefore, be arbitrary and the method followed for this purpose relies upon an arbitrarily chosen set of economic indicators. The need is to have a set of countries and com- modities that will cover the bulk of manufactured exports of 61 62 the developing countries. SELECTION OF COUNTRIES The rationale underlying the selection of developing countries is straightforward. Countries with a small amount of and probably erratic manufactured exports, or those spec- ializing in strongly resource-intensive manufactures were not considered. The economic indicators used to help in the selectionprocess were the gross domestic product per capita, the size of the manufacturing sector in the economy, the share and the value of manufactured exports in total exports, and the growth rate of manufactured exports from developing countries to the developed market economy countries (DMEC). These economic indicators are shown in Tables 2-6, Appendix II. Certain arbitrary critical values were assigned to these indicators in order to select the countries; these are pre- sented in Table III-l. Taiwan, Hong Kong, Pakistan, India, Mexico, and Brazil are the developing countries that meet these conditions and thus were selected for our analysis. These countries are fairly advanced among developing coun- tries, produce a large variety of manufactured goods, and are the major exporters of manufactured and semi-manufactured products among developing countries as indicated by Table III-2. Therefore, they are likely to be the main recipients of gains from generalized tariff preferences. Moreover, as these countries are already semi-industrialized rather than non-industrial, they will lend themselves to the application V‘s-‘h_ 1(v 63 TABLE III-l CRITICAL VALUES ASSIGNED TO ECONOMIC INDICATORS No. Economic Indicator Critical Value Assigned 1 Gross Domestic Product per capita Below $550.00 2 Manufactured Exports to DMEC's Total Exports 15% and above 3 Income Originating in Manufac- tures 12% and above Gross Domestic Product 4 Value of Manufactured exports ' to DMEC's $50 million and above 5 Growth rate of Manufactured Exports to DMEC's 15% and above P\U 64 TABLE III-2 EXPORTS OF MANUFACTURES AND SEMI-MANUFACTURES OF SELECTED DEVELOPING COUNTRIES TO THE DEVELOPED MARKET ECONOMIES, 1966 Value of Exports Percentage (Millions of of Country Dollars) total 1. Hong Kong 859.0 20.9 2. Taiwan 185.9 4.5 3. Pakistan 120.7 2.9 4. Mexico 229.8 5.6 5. India 451.6 11.0 6. Brazil 160.4 3.9 Other Developing Coun- tries 2,107.7 51.2 Total 4,115.1 100.0 Sources: U. N., "Commodity Trade Statistics," Series D; OECD, "Foreign Trade Statistics," Series C; "Exports of Manufactured Commodities from the Developing Countries," UNCTAD, Vol. IV, (New York: U. N., 1968), Table 15, p. 21. H. -. r Y I‘. (I) v.1 in f I l) ’1 65 of the comparative advantage models analyzed in Chapter II. The United States and the United Kingdom were selected as representative developed countries as they absorb about 70% of the manufactured exports of the developing countries. SELECTION OF COMMODITIES In reality the selection of commodities cannot be divorced from the choice of countries, but for the sake of analytical convenience they were considered separately. For the purpose of commodity selection, the manufactured and semi- .W- nan-.mp- ‘. “.mfil u... manufactured commodities were defined according to the methodology followed by UNCTAD which includes many commodi- ties lying in classes 0-4 according to SITC in manufactures and semi-manufactures.2 Table 7, Appendix II, lists forty- five commodities classified according to the SITC three- digit classification covering the bulk of manufactured and semi-manufactured exports of the developing countries to the developed market economy countries. Of these, thirty-six commodities were selected for our analysis. They are listed in Table III-3. The list includes almost all the manufactured and semi-manufactured exports of the selected countries. More- over, the growth rates of imports of manufactured products into the DMEC's from developing countries were also used in the selection process because the growth rates show the com- petitive ability of the developing countries in the export of manufactures. Commodities with more than a 10% annual fl. 'yFLFh r; . h - < .h. 66 .nuHHug magma "mousom moflunmm mafluxma Hmfiommm mmm moflunmm mDSh «.mmm npflsmoaou 6am suamsms ham monunmm amaooz N.mmo .oum .mcoou mcfluuomm .mmoe «mm moflubmm couuooncoz cm>03 mmm muosooum oaummam mom moflunmm couuoo mmm Hmuumz ompcwnm mom commas w cumw mafluxme Hmm nmm3uoom Hmm muosooum momma Nvo mcflnuoao Hum cumonummmm 6cm ummmm Hem mmmnocmm .mpooo Hm>mu9 Hmm .m.m.c .mmucuomMscmz @003 Nmm muouflcusm Hum U003>Hm tam ummcm> Hmm .m.m.: .moooo Hmoflunomam amp muosuoum “manna mmm msumummmg mcoaumoflcseeoomame «Nb Hmcpqu mo monsuomuscmz mam humcasomz Hmofluuomamucoz Han Hmcummq Ham .oum .sumauso mam .woum 6cm mamflumumz HmoHEmno mom couH mam Hum medmumm can mHHo Hmwucmmmm Hmm mnoswoum mumpuom mom mumnflm ofiumnucmm mom mumBmmme moo om>ummmum .mmanmummm> mmo mmmao vow Umummmum no om>nmmmum .pflsum mmo mmCHHm>ou Hooam nmm omnmmmum no Umccfla .nmfim Nmo .m.m.c .cmcwq .mxomm .mmmm mmm omummmum Ho Uwccfla .ummz mHo mufiooseoo mo coflumfluommo .oz muflooesou mo coflumfluommo .oz UBHm UBHm mmmbfiudmbz¢z m0 mamz¢m QMBUqum MIHHH mqm<8 67 rate of growth of exports were included in the sample unless the developing countries already accounted for more than 10% of the total imports of a particular commodity in the devel- oped countries. No commodity with less than $5 million in export value in 1966 was included in the sample in order to avoid incorporating the commodities with erratic and un- stable markets. NOTES TO CHAPTER III lDeveloped market economy countries include all coun— tries of the European Common Market, European Free Trade Area, U.S.A., Canada, Japan, Australia and New Zealand. Proceedings of the United Nations Conference on Trade and Development, Problems and Policies of the Manufac- tured and Semi-Manufactured Exports of Developing Countries, (New York: U.N., 1969). 68 “’4' -h z A D. e ,i \ .‘:‘ CHAPTER IV DERIVATION OF THE VARIABLES USED IN THE ANALYSIS INTRODUCTION This chapter estimates the effective tariff rates in the developed countries and develops alternative indexes to rank industries in the developed and developing countries by the degree of comparative advantage. We have already seen that only the factor proportions, human skills, R & D, scale economies model, and "revealed" comparative advantage, ade- quately approximate the principle of comparative advantage. While "revealed" comparative advantage forms the subject matter of Appendix I, and economies of scale are discussed in Chapter VII, other indexes will be developed in this chapter. 1 THE EFFECTIVE TARIFF PROTECTION The exports of processed goods from less developed countries to the developed countries are affected by the tar- iff structure in the developed countries, the transportation costs, the capital intensiveness and the technological re— quirements of the productive processes in certain products. The transportation costs provide "natural" protection to the importer and permit the remuneration of domestic factors in 69 AP «in? mun--‘krw .. I T 70 the industrial nations to exceed those in the developing countries. As capital is limited and expensive in the devel- oping countries, it provides them a disincentive for eXport- ing capital-intensive commodities. Similarly due to a shortage of skills and a lack of an advanced industrial structure, the developing countries are rarely able to export commodities at a high level of technological sophistication thereby providing "technological" protection to producers in developed countries. However, in recent years, considerable attention has been given to the effects of the industrial countries' tariff structure on the exports of processed goods from developing countries. This is due to the difference between the "nominal" and "effective" or "implicit" nature of protection provided by the tariff structure. The ordinary nominal tariffs apply to commodities, but resources move as between economic processes or activi- ties. Therefore, the resource allocation effects of a tar- iff structure depend upon the effective protection provided for each activity. The effective protective rate is the per- centage increase in value added per unit in an economic activity which is made possible by the tariff structure rela- tive to the situation in the absence of tariffs but with the same exchange rate. Alternatively, effective protection may be considered as the excess remuneration of domestic factors of production, obtainable by reason of the imposition of tariffs, as a proportion of value added in a free trade position. Therefore, in order to be able to sell in the 71 protected markets of developed countries, producers in developing countries have to operate with a value added smaller than that in the importing, developed, nations-- that is, only if the remuneration of domestic factors in the developing country is considerably lower than in developed countries. The simple theory of effective protection is based upon some highly restrictive assumptions as enumerated by Leith: (a) (b) (C) (d) (e) (f) (g) The rate of divergence between the free trade and protected prices of a tradeable measure the nominal tariff rate. The production functions are linear homogeneous. Strict input complementarity exists, implying zero elasticity of substitution between inputs. The elasticity of foreign supply of imports is infinite. The supply of domestic non-tradeable inputs is infinitely elastic. The elasticity of supply of other factor inputs to the domestic industry is less than infinite. Trade and production in protected industries exists both before and after the introduction of protection. Given these assumptions, consider j as an importable product, and i an importable input for commodity j. Now let Vj = value added per unit of jth activity in the absence of tariffs, 72 V'j = value added per unit of jth activity after tariff is imposed, tj = nominal tariff rate on j, i'= input coefficient, i.e., share of ith input 3 in the cost of jth output, net of tariff duty, t. = nominal tariff rate on input i, gj = effective rate for jth activity, Pj = pre-tariff price of j. Now 1 ....... V. = P. 1 - .. H J J( an) I = _ . (2) ....... V P [(1 + tj) aij(1 + tl)] V'. - V. P. 1+t. — .. 1+t. - l- .. 3 3 = J[( 3) a13( l) ( alJ)] (3) ....... gj = V. P. 1 - .. J J( an) (4) ....... gj = tj - 13 1 1" aij and . > t if t. =..>.. t. 9] if 3 J <: i Furthermore, for more than one input, n tj - .El'aijti (5) ....... g, = 1:1 3 n 1" E a.. i=1 13 and t. (6)°°°°"° ____2___.= gross subsidy per unit of value added 1 in j due to tariff structure, - a 73 a t iji ... (7)....... = impliCit tax rate (average) on value added in j due to tariff structure. l-aij Therefore, (8)....... . = S. - T.. 93 J 3 If j is the manufactured exportable of the LDC's to a developed country, then the granting of a preference to the m it... LDC's on jth product alone, other tariffs remaining unchanged, SKI. would reduce the subsidy--so far as competition with the F_ “a preference-receiving countries are concerned-~without reduc- ing the implicit tax on production. In the extreme case of 100% preference on all the manufactured products for the developing countries, the producers in the non-preferred, "third," countries would have a disadvantage by comparison with their competitors in the preference-receiving countries, in the markets of preference-giving countries, to an extent measured by the implicit or effective rate of protection. This would be the most likely outcome if a generalized tar- iff preference system is introduced, considering that the tariff structures of developed countries which are escalated by the stage of production and obviously bias trade towards raw materials, fuels, and semi-fabricated goods, towards producers’ goods rather than consumers' goods, towards goods of a luxury nature capable of bearing high tariffs, and to- wards both producers' and consumers' goods distinguished by technical superiority sufficient to overcome the competitive disadvantage imposed by tariffs (which are, of course, not . . .. .C x . (I vs. V». t .1. .fl 2. . . Pu tt v . FFL \: a. x n ... ( .qu 74 produced by the developing countries). Introduction of more realistic assumptions like international mobility of resources, the violation of assump- tions implying zero elasticity of input substitution, non- traded inputs and production for exports, will affect the magnitudes and the ranking of the effective protective rates.2 In the absence of a unique method of exactly estimating the actual effective rates, we will use the estimates prepared by B. A. Balassa in our analysis. 3 . . Professor Balassa has estimated the effective pro- tection in the developed countries in terms of protection provided to the domestic factors of production by using the pre-tariff input coefficients from the Common Market coun- tries. These estimates are presented in Table 8, Appendix II. These effective rates are "normalized" by obtaining a weighted average of effective rates of EEC, Japan, U.K., and the U.S.A., the main importers of manufactured goods from the developing countries. The effective rate of each devel- oped country (region) was weighted by the value of its total trade in the commodity to which the rate applies. The results are presented in Table IV-l. ESTIMATION OF COMPARATIVE ADVANTAGE INDEXES The input-output technique was applied to the esti- mation of total requirements of primary inputs to determine (1) capital/labor ratios and (2) the human skills ratios for ranking industries. The input-output technique is a method 75 mmumm m>Huommmm ocsom womccmxlwum« mon.om mOHunmm mIHHI< manna "monsom mafluxms Hafiommm mmm mmm.mm mOHunmm vum.mm suamzms mam aouuooucoz cm>oz mmo emm.mm .oum mmm.mm moflunmm conuoo mmm .moooo mcfluuomm .msoa 4am mpa.mm ammune a cum» mafluxme Hmm www.mm muosooum oflummam mam mam.HH muoswoum ummmm New mn>.o umpumz omucflum mom mmm.HH cumonummmm 62m ummmm Hem mam.om ummznoom Hmm mmo.hm .m.m.c mmm.qm mchuoHo Ham .mmusuommscmz coo: «mo oaa.mmmmmnwcmm .maoom Hm>mue Hmm mmm.mm woozsam 6am ummcm> Hmm ~mo.s~ muspflcusm Hmm moo.em muoscoum umnnsm mam hflo.mm .m.m.: maa.mm umnummq . mmvooo HMOHHDOOHW mm“. MO meSuUMwDCMZ Nam .pom.am msumummmm mmv.mm nmnummq Haw mCOHDMUHCSEEOUOHOE VNB 0mm . ma muoflmvoum UGO mmm.ma sumcflsomz mHMHumumz Hmoflsmno mam Hmofluuomamscoz Han Amm.o~ mmssmumm mHN.mN muoscoum cmanHunmm mam cam maflo Hmflucmmmm Hmm . mas mom.mm mumnflm oflumnucsm mom mmm.m couH mum Hem mon.ma um>ummmum .mmflnmnmmm> mmo mvH.m~ muoswoum mumuuom mom mom.ma cmummmum mmm.mm mumBmmme mmm Ho om>ummwum .DHSHM mmo mmm.mm mmmao «we mnm.n~ cmummmum moe.m~ magnum>oo Hooam Ame no omacfle .nmnm «mo Hmo.am .m.m.: mmm.am nmummmnm .cmcHA .mxomm .mmmm omw no Umccfla .umwz mHo .m.M|mHm muHUOEEou .oz ¥.m.M|mHm mpfloOEEOU .oz .mmpmm mo coflumfluommo oeHm mmumm mo coflpmfluomma oeHm m>fluommmm m>auommmm cmunmflmz cmunmflmz mmHBHQOEEOU DMBUmqmm m0 mMBdm m>H50mmmm DmEmUHMZ HI>H mqm4B —.r I :. 0.: tioz tin 76 of analysis, which based on certain theoretical assump- tions, makes possible the tracing out of the repercussions throughout the economy of a given change even in the most remotely connected sector. It is this structural inter- dependence that makes the input-output analysis so useful for estimating the "total"--i.e., "direct" plus "indirect" requirements of labor, capital and skills for the indus- tries. An expansion in one industry, given the output of other industries, causes all other sectors to expand produc- tion calling for additional inputs of capital, labor and skills. Therefore, the "total" requirement of a primary input for an industry is determined not only by the increase in the output of that particular industry, but also the secondary increases in the outputs of all other industries as well as the industry under consideration to sustain the required increase in the output of the given industry. 4 Computational Technique. The computation of total requirements of primary inputs, therefore, requires the "total requirements" input-coefficients matrix. This matrix can be obtained by subtracting the "direct" input coeffi- cients matrix, A, from the identity matrix, I, and inverting the resultant non-singular, square matrix, [I - A]. Notationally: (9) alla12 ........... ...aln . . direct input- _ a21a22 °°°°°°°° ..a2n coefficient matrix A ‘ where ai. = input of ith inustry to the jth industry; anlan2 ....... . ...... ann i (10) 1 o .............. 5 0 l .............. 0 I = . nxn . . . identity matrix. 9 0 .............. lJ t Then _-1 " _ _ _ r (11) 1 all a12 .............. a1n L - l- .............. — a21 a22 a2n (I-A)-l= . . , ! [anl -an2 .............. l-anrjj , or A A A _ (12) Fall a12 .............. aln " n . A A A total input a21 a22 .............. a2n requirements -1 matrix. (I-A) = o o ” — 3 3 a A - n1 n2 .............. nnj Given the vectors of physical capital, labor and skills coefficients encompassing all the industries in the input- output matrix, the "total" primary input requirements can be obtained by pre-multiplying the total requirements matrix 1 [I-A]- , by the respective primary input coefficients vectors. Notationally, 78 A A A x l O O O O O O O O a (13) all an 1:11 k1 A A A X a21 a22 ........ a2n = k2 (fkl’ sz' oooo ' fkn) o : : : A A x Lanl anz o o o o o o o o ann L kn . I where (fkj) is the coefficient matrix vector of the primary input K, and [ij] is the vector of total requirements of kth primary input for all the industries in the input-output matrix. Similarly, ij is the total requirement of kth in- put in jth industry. Once these input requirements are calculated, the input ratios to rank industries can be obtained easily. ESTIMATION OF CAPITAL/LABOR AND HUMAN SKILLS RATIOS Capital/Labor Ratios. The total capital and labor requirements were obtained by pre-multiplying the inverse or total requirements matrices by vectors of capital and labor coefficients respectively. The capital coefficients were computed in the net form by dividing the net book value of capital stock and the inventories by the current value added of the respective industries. The inventory values included the inventories of finished products, work in process, and materials.5 The labor coefficients were estimated in man- years as the ratio of the number of workers employed in an industry and its value added, i.e., the number of man-years required to produce one unit of value added. 79 The data on capital and labor coefficients was not available in sufficient detail for the developing countries. We, therefore, used the capital and labor coefficients of the Japanese industries for 1954 in estimating the input re- quirements for the developing countries. It is expected that these coefficients adequately reflect the industrial structure of the selected developing countries of today.6 For the U. 8. economy, the data on capital stocks was made available by Professor M. Gort of the State University of New York, Buffalo,7 and inventory values were obtained from the Censuses of Manufacturing, Agriculture, Mining and Business. Mr. Alterman of the U. S. Bureau of Labor Statis- tics furnished the data on labor inputs for the U. S. economy (1963). The capital and labor coefficients for the British industries in 1960 were obtained from the studies prepared by the Cambridge University Growth Project.8 These coefficients are presented in Tables 9-11, Appendix II, and were used along with national input-output tables to obtain capital and labor requirements. Capital/labor ratios were used to rank industries in the developed countries--a higher ratio implies higher comparative advantage. How- ever, for the developing countries the ranking was done by labor/capital ratios--a higher labor/capital ratio indicates higher comparative advantage. These ratios are presented in Tables IV-2 - IV-6. The switch in ratios is aimed at determining the nature of and significance of correlation between effective tariff rates in the developed countries Db.» u~<.~. 80 .ooflxmz mo magma DDQDDOIDSQCH mmum>cH .HH xflocmmmm .m manme "mousom mnmoo.o mam.ammm ~m5.mmm .m.m.: .mmusuommscmz “mayo nmm.smm «maoo.o omm.omqa qu.eme whosooum oflummam 6cm umnnom mom.mmo mmmoo.o Ana.mmma moo.Hmn mcflnmflansm a mcflucflum mom movoo.o vom.onmm Hmm.mmm ummsuoom .mcflnpoao Hmm.avm mmmoo.o som.mmom mam.omm .m.m.c .mvoou Hmofluuomam amp Gomoo.o mmm.enma mmm.oam sumcfinomz Hmofluuomamucoz Ha» mmmoo.o Gmo.ommm moa.mmm muosooum sawduso cam .muosooum Hmumz omumoflunmm mom emaoo.o wem.vmha mam.mom couH mam Hem mnaoo.o mmo.mmma Hmo.mmHH .woum snmuuom .mumzmmmflo .mmmau mmmuvmm sovoo.o www.momm mmH.Hmm .m.m.c .muosconm mafluxma nmmummo mmmoo.o www.mmkm mnm.mmm moflunmm cam mmafluxme mmmuamo mmmoo.o mma.ma~m nam.m~m muoscoum ummmm a ummmm Nvm.avm emmoo.o mma.ma~m hmm.HemH mposwoum a mflmflumumz Hmonsmno mam Hemoo.o mam.mmom ooa.mmma mmesmumm a maflo Hmfiucmmmm Hmm mnaoo.o omm.memfi mmm.vmn umnflm oflumgucsm mom ammoo.o vmm.mmmm man.ew~a muonnoum wmmmmooum umspo mmo.mmo.~mo opmoo.o mmm.mmvm mm~.mmma muosconm ummz 66mmmooum mac m m6n»u60fi500 CH kuflmmo Amumom awe "\oooam mo \uonmq uonmq an Hmuflmmo coflumfluommo oeHm Ausmuso mo mumaaoa coHHHHz H ammo iommav oonmz 2H mmHmemsazH oszaeommazmz mo mazmzmmHsomm momma ozg qaeHmao qmeoe NI>H mama?“ 81 .uwocH can caumwxdm mo Amouum>ch mmofiuumz unannouusmcn «HH xaocomma .m wanna “mounom ommwoo.o oav.mvm mom.mm~¢ In . I: In mowunmm couuooucoz cmmu ma hmamoo.o Hmm.o¢m mmw.ohn~ Hbvoo.o mom.mhm mo.mma¢ Hmmzuoom Hmm «a mmmmoo.o Hmm.mwv Huh.mmoa mwmcc.o mom.mom.a ma.nm~v muoacoum Hmuoz omuooflunmm wmm.mmm ma In In In momoo.o hmh.omo.a mm.mmmm mousuoomscmz .omwz ham.vmm NH amhmoo.o omm.>~m omm.mm¢a mHmoo.o «oh.mm> H~.HHv~ mmuduonw Incas Hmowuuuoam mmh.v~h Ha mmomoo.o «mo.qmm www.mana memoo.o mma.mmm m~.¢~mm mocflnumz Hmownuomaolcoz Han OH mmwwoo.o Hmm.am¢ Huh.mmoa momoo.o mom.msw mmm.mno~ mmfluumSpcn macaw: owmmm Hum m oewmoo.o oam.mmo voa.~>na Homoo.o mmm.~mm H~.omam .oum .mumuuom .mumzmmmao .mmmHo moonvmm m mhmaoo.o mam.omv omm.ohm oomoo.o ~on.omo.fl Hm.snmmmmwuumsocH Hmowaono mmm.amm h omomoo.o Hom.moo mom.mmv~ ommoo.o mmm.mmo.a wh.mmommmusuommscwz Magnum mmo m FNHmoo.o Hmm.ovm mmm.onn~ mmvoo.o mho.mmm.a mq.mm~m muospoum nonummq . paw uwfiwg Serge m In In In humoo.o mmm.amm oa.mmm~ mcwucaum Nam .muosooum .uoooo .mvo.~¢o v I: n: In onmoo.o mam.omm.a ~m.mmom muduwcndm Hum . .muosooum oooz .mmo.amo m mnvqoo.o omo.mmm mmm.ahmm anvoo.o mmm.mnm mo.mmav mownuoflu.moafluxoa . Hem hmmuamm N moavoo.o mam.vmo.a hm~.b¢~¢ Novoo.o hmv.a~m vm.mm~v muospoum coon mmo.mmo . . .Nmo.mHo a x\q x A qu x q mofiuwoossoo oucma .02 Amy mucmE Amado» cozy Amy mucmE Amumox cozy no cofiumwuommo uu>wsvm Imuasvom mucwemuwnvmm Iwuasomm mucwEmHszwm UBHm Hoodooo Hanna Houoe Houflomo gonna Houoa Hoummw Hmuow mwch cmumflxmm uaouso no nuuoz ounaaoo coaaawz oco ammo «HozH oz< zcameam mo mazmzumnoomm «Oman ozc qcH “HH xflocmmmé .m magma "mUHSOm vommoo.o Hma.mmH.H mmmo.ovn.m mmusuomMscmz wsomcmaamomflz nmmlvmm ma meamoo.o mmo.osm.a Hmoo.flmv.o monnuoao Hem ma mmomoo.o maw.mmm.a mwvm.mbn.m nomzpoom Hmm va mmmaoo.o vom.mvo.a Hmov.vmm.~ moonwaamm< HMUHuuomHm mmh.vmn ma mmmwoo.o Han.mmm.a mmam.voa.m wumcfinomz Hmoflupooamncoz Han NH nomaoo.o mam.vmv.a ommn.mvb.m muosooum Hmuoz omumofluoom oaosfim omo.mmo Ha mammoo.o mo>.HmH.H mvmh.oom.m Hmmum Ugo couH Hum oa mmmaoo.o mom.smm.a momm.omm.a muooooum smao paw whosooum mmmaw .mmmao mmmuvmm m moaaoo.o nov.aom.a ommm.oam.m manuafiono msomcmaamomflz mmm.amm m ommooo.o mmm.mmm.m wava.mwo.m mausponm UHDmmHm mom n vmamoo.o wmm.mom.a mmmm.mmm.m muospoum Hmbnsm mmm m mmomoo.o mam.mmm.a mmvm.m>n.m muosooum umnpmmq .Hmnummq mam.aam m momaoo.o mhm.omm.a oooo.mmm.m moooooum momma .uoomm moo.aoo a mmomoo.o H¢R.Hmm mmma.omm.~ mooooouo ooozsao a Hones; ~mo.fimo m mhvmoo.o mmo.vnm.a Hmvv.amv.v mwafluxme nmwnamw m omemoo.o mam.mmv.a momm.mmm.m muosoonm ooom pmccmo mmo.mmo mmo.mao a Hmuflmmo Annmaaoov Amummmacmev mmfluflpoeeoo mo coaumfluommo mucma .oz \Honmq mpcoEmuflsvmm mucmEmHasqmm |m>Hsvm Hmuflooo Hobos i I! i I Monmq Hmuoa UBHm Apoouoo mo onuoz muoaaoo oofiaaflz moo Home mazmzmmHoomm qaaHmmo oza momma gases m.z¢3H¢a TE mamas 83 Eocmcflx ponds: may no manwa unannouusmcH mmum>cH “HH xwpcmmmm .HH manna “mousom thHN.o «Nm.amam Hmo.b>m vao.mN> mam.ommm mmm.¢HmN chnuoHu va HNqu.o No¢.moma mNh.mam mNHm.mv0H wmm.vaN mmh.thN muocooum .wumcwnome Hmowuuomamucoc paw Hmofiuuomam mN>.HHh mmNNm.o mom.mNNN www.man oooa.mHHH www.mva mmm.mmNm .ouo .aumH luau .muospoum Hmumz Umumoannmm mam.mmm ooNNm.o oom.mmma Hmv.omo Hema.mHHN omm.mmmN mav.oovm muoapoum .couH ohm Hpo Hanm.o on~.omma mmm.mmm mmnm.oNNH mna.movN mmm.NHom muouuom new mom mum3mmmHu .mmmHu .mmm.¢mm oovHN.o mva.mNmN Nam.mam oawm.bmm mmv.mmvm moo.omvm muosooum mawuxma hmmuamm Howmv.o mNH.mowH HH.¢MN mmmo.nvNH moa.nom~ moo.n>mN mew» . uofiuo .ouosooum Nam momma can Hmong .Nqo.aeo HomvN.o mva.¢mMN mmn.¢mm mooa.hmm wvo.ommN omm.hmma muauwcuam can HNm muoooouo woos .Nmo.amo Nvam.o mmv.mmHN moh.wmn mHNN.omHH mmw.¢mmN mNN.oavm muospoum Hmnnsm mNm thHN.o va.HmHm Hmo.hno oNvo.mNn mao.ommm wmm.vamN H663 . uuoom .mpooooum Hmw nonumoq .uoaummq .NHo.HHo mmmmm.o 5mm.mmva vvv.bmn mmam.hhoN lo.mmNN mom.wmhv mamoN50nU .mmm.amm 0Nmm¢.o Noo.m¢HN va.mNm emNh.OMNH Nmm.mNmN mmm.momm cock oommmooum mmo.mmo .Nmo.mHo IMMII whom» ammo» “my whom» Aoooawv wowuwooasou UBHm Hm N name name uonmq acme Hmuwmmo mo cowumwuommn mluonnq Hmluonmq \Hmuwmmu mound vmnamxmcs omaafixw maawxm caEdm mcoHuuomoum Honoum 35950 no 5.33 coNHHNE H uuot Lemma. zoaosz omeHza use zH mquxm z¢zo= oz¢ mom of Dollars Years (In S, K/L) o 13 3681.821 206.112 17863 03 2 3681.821 206.112 17863 05 3 3681.821 206.112 17863 05 5 3681.821 206.112 17863 26 6 2488.286 148.671 16737 55 1 2526.923 135.354 18669 59 9 2519.419 135.091 18649 6 J. 1 1496.531 173.730 8614 6 12 1496.531 173.730 8614 62 9 2272.782 179.950 12630 6 3 1 2600.308 298.907 8699 6 3 2 2848.840 393.147 7246 64 1 2686.389 181.721 14783 64 2 2962.879 206.105 14375 65 :1 3193.237 315.321 10127 65 2 3193.237 315.321 10127 65 3 3096.384 351.684 8804 65 5 3096.384 351.684 8804 656 2995.055 249.558 12001 65 '7 2995.055 249.558 12001 66 4 2132.415 163.480 13043 66 5 2132.415 163.480 13043 66 6 2308.820 174.625 13221 6 7 :1 2919.251 182.364 16007 6 9 6 2525.916 201.649 12526 711 2347.645 188.283 12468 '72 4 1853.726 177.800 10426 729 2040.115 192.105 10619 82 :1 1633.902 259.325 6301 84 :1 2119.687 342.340 6191 85 :1 1711.711 288.878 5925 39-2 1934.880 177.464 10903 89 3 2272.782 179.950 12630 89 4 2222.712 215.890 10295 89 '7 2222.712 215.890 10295 ,iOui-ce: Table A-II-lO and the "'Total' Input Requirements bqiilole, U. S. Economy (1963)," Survey of Current Business, QV'ember, 1969. 85 and the comparative advantage schedules of developed and developing countries, approximated by the relative capital (or labor) intensity. Human Skills Ratios. Three alternative human skills iJndexes were developed to rank industries. They are the Ikotal Skills requirements, the "direct" skills requirements (tar Keesing), and the Productivity indexes. (1) Total Skills Requirements Index. The occupational structure of industries was divided into the following five groups of ski 115: 1. Professional and technical workers, 2. Administrators and managers, 3. Clerical workers, 4. Sales workers, 5. Manual workers and unskilled service workers. le, sz, Xj3’ Xj4’ and xj5 represent respectively 'tllee quantities of different types of skills required to If I91?c>duce one unit of value added in jth industry, then the ccNefficients of skilled and unskilled labor for various in- diVISStries and sectors can be estimated in the following way: Skilled Labor Coefficient for jth industry = 4 X'. = J M“ 1.; II H ji and Unskilled Labor Coefficient for jth industry = X"j = 86 X. 35 F. i=1 X31 Given these skill coefficients we obtain the total require- ments of skilled labor (Sjl) and unskilled labor (sz) in the following way: 3" l - -1 - 3 (14) (x j) [I A] — [sjl] g and ‘ u _ _1 — : (15) (X j) [I A] - [SjZ] i; where (X'j) and (X"j) are the skill coefficients vectors and [Sj l] and [SjZ] are the total skill requirements vectors. We then obtained the ranking of industries by Sl/S2 for the developed countries and by 82/51 for the developing cOuntries. The higher the ratio, the greater is the compar- ative advantage. The switching of ratios was, once again, ain‘ied at emphasizing the significance and nature of correla- tion between comparative advantage of developing as well as developed countries and the structure of effective protec- ti On in the developed countries. The skill coefficients for industries in the United St—ates and Britain were obtained from skills matrices for the U. 8. economy and Britain developed by Harowitz, Zymelman and Herrenstadt.9 No information on skill coefficients was a“’ailable for the developing countries. Therefore, Japanese Skill coefficients for 1950 were used to determine total Sl‘ills requirements ratios for industries in the developing 10 . . - countries. These coeffic1ents are presented in Tables 11-13, 87 .Appendix II. The total skill requirements ratios are presen- ted in Tables IV-5, IV-7-10. (2) The Keesing (Direct Requirements) Index. This :Lndex is based upon Keesing's estimates of direct skill re- cruirements ratios for the U. S. economy in 1960.11 He divi- ékas the occupational structure of American industries into 5 tile following eight occupational groups: 1 1. Scientists and engineers, 1T7- 2. Technicians and craftsmen, "9".“ mm: -_ 3. Other professionals, 4. Managers, 5. Machinists, 6. Other skilled manual workers, 7. Clerical, sales and service workers, 8. Semi-skilled and unskilled workers. :Kfieesing assumes that managers, manual skilled workers and <33L€2rical workers, etc. do not have a significant effect on ‘tllea trade patterns of countries. If Xkl' sz, Xk3' Xk4, ){}:55, xk6' Xk7' and Xk8 represent the direct requirements of respective skills for the kth industry, then the Keesing s”(ill index, S, can be computed as: 2(Xkl + sz + xk3) + xk5 (1 = 6) 5k Xk8 . The higher this index, the more skill-intensive is ‘tl163 commodity. The American industries were ranked by this 111(iexzas shown in Table 14, Appendix II. The occupational 88 bvflri ._. .4, -u 6. I. r. .OOonz mo oanoe mommaonusmcH ommo>cH .NHIHHua oamoe ”oomsom mnmmh.N mmN.momH mmm.Nmo .m.o.c .momsuoowscoz monmo 6mm.vmm omHmH.H 6mm.vmm va.mmN mmosooma ommmoaa poo monbom Nam.mNm hmmhm.o mam.mmm mmm.maoa mcflnmmamda m mamucmma Nam mava.N nmo.mmmN mam.mnoa mooauooa .mcmsmoao Hmm.avm mmNmm.H mmm.vHNH moH.Nnm .m.o.c .moooo Noummmooam mNN mmmoo.N mmv.mmNH Nmm.mNm amocmnooz Hoommmooaoucoz Han mmmmm.m moo.mmnm emm.meemuoooomo smommso .mmosoomo oomoommomm moo NNon.H mom.nmoa mmv.man comH mma Ham mommv.a mum.N>HH MNm.>mn muosooma mmomuoa .omo3mmoao .mmoao momuomm HmNNv.m mmn.mamN NHN.oam .m.o.c .mmosooma oammxoe hmmummm 0Nwmm.m mmm.mNmN mmm.amm mommboa cam moamuxoe mmmuamm mmomv.a mmn.noma mam.mmm mmosooma momma w momma Nvm.aom mmmvH.N mom.mmNN NNH.omOH muospoma w maommouoz HoomEoso mam vommo.a Hmo.moma mom.moam moEdmmoa w mamo Hommcommm Hmm mvaNo.H Noh.mmm Hmm.avm momma camosucmm moN mmo mmmmN.m MNN.H>mN mom.mom muosooma commoooma monmo.mmo.Nmo MNmON.m mmm.mNom Nmo.vvm muosooma coo: commoooma mac Nm\am Ammoom cosy Ammoom cosy mumpoafiou mo c0mmmmmomoo UBHm Nmumoomm mmumoomq ooaamxmca ooaamxm (ll II‘ Ausmmoo mo mmoaaoo comaame oco momv Aommav OUHxME mo mmHmBmDDZH UZHmbfiodeZ¢2 m0 mAAHMm ZdZDm m0 mBZmzmmHDomm AdBOB I: Emma 89 .Hmma .cm3HmB mo mommo>ch xmmmmz mammso msmcH “HH xmocomm< .NH oanme "oomsom wamh.m ammo.oon mvom.nmm momsmommsmmz msoocmHHoomHZ nmmlvmm mm mmmNm.m moon.mmmm mmmb.mom mcmsmoHO How ma movnm.m mHNo.vmhv womw.amm mmozmooa Hmm «a omon.N anom.mmma mmVN.mmo moocmmammm Hmommmooam mNh.th ma NmNNm.m mNNm.mnmN mbam.mhm mmocmcomz Hmommmooaolcoz Han NH moaom.v mmao.vaN NmNo.mmv mmosooma mono: moomommomm omoEHm omo.mmo mm oovom.m HNNm.>NmH momh.mmv Hoomm mam comH Hum om mmomm.v homb.nowa mvvo.on muosooma mmau ocm muosooma mmmao .mmmaw mowawmm m Homem.m Ova.mHhH mmmm.oam mHmomEonu msoocmaaoommz mmm.amm m Nmmam.N mvwa.omma mom>.mvm mmosooma omummaa mam n mmmmw.v mooo.NomN QONm.omv mp05poma moonsm mNo o movam.m MHNv.mmnv momm.amm muosooma monumoq .monmmoq Nam.aam m moome.e meme.mmam ommm.mme mooooomm momma .moamm Neo.mem e thmm.N ovum.mmvm noon.nmm mmosooma ooozmaa pom mobadq Nmm.amm m mmmNm.m mown.mmmm mmmh.oom moammxoe hmmlamm N wmoom.v wmom.mmam mmvm.vvm muosooma oooa ooccmu mmo.mmo .Nmo.mao Ha momma Ammmom amev Ammmom cmEv mommmooEEoo muooa .oz ooaaflxm mucoEommsqom mucoEomHSqom mo COHumHmomoo Im>msvm momma momma poaamxm monmq poaamxm OBHm ooHmemco 1CD ammoe Hmuoe mosauso mummmoo commmmz moo moms zmzmma mo mezmzmmmoomm mqqaxm momma mu>m momma 90 .mamcm cam cmummxma mo Amommo>ch mooamumz usmusolusmcm «HH xaocommd .NH oamma “oonaom momvom.m omm.mahm mmN.mom 11 11 11 moamuxoa ouah v mmw ha 11 11 1| mmquo.o oNh.maaa vNo.mN>H .msa ocm .nmma Nam ma ommmmm.o www.mva maa.omm mvhomn.a av¢.NmMN moo.omma mmozuooa amw ma :1 11 11 MNNoow.a th.mMNN ooa.¢mma omauacmsa ANN va nmomma.m Nam.mmNa oom.aov ommmmm.o mae.aoma mmv.thN muosooma ammo: commommmma wam.mmo ma :1 II 11 vmsmmm.~ mm~.vmam mmN.ohma .ommsqmz .ommz ham.¢om Na ovamao.N mom.mvoa mam.mmm momnmw.o mNo.bmoa oam.oHNa .ommscmz .ooam mNh.¢NN Ha smegma.m Nam.mmNa oom.aov memNm.o omo.mmma mom.hNha amocamomz amommuooaonaoz Han oa Nmmmhm.m www.mON oom.ama thNom.o Nav.mNm mmv.mmaa mommmmsp 1cm mammoz ommmm mew m ommNNN.o vme.mooa omm.ana NommNN.a mom.N>HH nov.mmm Ammomuom .ommamv .voma ammo 1cmz omaamuozlcoz oomuvmo m NaNomm.m omn.wmn oNv.aNN Namnom.o mma.mmNa oom.mmma .msmcm HmomEomu mmm.amm h vsnova.m mNm.mmoN mvm.oov ovvmm.o mmo.mvha www.mvma mausooma mommsm mNm m ommoha.m mmN.m¢MN mNm.va moomvv.a «mm.Nmom aNN.mmaN muosooma momumoa a momumoa Nao.aao m 11 In 11 mmhvam.o ann.vo~a NNm.NmmH muoaooma .momma Nqo.avo w mmmth.m mmm.Nmma Nmm.MNN mommam.a nvm.booN mam.NNoN muosooma 0003 wa.amo m aommmm.n meo.Nomm m~m.mvv omamwm.a www.mm>N amo.ohma mcammoao Hem mom moamuxoe .Nmo.amo N hmomoh.m Noe.onom moa.mmm Novam.a ONm.ommN www.mbma muosooma oomsu mmo.mmo unmascmz .oooa .Nmo.mao a momma Ammmommmmmoa :mEv momma Ammmom Ammmom cmev momumooeeou mucoa .02 o a m cmzv mucoE moHHaxm :mz. mucoe m0 comummmomoo 1m>mswm poaaaxmcs mucoEommsv nommsvom moaamxmcs mucoEomasv Lommsvom OBHm tom momma momma Iom momma momma poHHaxmcD. moHmem woaaaxoco woawmxm «HQZH ZdBwada III (l‘ Amsauso «0 mummaoo comamaz moms «SE ea EEE .3 mazmzmmsgm 3.5m mesa mI>H mqmdfi 91 TABLE IV-lO TOTAL SKILL REQUIREMENTS PER MILLION DOLLARS OF OUTPUT IN U. S. INDUSTRIES (1963) Total Skilled Labor Requirements (in man years) Total Unskilled Labor Requirements (in man years) Total Skilled Labor Requirements Per Unit of Unskilled Labor (man years) output T EEEEi§J November SITC (1) (2) 1/2 ‘6. 013 66.23389 140.51726 0.47135 032 66.23389 140.51726 0.47135 053 66.23389 140.51726 0.47135 055 66.23389 140.51726 0.47135 266 58.48164 90.17667 0.64852 551 54.98552 79.84508 0.68865 . 1 599 51.64904 90.17667 0.61987 611 36.94547 136.07076 0.27151 612 36.94547 136.07076 0.27151 629 59.21207 120.58467 0.49104 631 56.51263 242.47513 0.23306 632 72.92902 320.39453 0.22762 641 51.84514 130.86914 0.39616 642 57.74442 146.73225 0.39353 651 58.73003 256.57239 0.22890 652 58.73003 256.57239 0.22890 65 3 65.77014 283.84129 0.23171 655 65.77014 283.84129 0.23171 656 53.46978 194.50789 0.27489 657 53.46978 194.50789 0.27489 664 44.81665 118.38448 0.37856 665 44.81665 118.38448 0.37856 56 6 46.98184 128.41057 0.36587 5'71 42.55332 142.42001 0.29878 696 60.95037 140.85676 0.43271 711 42.55332 129.13870 0.45101 72 4 61.45302 115.98638 0.52982 729 63.40664 128.12369 0.49488 82 1 62.92258 211.67381 0.29726 841 61.54647 280.81126 0.21917 851 57.71474 231.14629 0.24968 892 86.72599 90.38221 0.95954 89 3 59.21207 120.58467 0.49104 894 66.44263 149.35757 0.44485 897 66.44263 149.35757 0.44485 \ ESCDEErce: Table A-II-l3; "Total Input Requirements Input- able, U. S. Economy (1963)," Survey of Current Busi- 92 data for other countries was not available in detail suf- ficient to construct the Keesing skill index. (3) Productivity Index. There is a systematic and unique relationship between the productivity and the skill composition of an industry. Zymelman has shown that a higher level of productivity is systematically associated with a higher proportion of highly skilled workers in the final product and vice-versa.12 This holds for all manufac- turing industries in twenty-one countries analyzed by Zymelman. A ranking of industries by the level of produc- tivity will, therefore, be consistent with the principle of comparative advantage. The higher the productivity, the greater is the comparative advantage of skill-intensive, de- veloped countries, where productivity is estimated by fitting linear regression of the following form: .= +..X..+B.X.+ ....... +B.X.+6, (17) y] A 813 13 23 23 n] n] where y. = value added per person in industry j, i.e., pro- ] ductivity in industry j. xi. = The proportion of type of labor (or occupation i J (i = 1,2, ..... , n) in the total labor employed in industry j; Bij = Regression coefficients; A = Regression constant; 6 = Regression error. The productivity (yj) estimates were obtained from the study prepared by Zymelman for the U. N. Planning and Programming Series,13 and are presented in Table IV-ll. .661666 .mm .26666 ..z .6 ”xmom 3626 .6 .oz oommom acme IEmmmoma ocm momcamaa HmmmmmsocH .NmOHocmooB mam maaaxm poocm>©< m0“ acaccmaa 93 =.mo6mmmD©cH mcflmsuommscmz c6 cowmmooom Ucm maaaxm .mu6>6m05o0ma= .cmanEmu .2 "oomsom 666.6 666 666.N 66N.6 1.6.o.o ..omo .maoe .mooow mcammommv momsmommsmmz .0662 666.666 1. 666.6 666.6 666.66 oucoe IDHDmGH UHMflHGmHUm Cam HMCOHmmOMOHm How 666.6 666.6 666.6 666.6 oeoemmoaa m6ommoooma 666 666.6 666.6 666.6 666.6 6moomao6z 666666666 .666662 ooo6ommo6a 666.666 .1 1. 666.6 666.66 moomm 666 comm 666 666.6 666.6 666.6 666.66 moosooma 6660 6 ooomm .66666 666.666 666.6 666.6 666.6 666.66 mmosooma 66065660 6 666omEono 666.666 666.6 666.6 666.6 666.6 moooooma mooosa 6N6 666.6 666.6 666.6 666.66 memmommooa 6 mommcmma 666 666.6 666.6 666.6 666.66 mmooooma momma 6 momma 666.666 666 666 666.6 666.6 omoomcmoa memosmoom .mmooooma oooz 6 mooesa 666.666 666 666 666.6 666.6 oomomxma 6 omommomoa 666 666.6 666.6 666.6 666.6 m6o3mooa 6666 156066 moooooma monm6oa 6 moom6oa 666.666.666 666 666 666.6 666.66ommmxo6 oom6ommoma mommo 6:6 aemnmomo 666 666.6 666 66N.N 666.6 mmosooma 6662 oammxoe 666-666 666.6 666.6 666.6 666.66 6o66mo>om 666 mooa 6661666 266663 2666m6 266663 266666 amm6oeeoo 6o oommmmmoooo 6666 mcaucomma mmmmh .M.D .m.m.D amm666oo .6 .2 666 6666666626 ozHaoeomaDzmz ao 6666626 mam>mmooooaa HHI>H mam¢8 94 RESEARCH AND DEVELOPMENT There is no exact method of ranking industries ac- cording to their R & D intensiveness. One R & D index that has been used by Keesing for the United States14 is the ratio of engineers and scientists employed in R & D activities to the total skilled labor in the industry. Practically no in- formation is available on the employment of human skills in R & D for the developing countries. Therefore, we shall ......A‘ -6 confine ourselves to a theoretical discussion of the R & D and general preferences, to be presented in Chapter VII ‘E- below. AGGREGATION OF EFFECTIVE RATES While the estimated values of the effective protec- tion rates in the deve10ping countries are available at the three digit SITC classification level, it was not possible to obtain the ranking of industries in the developing coun- tries according to various indexes developed in this chapter at the same level due to the highly aggregated nature of developing countries' input-output tables. Therefore, it was necessary to adjust the effective protective rates by obtaining weighted averages of tariff rates for the three digit SITC industries consistent with the available input- output detail, with each rate weighted by the value of world trade in the product to which it applies. 95 CONCLUS ION In this chapter we have estimated the effective pro- tection rates in the developed countries and alternative methods of ranking commodities in developing and developed countries in the order of their comparative advantage. NOTES TO CHAPTER IV 1See B. A. Balassa, "Tariff Protection in Industrial Countries: An Evaluation," Journal of Political Econom December, 1965, pp. 573-94; Giorgio Basevi, "The United States Industries and Industrial Labor," Review of Economics and Statistics, May, 1968, pp. 147-60; W. M. Corden, "The Struc- ture of Tariff System and the Effective Protective Rate," Journal of Political Economy, June, 1966, pp. 221- 37; G. H. ijohnson, The Theory of Tariff Structure with Special Reference to World Trade and Development, " Trade and Devel- oEment (Geneva: Librairie Droz, 1965); Economic PoliCies Toward Less Developed Countries (Washington, D.C.: The Brooking Institution, 1967); Stephen R. Lewis and Stephen E. Guisinger, "Measuring Protection in a Developing Country: The Case of Pakistan," Journal of Political Economy, 1968, pp. 1170-97. 2For elaboration on the issue of assumptions and the implications of their relaxation, see: Clark J. Leith, "Substitution and Supply Elasticities in Calculating the Effective Protective Rate," Quarterly Journal of Economics, October, 1968. B. A. Balassa, Kennedy Round Estimated Effects on Tariff Barriers, Part II, UNCTAD, Table A, pp. 209- 13; Tariff Protection in Industrial Countries: An Evaluation," Journal of Political Economy, December, 1965. 4This formulation of total requirements input-output matrix is adapted from: W. I. Abraham, National Income and Economic Accounting, (Englewood Cliffs, N.J.: Prentice—Hall Inc., 1969), notably Appendix to Chapter 5, pp. 184-88. A rigorous proof is discussed bv Leontief in his In ut- Output Analysis, (Cambridge: Cambridge UniversiEy Press, 1966). The input-output tables and their inverses were obtained from the following: Wouter Tims, Input- Output Table for Pakistan, 1963- 64 (in current prices), mimeographed, Harvard Development Advisory Service and Planning Commission, Government of Pakistan, 1965; Council for International Econ- omic Cooperation and Development, Executive Yaun, Taiwan' 5 Interindustry Inverse Matrix Table for 1961 (37 sectors), September, 1964, Taipeih; Mexican Input- Output TablesiI960, (Mexico: Banco De Mexico, 8. A. ); A. S. Manne and A. Rudra, et. al. "A Consistency Model of India's Fourth Plan," M.I.T. Center for International Studies, Sankhya, Series B, September 1965, pp. 57-144. "Total Requirement (Inverse) 96 97 Input-Output Matrix, U. S. Economy, 1963," Survey of Current Business, November, 1969; British Input-Output Relationships 1954-1966, A Program for Growth, No. 3, Department of Applied Economics, University of Cambridge, Supervised by Stone, Richard; Champman and Hall, 1963. No input-output tables are in existence for Hong Kong and Brazil. 5The methodology followed in the estimation of capi- tal coefficients is derived from the following: John W. Kendrick, Productivity Trends: Capital and Labor, National Bureau of Economic Research, (New York: ColumbiE University Press, 1956); "Industry Changes in the Non-Labor Costs," Industrial Composition of Income and Products, (Kendrick, ed.) Brookings Institution and National Bureau of Economic Research, (New York: Columbia University Press, 1968), pp. 151-75; D. Creamer, S. P. Dolseovolsky and I. Borenstein, Ca ital in Manufacturing and Mining, Its Formation and Financ1ng, (Princeton, N. J. Pr1nceton University Press, 1960). The Japanese capital and labor coefficients were obtained from the following sources: National Income Accounts, 1957, Economic Research Institute, Economic Planning Agency, Japan; National Wealth Survey, 1955; Economic Bulletin No. 1, February, 1959; Capital Structure of the J_panese Economy, Council for Industrial Planning, T6kyo,gl958; Japan Labor Yearbook, Council of Industrial Planning; T. Watanake, "Approaches to the Problems of Inter- Country Comparison of Input-Output Relation: A Survey and Suggestions," International Comparisons of Inter-Industry Data, U. N. Industrial Planning and Programming Series, No. I, (New York: U. N., 1969), pp. 187-210. 7The Gort-Boddy estimates will appear in a forth- coming article on productivity changes in the U. 8. economy from 1945-46 to 1963-64. Professors M. Gort and R. Boddy were kind enough to let me use their estimates of capital stock. 8 A Program for Growth, Capital, Output and Employ- ment, 1948-1960, The Department of AppIied Economics, Univer- sity of Cambridge, Chapman & Hall, April, 1964. 9M. A. Horowitz, M. Zymelman, and I. L. Herrenstadt, Manpower Requirements for Planning: An International Com- parison Approach, (Roston: Northeastern University Press, 1966). PP 16- 17, 46- 47. loIbid., pp. 30-31. 11Donald B. Keesing, "Labor Skills and the Structure of Trade in Manufactures," The Open Economy, (Peter B. Kenen, ed. ), pp. 1- -l4, especially TaBIe I. 98 12 . . . . M. Zymelman, "Product1V1ty, Skills and Education in Manufacturing Industries," Planning for Advanced Skills and Technology, Industrial Planning and Programming Series No. 3, (New York: U. N. , 1969), pp. 103-38. l3M. Zymelman, op. cit., Annex I, pp. 119-37. 14 . D. B. Kee51ng, "The Impact of Research and Devel- Opment on United States Trade," The Open Economy, (New York: Columbia University Press, 1968). " 1E}. i,” “A“ Dm‘Lfl VI: CHAPTER V EFFECTS OF GENERAL PREFERENCES ON DEVELOPING COUNTRIES WHEN COMPARATIVE ADVANTAGE IS ESTIMATED BY THE FACTOR PROPORTIONS AND HUMAN SKILLS MODELS 1Q? INTRODUCTION y In this chapter we draw upon the factor proportions W77 l’ .l and human skills models to compare the schedules of manu- factured commodities in developing countries ranked according to their comparative advantage with the effective tariff rates in the developed countries to determine the usefulness of generalized tariff preferences. On the assumption that the less developed countries will respond to preferences and ex— pand production and exports of the preferred commodities, it is important to know whether the expansion will take place in industries in which these countries possess the greatest comparative advantage. The gains from generalized tariff preferences, therefore, depend upon the extent to which the structure of effective protection in manufactured products in the developed countries is positively correlated with the com- parative advantage schedules of developing countries in manufactures . FACTOR PROPORTIONS MODEL AND IE DEVELOPING COUNTRIES While analyzing the investment criteria and the 99 100 comparative advantage in the developing countries in the second chapter, we came across the problem of defining the prospective or "incremental" comparative advantage in the dynamic setting of changing factor endowments and technol- ogical change which obtains during the growth process. Growth is associated with the investment or capital formation; in- crease in the supply of labor through immigration, natural increase, education and training, and the increase in the quantity of natural resources through discovery of resources previously unknown. As the stock of factors of production changes, the specialization that is most advantageous changes as well. Furthermore, growth is accelerated by the growth of technology which may not necessarily be growing at the same rate and in the same direction in all the countries. This may give rise to differences in production functions from country to country and cause shifts in comparative advantage completely independent of the changes in factor endowments. The changes in factor~endowments and technological change may be expected to greatly limit the usefulness of the factor proportions model in determining the comparative advantage of a developing country.1 Nevertheless, a devel- «oping country's present comparative advantage provides the .best available guide to its appropriate choice between in- <1ustry and agriculture, complex and simple manufactures, :hnport-replacement and export expansion, as it develops. (”115 is so because the fact that economic development is pro- ceeading does not alter conditions sufficiently to alter the 101 conclusions. One should also note that growth is taking place in developed as well as the developing countries ac- companied by capital formation, etc., and such developments are likely to neutralize each other, leaving the relative fac- tor endowments substantially unchanged. Moreover, the coun- tries selected are semi-industrial rather than totally non- industrial so that some indication regarding their comparative advantage exists and it can be adequately reflected by the factor proportions model. However, this is also a limitation because past industrial development may not have been along comparative advantage. Furthermore, if unequal rates of change in productivity were the only allocational difficulties caused by economic development, relative factor endowments and calculations of comparative advantage based on it would still be the best single guide to allocation. Besides, if a country's techniques are advancing in one field, broadly speaking they tend to be advancing in other fields as well, and this widespread advance tends to leave relative resource endowment an important factor in comparative advantage. Moreover, even in a country in which technical progress is taking place and changing comparative advantage, previous comparative advantage is the point of departure from which allocation should be decided. It is, therefore, reasonable to use the ranking of Inanufactured commodities in developing countries according tn: labor/capital ratios as representing the degree of compar- ative advantage. Developing countries endowed with relatively 102 greater stock of labor than capital are expected to have comparative advantage in the production and export of rela- tively labor-intensive commodities_ i.e., those for which labor/capital ratios are relatively higher. If the structure of effective protection rates in the developed countries is significantly and positively correlated with the schedule of manufactured commodities ranked according to labor/capital ratios in the developing countries, then the generalized tariff preferences for the manufactured exports of develop— ing countries will promote the expansion of industries in which these countries possess comparative advantage. Chapter IV applied the input-output technique to rank manufacturing industries according to the ratio of total labor and total capital requirements. The input-output tables for the developing countries are not available in detail sufficient to estimate labor/ capital ratios for the commodities at the three digit SITC classification level. Input ratios could only be estimated for less than twenty broad groups of manufacturing industries. The effective protection rates were accordingly adjusted by [obtaining weighted averages of tariff rates for the three P. United States: Y = 43.7832 - 0.001568 x R2 = .268944 a .L(Z.60093) (—2.10391)* R =—.518598 6 Y = 47.3556 - 0.000663 16g x R2 = .09613 ' (2.92106) (-1.29271)* R =-.310049 16g Y = 26596.7 - .76224 log x R2 = .100023 Q (3.0000) (-l.0829) R =—.316264 , y Number of observations = 35. United Kingdom: Y = 41.69940 — 0.0388467 x R2 = .597245 (2.970132) (-3.182530)* R =-.772816 Y = 44.1546 - 0.0305807 log x R2 = .607387 (2.83510) (-2.77150)* R =-.77935 16g Y = 556.1830 — 0.512711 16g x R2 = .544482 (3.089881) (-2.54190)** R =-.73789 Number of observations = 13. = The effective protection rate on i-th activity in the = the fio= the developed countries, capital/labor ratio of the i-th industry, t- values are presented in the parentheses, indicates values that are significant at the 0.05 level. The regression analysis indicates very significant «and negative correlation between capital/labor ratios in the United States and United Kingdom and the effective pro- tection rates in the developing countries. This result con- forms with Basevi's conclusions that the U. S. tariff 118 structure provides especially heavy protection to the labor- intensive industries.2 This means that the U. S. and U. K.-- the two major importers of manufactures from the developing countries--tend to protect most heavily the activities in which they have least comparative advantage. A generalized preference system will, therefore, give the greatest incen- tives to developing countries to export those commodities to “P - mas—l ‘59: .I i K. ...-1 (a? developed countries in which the latter have least compara- tive advantage. Hence, the trade diversion effect will be minimum. Rank Correlation Tests. The results of rank corre- lation tests (Table VI-l) tend to reinforce the conclusions derived from regression analysis. Very significant negative rank correlation coefficients characterize the relationship between effective protection rates in the developed countries and comparative advantage schedules of the U. S. and U. K. TABLE VI-l RESULTS OF RANK CORRELATION TESTS-- FACTOR PROPORTIONS MODEL Coefficient of Signifi- Number of Coun- Rank Correla- t- cance Observa- tries tion (r) Value Level tions U. S. A. -0.61116 -4.227407 0.005 35 U. K. -0.86540 -5.4415 0.01 13 Sources: Tables IV-l, IV-5,6. 119 HUMAN SKILLS MODEL Developed countries with a relative abundance of professional personnel and highly trained labor are expected to have comparative advantage in the production and export of the relatively skill-intensive goods, and comparative disadvantage in the relatively less skill-intensive products. We can, therefore, rank commodities according to their skill content to represent comparative advantage. The schedule of ranked commodities can then be compared with the structure of effective protection rates. As develOped in Chapter IV, three alternative skill indexes--the Total Skills Requirements, the Keesing Skill and the Productivity Indexes--were used to rank the commodi- ties. Linear regression and rank correlation tests were carried out to determine the nature and significance of correlation between effective protection rates (Y) and the schedules of industries for the United States and United Kingdom, ranked by the human skills indexes (X). Regression Tests. U. S. A. The results of regression tests regressing human skills indexes (X) on effective protection rates (Y) in developed countries are summarized by the equations given below: (a) The Total Skill Requirements Index (81/82): (7) Y = 38.2583 — 0.0321996 x R2 = .281609 (4.10080) (-2.71901)* R = -.530668 (8) Y = 64.5978 - 0.027356 16g x R2 = .392586 (3.58991) (-3.00461)* R - -.626567 120 A (9) log Y = 8027.39 - 0.10056 log x R2 = .281173 (2.98000) (-2.27325)* R = -.530257 Number of observations = 35. (b) The Keesing Skill Index (S): A (10) Y = 37.0569 - 0.0033009 x R2 = .290301 (4.60032) (-2.1130)* R = -.59080 Number of observations = 32. (c) The Productivity Index (y): A (11) Y = 40.7779 - 0.0016975 x R2 = 0.264707 (2.00019) (-1.77725)** R = -.514497 (12) Y = 46.1640 - 0.0000908 log x R2 = 0.0912906 (2.2336) (-0.62013) R = -.302143 (13) 16g Y = 820411.0 - 1.16796 16g x R2 = 0.133697 (2.63000) (—0.69300) R = -.365645 Number of observations = 18. United Kingdom. The following equations summarize the results of regression tests: (a) Total Skills Requirements Index (Sl/SZ): A (14) Y = 42.3178 — 0.00729763 x R = .39177 (3.29781) (-2.235510)** R = -.625915 A (15) Y = 47.7023 - 0.0229742 16g x R2 = .434055 (3.49915) (-2.511943)* R = -.658828 A (16) log Y = 812.5690 - 0.294248 log x R2 = .227069 (2.095483) (-1.988101) R = -.476518 Number of observations = 13. (b) Productivity Index (y): A (17) Y = 42.7928 - 0.000209225 x R2 = .159585 (2.94116) (-1.55730) R = -.399481 A (18) Y = 41.2415 - 0.00858098 log x R2 = .17210 (2.87690) (-l.730496)** R = —.41485 A (19) log Y = 499.4999 - 0.0795775 16g x R2 = .0832986 (2.23009) (-1.00475) R = -.288615 Number of observations = 18. n'-'l__‘y "filfi- ‘8 0'.“ .251 I 121 *;**; indicate that the results are significant at the 0.05 and 0.10 levels respectively. Tests involving total skills requirements indexes show pronounced negative correlation between the structure of ef- fective protection rates in the developed countries and the comparative advantage of the U. S. A. and U. K. The parameter is rather small for the United Kingdom which could E be due to the high level of aggregation. However, the coef- ficients of correlation (R) are, in general, significantly high and negative. The Productivity Index, on the other ‘64.": hand, fails to show any significant results. This index is based on theoretically weak foundations as it considers the skill content of the "final" products only. This could be a reason for the weak negative correlation indicated by the Productivity Index. Rank Correlation Tests. The results (Table VI-2) of rank correlation tests tend to support the conclusions derived from regression analysis presented above. CONCLUSION We have used factor proportions and human skills models to analyze the relationship between effective protec- tion rates of developed countries and the comparative advan- tage of the U. S. A. and U. K. in manufactured products.2 The statistical investigation shows that the developed coun- tries tend to protect most heavily the industries in which they have least comparative advantage. Therefore, a general- ized preference system granting the greatest incentives to 122 TABLE VI-Z RESULTS OF RANK CORRELATION TESTS-- ALTERNATIVE HUMAN SKILLS INDEXES Coefficients Signifi- Number of Coun- of R Correla- t- cance Observa- tries tion - r value Level tions TOTAL SKILLS REQUIREMENTS INDEX (SI/82) U. K. -0.6950 -3.0536 0.05 13 U. S. A. -0.50061 -3.2065 0.01 35 KEESING SKILL INDEX (S) U. K. -- -- —- -- U. s. A. -0.59080 —3.33570 0.01 32 PRODUCTIVITY INDEX (Y) U. K. -0.54660 -2.35757 0.05 18 U. s. A. -0.65340 -3.22556 0.01 18 Sources: Tables IV-1,5,10,ll. developing countries that are subject to the greatest pro— tection in the latter, will minimize the trade diversion effect of preferences. Furthermore, it may help move re- sources in the developed countries toward industries in which they have relatively greater comparative advantage and away from those sectors in which they have least comparative ad- vantage. NOTES TO CHAPTER VI 1In view of the fact that the U.K.‘s input-output table was not available in required detail, the effective protection rates were adjusted by obtaining weighted averages of tariff rates for the three digit SITC classified indus- tries consistent with the input-output detail available for the U.K. The bulk of world trade in a product was used as weight for the tariff rate on that product. Giorgio Basevi, "The United States Tariff Structure: Estimates of Effective Rates of Protection of United States Industries and Industrial Labor," Review of Economics and Statistics, Vol. 48, May, 1966, pp. 147-60. 3Although no effort to aggregate human and non- human capital was made to obtain ranking of industries ac- cording to their overall capital intensity, such an aggre- gation is most likely to improve the results because both factor prOportions and human skills models yielded the same results. A multiple regression was, however, carried out to determine the degree of interdependence between the struc- ture of effective protection in the developed countries and the comparative advantage schedules of the U.S.A. as approxi- mated by capital/labor ratios and human skills (total require- ments) ratios. The results are summarized by the equation given below: A Y = 45.579 - 0.002270 X1 - 0.0092224 X2 (0.0109364)* (0.005226) t = 1.7646 t = 2.07602 1 2 R§Xl = -.535088 R2 = .3555 RYX = -.509648 F-Ratio = 7.9988 2 (* values in parentheses are standard errors) where, Yi = Effective tariff protection rate on i—th activity in the developed countries; X11: K/L of i-th industry in the U.S.A.; Xi2= Sl/S2 of i-th industry in the U.S.A. 123 CHAPTER VII RESEARCH AND DEVELOPMENT, TECHNOLOGY, ECONOMIES OF SCALE AND THE GENERALIZED TARIFF PREFERENCES INTRODUCTION The factor proportions models of trade assume that tastes are given, knowledge is or ought to be a free good and that scale effects are absent. However, these assump- tions are often not satisfied. It is, therefore, possible to cut down the unexplained variance in international divi- sion of labor by incorporating the influences of such aspects as research and development (R & D), technology and the economies of scale. In this chapter we will try to analyze the R & D and the economies of scale in the context of generalized tariff preferences. R & D AND THE STRUCTURE OF TRADE One way of analyzing the impact of technology on the pattern of trade is to look at the relative intensiveness of a country in "research and development" expenditures. Research and development brings the technical progress that either creates new products which confer temporary "avail- ability" advantage on the economy, or helps to develop new processes, lowers Costs, attracts resources into these indus- tries and confers comparative advantage on the economy in 124 .. 1H5"? 125 research-intensive industries. The influence of R & D on the trade pattern can be analyzed in three alternative ways: neo-factor-proportions formulation, the market-scale model, and the product-cycle approach. . l Neo Factor-Proportions Approach. Harry Johnson, in his Wicksell lectures, introduces a "dynamic" theory of com- parative advantage that draws upon a neo-Schumpeterian inter- pretation of innovation as the basis of future patterns of 1~"'-"““““Hw trade. In this analysis, human capital engaged in research and development is considered as a component of a nation's factor-endowment and influences its trade pattern in the fashion indicated by the factor-proportions model. This would imply that nations well-endowed with human capital resources suited for R & D--i.e., developed countries--are expected to produce and export R & D-intensive products and import technologically simple products.) On the other hand, developing countries, scarce in R & D resources, are likely to export technologically simple products and import R & D- intensive products. This would require, inter alia, that the productivity of R & D resources in a given line be des- cribed by a production function that is invariant among countries, i.e., the phenomenon of factor-intensity reversals is absent. Keesing; Gruber, Mehta and Vernon; and Baldwin2 use models similar to the one presented above to determine the comparative advantage and export-competitiveness of the 126 U.S.A. versus the OECD and other developed countries. It is possible to use either the ratio of R & D scientists and engineers in total employment in each industry, or the ratio of R & D expenditures to value added in each industry as the R & D index, to rank commodities according to their relative R & D intensiveness which would approximate the principle of comparative advantage. If effective tariff protection rates in the developed countries are relatively higher on commodities with a relatively lower R & D-intensiveness in- dex, then a preferential treatment of manufactured imports from developing countries into developed countries will have desirable economic effects on developing countries. Rank and linear correlations were computed for the commodities arrayed by their respective R & D indexes in the U. S. A. (per cent of engineers and scientists in total em- ployment), and the effective protection rates in the devel- oped countries. The results as presented in Table VII-l below indicate significant negative correlation between tariff rates in the advanced countries and the R & D index of U. S. industries, suggesting that a generalized preference system would produce very little trade diversion. In the absence of any information on an R & D index for developing countries, it is not possible to find out the usefulness of generalized preferences for developing countries. It is, however, expected to encourage the exports of products, not intensive in R & D, from developing to developed countries. 127 TABLE VII-l RESULTS OF CORRELATION TESTS ON EFFECTIVE PROTECTION RATES AND THE R & D INDEX Values of Type of Test test statistics t-value, and sig- Spearman Rank Linear nificance level Correlation (r) Correlation (R) Value of the Correla- tion Coefficient -0.4942 -0.511209 t-value -l.60797 -l.83274 Significant at 0.12 0.10 Source: Footnote 2, Table IV-l 3 Market-Scale Approach. This approach recognizes the ex ante uncertainties on both the supply and the demand sides of the innovative process involving expenditures on R & D activities. Uncertainty regarding the returns for different R & D expenditures helps us to develop a stochastic model of R & D in which it is assumed that an investment of given size in R & D has the chance of lowering unit costs of production by, say, r-per cent in every industry. The ex- pected value of the return on R & D investment would then be directly proportional to the initial size of each industry's sales. As the sales of a country's exports are greater than those of its import-replacements, each country would tend to place its R & D factors in its export industries. In other words, each country's exportables will be the ones that are relatively more intensive in each country's R & D factors. ‘- M '0' HOW b—lu ETD-iii}.- da‘. pr 128 This model may not give the same conclusions as the previous, neo-factor-proportions model because the mode of ranking the industries is different. Moreover, it has not been tested for validity and general applicability. How- ever, it implies that the best way to turn a potential com- parative advantage into actual trade flows is to concentrate R & D activities in the sector or industry with the greatest potential comparative advantage where such a potential is determined by something else such as market structure. No definite conclusions are derivable from the Market-Scale Model regarding the generalized preference system except that generalized preferences will help the developing as well as developed countries to allocate their R & D expenditures towards exportables and result in gerater efficiency. Product Cycle Approach.4 Vernon and Hirsch have de- veloped an alternative approach to incorporate R & D in the study of trade structure. It is assumed that the development and production of manufactured commodities undergoes various changes and phases with respect to the type of production function, the factor inputs, and prices. The pattern of pro- duction overtime can be broken into various phases or stages such as the early or "new" stage, the "growth" or the middle stage and the latter or "mature" stage. In the early stage, the production process is characterized by stress on problem- solving; research and development to determine the most ef- ficient method of production and distribution. It is the stage that involves strong R & D-intensiveness and relatively 129 less of other factors. The growth stage involves more innova- tion and establishment of more capital-intensive plants aimed at mass production and distribution. The skills required are essentially efficient management and marketing ability rather than highly trained human capital for R & D. In the third stage, or mature stage, the product and the process of produc- P tion become fairly standardized and capital-intensive, but the composition of the labor force changes as the proportion of unskilled and semi-skilled workers rises and the R & D- 1 intensiveness is minimized as there is very little basic 9 problem-solving at this stage. The product-cycle model indicates that the less de- veloped countries can eXpect to have competitive advantage in mature commodities because the need for R & D-oriented scientists and engineers is comparatively limited. Commodi- ties can be ranked according to the stage in their develop- ment as suggested above with developing countries expected to have comparative advantage in the mature commodities. Typical mature products include food products, tex- tiles, wood products, simple fabricated products and machinery.5 We have already seen that effective tariff rates are higher for textiles, wood products and other simple manufactures (Table IV-l). Therefore, heuristically, it is possible to suggest that generalized tariff preferences for the manufactured exports of the developing countries will be in keeping with an efficient system of trade structure as determined by the product cycle model. 130 ECONOMIES OF SCALE, COMPARATIVE ADVAN- TAGE, AND GENERALIZED PREFERENCES The received theory of international trade states that the forces of the market will lead every country to pro- duction which utilizes its full comparative advantage as de- termined by factor endowment and thereby results in maximum welfare in each country and in the world, if there is full employment, perfect competition and no economies of scale and externalities. However, the violation of these assumptions adversely affects the predictive ability of the theory, especially if economies of scale exist. This poses much more severe prob- lems to the theory than the violation of, say, perfect com- petition. The inclusion of economies of scale--at the plant, company and market level--adds an additional dimension be- sides factor endowments, however defined, in the determination of comparative advantage. Moreover, the scale effect at the economy or the market level is more relevant in the case of generalized preferences involving the extension of market size. One way of looking at the economies of scale effect on the structure of trade is to state that a large home market is conducive to the export of goods produced under increasing returns to scale and a small home market is conducive to the export.of goods produced under constant or decreasing returns ‘UD scale.6 This is so because the large home market enables firms to install special purpose equipment, take advantage of bulk purchases, spread overhead and other fixed costs over 131 a large number of units, and economize on the use of trans- port so as to minimize unit manufacturing costs. Great emphasis has been placed by the proponents of generalized tariff preferences, like Prebisch,7 on the role played by the size of the market in the efficient allocation of resources and the realization of potential comparative advantage of the developing countries in manufacturing. Ac- . 6 7791—...“ 71.17.. 5’ cording to Prebisch,8 the small size and the low per capita t n 9» “in!" - income of most of the developing countries has led to shel- tered, small-scale domestic industries unable to compete in the world market. A real vicious circle has been created as regards the export of manufactured goods of the developing countries. These exports encounter great difficulties because internal costs are high, and internal costs are high because, among other reasons, the exports which would enlarge the markets are lacking. Therefore, preferential export trade for developing countries is one avenue of escape from the im- passe of the limited market. What this means is that as over- all size of the market increases, an industry's capacity to lower its costs by moving towards its optimum scale of output also increases. Thus the only thing standing in the way of a stagnant industry and its optimum scale in production is the size of the market, if the industry is technically feasible in the developing economy. This argument, however, overstates the case for scale economies. The real point is that, if scale economies cannot be exploited, the real return to investment in industrial activity will fall, raising 132 thereby the resource costs necessary to achieve the same level of industrialization. In order to determine the rele- vance of the scale economies in the context of "dynamic" comparative advantage of the developing countries and the need for tariff preferences for their manufactured exports, we present a simple model below. Mgdel. The model brings the market, technology and growth together to explain the role of market size in the efficient allocation of resources and growth.9 The size of I the market in a particular country depends upon the level of I per capita income, population, and its technical capability. Moreover, the size of economy needed to support a given in- dustry changes as the quality of the industry's product changes. The model is based on the assumption that as techni- ques advance in any industry, the scale of production needed to employ those techniques most economically also increases. It does so because economic use of advanced techniques in- volves plants of increased capacity and increased specializa- tion among productive units. It is further assumed that the state of techniques in use determines per capita and aggregate income in the economy and thereby partly influences the determination of the market for each type of product. As technology advances, the "required" size of industry increases, where "required" size of an industry is defined as the size required to achieve the minimum unit cost at any given state of technology. The size of the industry justified by the 133 demand for the product at that minimum unit cost will be termed "justified" size of the industry. Both technology and market sizes have a positive influence on the "justi- fied" size, while the "required" size is a function of state of technology alone. The interaction between the required and the justi- fied size determines the minimum justified size of the in- dustry. This is shown in Figure VII-l where P indicates the “can“ ‘3'. minimum justified size of the industry. However, changes in technology and market will affect the justified and the re- quired sizes of the industry, and it is not necessary that they will increase at the same rate. If the "required" size of the industry grows more rapidly than the "justified" size, then the system will not permit an industry size greater than where required size is equal to the justified size (Figure VII-la). As a matter of fact the minimum justified size of -the industry is also the maximum size. If the "justified" size grows more rapidly, then the system will be self- propelling (Figure VII-lb), because, (a) if the industry was initially justified, it will always be justified, and (b) even if the industry was not initially justified, it may be- mmeso. Let us concentrate on the relationship between the size of the industry and the state of technology in the system where "required" size grows more rapidly than the "justified" size in order to determine the significance of market size and scale effect in the context of trade 134 THE "REQUIRED" AND "JUSTIFIED" FUNCTIONS )— o: * (— a (D "' a r: E 53 (:1 D.‘ B a) A m P F- ———————————— ——-—-—-————-—--— ' E : I 8 5 A : ' l ' B I A E : i 5 STATE OF STATE OF '3’ TECHNOLOGY (‘9) TECHNOLOGY FIGURE VII~1 "Required" size function, given the size of the market. "Justified" size function, given the size of the market. THE SCALE EFFECT OF PREFERENCES SIZE OF INDUSTRY STATE OF TECHNOLOGY FIGURE VII-2 135 preferences. Consider Figure VII-2. AA and BB show the "required" and the "justified" sizes of an industry respec- tively for different levels of technology, given the size of the market. In such a system, if the industry was not ini- tially justified, it will never become so, and even if it was initially justified, it may cease to be so after some point (P) in the level of technical progress. In a typical developing country, growth is associated with new industries and, hence, better technology. With technological changes the required size may increase more rapidly (from P to P') than the justified size (from P to P"). Such a situation cannot be sustained and will force the developing country to continue to use the inefficient technique and be content with a lower growth rate, associated with P in Figure VII—2. If, however, the market size can be increased through the extension of exports into the developed countries, then the justified size for the given level of technology may rise sufficiently (from BB to B'B') to be equal to the required size, P'. Different industries will have different justified and required sizes given the state of technology and market. Since the resource base and the income level determine the size of the market, different countries will have different market sizes and hence justified sizes for various industries, larger for the countries with bigger markets and smaller for the countries with smaller markets. Therefore, for bigger countries, the point of intersection between required and 136 justified sizes, that is, the minimum justifiable size of the industry, will be larger than for a smaller country, and a relatively greater ability to produce commodities with greater economies of scale effects than the smaller countries. Moreover, the scale effect is not equally important for all industries. Industries with a relatively moderate or smaller scale effect will be characterized by a flatter "required" size line. It is, however, not easy in practice to specify industries which have relatively greater economies of scale effects. This model suggests that economies of scale are very important for the developing countries of relatively small size at the intermediate level of industrial- ization and growth. ESTIMATES OF THE ECONOMIES OF SCALE 10 11 Chenery and Maizels estimated the impact of econ- omies of scale (market) on manufacturing by computing multiple regressions of the effects of per capita income and popula- tion--two determinants of market size-~on manufacturing production for about fifty countries. Dividing all manufac- turing into fifteen groups, Chenery found a significant and positive correlation between manufacturing production and size of population. This implies that in a country with larger population, manufacturing production is a larger share of total production, presumably due to the economies of scale. This relationship was more significant for products like automobiles, aircraft, chemicals, electronics, paper 137 and paper products. On the other hand, Maizels found a significant positive correlation with population for only one of the six manufacturing groups he considered, i.e., basic metals, and found a negative correlation for food, beverages, and tobacco. Even though these studies fail to incorporate tech- nology in the analysis of the economies of scale, they indi- cate roughly that commodities in which developing countries v in" .4 my a are likely to have comparative cost advantage do not exhibit ‘ W. 1" strong economies of scale effect and that a generalized preference system cannot be expected to be helpful just on the grounds of the scale effect. In contrast to these studies, Keesing12 used value added of the average plant in each industry as an indicator of economies of scale in the absence of a better indicator. Higher value added per establishment implies relatively greater economies of scale. He found that more sophisti- cated industries such as aircraft, drug and other chemicals, petroleum refining, motor vehicles, electrical equipment, paper and paper products, etc., have larger economies of scale. The simple manufactures such as textiles, glass, stone and clay products, fabricated metal products, simple machinery, lumber and wood products, and rubber products, etc., were found to have very low economies of scale. If it is assumed that the structure of industries is the same in developing countries as in the United States, then Keesing's estimates suggest that the scale effect is not strong for 138 commodities in which less developed countries are likely to have comparative advantage. Using a product-cycle model, Seev Hirschl3 analyzed the relevance of economies of scale for new, growth, and mature stage commodities in order to explain the structure of trade. He found that a relatively larger market gives com- petitive advantage in the production of all types of manu- factured commodities. However, the economies of scale ef- fect is the strongest in the newly developed products like electronics and weakest for the mature commodities like “1". 2m -Jfin- wood products, textiles, machinery and other fabricated products. Moreover, according to this model, developing countries have comparative advantage in the production of some mature commodities that enjoy relatively weak scale effects. In order to determine the role of economies of scale in the determination of comparative advantage of a country in manufactured products, Hufbauerl4 has also presented an extensive analysis. He estimates the scale economies as the 5 regression exponent (d)1 in the regression equation of the following type: V = Knd where n = number of workers employed, V = ratio between value-added in plants employing n persons and average value-added for the four-digit U. S. census bureau industry, and K = constant. 139 The estimates indicated significantly high scale effects in transport equipment, synthetic fibers, paper and paper products, chemicals, iron and steel, and electrical machinery. Relatively low (even negative) scale effects were found in manufactures like man-made textiles, clothing, leather products, glass and mineral non-metallic products, and miscellaneous manufactures. Table 15, Appendix II pre- sents these estimates. If it is assumed that the structure of production is similar in all the countries including the developing world, i.e., industries showing a higher scale economy effect in the United States are also the ones with a relatively high scale economy effect in developing countries, then the U. S. data developed by Hufbauer can be used to rank commodities by a scale economy index that may represent potential compara- tive advantage. Then the tests for the relevance of genera- lized tariff preferences in the context of scale effects can be conducted. Spearman rank and linear correlation tests between the schedule of commodities ranked by Hufbauer's index of economies of scale and the structure of effective protection rates in the developed countries were carried out by drawing the data from Table 1, Chapter IV, and Table 15, Appendix II. The rank correlation test shows a modest negative correlation of -.2506 (t = -l.52056) between the effective tariff rates and the index of economies of scale for the industries in- cluded in this study. 140 The linear correlation test gives a similar result as indicated by the correlation equation given below: A Y = 40.130183 - 0.0099328 X R = -.l48508 T = 1.97723 where Y. = Effective tariff rate on i-th activity in developed 1 countries, Xi = scale economy index (d) for industry i. The rank correlation test appears to suggest that the tariff structure of advanced countries does not discrimi- nate against the actual or potential manufactured exports of develOping countries characterized by relatively higher economies of scale. However, these results should be inter- preted with caution because of various limitations of this kind of analysis. The estimator of scale economy used in this analysis does not encompass all the dimensions of scale effects, and, therefore, may not be accurate.16 Therefore, it will be misleading to use the analysis given above to draw any definite conclusions regarding the influence of tariff policies of the developed countries on the growth potential of manufactured exports of the developing countries, especially where economies of scale have a singificant in- fluence. The generalized tariff preferences may, however, help a developing country that is so small that the domestic market cannot permit the efficient utilization of its technical capability and the industrial capacity. For such a country the extended market into the developed world will be of 141 great importance. However, most of the countries that are going to gain substantially from a generalized tariff pref- erence system such as Pakistan, India, Mexico, and Brazil, already have fairly large markets that can sustain the ef- ficient utilization of most of the industries within the reach of their technical capability. CONCLUSION A high economies of scale effect is not enough to justify tariff preferences. Most of the manufactured products ‘mr; in which the developing countries can be expected to have potential or actual comparative advantage—~i.e., those that are within the reach of their technical capability--do not exhibit very significant scale economies. However, for very small developing countries, the importance of an extended export market may be great even for commodities that are char- acterized by moderate or even low scale economies. Most of the products with higher economies of scale effect are highly sophisticated ones in whose production and trade the advanced countries are already highly competitive. Therefore, even if preferences are granted to the developing countries for such products, they will not be able to compete with the developed countries and may actually further distort their own resource allocation. The generalized preferences may, however, help the developing countries to get out of the straight-jacket of domestic institutional rigidities that are only indirectly related to the scale effects. The 142 rigidities are the outcome of extensive import-substituting industrialization in the highly insulated and protected de- veloping economies. A small domestic market and monopolistic structure of industries did not allow the "infant" industries to become competitive. The generalized preferences may im- prove the situation by increasing competition among develop- ing countries for the markets in the developed countries besides competing with the domestic producers of develOped countries. This argument apparently lies behind Dr. Prebisch's plan for preferences though he seems to have confused the scale economies with the "infant" industry argument, because the infant industry argument assumes a fixed market which is gradually taken over by the infant industry as it becomes competitive through improved efficiency.17 ‘umfl--“ NOTES TO CHAPTER VII Harry G. Johnson, Comparative Cost and Commercial Policy Theory for a Develpping World Economy, Wicksell Lec- tures 1968, (Stockholm: Almquist & Wiksell, 1968); "The 9 Theory of International Trade," International Economic Rela- tions (Paul E. Samuelson, ed.), InternatiOnal Economic As- sociation (New York: MacMillan, 1969), pp. 55-66. i Donald Keesing, "The Impact of Research and Devel- . Opment on United States Trade," Journal of Political Economy, "” Vol. 75, No. l, 1967, pp. 38-48, reprinted in The Open g Economy (New York: Columbia University Press, 1968), pp. 175-89. W. Gruber, D. Mehta, and R. Vernon, "The R & D Factor in International Investment of United States Indus- tries," Journal of Political Economy, Vol. 75, February, 1967. R. E. Baldwin, Determinants of the Commodity Structure of U. S. Trade (mimeographed), Social Systems Research Insti- tute, University of Wisconsin, 1969. The data on R & D for the correlation tests in the text was drawn from the following table. R & D INDEXES FOR U. S. INDUSTRIES Percentage of Scientists and Engineers in the Description of Total Employment SITC Commodities of the Industry 266,551, 599 Chemicals 9.5 629 Rubber Products 2.1 641,642 Paper and Allied Products 2.0 651,653, 655-657 Textile Products 0.7 631,632 Lumber and Wood Products 0.5 664-666 Stone, Clay and Glass Products 2.0 671 Iron and Steel 2.5 695,696 Fabricated Metal Products 2.4 711 Other Non-electrical Machinery 4.3 729 Electrical Equipment, n.e.s. 8.2 143 144 The table was taken from Donald Keesing, "The Impact of Research and Development on United States Trade," The Open Economy, Appendix Table I, pp. 186-87; Table 2, Appendix I. 3The market-scale model is in its early stages of development and an extensive research is still required. For details, see, however: H. G. Johnson, pp. cit.; R. E. Caves, "Comments on Professor Johnson's Paper: I," International Economics Relations, pp. 66-70. 4The product-cycle doctrine has many champions. For detailed analysis of the background, see: R. Vernon, "In- ternational Investment and International Trade in the Product Cycle," Quarterly Journal of Economics, May, 1966; Seev Hirsch, Location of Industry and International Competitive- ness (Oxford, Clarendon Press, 1967); and G. C. Hufbauer, Synthetic Materials and the Theory of International Trade (London: Duckworth, 1966). 5See Table 8, Appendix II. 6G. C. Hufbauer, "The Commodity Composition of Trade in Manufactured Goods," The Technology Factor in Trade, Universities-NBER Conference Series No. 22 (New York: Columbia University Press, 1970), pp. 250-90. 7The case for tariff preferences based upon the so- called economies of scale has many proponents. For an extensive, though somewhat misleading formulation, see: Raul Prebisch, Towards A New Trade Policy for Development, Report by the Secretary General of UNCTAD (New York: U. N., 1964). For a critical evaluation of this argument, see: S. Weintraub, Trade Preferences for Less Developed Countries, Praeger Special Studies in International Economics and De- velopment, 1966, Ch. 3, pp. 55-90. Harry B. Johnson, Economic Policie§_Toward Less-Developed Countries (Brooking Institution, 1967), Chapters I, VI. 8 R. Prebisch, op. cit. 9E. E. Hagen, The Economics of Development (Homewood, Ill.: Irwin, Inc., 1968), Chapter 11, pp. 236-46. The model presented in this chapter is a modified and extended version of Hagen's model. 10H. B. Chenery, "Patterns of Industrial Growth," American Economic Review, September, 1960, pp. 624-54. 11A. Maizels, Industrial Growth and World Trade: An Empirical Study of Trends in Production, Consumption and Trade in Manufactures from 1899-1959 with a DiScussion of Probable Future Trends (Cambridge: Cambridge University Press, 1963). , L b 145 12Donald Keesing, o . cit., also especially Table 4 column 7 . l3Seev Hirsch, op. cit., Chapter III, especially pp. 52-58. 14 G. C. Hufbauer, Synthetic Materials and the Theoqy of International Trade, pp. 73-78; Commodity Composition of Trade in Manufactured Goods, pp. cit. 15A somewhat similar formulation of economies of scale was presented by Baldwin, op. cit. 16For further elaboration on this point, see: S. Weintraub, op. cit., pp. 63-69; H. G. Johnson, op. cit., Chapter VI. 17For a detailed interpretation, see: Hla Mynt, "International Trade and the Developing Countries," Inter- national Economic Relations (Paul E. Samuelson, ed.), International EconomiC Association, pp. 15-35, especially section III. CHAPTER VIII SUMMARY AND CONCLUSIONS Economic develOpment calls for a substantial volume of imports of materials, capital goods and technical services in addition to a substantial increase in national investment. If exports are insufficient to finance required imports-~i.e., 1411. ‘ a trade gap exists--and/or domestic savings are insufficient to finance an increased volume of investment--i.e., a savings gap exists--rapid economic development cannot take place. Since most of the developing countries have a "dominant" trade gap, it is not possible to attain a target growth rate without filling this gap. One way to fill the trade gap is through economic aid from the developed countries. However, even if developed countries contribute one per cent of their respective GNP's in public capital flows and about $4 billion in private capital movements, a residual gap of over $7.5 billion will still exist if an annual average growth rate of 6.1% is to be maintained in the developing countries. This residual trade gap can be met by increased import substi- tution and eXport acceleration in the developing countries. Import stustitution as a development strategy has certain well known structural limitations, reference to which has already been made, such as the size of the domestic market 146 147 and the technical requirements for efficiency. Export acceleration appears to be the only means available to fill the residual trade gap. There is little chance for the de- veloping countries to generate a dramatic expansion in their primary goods exports because of low-income elasticity of demand and the development of synthetic substitutes for r” _. natural materials in the developed countries. Thus there remains the expansion of manufactured exports to the developed fimw‘ " i L‘W world as an alternative method to accelerate the growth ‘mET process. One of the main obstacles in the way of manufactured exports of developing countries to the developed countries is the effective protective tariff barriers in the latter countries against manufactured imports. The structure of ef- fective protection in developed countries tends to discrimi- nate against the manufactured products of special interest to the developing countries as potential or actual exports. A general preference system eliminating tariffs in developed countries on manufactured imports from developing countries will grant the greatest preference to the commodities that are subject to the highest effective protection. This will give the developing countries a maximum incentive to estab- lish those industries which are subject to the highest ef- fective protective duties in the developed countries. General preferences will be useful only if these are the industries in which develOping countries have relatively greater and developed countries relatively lower, comparative advantage. 148 This is what the present study sought to find out. To do this, we compared the structure of effective protection rates in the developed countries with the comparative advantage schedules of developing and developed countries in manufac- tured products. There is no single determinant of a country's com- parative advantage. It was, therefore, necessary to consider various theoretical models to determine a ranking of manufac- tured commodities in developing and developed countries in the order of their comparative advantage. In Chapter II, we considered many alternative formulations to approximate comparative advantage. One could apply various criteria for resource allocation in developing countries to rank commo- dities, but due to serious data problems and analytical limitations, we were unable to use any of these criteria in our analysis. Instead we drew upon the factor proportions, human skills, research and development (R & D), and scale economies models, in addition to Balassa's "revealed" com- parative advantage to rank the manufacturing industries by the degree of their comparative advantage. We selected six developing countries--Taiwan, Hong Kong, Pakistan, India, Brazil and Mexico--two developed countries—-the United States and United Kingdom--and thirty- five manufactured commodities to conduct our investigation. The selection was aimed at having a set of countries and commodities that covers the bulk of manufactured exports of the developing countries. It was hoped that the conclusions 149 derived from the study of these countries will also apply to other countries. The input-output technique was used to obtain total requirements of capital, labor and human skills in order to estimate the input ratios for the ranking of industries. Capital/labor ratios in the developed countries and labor/ capital ratios in the developing countries were used to rank the manufactured commodities according to the factor propor- tions model. Similarly, ratios of total requirements of skilled to unskilled labor in the developed countries and ratios of unskilled to skilled labor in the developing coun- tries were used to rank the industries according to the human skills model. Accordingly, the higher the input ratio, the greater is the comparative advantage. We also used the Keesing Skill and the Productivity indexes to represent the skills model. No detailed information was available on R & D and economies of scale in the developing countries, therefore, most of the analysis involving R & D and scale effects was carried out in theoretical terms. The effective protection rates in the developed countries were obtained from the estimates prepared by Pro- fessor Balassa. These rates were "normalized" by obtaining a weighted average of the effective rates of BBC, Japan, the U. K. and the U. S. A. with the rate of each developed region weighted by the value of its total trade in the product to which the rate applies. Linear regression and rank correlation tests were 150 used to determine the nature and significance of correlation between the structure of effective protection rates in the developed countries and the schedules of manufactured com- modities in the developing and developed countries ranked by the various indexes developed in this study. We found significant and positive correlation for the developing countries, and significant and negative correlation for the '0 Ll." ,u. developed countries under the factor proportions and the human skills models. fi—V . .-. The R & D factor in the context of general prefer- g ences was analyzed by considering three alternative formula- tions of R & D--neo-factor proportions,market scale, and the product cycle models. The neo-factor proportions model indicated a negative correlation between tariff rates in the advanced countries and the R & D index of the U. S. industries. Since no such information Was available for the developing countries, it was not possible to determine the impact of general preferences on the developing countries according to the R & D factor. The analysis of the product cycle model suggested that generalized preferences for the manu- factured exports of developing countries will be in keeping with an efficient system of trade structure. No conclusions could be drawn from the market scale model. A theoretical model was developed to bring market, technology, and growth together in order to explain the economies of scale effect. It was found that the scale ef- fects are not equally important for all countries and 151 commodities; they are, however, Of great significance for the developing countries of relatively small size at the intermediate level of industrialization and growth. In order to determine the importance of economies of scale in the context of general preferences, estimates of the scale effect in manufacturing industries by Chenery, Maizels, Keesing and Hufbauer were drawn upon. These studies in- ! dicate that most of the industries subject to a higher scale effect are highly SOphisticated and are beyond the technical capability of the developing countries. Moreover, they are 5 subject to relatively lower effective protection in the developed countries. Therefore generalized preferences may not be in conformity with the comparative advantage of the developing countries. However, these estimates of scale effect are essentially partial and very Crude approximations; therefore no definite conclusions could be derived. In Appendix I we estimated the "revealed" comparative advantage of developing countries by ranking the manufactured products according to their export performance indexes. Un- like the factor proportions and human skills models, the results of country-by-country studies show rather weak positive rank correlation between the structure of effective protection in the developed countries and the "revealed" com- parative advantage of the developing countries in manufac- tures. However, these results should be interpreted with caution because revealed comparative advantage is not an "eXplanatory" model as are the factor proportions and human 152 skills models. Our investigation appears to show that, in general, the structure of effective protection rates in the developed countries is significantly and positively correlated with the comparative advantage schedules of developed countries and negatively correlated with the comparative advantage schedules of developed countries. Therefore a generalized preference system will give the greatest incentive to the establishment of industries in the developing countries in which they have the greatest comparative advantage. Further- more, preferences will encourage the developed countries to import those commodities from the developing countries in which the former have least comparative advantage. Hence, the trade diversion will be minimized. APPENDIX I mun-n n‘ .vvw— APPENDIX I "REVEALED" COMPARATIVE ADVANTAGE AND GENERALIZED PREFERENCES INTRODUCTION Comparative advantage in manufactured products ap- pears to be the outcome of a number of factors such as fac- tor proportions, human skills, level of technology and the extent of scale economies in production and distribution besides a host of other price and non-price variables.1 More- over, some of the assumptions underlying these explanations of trade structure are very limiting; some of the explana- tions involve factors that are not easily measurable. There- fore, instead of enunciating general principles and trying to apply these to explain actual trade flows, it may be de- sirable to take the observed pattern of trade flows as a point of departure, and subsequently use them to find the main influences that have determined the pattern and perfor- mance of trade for a country.2 In other words, the trade performance of a country in regard to manufactured goods can be drawn upon to determine what Professor Balassa calls the "revealed" comparative advantage of that country in manufac- tures.3 This is possible because the commodity pattern of trade is expected to reflect relative costs as well as 153 E 154 differences in non-price factors. It is possible, therefore, to rank commodities by their trade performance over a period of time that will be consistent with the "revealed" compara- tive advantage of the country. In this appendix we rank the commodities according to their "revealed" comparative advantage in the developing countries included in our investigation. Simple Spearman rank correlation tests are used to compare the structure of effective protection rates in the developed countries with the "revealed" comparative advantage schedules of developing coun- tries included in this study to determine the desirability of generalized tariff preferences for the manufactured exports of the developing countries. ESTIMATION OF "REVEALED" COMPARATIVE ADVANTAGE Various indexes have been used to estimate the com- parative advantage as revealed by trade flows. Liesner has used relative export performance as an indicator of compara- tive advantage.4 A similar method was adopted by Balassa in order to indicate the possible consequences of trade liberali- zation among OECD countries.5 Kreinin used the Balassa index to analyze the restrictive effect of the tariffs in the OECD countries.6 One could also use the changes in export/import ratios to reflect the relative advantages.7 There is a pos- . sibility of extending the market-share analysis of export growth to determine the "competitiveness effect" and rank the commodities by using their respective competitiveness effect 155 indexes.8 We have chosen export performance as the index by which to rank commodities. It was found necessary to concentrate on export performance rather than export/import ratios because most of the developing countries are net im- porters of many of the manufactured products incorporated in our study, and because imports are subject to restrictions. THE EXPORT PERFORMANCE INDEX The export performance of an individual industry in a particular market can be evaluated by (a) comparing the relative shares of a country in the world exports of indivi- dual commodities, (b) indicating changes in relative shares over time, and (c) a combination of (a) and (b). Following Balassa, export performance indexes are estimated by combin- ing (a) and (b).9 This was done by dividing a developing country's share in the exports of a given manufactured commo- dity by its share in the combined exports of manufactured goods of all the developing countries to the developed coun- tries, and expressing the results in index number form. The higher the index number for an export commodity, the greater is the revealed comparative advantage of the country in that export product. The export performance index is developed in symbols below: Definitions: Vi. = Value of developing country A's ex orts of manufactured commodity i in period 1 (1960-61)1 to the developed countries. V'i.= Valgs of A's exports of commodity i in period 2 (1967- 68) to the developed countries. A -_».'...114 4463i! 156 V.j = Value of A's manufactured exports to developed coun- try j, in period 1. V.'j = Value of A's manufactured eXports to the developed country, j, in period 2. Vij = Value of A's exports of manufactured commodity, i, to the developed country, j, in period 1. V'ij = Value of A's exports of manufactured commodity, i, to the developed country, j, in period 2 Then: 9 (1) Vij = Vi. Z . I j (2) :ES V'ij = V'i. . j 9 (3) Z Vij = V.j i ll <‘. u. (4) Z V'ij 1 Vij 2: ll < H II < LJ. II < V.. = Total manufactured exports of A to developed market economy countries in period 1. (6) EZV'ij = EV'1.= Zvuj =V'.. i j i j V'.. = Total manufactured exports of A to the developed coun- tries in period 2. (7) 2 Vi. = Total eXport of i-th commodity from all the A developing countries of the world to the developed countries in period 1. (8) 2 Vi'. =Total exports of i- -th commodity from all the developing countries of the world to the developed countries in period 2. 22...: A 157 V Total manufactured exports of all develop- ing countries to all the developed countries A in period 1. (10)ZZV'i. = V'., 2 V' . . = Total manufactured exports of all develOping A countries to all the developed countries in period 2. (11)———EE6—- —% ——YL;—— = Relative share of A's ex- :EE Vi :E: V port of i-th commodity to ' ’° the developed countries in A A . per1od l. (12) Vi.‘ . V'.. Relative share of A's ex- 7' = port of i-th commodity to ZV'i. ZV'.. the developed countries in A A period 2. (l3) V'i. V'. ;_ Vi. I V.. Growth rate of A's ex- ‘ = port of i-th manufac- EV'i. A.V 2V1 V.. tured commodity to the developed countries. Equations (11) and (12) represent approach (a) to the estimation of the export performance index while equation (13) represents approach (b). A combination of the two ap- proaches, approach (c), assumes "that while past trends in relative shares of exports can be expected to continue, this "12 will take place at a declining pace as compared to the past. Such a combination can be approximated by the following: (14) :[ivi;.v Z, “:61: 2:} 32211.22” ]= X' X = Export Performance Index I.“ H'Q-l. VIE J" - 158 where, i = 1 ...., n = all manufactured goods as defined by UNCTAD (see Chapter III), j = l ...., m = all the developed countries (see Chapter III). This export performance index was estimated for each commodity and for each of the six developing countries inclu- ded in our investigation. The results are presented in Table 16, Appendix II. The export performance index was used to rank the commodities in terms of their comparative advan- tage. EMPIRICAL RESULTS Rank correlation coefficients were calculated between the structure of effective tariff protection in the developed countries and the "revealed" comparative advantage schedules of the six developing countries. The results are presented in Table l. The usefulness of generalized preferences for a developing country will depend upon how strongly the re- vealed" comparative advantage schedule is positively corre- lated with the schedule of effective protection rates. The investigation was carried out on two levels: (a) All commodities with positive export performance indexes were included in the correlation test. (b) If the results failed to show a significantly positive correlation coefficient in (a), then some commodities with the highest divergence between ranks of the export per- formance index and effective tariff rates were dropped and RESULTS OF COUNTRY-BY-COUNTRY TESTS FOR CORRELATION 159 TABLE 1 BETWEEN EFFECTIVE TARIFF RATES IN DEVELOPED COUNTRIES AND EXPORT PERFORMANCE INDEXES OF DEVELOPING COUNTRIES Commodi- Spearman Rank ties Ex- Correlation t- Signifi- cluded Coefficients Statistic cance Country (SITC) (r) Value Level . None -0.3437 -l.89747 .10 Taiwan 055,653, 711,724 0.3214 1.72101 .10 (7 None 0.110 0.06384 Not Signi- 5 ficant at .10 H n Kon 0 g 9 642,653, 655,666 0.3711 1.94101 .10 None 0.3110 1.34849 Not Signi- Pakistan f1cant at .10 656,711 0.3891 1.75592 .10 None 0.0561 0.29148 Not Signi- India ficant at .10 655,666, 892,894 0.3520 1.73201 .10 None -0.2070 -0.9257 Not Signi- ficant Brazil at '10 641,652, 711,851, 599,729 0.3010 1.1340 Not Signi- ficant at .10 None 0.1670 0.972274 Not Signi- ficant Mexico at '10 551,642, 653,657 0.4080 2.3207 .05 Source: Table 1, Chapter IV, Table 16, Appendix II. 160 correlation coefficients were recomputed. The commodities with relatively high divergence between the ranks of the ex- port performance index and effective tariff rates are, either the ones in which a developing country has relatively higher revealed comparative advantage but subject to very low duty in the developed countries, or vice versa. When all commodities with a positive export perfor- mance index were included in the test, no country showed a significantly positive correlation coefficient. In fact, Taiwan and Brazil showed negative correlation. This rather surprising result implies that the best performing manufac- tured exports of Taiwan and Brazil are not hampered by the tariff system of the developed countries. Five out of six developing countries investigated showed significantly positive correlation coefficients when up to four commodities with the highest divergence between the ranks of export performance index and effective tariff rates were excluded from the test. Brazil failed to show a significantly positive correlation even when six commodities were excluded. In the case of Pakistan only two commodities had to be dropped to obtain a significantly positive correla- tion coefficient. CONCLUSION The results of country-by-country studies show that, unlike other models we have used in this investigation, there is weak positive correlation between the structure of effec- tive protection in the developed countries and the "revealed" 161 comparative advantage of the developing countries in manu- factured goods. These results may be used to infer that there is not a very significant relationship between the structure of effective protection rates of developed, and the comparative advantage of developing, countries in manu- factures, implying that generalized preferences for the manufactured exports of developing countries to developed countries may not be so productive after all! However, these results should be interpreted with caution because "revealed" comparative advantage is not an "explanatory" model that de- pends upon the structure of the economy to explain the pat- tern of production and trade. Moreover, the existence of differential effective tariff rates in developed countries may have affected adver- sely the export performance of some of the commodities in which developing countries may have relatively higher (po- tential) comparative advantage. However, our investigation in this appendix appears to show that generalized preferences in developing countries for the manufactured exports of de- veloping countries may not be as productive as indicated by other models used in this study. Instead, preferences for specific commodities and countries may be more productive. NOTES TO APPENDIX I See W. M. Corden, Recent Developments in the Theory of International Trade, Special Papers in International Econ- omics, No. 7, International Finance Section, Princeton Uni- versity, 1965. B. Balassa, "Trade Liberalization and 'Revealed' Comparative Advantage," The Manchester School Economic Papers, May, 1965, pp. 99-117. 3 Ibid., p. 103. W” "" 4H. H. Liesner, "The European Common Market and British Industry," Economic Journal, June, 1958, pp. 302- 16. 5B. Balassa, op. cit., p. 100. 6Mordechai E. Kreinin, "On the Restrictive Effect of the Tariff-A Note on the Use of the Balassa Index," The Manchester School Economic Papers, January, 1966, pp. 75-80. 7B. Balassa, op. cit., p. 104. 8 E. E. Leamer, and R. M. Stern, Quantitative Inter- national_Economics (Boston: Allyn & Bacon, Inc., 1970 , Chapter 7. 9 B. Balassa, op. cit., pp. 103-104. 10The year 1960-61 was selected as the base year in our study due to the fact that it was the first year of the First Decade of Development. 1967-68 was considered the cur- rent year as it was the latest year for which detailed data on commodity trade is available. 11The developed countries included in this investi- gation and the E.E.C., EFTA, U.S.A. and Canada, Japan, New Zealand and Australia. 12B. Balassa, op. cit., pp. 106-107. 162 APPENDIX II w- 'fl'l.—— ._. NOTES TO TABLE A-II-l Table A-II-l shows the projections of the export/ import gap for thirty-seven and savings gap for thirty-five major developing countries. These gaps as defined here are comparable in terms of national income accounting concepts. The export/import gap is defined as the excess of projected “T-.-nun- .12: In nhfifm ! y ‘ imports of goods, services over the corresponding projection of export earnings. The savings gap is the excess of invest- ment requirements over domestic savings at the indicated projected growth rate. Actual estimates for the two gaps were available for 1963, the base year, while the gaps were projected for 1975 under the assumptions of "low" and the "high" target rates of growth. Whereas the "low" target rate (5.2%) is based upon the previous development experience of the developing countries, the "high" target rate (6.1%) is based on the assumption that requisite domestic and ex- ternal resources will be mobilized, and that the develOped countries will experience rapid growth during 1965-75. 163 164 m.hmH NHom m.hm IIIIIIIIIIIIIIIIII Hmmmcmm ON 0 th o mHHI O.HmH N.mNm h.mwv o.mNN Mahmmflz ma 6.66- 6.66- 6.66- 6.666 6.66 6.66- 66606 66 6.666- 6.666- 6.666- - uuuuuuuuuuuuuuuuu 66600 66066 66 6.666 6.66 6.66 6.6 6.666- 6.66 66660 66 6.6 6.66- 6.66 6.66 6.6- 6.66 66606666 66 606666 6.6666- 6.6666- 6.6666- 6.666- 6.6666- 6.6666- 660660666 66 6.666 6.666 6.6- 6.666 6.66 6.6- 6666666 66 6.666 6.666 6.66 6.666 6.666 6.66 6666 66 6.66 6.66 6.6 6.66 6.66 6.6 66666666 66 6.666 6.666 6.66 6.66 6.66 6.66 666666 66 6.666 6.666 6.666 6.666 6.666- 6.666 006x02 6 6.66 6.66 6.6 6.66 6.6 6. 6066666 6 6.66 6.66 6.66- 6.666 6.66 6.66- 6066606 6 6.666 6.66 6.66 6.666 6.666 6.66 66660600 6 6.666 6.666 6.66 6.666, 6.66 6.66 66660 6 6.666 6.666 6.66 6.666 6.66 6.66 6066666 6666660 6 6.666 6.666- 6.66- 6.666 6.666- 6.66- 666666 6 6.666 6.666 6.66 6.666 6.66 6.66 66>6606 6 6.666 6.66 6.666- 6.666 6.666- 6.666- 666066666 6 6066626 26666 6666 306 16660066 6666 306 66666066 6666600 6666 6666 6666 6666 ommo unomEH\uuomxm Q mmo mmc6>mm 1660666 6666 66 6666606 60 660666666 mmmHmBZDOU UZHmOAm>mD m0 mmmw BmOmEH\Bm0mxm 024 mDZH>6m56ox6 6606>umm 6cm 66006 60 muuomxm 6666 mp6omEH 60 .mmcw>mm 06666506 6666 ucmaumm>c6 An .muHomE6 66>0 manomxw 60 .pcmfiumm>c6 66>o mmc6>66 mo mmmoxm mmuMUHUGH cm6m 65:62 66 .66 .6 .66 66666 .66666 ..z .6 "6606 3626 66666666066 94602: .666uucsou mc6mo66>6o 60 66662 6666m60 6cm muommmonm 66669 8066 66:66630 66369 165 6.666 6.666 6.666 6.666 6.666 6.666 66666666 66 6.66 6.666- 6.666- 6.66 6.666- 6.666- 66666666666 66 6.6666 6.6666 6.666 6.666 6.66 6.666 66666666 66 6.666 6.66 6.66- 6.666- 6.666- 6.66- 66666666 66 6.66 6.666 6.666 6.66 6.66 6.666 666606 66 6.666 6.666 6.666 6.6666 6.666 6.666 666666 66 6.666- 6.666- 6.666- 6.666- 6.666- 6.666- 6666 66 6.6666- 6.6666- 6.666- 6.6666- 6.6666- 6.666- 6666 66 6.666 6.666 6.66 6.6666 6.666 6.666 666660666 66 6.6666 6.6666 6.666 6.666- 6.6666- 6.666 66666 66 6.66 6.66 6.66 6.66 6.66 6.66 66636666 66666 66 6.666 6.666 6.6 6.666 6.66 6.66 606666 66 6666 6.66 6.66- 6.66- 6.66 6.66- 6.66- 666666 66 6.666 6.66- 6.666 6.666 6.66 6.66606666666 6666 666666 66 6.666 6.666 6.66 ------------------ 6666666 66 6.66- 6.6- 6.6- 6.66 6.66- 6.66- 66666666 66 6.66 6.66 6.66 6.66 6.6- 6.66 66666 66 6666 306 66666666 6666 306 66666066 6666606 6666 6666 6 6 6666 Q60 660m96\uuomxm mum mm26>6m 66666666066 6-66-4 66666 NOTES TO TABLES A-II-2-6 Tables A-II-2 to A-II-6 demonstrate the economic in- dicators used in order to select the developing countries for our investigation. Economic indicators like gross domestic product (GDP) per capita, the size of the manufacturing sec- tor, the share of manufactures in total exports, and the growth “a" ...“... finfi‘m rate of the manufactured exports, can be drawn upon to deter- mine the extent of an economy's development. GDP per capita is the output of goods and services per person within the geographical limits of the country. The U. N. classification system was followed to define manufacturing activities and exports. The competitiveness of a country in manufactured exports is approximated by the annual trend rate of growth of its manufactured exports to developed countries between 1960 and 1966. The annual trend rate indicates the percen- tage rate at which the manufactured exports grew from one year to another over the period 1960-66. 166 167 TABLE A-II-Z PER CAPITA GDP AT FACTOR COSTS 1967 GDP Per Capita GDP Per Capita Coun- at Factor Cost Coun- at Factor Cost tries in $ tries in U.S. $ 1 Argentina 758 20 Senegal 195 2 Bolivia 176 21 Sudan 91 3 Brazil 273 22 Tanzania 66 4 Cen. America 251 23 Tunisia 186 5 Chile 539 24 U.A.R. 167 6 Columbia 313 25 Uganda 87 7 Ecuador 218 26 Ceylon 140 8 Jamaica 493 27 China(Taiwan) 238 9 Mexico 520 28 Hong Kong 351 10 Panama 553 29 India 81 11 Paraguay 211 30 Indonesia 94 12 Peru 263 31 Iran 287 13 Uruguay 551 32 Iraq 292 14 Venezuela 935 33 Israel 1344 15 Ethiopia 61 34 Jordan 230 16 Ghana 231 35 Malaysia 280 17 Ivory Coast 223 36 Pakistan 123 18 Kenya 111 37 Philippines 259 19 Nigeria 75 38 Thailand 140 Source: Handbook of International Trade and Development Stat- istics, UNCTAD (New York: U. N., 1969), Table 6-2, pp. 168-70. 168 TABLE A-II-3 SIZE OF THE MANUFACTURING SECTOR IN THE DEVELOPING COUNTRIES Percentage Value of Income Originating in Manufacturing Countries GDP 1. Argentina 34.2 (1967) 2. Bolivia 14.7 (1966) 3. Brazil 18.1 (1966) 4. Central America 16.0* (1967) 5. Chile 27.1 (1967) 6. Columbia 17.4 (1967) 7. Ecuador 17.0 (1967) 8. Jamaica 14.9 (1967) 9. Mexico 29.7 (1967) 10. Panama 16.3 (1967) ll. Paraguay 15.4 (1966) 12. Peru 18.4 (1965) 13. Uruguay 17.0 (1965) 14. Venezuela 14.6 (1967) 15 Ethiopia 7.7 (1966) 16. Ghana Not available 17. Ivory Coast 9.2 (1966) 18. Kenya 11.0 (1967) 19. Nigeria 6.1 (1966) 20. Senegal 9.0 (1966) 21. Sudan 5.8 (1964) 22. Tanzania 4.6 (1967) 23. Tunisia 15.0 (1967) 24. U.A.R. 18.8 (1967) 25. Uganda 8.5 (1967) 26. Ceylon 7.6 (1967) 27. China (Taiwan) 20.8 (1967) 28. Hong Kong 39.5 (1966) 29. India 15.0 (1967) 30. Indonesia 12.4 (1967) 31. Iran 1 35.6 (1967) 32. Iraq 7.1 (1967) 33. Israel 20.5 (1966) 34. Jordan 8.8 (1967) 35. Malaysia 11.1 (1967) 36. Pakistan 12.1 (1967) 37. Philippines 17.2 (1967) 38. Thailand 13.0 (1967) Source: Year Book of National Accounts Statistics 1968, Vol. I, Individual Country Data, U. N. * Obtained from Economic Survey of Latin America, 1967 (New York: U. N., (l) Extraction of Oil (crude) included. 1968). mew-'3‘“: mifimfiw 169 TABLE A-II-4 PERCENTAGE SHARE OF MANUFACTURING EXPORTS TO DMEC IN TOTAL EXPORTS Xm(DMEC)* Countries X (100) 1. Argentina 8.0% 2. Bolivia 0.8 3. Brazil 9.2 4. Central America Negligible -» 5. Chile 3.8 - 6. Columbia 4.3 7. Ecuador 3.8 8. Jamaica 20.9 9. Mexico 19.1 10. Panama 18. 11. Paraguay . 12. Peru . 13. Uruguay 1 . l4. Venezuela . 15. Ethiopia . 16. Ghana 1 . 17. Ivory Coast 1 . 18. Kenya . 19. Nigeria . 20. Senegal . 21. Sudan . 22. Tanzania Not v ilable N 23. Tunisia 24. U.A.R. 25. Uganda 26. Ceylon 27. China (Taiwan) 3 . 28. Hong Kong 5 . 29. India 2 . 30. Indonesia . 31. Iran . 32. Iraq H 33. Israel 34. Jordan 35. Malaysia 36. Pakistan 37. Philippines 38. Thailand Sources: Commodity Trade Statistics, Series—B, U.N.; Foreign Trade Statistics, Series C, OECD; Year Book of International Trade Statistics, 1967, U.N. *Xm(DMEC): Manufactured exports to developed market economy countries (0.8.; BETA; EEC: Japan, Canada--19 in all). Xm (Total A) includes all exports of manufactures minus SITC 667, 711, 735, 332 (petroleum) and 682.1, 683.1, 685.1, 686.1, 687.1, 689. FJN bl—IOUIOGJOoameAOOMOSD OANQONNOI—Iwb oo o o 0 \oNHmI-M-uor-awmmqmwwqm NmemmmNHmom 170 TABLE A-II-S VALUE OF MANUFACTURED EXPORTS (TOTAL - A) OF DEVELOPING COUNTRIES TO DMEC Value of Exports of Developing Countries (Total - A)* in $ Millions Countries 1960 1966 1. Argentina 93.5 121.6 2. Bolivia 0.11 0.93 3. Brazil 98.7 160.4 4. Central America --- —-- 5. Chile 13.4 34.8 6. Columbia 3.6 21.8 “ 7. Ecuador 3.0 7.0 8. Jamaica 25.0 48.1 9. Mexico 99.6 229.8 10. Panama 3.2 16.7 11. Paraguay 13.3 19.8 12. Peru 21.3 29.4 13. Uruguay 12.1 20.8 14. Venezuela 3.7 5.0 15. Ethiopia 1.2 2.7 16. Ghana 17.7 40.4 17. Ivory Coast 3.7 34.1 18. Kenya 6.1 13.6 19. Nigeria 10.3 17.7 20. Senegal 3.5 6.5 21. Sudan 0.5 0.4 22. Tanzania 5.2 12.8 23. Tunisia 33 1 29.2 24. U.A.R. 24.1 20.3 25. Uganda --- 0.5 26.‘ Ceylon 1.5 2.0 27. China (Taiwan) 26.3 185.9 28. Hong Kong 319.0 859.0 29. India 353.5 451.6 30. Indonesia 12.0 22.6 31. Iran 57.6 106.5 32. Iraq 7.2 8.9 33. Israel 38.1 80.9 34. Jordan 0.1 0.3 35. Malaysia 41.7 69.8 36. Pakistan 39.6 120.7 37. Philippines 41.4 97.1 38. Thailand 31.3 32.8 Sources: Commodity Trade Statistics, Series D, U.N.; Foreign Trade Statistics, Series C, OECD. *Total - A excludes petroleum and certain other products as indicated in the previous table. 171 TABLE A-II-6 ANNUAL TREND RATE OF GROWTH OF MANUFACTURED EXPORTS OF DEVELOPING COUNTRIES TO DMEC Trend Rate of Growth of Manufactured Exports Countries 1960-66 in % 1. Argentina 2.6 2. Bolivia 32.0 3. Brazil 8.5 4. Central America ---- 5. Chile 29.7 6. Columbia 31.5 7. Ecuador 14.9 8. Jamaica 13.7 9. Mexico 14.4 10. Panama 25.4 11. Paraguay 4.8 12. Peru -6.3 13. Uruguay 12.7 14. Venezuela 7.8 15. Ethiopia 13.5 16. Ghana 13.5 17. Ivory Coast 30.1 18. Kenya 13.9 19. Nigeria - 8.6 20. Senegal 13.7 21. Sudan -4.5 22. Tanzania 14.0 23. Tunisia -8.7 24. U.A.R. 0.9 25. Uganda ---- 26. Ceylon 4.6 27. China (Taiwan) 38.8 28. Hong Kong 19.7 29. India 4.8 30. Indonesia 11.4 31. Iran 11.6 32. Iraq 3.9 33. Israel 15.5 34. Jordan ---- 35. Malaysia 8.5 36. Pakistan 17.9 37. Philippines 14.1 38. Thailand 5.9 Source: Table A-II-S. NOTES TO TABLE A-II-7 Table A-II-7 lists the value and the annual average growth rates of major manufactured imports of developed market economy countries (DMEC) from the developing countries and the rest of the world in 1966. The manufactured commo- dities were defined according to the SITC system of class- ification as followed by the United Nations and the UNCTAD. The growth rate of imports and the value of imports in mil- lion dollars and in percentages of the total, served as the two basic economic indicators used in the selection of manu- factured commodities for our investigation. The annual average growth rate of imports indicates the change in imports of a commodity from one year to the next over the period 1960-66. 172 173 m.m v.au m.vm m.ma m.mq mpomnuxm mcflccme cam mcflmmo Nmm m.ma m.Hm H.mH o.m~H a.mmm mHmUHEmgo oflcmmuocH mam n.n n.m| m.nH o.ma m.vm mHHO mahmummm> tam HmEflcd commmooum Hmw m.m >.om H.H m.ma m.mmma umnfim oaumnuc>m mom m.v m.n m.mH «.mmm h.vama U003 Ummmzm mvm m.vH m.N m.m o.m o.amammusuomMscmz OOUMQOB NNH 5.0 m.m m.oa n.eom H.55mammmmum>mm oflaocooam NHH m.o m.mfi o.vv v.mm N.H~H mummm cam Hmuusm moooo m.~no m.n «.ma m.mm n.NH H.mv Downbxm mmwmoo m.Hho >.HH o.mH 0.5m H.nm H.mmm pw>nmmmum .mmanmummm> mmo H.m m.~a m.mm n.ena «.mmm omnmmmum uo pm>ummmum .uflsum mmo o.m m.mu v.0a m.nH >.Hna uflsum Umfluo mmo o.m m.m m.mH mzvo m.mmm cmummmum no coccfle .nmflm mmo m.n m.H H.0m m.oma m.hmm cmnmmmum Ho omccfle .ummz mHo cauoz mmfluucsou mmfluucsou mCOHHHHz CH mcoHHHHS mufl©OEEoo mo .02 Eoum mcflmon> mcflmon> mmflnucsou CH coflumflnommo OBHm two Eoum too Eoum mcflmon> panoz Eoum mmmucmoumm loo Scum Amumm nvzouwv oomauamma manomeH CH mmcmco w mmmum>< Hmsccd ommfi ca omzo an muuomeH mmHmBZDOU wzozoum Emmmdz Dmmoqm>mo wm mBUDQOmm QmmDBU03 mmm m.m v.HH m.vm m.aom m.m~m moflunmm coguoo cm>oz mmm m.oa n.mH m.m v.nn m.momH ommuna ocm onmw mafluxme Hmm m.m 5.0: m.o H.m o.mmvm UHMOQ unmmmm cam Hmmmm Hem m.m m.HH o.m 0.0m m.mmm .m.m.c .m@H5D0Mmscm2 @003 «mm 8.4H 4.mm o.m~ 4. 4H m.som voozsam can Hmmcm> Hmm m.m m.m 0.0 m. H.oon whoscoum nonnam mmm «0.x m.m H.h H. m.mm .m.m.c.mmu9u nommscmz “maummq Nam H.m m.va o.mm o.hma m.vnv Hmcpmmq Ham o.m o.mu m.m m.mm m.omoa .m.m.c.mamoHEm£O mmm m.oa m.m m.am m.om >.mmm mHHo Hafipcwmmm Hmm m.ma m.mm h.m o.mm m.mmm muosooum HMUHusmomEHmnm can Hmcfloflpmz me cane: mmfluucsoo mmauucsoo mcowaaflz 2H mcofiaaflz auflcoaeoo mo .02 Scum mcflmon> mchon> mmfluucsou Ga coHumfiuommo OBHm Imo Eoum loo Eoum mcflmoam> cauoz Eoum mmmucmoumm loo Eoum Ampmm nuzonov womauammfl muuomeH 6H mmcmcu m mmmum>< Hmsccd mead cu omza an muuomeH Apmscflucoov NIHHI< mqmée 175 ~.mH G.om m.NH m.me m.m~o mcooo mcfluuomm .mmoe vow m.mm m.om v.o v.e~ m.vmm muosooum oflummam mmm m.mH H.0m o.m m.mm n.0mm umm3uoom amm m.mH H.vm m.Hm m.mmm H.mmmm @cflnuoao mom mcflosaoxm mcflcpoHo qu N.FH v.54 m.vH m.- «.mma mmmavamm .mpoow Hm>mue Hmm n.mm n.aa v.m m.am m.mmv mHSDchsm Ham m.om H.mm H.o o.am m.omm .m.m.c moooo HmOAMDUon amp m.ma m.mv m.v m.mo n.0mva msumummma mcoHDMOHGSEEoomHmB vmn m.va m.ma m.v m.mm o.mHHN wumcflnomz Hmofluuomamncoz Han m.oa m.mH m.m m.mv v.HvHH pmxnozco omxuoz .EscHEsHa vmo «.mm «.mo N.@ n.em m.omm msoHHa a ummmou N.mmo v.mH o.ooH m.o m.ma m.vmma .ooum Hmmum cam couH umcuo who n.m m.mm v.0 m.m m.amma .oum .moom .mumm Hmmum cam couH mum F.HH e.m o.m o.mm p.046 .oum .couH mum Hem m.NH m.ma m.mm v.mva n.mom mcflum>oo Hoon 5mm canoe mmfluucsoo mwfluucsoo mcoHHHHz cu mcoflHHHz suHBOEEOU .oz Eonm qcflmon> mnemoam> mmfluucsou ca mo coaumfluommo OBHm loo Eoum loo Scum mcflmoam> cauoz Eoum mmmucmoumm moo Eoum Amumm guzouov momauamma manomeH cfl mmma ca omzo >6 muuoasH mmcmno w mmmum>< Hmzcc< Awmscflucoov nnHHud mamas 176 .Hmm .mmm .mmm .mmm .mmm .Hmm .mmm .Hmm .meo .Hom .Hmm .Hmm .mam u.moz oeHm mosaonfl mmusuommscme umsno *« .mmma 4 .0 mmflumm .moflumflpmum momma cmflmuom .oomo no mmflumm .moflumflumum moans muflooaaoo .z .D "mousom m.HH o.HH ¢.m H.mHHv m.vmmmh Adeoe Inn: 1:1: nun: m.mmm omoa m®H5p0Mmscmz umcuo ** m.m m.mH v.om m.mm m.amv mmauaunm whamzmb 5mm UHHOK mmfluuccou mmfluucsoo mcoflaaflz ca mcoflaaflz huHUOEEOU .oz Eoum mcfimoam> mcflmoam> mmflnucdoo ca mo coflumwuommo UBHm loo Eoum two Eonm mcflmon> wauoz Scum mmmucmoumm Imo Eoum Ampmm Busouov woma-ammfl mpuomeH ca mead cu omzo an muuomeH mmcmnu w wwmum>m Hmscc« Aomscflucoov nuHHud momma ’. e Ki NOTES TO TABLE A-II-8 Table A—II-B lists the nominal and effective tariff rates on manufactured products in the developed countries (and regions) before and after the Kennedy Round (KR) talks for mutual tariff cuts. These rates were obtained from the calculations made by Professor Balassa for the Atlantic Trade Project and the UNCTAD. According to him, the effec- tive rate of duty indicates the degree of protection of value added in the manufacturing process due to the nominal tariff structure, given international immobility of labor and capital. If input coefficients are constant in the rele- vant range, the effective rate of duty (9) for any commodity can be expressed in the framework of an input-output system. Let t denote the nominal rate of tariff, M the material in— put coefficients, and V the proportion of value added to output, all measured at world market prices. For commodity j we have, In order to calculate the effective rates of protec- tion, Balassa obtained comparable data on nominal tariff rates and input-output coefficients net of duties. For this purpose "standardized" input-output coefficients (1959) 177 178 for the Common Market countries were used. With regard to tariffs, Balassa used the Brussels Tariff Nomenclature, as it was followed by all the developed countries. ‘._ 1 179 Isl---.--- OO Om OOH HOH am mm mm «O HO mm oflumm ~.m~ m.HH O.ma 0.0 m.OH h.O ~.mH H.O O.MH 0.0 mmm mx-umom mmusuomH m.mm m.OH m.OH 0.0 0.0~ H.ma m.m~ 0.0H v.O~ O.~H mx-mum -s:mz noosz pm Oh OH om OO mu mm om Hm om ofiumm . O.m~ 0.0H O.~ m.m 0.0H m.HH O.NH 0.0 O.ma m.O ~.HOO mx-umom ~.OO 0.0~ 0.0H 0.0 m.~m O.ma e.mm m.na O.m¢ H.OH mx-mum Ooozsamxm «O OO -- -- Om om Om OO O MO ofiumm m.m m.~ -- O 0.0 O.H 0.0H m.O O m.O MON mx-pmom Omxuoa m.mH O.m -- O -muw m.m m.a~ 0.0 H.H 0.0 mx-mum samsfim voozxm -- -- -- -- -- OO -- HO -- -- oflumm zmsou map -- O -- O -- O.H -- O.H -- O m.~¢~ mx-umom cw noozxa -- m.~ -- O -- H.~ -- m.~ -- O.m mx-mum Oooz HO On OOH mm mm OO OO mm OOH OO oflumm m.om m.- O.- O.HH m.OH O.HH O.~m 0.0~ m.ON 0.0H Hmm mx-uwom H.me m.m~ O.- O.va O.mm 0.0H ~.Om 0.0~ m.m~ 0.0H mx-mum mmonmxv me Om OOH mm mv om Hm om OO Om ofiumm mmonm can» O.mH O.HH H.- v.OH 0.0H m.O H.O m.m O.HH 0.0 HMO mx-umom umnuo mooom O.mm O.m~ 0.0~ m.~H m.v~ 0.0H 0.0~ 0.0H m.O~ m.ma NHO mx-mum umgumwubm mm mm ON ON no mo mm on 54 OO oflumm 0.0m O.HH m.v O.H m.mH 0.0 m.om O.HH O.~H O.v HHO mx-umom 0.0m 0.0H O.H~ O.s m.OH m.h m.Om 0.0H O.m~ 0.0 mx-mum umnummqflm -- -- -- -- -- -- -- O O OOH oaumm mcflxm -- O -- O -- O -- 0.0 -- H.H Ham mx-umom 6cm mmofimpw -- O -- O -- O -- O.m -- H.H mx-mum maxequ m>flu Han m>flu Hm: m>fiu Hmc m>flu Hm: m>fiu am: .02 aonmm weHOozzoo -omwum -Heoz -ommmm -HEoz -ommum -Heoz -umumm -flsoz -ommmm -usoz oeHm zaman zmomzm weHzazzoo onoz zooosz omshza mmeH90mmmm 024 AflZHEOZ mIHHI< Mdmflh 180 -- -- -- -- -- -- -- -- -- OOH oHumm -- O -- O -- O -- O -- ~.O mx-umom 3mu -- O -- O -- O -- O -- ~.O H.MO~ mx-mum .couuooAH 209906 am Oh mm mm mm ms mm OO 8 ma oHumm ~.O~ «.mH O.FH O.HH ~.OH «.mH «.mm O.OH O.~ O.OH mx-umom O.Om O.- O.O~ O.vH «.mm m.O~ s.m~ O.ON «.m- H.HNANVH.HOO mx-mum OcHnuoHu X0 H0033. O4 Om pm pm mOH OOH On OO HO HO oHumm cm>oa m.H~ O.OH «.mm H.HH H.Om O.OH e.n~ O.OH O.OO ~.O~ mx-umom .moHu H.me O.O~ H.ON O.~H O.Om O.OH m.om m.hH H.OHH H.Om ~.mmO mx-mum -n8m Hoozam om Om av Om OOH OOH as up mm OO oHumm n.eH O.m 0.0 O.~ m.aH s.m O.m~ 5.5 m.OO A.O~ mx-umom cum» O.m~ O.OH O.OH O.m H.OH n.m O.Hm O.OH «.mm O.HN ~.HmO mx-mum .Hoozxm -- -- -- -- -- -- -- -- -- mm oHumm -- O -- O -- O -- O -- n.m H.~Om mx-umom -- O -- O -- O -- O -- O.HH mx-mum 38H.HoozAH Hoox mm me O4 Om OO am «O mm mm HO oHumm mmHoHu N.~H ~.e O.H O.H O.HH H.O F.NH O.m O.m m.~ NOO mx-umom -Hm ummwm m.OH m.OH N.O O.“ O.OH m.OH O.HH 0.0 O.m ,H.m H4O mx-mum 6cm umammam OHH OOH mm -- NO OF Om -- HO -- oHumm .mp O.m m.O- O O.m m.m «.m- O H.H- O Hmm mx-Hmom m.O O.m O.O- O O.m A.O m.O- O O.H- O mx-mum mHsmOoozAm -- -- ---- -- -- O -- -- -- -- oHumm -- O ---- O ---- O -- O -- O H.HON mx-umom -- O -- O -- H.N -- O -- O mx-mum OooamHsmxH mmmdm O24 OHOE m>Hu Hm: w>Hu Hm: m>Hu Hm: m>Hu Hmc m>Hu Has .02 oonmm weHQOZZOO -umuum -Heoz -omuum -Heoz -omuum HEoz -omuum -Heoz -omuum -Heoz oeHm zo3 v.em 0.0m O.HN m.> m.mm 0.0H h.vm 0.0~ 0.0- O mx-umom .moHu m.hO O.mm m.Om O.HH ~.~O O.m~ ~.vm 0.0~ 0.0 O.~ «.mmm mx-mum -nmw muSOHN -- -- -- -- -- -- -- -- -- -- oHumm -- O -- O -- O -- O -- O mx-umom -- O -- O -- O -- O -- O VON mx-wum mush ammHH mesh On mm mm mm no mm mm mm mm on oHumm pmuuch H.5q m.mH H.Om O.mH 0.0~ H.NH H.Ov H.HN «.mm 0.0H mx-umom.mmHu0mmwoom m.OO 0.0~ «.mv 0.0H m.Hv man- O.m¢ v.m~ O.mv O.m~ O.HOO mx-mum mammuoHon H.mO Oh OO mm vs Oh HO OO mm mm oHumm m.nm 0.0H 0.0H v.~H m.ON 0.0H O.m~ 0.0~ v.mm 0.0m mx-umom m.Ov O.Hm O.HN 0.0H H.Om m.mH m.OO O.mm H.mv 0.0NH~OH.HOO mx-mnm mcHnuoHo xw couuooHO mm mu Oh Om OO om mm mm Oh Oh oHumm 0.0H 0.0 H.OH m.OH H.mH O.~H 0.0v 0.0H O.¢~ m.mH mx-umom cuoz moHu 0.0~ m.OH m.m~ H.mH m.hm O.mH 0.0v 0.0H N.Hm m.OH NmO mx-mum -28m coupouAm av Om Om Om mOH OOH mm Hm Oh OO oHumm pmmucu 0.0 m.m v.Om O.m O.Nm 0.0H m.OH H.O O.m~ m.OH mx-umom new cum> m.mH O.m O.vm 0.0H O.Hm 0.0H m.- m.> m.mm H.MH m.HmO mx-mum .couuouHm m>Hu Hmc m>Hu Hmc m>Hu Hmc m>Hu Hmc m>Hu Has .02 OOHmmm meHOozzOo -omwmm -HEoz nommmm -wEoz -ommwm -HEoz nommmm IHEoz -ommwm -HEoz oeHw z¢m¢h zmomzm WBHZDZZOU UHZOZ ZOQOZHx omBHZD mmeéem QmBHZD -oom zammomam- AUOSCHUCOUV mIHHI< mamdB 182 0.0m H.mH N.OH v.m O.mm 0.0H O.mm O.mH m.mm v.vH mx-wum mmusu nomwscwz HmuwzHMH m.~m 0.0m h.vm O.m O.HN «.mH 0.0~ 0.0H 0.0H 0.0 mx-mum mmcHumwO Hmuwzwa H.vH O.MH m.m v.m m.ON m.m 0.0v 0.0H m.O H.m mx-mum muospoum Hmmum umzuoHHH m.mm v.mH m.mH ~.m m.OH N.O v.5 m.m N.N- H.» mx-mum .poum HHHE OcHHHoEOH v.n~ m.mH O.NN m.mH 0.0m v.vH N.Om m.mH m.m~ O.mH mx-mum muospoum mmmHm .mmMHoAm m.ON m.mH 0.0H 0.0 m.mH m.mH m.Om O.mH v.om N.mH mx-mum mausooum HmumCHz oHHHmume-cosz m.mm m.OH 0.0 m.m H.MH O.HH 0.0H v.mH O.mH O.NH mx-mum muospoum HmoHEmcu .ommzAh m.HO ~.Om 0.0m O OH n.em m.MH N.HH H.HH m.mH N.HH mx-mum medwumm .mucmod OchmmHOAO O.mm N.~H O.m m.v m.ON m.HH m.mm 0.0H 0.0m m.mH mx-mum .poum HmoH 1:020 “$5on H.mm H.mH O.NH m.n 0.0H O.NH H.5H O.NH m.mm 0.0H mx-mum mHmHumumz 0Humauc%M%v m.mm 0.0m m.mm O.mH 0.0m 0.0m H.Om 0.0H 0.0m O.Hm mx-mum mmHoHuu< oHummHmHm m.v- O.H 0.0 0.0 0.0- m.m m.O O.m m.m m.m mx-mum umuuwe pmucHumHm m>Hu Hm: m>Hu Hm: m>Hu Hm: m>Hu Hch w>Hu Hm: .oc oonmm weHoozzoo -ommmm -HEoz -ommwm-IHHE02 -omwwm -HEoz -ommmm -HEoz -umwwm IHEoz oeHm z¢mdn zmomxm >BHZDEZOU UHZOZ EOQUZHx DMBHZD mMBdem omeHZD Ioum Zmm\O\Oe .Ocsozo .HH uumm uxhow 3m¢~.Dwo ou ummumucH mo mmusuommscwEIHEmm paw mmH5908mscmz cH mpmue mo connmmmm cam :oHumeHmuwnHH on» new meHmoum =.mcoHumz mchon>mo Scum mpooo memwooum co wuowm m muH Eocoom Hmo sHuHHom mo Hmcuson =.c0HumsHm>m ad ”mmHHucsou HmHuumsvcH CH coHuomuoum MMHHmeg .mmmMHmm .< .m "mmousom N.Hm O.Hm 0.0H 0.0H 0.0m m.nH O.mm m.m~ O.Hv O.mN mx-mum .oum .muHmzmn .mxou .mooom mcmuuomman m.mm N.mm 0.0H 0.0 N.Om m.MH N.OO O.mm m.~m O.HN mx-mum mucwe usuumcH conHumHmHmH .<.z .<.z .«.z .<.z .<.z .¢.z .¢.z .¢.z O0.0N m.OH mx-mum mcooo new mmocmHHmm< HMUHuuomHmHhH m.mm H.mH 0.0H 0.0H m.H~ m.vH 0.0m n.mH H.mH ~.NH mx-mum mumcHnomz HmoHuuomHmHOH O.HN m.OH O.HH m.m «.mH m.OH ~.HN H.OH H.OH O.HH mM-mum wumcHnomz HmoHuu uomHm-coszH m.mm 0.0N 0.0H 0.0H O.mH O.mH m.H~ v.mH m.O- 0.0 mx-mum wwmcHnomz HmusuHsoHuoHu Hmc m>Hu Hm: m>Hu HOc w>Hu Hmc m>Hu Hm: .oz oonmm wBHoozzou -omuwm -HEoz nommwm -Heoz -owmwm -HEoz -ommwm -HEoz nommmm -HEOZ OBHm z¢mdh zmamEm wBHZDEZOU UHZOZ ZODOZHx QmBHZD mm9<9m QMBHZD loom demomam inmscHucouO mIHHI¢ mdm<9 NOTES TO TABLES A-II-9-l3 Tables A-II-9-l3 list the capital and labor coef- ficients per one million dollars (or pounds sterling) for industries in Japan, the U. S. and U. K. The capital co- efficients as presented here are computed in the net form by dividing the net book value of capital stock and inven- tories by the current value added of the respective indus- tries and multiplied by one million. Similarly the labor coefficients were computed in man years as the ratio of the number of workers employed in an industry and its value added, multiplied by one million, i.e., the number of man years required to produce one million dollars (or pound sterling) worth of value added. The occupational structure of the labor force was divided into two major groups of skilled and unskilled labor in order to define the human skills coefficients in the same way as the capital and labor coefficients defined above. 184 185 TABLE A-II-9 JAPAN'S CAPITAL AND LABOR COEFFICIENTS, 1954 (Per Million Dollars Worth of Output) Labor/ Capital/ Output Output in No. Description of Industry (man years) Thousands of $ 1 Agriculture 5600 1430 2 Forestry 2100 433 3 Fishing 2400 691 4 Coal Mining 860 563 5 Crude Oil 468 960 6 Other Mining 524 535 7 Processed Foods 841 140 8 Textiles 1545.6 174 9 Lumber & Wood Products 1645.5 282 10 Paper & Paper Products 701.2 195 11 Printing 728.8 179 12 Chemicals 493 284 13 Coal Products 415.2 287 14 Petroleum Products 415.2 287 15 Rubber & Plastic Products 664 162 16 Leather, Leather Products including Footwear 1139 93 17 Non-metallic Mineral Products 807 409 18 Glass & Glass Products 777 421 19 Iron and Steel 649 203 20 Non-ferrous Metal Products 641 203 21 Metal Goods Industries 1289 362 22 Non-electrical Machinery 1280 298 23 Electrical Mach. & Equipment 994 346 24 Precision Tools 843 229 25 Transport Equipment 714 324 26 Other Industry 1506 89 27 Transportation, Communication & Public Utilities 714 2610 28 Land Transportation 747 2698 29 Water Transportation 575 2078 30 Trade and Services 5542 2217 31 Financial Services 699 774 32 Electricity & Water 438 3304 33 Trade 1390 556 34 Other Services 2760 1736 Sources: National Income Accounts (Japan), 1957; National Wealth Survey; Economic Bulletin No. 1, February, 1959; Capital Structure of Japanese Economy, Economic Planning Agency and Council for Industrial Planning, Tokyo, Japan; T. Watanabe, "Approaches to the Problem of Inter-Country Comparison of Input-Output Relations: A Survey,: Inter- national Compgrison of Interindustry Data, (New York: U.N., 1969), PP. 187-210. 186 .m.D mmB ZH Ammfimw Zdz ZHV BDmBDO ZOHHHHE Hm mmm AO\AV Amm4HHOD .m ‘I’ilHllLl .O mo moz¢maome zHO 1|” Hlvr) Di I'M, ll-i 1|"! oHIHHI< mqmdfi hflfifl- - ..fiwr om0.0m omm.OHHH mposcoum omHHH< 6cm mchmm Om omm.mm OOO.mmm mcoHumummmum umHHOB paw OCHcmmHO .mmsuo ON OOO.HO omm.vmm mHMHnmumz 0Humcuc>m cam 0HummHm mm OHH.Hm OON.mmm muospoum HMUHEmcu omuomHmm paw mHmoHEmcu Om OO4.OO OOO.mOO mcHsmHHnsm Ocm OcHucHuO mm omm.>m omv.ONMH mmxom paw mumchucou UHmOQHmmmm mm ovm.mn omm.mmmH mumchucou pmmoxm mausooum UmHHH¢ cam Hmmmm Hm OO0.0HH OOh.H>O mmusume Cam musuchsm Hmcpo ON NO.mOH OO0.000 musuchsm cHonmmsom OH owm.mmm OO>.mHHH mumchucou cocooz mH OOH.>OH OO0.0mm mumcHMHcoo ummoxm .mposOoum Oooz O HmnssH RH mm.OOH OmO.NHm muoscoum mHHuxma UmumoHHnmm .omHz OH omm.~OH OO0.0mm Hmuummmm mH OOO.MOH omm.mOm mOCHum>ou HOOHm Ucm mnoow mHprwe .omHz OH omm.mmH OOO.mmm mHHHz pmmuca paw cum» .moHHQmm Bouumz O pmoum MH OOO.mm omm.>OOH mmusu08mscmz ooomboe NH OOH.mm OHH.NOO muosooum omnchm paw poom HH OOm.HO omm.mOOH OchHz HmumcHz umNHHHuumm 6cm HmoHEmno OH ONA.OO Osm.mmmH mchHz smHO cam mcoum m OOO.Hm omH.OmOH mmo Hmnsumz paw EsmHouumm mpsnu m ONO.mmOH oom.mmom OCHcHz Hmoo n Omm.>O OOO.mOmH OCHCHZ mono Hmuwz msonumm-coz O Oev.mv OOO.OOmH maHaHz mwuo HoHHa-ouumm 6cm couH m OOm.OHH OO0.0~OH mmOH>umm HumgmHm Ocm Hupmmuom .HmusuHOOHuOO O omm.mm om0.0~ON muosooum mnmanm cam muummnom m OOO.mv omm.mmhm muosooum HmnsuHDOHHmm umcuo N OHv.Oh onm.mmmm muosooum xoopmm>HH cam xuoumm>HH H mucmHonmmou mucmHonmmoo wuumsch mo cOHHQHuommo .oz HOQMH HmuHmmo BZMHUHmmmOU momdq mmB sz BDmBDO ZOHHHHZ Hm mmm AO\MV mBZmHUHmmmOU HfifiHmfiu 187 0mm.Hm 0mm.omn muuma pcm uwmuouH< mm OmO.mm OH~.OOm ucoemqum Ocm mmHoHnm> Hobo: om OOO.mm OHm.mHO ucmEmqum .xumcHnomz HmoHHpomHm .omHE mm OO0.00H Omv.wmw meHOmmmoo< cam mucmcomeou oHHuomHm Om OO0.00 OO0.000 pcmEmHsOm coHOMOHcsseoo Ocm .>.e .oHOmm mm OOO.mm OHO.HOO ucmEmqum mHHZ pcm mcHuanH UHMHOOHm mm omm.OO OON.NHO mwocmHHmm¢ OHogmmsom Hm OOH.Om OOO.OOO msuwummmd a ucmEmqum HmHuumsch HMUHuuomHm om oom.vn ONm.mmm wumcHnomz mnumsch moH>umm mv omh.mm Ohm.mmm >HmcH£omz mcHucsooo< cam mcHusmEOO .monwo mv Omm.mmH OOO.mNO muoscoum mocm mcHnomz Ow OHm.mm HOH.ONHH wumcHzomz HMHHumsch Hmumcmw OO omm.Hm OM0.000 mumcHnomz muumSUGH HMHommm mv OOm.mm omm.NmO wumCHsomE mcquoz Hmumz Ow Ohm.H> omv.mmm ucwEmqum Ocm wuochomz mcHHpcmm mHMHumumz mv OHO.OO omm.vqm Humchomz OHmHm HHO Ocm OOHcHs .coHuosuumcoo N4 ONO.NO ONH.OOO ucmEQqum Ucm mumchomz Eumm HO oom.OO OOO.mmm mmcHnuse Ocm mmcHOcm OO OOH.OOH OHO.Hmm mp09poum Hmumz OwumoHuhmm Hmzuo mm OO0.0HH Omv.mmm muHom Ocm muosOoum mcHnoms zmuom .OCHmemum Om Omm.OOH OO0.000H .Ooum Hmpmz Hausposnum Ocm OOHQEOHO .OcHummm Om OOO.m> omm.vmm mHmchucou Hmumz Om Ohm.mm OOO.HomH OCHHSHOMMDcmz Hmumz muonummcoz wumEHum mm Omv.mn OOO.mOvH mcHHspommscmz Hmmum cam COHH mumEHHm Om Ovm.Om omm.Omm muospoum >MHU cam mcoum mm OHN.Hm OOO.NMOH muospoum mmMHw 6cm mmMHw mm omm.mnH OM0.000 macawoum Hmcummq chuo cam ummzuoom Hm OHH.HHH OO0.000 muosoonm Monummq HmHHumsocH a mchcme umnummq om ovm.Om omm.nHm mposooum UHDmMHm .omHz cam umnnsm mm oom.Om omv.mvom mmHHumSUcH Umpmem cam ochHMmm Edeouuwm mm mucmHonmmoo mucmHonmmou wuumsch mo coHumHuommo .oz uogmH HmuHmmo HOmschcooO OH-HH-a mHmme humus ,nmw ‘1‘ .mn-HmH .Qm .OOOH .mmconoum .uospoum paw mEoocH mo coHuHmomEou HmHHu -mmmmH .onHOamO .3 anon uHHH xHOcmmm< mmm .OcHeoosunoOO HOOom Ocm pace .2 “AcmEHmuH< xomhv I .m.m.o .moHDmHumum HOQMH mo smmusm uHmOOHO mmmchsm 0cm HMOOHO mchHz .HOOOHO musuHsoHnmd .HNOOHO mcHusuommscmz mo momsmcmu "mmousom 188 ONO.mmH OOO.OO>~ mmoH>umO HmcoHumosOm .HmoHOmz HO OOO.HNH OOO.OOOH mmoH>umO mmmchsm Ob OOO.OOH OO0.000 mmoH>HmO HHOOOO new Hmcomumm “mHmpom OO HmO.HN OOO.ONOO Hmucmm Ocm mpmumm Hmmm OO OOO.MO OON.Om mocmuumcH cam mocmcHO OO OOO.HO OOO.HOHH mwmue HHmumm new mHmmmHonz OO HOH.OO OMO.OOOH mm0H>an “mum: .mmo .oHupomHm OO OO0.00 OO0.0000 OOHummocmon .>.a .oHOmm OO OHH.HOH OOO.~OOm .>.e Ocm oHOmm ummoxm .coHumoHcsasoo mO OOO.mNH OOO.OOOH mchsogmumz new coHumpuommcmue NO OOm.OHH OOO.mOO OcHuspomOscmz msomcmHHmomHz HO OOO.~O OO0.0mOH ucmEOHOOm oHnmmuOouozm Ocm oHsHmnunmo .HmoHpmo OO OO0.0m OHm.vaH mucmesuumcH mcHHHouucou can UHMHucmHom mm OHH.OOH OH0.000 ucmsmHsOm coHumuuommamua umnpo Om mucmHonmmou mucmHonmmoo wuumsccH mo COHpmHuommo .oz gonna HmuHmmo HOmscHucooO OH-HH-a mHmOO~ om.mmm mmH.Omm.H O0.0omm xUOpm mcHH IHom cam mOUOH >m3HHmm OH mm.HHm On.mmm OOO.m>m mm.mmOH uwmuoHHd mH OO.OOO OO.OO~ OOH.OOH.H O~.OOOH mmHomu Ocm muogoz OH OO.OOOH OH.OHO NOO.OOO OO.OOOH OOHnmmchcm mcHumz Oam OOHOHHOOOHOO mH OH.OOO mm.mHO oom.mmm mm.OONH mcoow HOUHHuomHm Ocm mcHHmmchcm NH Hm.mOm Om.mmm OOO.mmH.H O0.000H mHmumz msouumwlcoz HH Om.mOO Om.HOH OO0.00H.H OO.MOO H.u O .uO Hmmum O couH OH O0.000 OO.HON ONO.NNO.H H~.mmh .m.m.c mHmoHEmno m MH.OOO OO.mOH HHH.HOO.N O0.000H H.0 O .M.EO Hmmum O couH O OO.HOO mm.mON OH0.000.HH OH.Omm mchHmmm HHo HmumcHz n O~.OOO mm.mON OHH.OON.O O0.000 .uum .mcm>o mxou O ON.OOO Om.HOm OOO.NOH.H . O0.00m ooomnoe paw xCHuo m mn.OOOH O0.0mm mO0.00m.H mm.HOOH Ochmmooum boom O OO.MHMH OO.OOH OOH.OOO.H OO.HNOH.m.m.c.OOH>uHOsO wan OchHz m Om.mmO O0.0m OO0.0MO.H mO.mOO OchHz Hmoo N NO.NOHH OO.OOH OOO.OOO OO.OOMH OOHOOHO Ocm xuummuom .wHSDHdoHHmd H Hmummw cmEO Hmummwcmev Hmummx muumsOcH .oz usmuso usmuso H mocsomv cmEO \HOQOH \uonmq usmwso usmuso OmHHmecO OmHHHxO \HOOHOOO \uoan Hummuso Oo nuuoz u coHHHHz mco HOOO AOOOHO zOOOzHO OOOHZO mme mom OezmHonmmoo OHHHOO oz< .momOH .HOBHOOO HHIHHI< mqmfifi 190 .NOIHO .mm .AOOOH .OUHmum>mcD cumummmnuuoz "coumomv N .Ho> .somoummm COOHummEoo HmcoHuflmmvucH ad .mchcmHm H0O mucmEmHqumm Hmzomcmz .Dcmumcmuumm .H .H pcm .CMEHmEON .z .NDHBOHom .2 “OOOH .mmmum 9H2 .Ovmmum>HcD mOUHHnEmu .moHEocoom pmHHmm¢ mo ucmEuHmmmo .HOQOH paw HmuHmmu .GOHuUSUOHm =.£u30uw H0O Emnmonm 4: ammonsom O0.0HOH O0.000 OON~mmm OO.HOOH .m.m.c .mmoH>Hmm Hm mH.mmm O0.0mOH HMO.HOO m0.00MH mmpmne m>HuanupmHo Om Om.m>O mm.HMO OOO.HNN.O NO.HHMH mGOHp IOUHCOEEOU cam “Hommcmue ON OH.HNm Om.OmN OOH.OH>.ON Hm.mbh nmumz ON mn.OOm OO.>HN OOO.mmm.mH OO.HOO mpHOHuuomHm ON OO.mOm mH.m>O OHO.mmH.O HO.HmOH mmo ON m0.0mNH Om.OON OOm.NHO ON.MNmH COHposuumcou ON Om.OmOH N0.00m OO0.00N.H O0.000H umnbsm mcHosHocH mmnsuomwsamz Hmnpo ON OO.NNO OH.HOm OOO.OOH.H OO.m~HH OOHOOHHO .HOOOO mm O0.000H HO.>ON HHN.mmO OO.HO>H .oum .mHDuchsm .quEHB NN H0.00HH Nm.OON OOOONmm MN.mmOH mmmHo .wumupom HN Amummm cmEO Hmumwm cmEO Hmummm wnumnch .oz usmuso psmuso H mccsomv cmEO \HOQOH \HOQMH usmuso psmuso OOHHHOOOO OOHHHxO \HmuHmmo \uoan HOmscHucooOHH-HH-a mHOOB I: ..... D n O. ..- lll' ‘ l 'Ollv.‘ ...? . .... ..u......- ,7: ..u .v. v. \o. -.u.~¢.H ..rr... r.... O...II -. O , ... . .I a t n . .. - A 6»qu \11v3/ /¢:~.f - ..I . - \ \w ... . l Auv «fun. Or v- A..:. \.\-.&.\- ..L ... O A. .— 4 r ra.-u.~ - «Is A , - .D m x. N. COO UNI. .- -. - - «.Cr mflAUN/U 1:... 1-». . :12 m3 H Fm Ca -3. N a..».H ...» H 7:4»-..ZHHOLOHO Huh. 0 Ann-“- L . . .I A M m 5.90 n-K-Nl-H F.fflddvf fAJfl/nw/fi#nuw . . . ..CC ~ L . . . .55 NHL .D..fi A v H NIH? - .a»—H..u‘ «(H H .,..«..P;O.,O 191 mm.mOO OO.mNH muosOoum OOHO Ocm mcoum .OOOHO ON O0.0HO m0.00 muosvoum OMHO Ocm mcoum ON O0.00N NN.ONH muOSOOHO Hmou mam Edeouumm ON O0.0mm Hm.OmH mu05poum .mHOOHEmnu NN m0.0Hm O0.00H mposcoum Ocm nmnasm HN OOO.NNO OO0.000 mcHsmHHhsm can mcHucHum ON OO.HOO NO.ONH muosmoum ummmm .ummwm OH O0.000H HO.mOH mnsuchsm cam muospoum @003 OH NO. mOH O0.00N muosnoum 0003 van HGQESH OH Om.OmmH O0.00H mmusume cam musuchsm OH N.HHO 0.0NN muosooum uwsummq can Hmcummq OH HO.mHOH ON.ONH Hmm3uoom OH OHN.OOmH NOm.OOH mmHHuxmB OmumoHunmm umcuo .OCchoHU mH mom.NOmH OmO.mOH muospoum HHHz mHHuxmB OOm.ONm mmO.mHm muoswonm can ooomnoe OO0.000 MHm.NOH mmmmum>mm .Ooom OOO.OOO OOO.OO mumnuo cam OOHOHHOOO m NOO.Nmm ONH.mHH woo Hmusumz .EsmHouumm O N.OHO 0.00H . mmuo Hmumz O O0.0MO NH.NNH Hmoo O Om.omm mo.HOH OCHOHMODO paw UGHGHZ m 0.0HmN 0.00 mmHHmanm O O.mOOH 0.0mH whammuom m N.ONOO O.NOO mHDuHDUHHmé N 0.00mm N.NmO meumanm Ocm Oupmwuom .musuHsoHHO< H mycmHonmmoo mucmHoHOmmou OuumsvcH mo coHumHnommo .oz uonmq HOQMH OmHHmecD OwHHme Ausmuso Oo Sumo: OHOHHOO coHHHHz mao ummv OOOH AOOOOO zazv OezmHoHOOOOO HHHOO O.z«mOO NHIHHLQ mqmfimB .Hm-Om .mm .HOOOH .mmmum OuHmHm> IHcD cumpmmmzuuoz "coumomv comoummd COOHHOQEOO HmcoHpmcumucH c¢ ”mchamHm How mucmEmuHsmwO um3omcm2 .pomumcuumm .H .H van cmEHwEON .2 .NuH3ouom .4 .2 "mousom 192 NO0.0HN Om0.0HN umumz cam OOHOHuuomHm mm Om.OOmH O0.0mHO .oum .mmoH>umm HmHocmch .mocmusmcH .mpmne Om Om.mON OO.mOHH mOmHB Om ON.Omm H0.0mm mmeHHHuD Om ON.ONO O0.00N mCOHumoHCDEEOO mm OO.OOO HO.OMN OchsocmHm: OOHumuHommcmne Om OH.OONH OO.mNN OCHHsuomwscm2 mSOOGOHHmomHz mm Hm.mNO OO.OHN mucmasuumcH UHOHucmHom Ocm HMOOHmmmmoum Nm NO.HOm OO.NmH ucmEmqum uuommcmue Hm O0.000 HO.mNm ucmEmstm cam mumchom2 HMUHHuomHm om ON.HOO NO.OHm OumcHnomz HmoHuOomHm-coz ON NO.OOOH OO.OOHAOumcHnoOz OOHOOHome mposnoum Hmumz OmumoHunmm ON O0.0HO Om.mmH mHmumz OHOEHum ON O0.0Nm OOm.OHH muosconm Hmumz cam Hmum2 ON mucmHonmmou mgcmHonmmoo muumsccH mo coHumOuommo .oz HOQMH uonmq OmHHmecD OmHHme AcmscHucooO NH-HH-O mqmme 193 OO0.0m O®H.@m om OOH.mN owm.mm mm oom.mm oom.Hm av OOH.HN OHo.om ON OO0.0m om0.00 OO oom.mm OOO.NO ON OOO.mO ON0.00 OO OO0.0N OOO.mO NN OOH.OO OHN.Om OO OON.HN OO0.0m HN OO0.0m OO0.00 mO OON.HN OOO.mO ON OO0.0m OOO.HO OO OON.ON ON0.0HH OH OOO.mN OO0.00 MO OO0.0m OOH.OOH OH OO0.0N OHO.Nm NO OOH.ON OOO.NOH OH OO0.0m ONO.HO HO OO0.0N OO0.0NH OH OOO.NN oom.mm OO OOO.mN OOO.mOH mH OO0.0m OOH.OO Om OO0.0H OO0.00 OH OO0.0m OOO.HO Om OO0.0H OOO.mOH mH OOO.NO OOO.mO Om OO0.0H OO0.00 NH OOH.ON OOO.mm Om OOO.HN OOm.OO HH OO0.0N OOO.HO mm OOO.HH OO0.0m OH OOO.mH ONH.NO Om OOO.HH ONN.mO O OO0.0N OON.OO mm OOH.NH OO0.0H O OO0.0N OHm.OO Nm OOm.OOH ONO.HmO O OO0.0N OOO.NmH Hm OO0.0H OON.OO O OO0.0H OHN.NO om OON.O OON.Om m OO0.0N OOO.NO ON OOO.NO OOO.mO O OO0.0N OO0.0m ON OOH.OH OON.OO m OO0.0N OOO.Nm ON OO0.0H OO0.0N N OOO.HH OO0.0N ON OOO.NN OHO.mm H Hmummw cmzv Amummw cm2v .oz Hmnmma cm2v Hmummw cm2v «.oz mucmHonmmou wucmHOHmmmou mucmHOHmmmou mpcmHOHmmmou NOQMH noan HOQMH HOQOH OwHHme OmHHmeCD OwHHme UmHHmeaD mmdmw Z<2 ZH HOOOHV BDmBDO mmHcs cumummw Inuuoz ”coumomv N .Ho> .comoummd COOHHmmEOO HmcoHumcumwcH ad "OCHQGMHm Mom mucmEmHszwm umzomcmz .uvmpmcnumm .H .H can cmEHmEON .2 .NpH3onom .2 "mousom OOH.OO OON.OO HO m.mO OO0.0m HO OOH.ON OO0.00 OO 0.00 OOO.mN OO OOO.NO OO0.00 Om OO0.00 OO0.0mH OO OON.Om OH0.00 Om OO0.0H Omm.N OO OOO.NN OO0.00 Om OO0.00 OOH.O OO OO0.0H omm.OO Om OON.Om OON.OH OO OOH.Hm OOO.mO mm OOO.NN OOH.OO mO OO0.0m OO0.00 Om OO0.00 OO0.0N OO OO0.0N OOm.Om mm OOO.HO OO0.0N MO OOH.Nm OO0.00 Nm OON.Om OO0.00 NO OOH.ON OON.OO Hm Hmummw cm2v Hmummm cm2v .oz Hmummm cm2v Amummm Gm2O .oz ucmHonmmoo ucmHonmmou ucmHonmmou ucmHonmmou uonmq uoan HOQMH Honmq OmHHme OmHHmecD UmHHme UmHHmecD “OmsaHucooO mH-HH-< mqmme m ”a H i. . 3,. .. a.....» - O? um,» a..... ... A. 195 Illllll LII III .III‘ I mOHH.O NOH.OO NO0.0H OO0.0H HOm.H OO0.0 ONO.H HO0.0 OO0.0 mmcHum>oo uoon OmO mHO0.0 HOm.mO NmH.O OHO.HH OOH.H OON.O OO0.0 OO0.0 Nmm.O .m.m.c .cmcHH .wxomm .mmmm OmO OHOO.O HOm.mO NOH.O OHO.HH OOH.H OON.O OO0.0 OO0.0 NO0.0 moHuan OHHuxme HOHuwam OOO OHOO.O HOm.mO NOH.O OHO.HH OOH.H OON.O OO0.0 OO0.0 NO0.0 moHuan coupon-co: cm>03 mmO mHO0.0 HOm.mO NmH.O OHO.HH OOH.H OON.O OO0.0 OO0.0 Nmm.O OOHnnmm couuoo NOO OHO0.0 HOO.OO NOH.O OHO.HH OOH.H OON.O OO0.0 OO0.0 NO0.0 ammunp O cum» wHHume HmO ONOH.O NHH.OO OOO.OH OOH.OH OOm.N OO0.0 HOH.N HO0.0 OMH.H muosOoum umOmO NOO NOON.O ON0.00 OON.HH Om0.0H OOO.N HOO.m OHm.N OOO.N OOO.H Oumon Iuwmmm paw ummmm HOO HOOO.O OO0.00 mmm.O OON.OH OOO.H HO0.0 OO0.0 Omm.O OOm.O .m.m.c .mwusuomwscm2 @003 NMO NOON.O OOO.OO HO0.0H OOO.HH HNm.N OO0.0 OOO.H OOO.H OOO.N muoswoum Omnnsm ONO OOO0.0 OOO.NO OHO.HH OMO.O OOH.H HON.O MO0.0 OOH.O HO0.0 uwsuqu O0 mwusuomwscm: NHO OOOO.O OOO.MO OOO.O OOO.OH NHo.O HNH.O OOOHO HOO.c OOO.O OOOOOOOO .umcummq HHO HOON.H HHO.Om NN0.0H ONO.OH OOO.N OO0.0 OHN.m OOO.O MO0.0 mausooua ocm mHOHumum2 HmoHEmnO OOm NMO0.0 HH0.0m NNO.OH ON0.0H OOO.N Om0.0 OHN.O OO0.0 MO0.0 mmasmumm O HHo HmHucmmmm HOO OOO0.0 MOH.OO Om0.0 OON.OH ONO.N OOO.H NNO.N OOm.O OH0.0 mmunHO oHumcucxm OON OOHH.O OO0.00 OOm.mH HHN.HH OOO.H NOm.O ONO.H OO0.0 NO0.0 = .mmHQmuwmm> mmO OOHH.O OO0.00 OOm.mH HHN.HH OOO.H NOm.O ONO.H OO0.0 NO0.0 Um>nmwmum .uHsum mmO OOHH.O Om0.00 OOm.mH HHN.HH OOO.H NOm.O ONO.H OO0.0 NO0.0 pwummmum No OwccHe .LOHO NOO OOOO.O OOO.OO OOm.OH OHH.O NOO.O OOO.O ONO.H OOH.O OOO.O Omummmum uo pmccHu .umwz OHO . OHH> HHH> HH> H> > >H HHH HH H .02 >+HHHH+HH+HON ODHUOEEOO mo wake UBHO m xmozH AOOOOO OCOOA HOOOO mo OO AOOOHO onOOOHmemHO AHHOO quoHOOOOooo Oszmmx UZHmDBU¢MDZmue HmO OHOO.O HON.OO ONO.NH OO0.0N ONO.H OOm.O OO0.0 NO0.0 NO0.0 muzuchsO HNO OOm0.0 OM0.00 Om0.0H OOH.NH OO0.0 OON.O OOm.m OO0.0 ONN.O .m.m.C .mpoom HmoHuuome ONO OON0.0 OO0.00 NO0.0H OOO.MH NOm.OH OO0.0 OOO.N OOO.N OO0.0 OumCHsome HMUHHHUOHOICOZ H.HH. OOON.O ON0.00 OO0.0H OHO.mH Nm0.0 ON0.0 Hmm.H OO0.0 OOO.H .oum .OCmHuso OOO OOON.O NOm.OO OO0.0H OON.ON HN0.0 OON.N OOO.H OmN.H OON.N Hmmum O couH OHO HOO OOOH.O OOO.HO HN0.0 OOO.O OOH.N HHO.m OHO.H HNN.H OOO.H muosuoua Ouwuuom OOO HOHN.O OOH.OO ONO.HH OOH.NH HMN.m OOm.M HHO.N HMH.H OOO.H mum3mmmHo OOO HoHN.c mmH.OO ONO.HH OOH.NH HMN.m OOm.m HHO.N HMH.H OOO.H mmmHO OOO HHH; HHH> HH> H> > >H HHH HH H .02 >+AHHH+HH+HON OquoEEou wo wmxe UBHm O xuozH Hmoaom OOOOO HOOOO no OO AOOOHO onOOmHOemHO OHHOO HOZOHOOOOOOO qummmx ..P. n u .- 1... On.» . .... a. . ..—\As.n~ .....ck. ... y H .H O . O x- .. .. _ AOOOOHOOOOO ftp-JO ll. :13 1:: OHIHHld mqmdfi ... nu. .C .3 .....v..... ;.#t v.1...v. H— H—.-.v JHOHO‘H—.Or,u ...... a . .r—.a.lf .— O.H..H ~.Hp. a a a.. -..-.v.-u 4v.o\t.¢7/ r r.. J NOTES TO TABLE A-II-lS This table lists the Scale Economy Index (d) as developed by Hufbauer. The scale economies were equated with the exponent in the regression equation, v = knd, where v is the 1963 ratio between the value added in plants employing n persons and average value added for the four- digit U. 8. Census Bureau industries (1966), and k is a constant. Four-digit industries were reclassified accor- ding to the three-digit SITC prior to running the regres- sion analysis. 197 NOO.O .oum .coOH OHO HOOO HH0.0 OOOOOO Oo OOHUHOCO ONOO OO0.0 Oumuuom OOOO HH0.0 Cmnnsm Oo OHOHCOOOO HNO NHH.O 8HOszOHO OOOO OO0.0 umnummq Oo mmusaommscmz NHOO OM0.0 mmmHo OOOO OO0.0- umnummH HHO« Nm0.0 .oum OO0.0 .m.m.C.mHmHnmum2 HOOHEmCO OOO« .mOCHCm>oo Hoon OOOO OOO.O mumNHHHuumm UmnspUOOSCm2 HOm OHO.O- mmHoHuH< OON.O OOHumEmou UCm OHOESMHOm Mmm OHHOxme Os-mumz OOOO OOH.O .opm .mmssO HHO.O moHunmm numm .mHHo HmHqummm HOOO OHHuxme HmHomOO OOOO OO0.0 muosOouO OOO.O- OCOOHoun HOUHOOOOOECOOO O HOcHoHOmz HOO -Em .momH .mHHsB OOO OO0.0 muomuuxm OCHCCOB O OCHOOO Nmm OMO.O- Cm>03 .moHH OO0.0 mmwsummwa UHCmmHO OHumCqum HMm -bmm mHHuxmB H9.30 MOO« ON0.0 mHmoHEmCO Nm0.0- Cm>oz mpsuu OCm HOB HmumCHz HNO .moHHQmm .Couuou NOOO OO0.0- mHmoHEmCU UHCmmHOCH MHm OO0.0 Ommuse O Cnmw mHprmB HOOO OO0.0 mHmoHEmCU oHCmm-Ho NHO OH0.0 Cmmmm .OHOO HO0.0 mumnHO 8 Hmmmm Oo mmHoHOCO NOOO OmpmumcmOmm .oHumOpCOm OONO 9 HOH.O CHOOQHOQOO O Cmmmm HOOOF .m.C ummmm mummz ~mHan HON l OOO.O mmusuommsCm2 xuou MOO OOO.O mmmmum>mm OHHOCoon-Coz HHH OO0.0 m.m.c OHH.O mmHnmumOm> om>ummmum OOOO .OOHHOOUOOOOOE Oooz NOOO HNH.O Ooom Om>ummmum OOOO ON0.0 OOCmom HOO.O OOHO Omummmum can Omccmo NmOO OOOBOHm OmummCm> HMO« ONO.O Hmm2 cmummwum UCm OOCCMO OHOO HOV mmHuHOOEEOO .oz HOV mmeHOOEEou .02 meCH mo CoHuQHHommo OBHO meCH mo COHuQHHommo UBHO OEOCoom OEOCoom mHmom mHmom EMBmMm ZOHBflUHMHmmfiqo UBHm OB GZHDmOUU¢ mmHmBmDDZH OZHmDBU¢MDZ¢E mom mMUHQZH mmHEOZOUM quum m.mmDHmCHCom2 HMUHHuomHm ONOO OO0.0 quEmHCOm OOO.O .m.m.C HmoHuuomHm OHCOmEoo ONO mmHoHpC< OwusuommCsz OOO ONO.O msumumm NmO.O mmumz -m4 mCOHuOOHCCEEoomHmB ONOO O.OOHEOOH0O .OCHOBOO OOOO OOO.O OuHoHCuomHm OcHu OOO.O mmHHOOsO -:nHCumHO COO ucmEOHsOm ONO .NCOOoHOOOO ~moHOOo OOO OOO.O OumcHnomz Cmgom oHCuomHm NNO OO0.0 mOoow OOHOCOOO OON.O m.m.c .mwoe .mHOCMHCCEmHmm OOO« .mmoCMHHmmd O >H®CHC0m2 OHO OO0.0 mmHoHuHm oHCOMHm mOO« OO0.0 mmCHCom2 monmo OHO OMO.O Cmuum2 UmuCHum NOO« OMO.O OHmCHCom2 HMCOCHCUHHOC NHO OMO.O mquECHuwCH HOOHumO OM0.0 OHmCHCom2 OCHCmHmme Hmzom HHOO .Hmonm2 .OHmHquHom HOO OMM.O pCmEmHCvm CHOCmmsom OOO NO0.0 Cmmxuoom HOOO OO0.0 OCmHuso OOOO OO0.0 H5O OO0.0 mCHCom2 no UCmm now mHooe OOOO ummoxm .OCHCuoHO HOO¢ OM0.0 .oum .musz .mBOHom .mHHmz OOO HM0.0 .oum .mOOOC OO0.0 muumm kusuosuum omCmHCHm HOO -Ocmm .mOooo Hm>mCe HmO« OO0.0- mHmpmz msoCCmO-Coz HOO NOO.O musuHcCsO HNOO OOO.O- mOcHOCoO new mOcHummo OOO OOO.O mmnsume OCHCCOHH OH0.0 com OHHB OCHUCHoxm .mqu OOO .OCHummm .OHMCHCOO NHO OOO.O OOQMCO .meOCd .mpom .mumm MOO HOO mmHuHOOEEOO .oz HOO mmHuHooEEOO .oz xmpCH CoHumHnommo UBHO xmeH mo CoHumHuommo UBHO OEOCoom OEOCoom mHmom mHmom AOOOCHpcooO OH-HH-< OHOOO v . c . C I .-. H O O XHWH_..O:... .. ...Hfl... Z c .... : ... .m Ar .. fl “aha, ,. a ...:u.........,. Arm...’ L... .....M......H....,.H.E.M.H 0.03.3) ...HHZ.H. «I... J HOSTS} :7» 12.2 s....-..— l . -,DI ‘nh'vfllfl ~.¢3.3 fly. . H .3 :r...:.:21:> .O NOTES TO TABLE A-II-16 Estimates of the "revealed" comparative advantage index (X) are listed in this table. The revealed comparative advantage index is approximated by the export performance of a developing country in a particular manufactured commo- dity in the market of develOped countries. The export per- formance could be evaluated by (a) comparing the relative share of a country in the world export of individual commo- dities, (b) indicating changes in relative shares over time, and (c) a combination of (a) and (b). The revealed compara- tive advantage index as demonstrated in Table A-II-l6 was developed by combining (a) and (b). This was done by divi- ding a developing country's share in the exports of a given manufactured commodity by its share in the combined exports of manufactured goods of all developing countries to devel- oped countries, and expressing the results in index number form. 200 2(31 OOO.HOO. OOO.NHO HOO.HO - OO0.00 O HON.OOO OOO-a ..suHauOHoO .OuHonuO HOO - OOO.H OOO.OOO. OHO.ONH OOO.NOO. OHN.OOO .Ooou OaH quona . nOOu . cucuaHanI-uom OOO - - OOO.O. OOO.NO. ONO.OHOOO OO0.00NH .oHuHuua UHOOOHO OOO OHN.O HOO.OON HO0.00 - - HOO.HOH Couuux OouuHuO NOO OO0.0 NNO.OOH OOH.OO OOH.OOH. HO0.00N NON.HOO uuoauoou HOO - OO0.0H NH0.00 NON.NH OOH.OOH OH0.0HN OOO uOuuxo OaHOuoHO HOO - NOH.O. OOO.ON - OO0.00. ONO.OONH .uuu .Oxooauaan ..OooO Ho>aua HOO HNO.ON¢ ONH.OH. NON.OON - NOH.HO. NHO.NO «COOHOCOO HNO OOO.HO. OOO.O OOH.OOHOO NOO.HO HOO.OOH.. HOO.NOO .O.o.c .OuoaHnuaz HOOHuquHO ONO - OOO.N NOO.NHN.. NHO.N NNO.OHOH. OOO.OOH .OOOCOOO< OOOHOOOHasaaouoHua ONN OON.OOOOH OHO.NOH OOO.NNH OOO.ONO NOO.NOHN OOO.HNOH OuocHquz OaHuauoaoo uoaoO HHH NHN.OO OOO.OOO OHN.O OHO.HO. - - .uuo .aouH OHO HHO - OOO.ONOO OON.OHHOO - HOO.OO HNO.HHO OuoOuoO OOO - - OHN.OHH - OOO.NNOO OHO.OOO unaunOOHO OOO OOO.OOOO - OOH.ON - NHN.OOO ONN.OO OOOHO OOO - OOO.OOH OHO.O OHO.OON ONN.O OHN.OH .uOo .OOOHCo>oO CooHO COO OON.OH OOO.NOO NOH.H NOH.OOO OOO.HO OOO.OOO OOHUHOC< OHHOxoa Oa-uuqx OOO OON.HOON OOO.N OOO.OOO - OHO.NO HOO.O OUHCOOO OHHOxOO HOHOOOO OOO - OOO.NOHH NOO.O ONO.ONO HOO.O O OON.O co>os .OUHCOOO OHHOxOO COOOO OOO OOO.O HNN.OOO OOH.OO NOO.OHO ONO.OO ONO.HON co>oa .OUHHOOO .coOuoO NOO NOO.HNN HOO.OON OOO.NOO NHO.NOOO ONO.OO OON.NO OOOCOO O cue» OHHOOOO HOO - OOO.OON OON.ONOO. - OHO.OO OOO.OON COOOO .OHOO COOOO Oo OOHUHOCO NOO OOO.OOOO - ONO.O - - OOO.O OCOOOCOOOO O OOOOO HOO - - HHO.OOO - ONN.OOO OOO.ON .O.o.c .OcHusuuaOacmz Oooz NOO HHN.OHN - OOH.OO - ONH.NON - OOCOOO OooaOHO .Ouoo:o> HOO - NOH.NO - - NCO.ONHO. - COOOOO Oo OOHUHOCO ONO OOO.OH OON.OOHO. OOO.OOO HNN.OOO« HOO.OON.O OOH.NNH COOOOOH Oo Ouusuqusauz NHO OHO.OHH OOH.NHO OOO.NN OON.OOON - - COOOOOO HHO OHO.OHN OOH.HON OOO.ON - - - .O.O.O.OHOHCOOO= HOOHaOOO OOO OOO.OON NOO.OHH OOH.OHO - OOO.OO NHO.OO .uOo.O~a=OCOO .OHHO HOHOOOOOO HOO - ON0.0N NOO.N - OO0.00H OHO.HON OCOOHO vmuauucowmm .uHOOOOcOO OON NOO.ON O OOO.O OON.ON OOO.OO OOO.OOOH OOO.N OOHOOOOOO> Oo>COOOCO OOO OOO.OOH OOO.O ONO.HOH - OOO.OHH ONO.O OooO Om>uumoum OOO - OOO.OOON OOO.ONO OON.NOO OOO.O .O NHO.OH OOHO OOOOOOCO OOO Oucauo NOO NOO.HOH - - HON.H - OOO.O OOO: OOCOOOCO OOO OOOOOO OHO OHNOOO «HOZH OOHsz zOOOHOOO zOOHOH OzOO OOO; OOHHHOOzzOO OO OOHOOHOOOOO .Oz OHHO x. OOH I HOIOOOH .AwOIOOOHIIHoIOOOHV mmHMHZDOU qumOHm>ma QmHum-Hmm mom MMO§HHmm NmozH mUQ< m>HH<~Hmd.. O OHIHHId mqmdfi 202 lumum mpmne COHOHOO: a -5 Q aqua-....IN .nu-Mw o ”v .5“! O N\-.?.-- R-AO- .n'oi I- ,.uf.:- NH...P.\ l \f—.n..l 13.5.}..1 HHO.O-.NC COH.I—i \.ON.:A.N n.f\u.HCC 32¢.O «a..—.Cn-HCC | ..3.\ N31..~HN-CC nON.u:O.uH ......NrH 1?»..qu d ...N film. 3‘. -:.\—C SIN O— OH: AC 5 H$.N 4 .sw . .H> .MOINOOH mH .NOIHOOH mH O O ..H.>N + ...>W .H> ...> "OH meCH mCu COM Cam: mHCEHom OCH =.O mmHme HOmsCHucooO OH-HH-O mHmOO -...... ...n. d, v 0‘.- !‘AI-IUII\ sch-taut .v.~‘.3.-H :—3.f—~. Ava-a .‘f—C 4.5,.nuo-N.‘ \HvO.—ZC NHJ.N: HHO.O.aOosHC. fi~¢.:Ho~HC «...: . nvnad NAOP . H.9H N OOO.O.HOA-d -I)I--'l at . "v..- I 'Jhinundfi I' - uuuul o '- ... U l". ’ I .o..’ ..- . Ini'yv“ ’5- ‘ .p .. - tu'aa a 'odAdt.‘ o .9 flII\/l- l-fl. ..u-qduanuuaah— Phanud‘aG': anOA h a 1-4-3 In: vauiinuaa‘ In—:d adv Jan-145! .I.I.= - Pablvdadu. >h8p—«pa.alx ISO wa.b95....a V..-33.! ...ui .r—Shn 5H5 >n3ud...’ Ham» mmmb mCO «« umm» mmmn OCH « d -..le-W- N "x -+ .H.> H .mOH mH ~Qumo =.Q mmHHmm .moHumHumum mcmue OCHOOEEOO= .p . ..z.D "mwonsom BIBLIOGRAPHY BI BL IOGRAP HY Abraham, w. I. National Income and Economic Accounting, Engle- wood Cliffs, N. J.: Prentice-Hall, Inc., 1969. "A Program for Growth," Capital, Output and Employment, 1948- a 1960, The Department of Applied Economics, University 33 of Cambridge, Chapman and Hall, April, 1964. Balassa, Bela A. "The Capital Needs of the Developing Coun- tries," Kyklos, 1965, Vol. 2. . The Kennedy Round, Estimated Effects on Tariff 35 Barriers, Part II, New York, UNCTAD, 1968. . "Structure of Protection in the Industrial Coun- tries and its Effects on Processed Goods from Developing Nations," Programs for Liberalization and Expansion of Trade in Manufactures and Semi-manufactures of Interest to Developing Countries, UNCTAD, TD/B/C,2/36, New York: U. N., 1967. . "Tariff Protection in Industrial Countries: An Evaluation," Journal of Political Economy, December, 1965. . "Trade Liberalization and 'Revealed' Comparative Advantage," The Manchester School Economic Papers, May, 1965. . Trade Prospects for Developing Countries, Economic Growth Center, Yale University, Homewood, 111.: Richard D. Irwin, 1964. and Schdlowsky, Daniel M. "Effective Tariffs, Domes- tic Cost of Foreign Exchange and the Equilibrium Exchange Rate," Journal of Political Economy, May, 1968. Baldwin, R. E. Determinants of the Commodity Structure of U. S. Trade, (mimeographed), Social Systems Research Institute, University of Wisconsin, 1969. Banco De Mexico. Mexican Input-Output Tables, 1960, Mexico, 1966. 203 204 Basevi, Giorgio. "The United States Industries and Indus— trial Labor," Review of Economics and Statistics, May, 1966. Bhagwati, Jagdish. "The Pure Theory of International Trade: A Survey," Economic Journal, Vol. 74, 1964; reprinted in Survey of Economic Theory, Vol. 2, New York: Macmillan, 1965. Bharadwag, R. and Bhagwati, J. "Human Capital and Pattern of Foreign Trade: The Indian Case," Indian Economic Review, October, 1967. British Input-Output Relationships, 1954-66, A Program for Growth, No. 3, Department of Applied Economics, Univer- sity of Cambridge, Supervised by Richard Stone, Chapman and Hall, 1963. m.lwr.m. n 1': rra‘lm—I“Wfll . . H. . V 'O . ’ I ‘ I .; Bruno, Michael. Interdependence, Resource Use and Structural Change in Trade, Jerusalem: Bank of Israel, 1963. "The Optimal Selection of Export Promoting and Import Substituting Projects," Planning the External Sector, Techniques, Problems and Policies, New York: U. N., 1967. Caves, Richard E. "Comments on Professor Johnson's Paper: I," International Economic Relations, Paul E. Samuelson (editor). Chenery, Hollis B. "Comparative Advantage and Development Policy," Survey of Economic Theory, Vol. II, St. Martin's Press, 1965. and Strout, Alan M. "Foreign Assistance and Econo- mic Development," American Economic Review, September, 1966. . "Reply," American Economic Review, September, 1968. Chipman, J. S. "A Survey of the Theory of International Trade, Parts I, II," Econometrica, Vol. 33, 1965. Clement, M. O., Pfister, Richard L. and Rothwell, Kenneth J. Theoretical Issues in International Economics, Boston: Houghton Mifflin Company, 1967. Corden, W. M. "Patterns of Industrial Growth," American Econo- mic Review, 1960. . Recent Developments in the Theory of International Trade, Special Paper in International EconomiCs No. 7, International Finance Section, Princeton University, 1965. 205 . "The Structure of a Tariff System and the Ef- fective Protective Rate," Journal of Political Economy, June, 1966. Council for Industrial Planning. Capital Structure of the Japanese Economy, Tokyo, 1958. . Japan's Labor Yearbook, 1959. Council for International Economic Cooperation and Develop- ment, Executive Yaun, Taiwan's Interindustry Inverse Matrix Table for 1961, Taipeih, 1964. Creamer, D., Dobrovolsky, S. P. and Borenstein, I. Capital in Manufacturing and Mining, Its Formation and Financing, Princeton, N. J.: Princeton University Press, 1960. Diaz-Alejandro, Carlos F. Exchange Rate Devaluation in A Semi-Industrial Country, Cambridge, Mass.: M.I.T. Press, 1965. Economic Commission for Asia and the Far East. Economic Sur- vey of Asia and the Far East, "Appendix to Chapter 1: Factors in the Asian Export Performance," Bangkok, 1967. Economic Research Institute, National Income Accounts, Economic Planning Agency, Japan, 1957. . National Wealth Survey, 1955. . Economic Bulletin No. 1, February, 1959. Fei, John C. H. and Paauw, Douglas. "Foreign Assistance and Self Help: A Reappraisal of Development Finance," 33f view of Economics and Statistics, 1964. Felix, David. "The Dilemma of Import Substitution--Argentina," Development Policy, Theory and Practice, Gustaf F. Papanek, editor, Harvard University Press, 1968. Fuchs, V. "Capital-Labor Substitution: A Note," Review of Economics and Statistics, Vol. 45, November, 1963. Glassburner, Bruce. "Aspects of the Problems of Foreign Exchange Pricing in Pakistan," Economic Development and Cultural Chagge, June, 1968. Gruber, W., Mehta, D. and Vernon, R. "The R & D Factor in International Investment of United States Industries," Journal of Political Economy, Vol. 75, February, 1967. Guisinger, Stephen. "Effective Protection and the Concept of Negative Value Added," Quarterly Journal of Economics, November, 1969. 206 Hagen, E. E. Economics of Development, Homewood, Ill.: Richard D. Irwin, 1968. . "Economic Justification for Protectionism," Quarterly Journal of Economics, Vol. 72, November, 1968. Harrod, R. F. "Factor-Price Relations Under Free Trade," Economic Journal, June, 1958. Hirsch, Seev. Location of Industry and International Com- petitiveness, Oxford: Clarendon Press, 1967. Hirschman, Albert. The Strategy of Economic Development, New Haven, 1958. . "Political Economy of Import Substituting Indus- trialization," Quarterly Journal of Economics, Septem- ber, 1968. Horrowitz, M. A., Zymelman, M. and Herrenstadt, I. L. Man- Power Requirements for Planning: An International Com- pariSon Approach, Boston: Northeastern University Press, 1966. Hufbauer, G. C. "The Commodity Composition of Trade in Manufactured Goods," The Technology Factor in Trade, Raymond Vernon, editor, New York: Columbia University Press, 1970. . Synthetic Materials and the Theory of International Trade, London: Duckworth, 1965. Islam, Nurul. "Comparative Costs, Factor Proportions, and Industrial Efficiency in Pakistan," Pakistan Develgpment Review, Summer, 1967. Johnson, Harry G. Comparative Cost and Commercial Policy Theory for a Developing World Economy, Wicksell Lectures, 1968, Stockholm: Almquist & Wicksell, 1968. . Economic Policies Toward Less Developed Countries, New York: Praeger, 1967. . "The Theory of Tariff Structure with Special Ref- erence to World Trade and Development," Trade and Devel- opment, Geneva: Librarie Droz, 1965. . "Trade Preferences and Developing Countries," Lloyds Bank Review, Vol. LXXX, April, 1966. . "The Theory of International Trade,‘ International Economic Relations, Paul E. Samuelson, editor, New York: Macmillan, 1969. 207 Jones, Ronald. "Factor-Proportions and the Heckscher-Ohlin Theorem," Review of Economic Studies, Vol. 24, 1956-57. Keesing, Donald B. "The Impact of Research and Development on United States Trade," Journal of Political Economy, Vol. 75, February, 1967. . "Labor Skills and International Trade: Measuring Many Trade Flows With a Single Measuring Device," Review of Economics and Statistics, February, 1959. . "Labor Skills and the Structure of Trade in Manu- factures," The Open Economy. Kenen, P. B. "Nature, Capital & Trade," Journal of Political Economy, October, 1965. Kendrick, John W. Productivity Trends: Capital and Labor, National Bureau of Economic Research, New York: C61umbia University Press, 1956. . Industrial Composition of Income and Products, Brookings Institution and National Bureau of Economic Research, New York: Columbia University Press, 1968. Kemp, Murray C. The Pure Theory of International Trade and Investment, Prentice-Hall, 1969. Kravis, Irvin. "Availability and Other Influences on the Commodity Composition of Trade," Journal of Political Economy, Vol. 64, April, 1956. . "Wage and Foreign Trade," Review of Economics and Statistics, February, 1959. Kreinin, Mordechai. "Comment" on Hufbauer's Paper, R. Vernon, editor, The Technology Factor in International Trade, Universities--Nationa1 Bureau of Economic Research Conference Series No. 22, New York: Columbia University Press, 1970. . "On the Dynamic Effects of Customs Union," Journal of Political Economy, April, 1964. . "On the Restrictive Effect of the Tariff-—A Note on the Use of the Balassa Index," The Manchester School Economic Papers, January, 1966. Krueger, Ann 0. "Some Economic Costs of Exchange Control: The Turkish Case," Journal of Political Economy, October, 1966. 208 Lal, Deepak. Foreign Exchange Bottlenecksy Balance of Pay- ments, Investment Criteria and the Optimum Pattern of Trade and Production, Oxford University, January, 1967, (mimeographed). Lary, Hal B. Imports of Manufactures from Less Developed Countries, National Bureau of Economic Research, New York: Columbia University Press, 1968. Leamer, E. E. and Stern, R. M. Quantitative International Economics, Boston: Allyn & Bacon, Inc., 1970. Leith, Clark J. "Substitution and Supply Elasticities in Calculating the Effective Protective Rate," Quarterly Journal of Economics, October, 1968. ; Leontief, W. W. "Domestic Production and Foreign Trade: 5 The American Capital Position Re-examined," Proceedingg of the American Philosophical Society, Vol. 97, 1953. Input-Output Analysis, Cambridge: Cambridge University Press, 1966. ________ "An International Comparison of Factor Costs and Factor Use," American Economic Review, Vol. 54, June, 1964. Lewis, Stephen R. and Guisinger, Stephen E. "Measuring Pro- tection in a Developing Country: The Case of Pakistan," Journal of Political Economy, 1968. Liesner, H. H. "The European Common Market and British Industry," Economic Journal, June, 1958. Linder, Stephan B. An Essay on Trade and Transformation, New York: John Wiley & Sons, 1961. Little, 1. M. D. and Clifford, J. M. International Aid, Chicago: Aldine Publishing Company, 1965. and Mirrless, J. A. Manual of Industrial Prgject Agalysis in Developing Countries, Vol. II, OECD, Paris, 1969. Maizels, A. Industrial Growth and World Trade, Cambridge: Cambridge University Press, 1963. Manne, A. S. and Rudra, A. et. al. "A Consistency Model of India's Fourth Plan," M.I.T. Center for International Studies, Sankhya, Series B, September, 1965. Mikesell, R. F. Economics of Foreign Aid, Chicago: Aldine Publishing Company, 1968. 209 Minhas, B. S. "The Homohypallagic Production Function, Factor-Intensity Reversals and the Heckscher-Ohlin Theorem," Journal of Political Economy, Vol. 72, 1962. Mynt, Hla. "International Trade and the Developing Countries," International Economic Relations,Pau1 A. Samuelson, editor, International Economic Association, New York: Macmillan, 1969. Myrdal, Gunner. An International Economy, Problems and Prospects, New York: Harper and Brothers, 1956. Nurkse, Ragnar. Problems of Capital Formation in Under- developed Countries and Patterns of Trade and Develop- ment, A Galaxy Book, New York: Oxford University Press, 1967. Ohlin, Bertil. Inter-regional and International Trade, Har- vard University Economic Studies, Vol. 39, Cambridge, Mass.: Harvard University Press, 1933. Prebisch, Raul. Toward A New Trade Policy for Development. Proceedings of UNCTAD, New York: 1964, Vol. II. Ranis, Gustaf and Fei, J. C. H. ”Comment," American Economic Review, September, 1968. Robinson, Romney. "Factor Proportions and Comparative Advan- tage, Part I," Quarterly Journal of Economics, Vol. 60, 1956. Rosenstein-Rodan, P. N. "International Aid for Underdeveloped Countries, Review of Economics and Statistics, 1961. Roskamp, Karl W. and McMeekin, Gordon C. "Factor Proportions, Human Capital and Foreign Trade, The Case of West Germany Reconsidered," gparterly Journal of Economics, February, 1 1968. Soligo, Ronald and Stern, J. J. "Tariff Protection, Import_ Substitution and Investment Efficiency," Pakistan Devel- Opment Review, Vol. 5, 1965. Survey of Current Business, "Total Requirements (Inverse) Input-Output Matrix, U. S. Economy, 1963," November, 1969. Swerling, Boris. Current Issues in Commodity Poligy, Essays in International Finance, No. 38, Princeton University Press, 1962. Tims, Wouter. Input-Output Table for Pakistan, 1963-64, (in current prices), Mimeographed, Harvard Development Advisory Service and Planning Commission, Government of Pakistan, 1965. 210 U. N. The Outlook for Debt Service, TD/97, Vol. IV, New York, 1968. . A Study of Industrial Growth, New York, 1963. . UNCTAD Secretariat, Trade Progpects and Capital Needs of Developing Countries, TD/34/Rev. 1, New York, 1968. . UNCTAD, Second Session, Commodity Problems and Policies, Vol. II, 1968, New York. . United Nations World Economic Survey, Vol. 1, New York, 1963. . Trade and Development: Trends, Needs and Poli- cies, New York, 1964. . Trade in Manufactures, Vol. IV, New York, 1964. Vernon, Raymond. "International Investment and International Trade in the Product Cycle," Quarterly Journal of Economics, May, 1966. . "Problems and Prospects in the Exports of Manu- factured Goods from the Less Developed Countries," Trade in Manufactures, New York: U. N., 1964. Waehrer, Helen. "Wage Rates, Labor Skills, and United States Foreign Trade," The Open Economy: Essays on International Trade and Finance, Columbia Studies in Economics, Vol. I, P. B. Kenen and R. Lawrence, editors, New York: Columbia University Press, 1968. Watanabe, T. "Approaches to the Problems of Inter—Country Comparison of Input-Output Relations: A Survey and Suggestions," International Comparisons of Inter-Industry Data, U. N. Industrial Planning and Programming Series No. 1), New York: U. N., 1969. Weintraub, Sidney. Trade Preferences for Less-Developed Countries, An Analysis of U. S. Policy, Praeger Special Studies in International Economics and Development, New York: Praeger, 1966. Yahr, M. Estimating the Elasticity of Substitution from International Manufacturing Census Data, Ph.D. Thesis, (unpublished), Columbia University, 1967. . "Human Capital and Factor Substitution in C.E.S. Production Function," The Open Economy, Yudin, E. B. and Kenen, P. B. Skills, Human Capital and the U. S. Foreign Trade, International Economics Workshop Paper, Columbia University, New York, 1965 (mimeographed). 211 Zymelman, M. "Productivity, Skills and Education in Manu- facturing Industries," Planning for Advanced Skills and Technology, Industrial Planning and Programming Series No. 3, New York: U. N., 1969. MICHIGAN STATE UNIV. II 2IIISII III II IIII327 LIBRRRI ES