llllill'llllllllllllll'llllllllllllflllllllllll 1 3 1293 00569 1906 r LIBRARY Michigan State University This is to certify that the dissertation entitled CAPITAL MARKET DISTORTIONS & DEVELOPMENT OF SMALL SCALE ENTERPRISE IN KOREA WITH REFERENCE TO TAIWAN presented by BOO CHUN SU has been accepted towards fulfillment of the requirements for Ph.D . Economics degree mid Major professor Dateéfu ! l1 57 [ 9f5/ Man“... ”.1- o' ‘ ' . "“ ' ' ' ‘ 0-12771 MSU RETURNING MATERIALS: Place in book drop to LIBRARIES remove this checkout from V your record. FINES will be charged if book is returned after the date stamped below. ”v -.r. (“:9 1 Will) 61992 CAPI'I OF 11'] pa: OF SMALL SCALE ENTERPRISE IN KOREA I i CAPITAL MARKET DISTORTIONS & DEVELOPMENT l ‘ WITH REFERENCE TO TAIWAN l I l BY ! Boo-Chun Su l A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Economics 1988 This t] Korea and Ta: over the last economic grol newly industr: resource EDdOl industrial st; Portion of th. dominate impoj differences 5.: This t “Pital marke policY 0f f study Contin- market disto of Slall firm We °°ml>are t Ferromance o The Kor Presumed hen world export ABSTRACT CAPITAL MARKET DISTORTIONS & DEVELOPMENT OF SMALL SCALE ENTERPRISES IN KOREA WITH REFERENCE TO TAIWAN BY Boo-Chun Su This thesis focussed on the contrasting record of Korea and Taiwan towards small scale enterprises (SSEs). Over the last two decades, both have experienced rapid economic growth, much more rapid than that of the other newly industrialized countries. Further, they share a poor resource endowment, but there are differences in their industrial structure. In Taiwan, SSEs contribute a major portion of the GDP. In Korea,on the other hand,large firms dominate important sections of the economy. These reflect differences in their respective economic policies. This thesis examines the nature of the Korean capital market distortions and the effects of the related policy of favoring large scale enterprises (LSEs). The study continues to examine the effect of this capital market distortion on the efficiency and performance of small firms by looking at total factor productivity. We compare this record with the industrial policies and performance of Taiwan with regard to SSEs. The Korean government favored LSEs because of their presumed benefits and the resultant competitive edge in world export markets. The lower interest rates and overvaluation the capital were unfairly status decline the Taiwanese 58135. There we SSEs fared bet Korea no 0f their grea egalitarian i] to expand its t0 growing r other LDCs. overvaluation of Korean currency(Won) resulted in reducing the capital cost of LSEs relative to SSEs. As the latter were unfairly treated in the allocation of capital, their status declined in terms of employment. On the other hand, the Taiwanese policy was more balanced between LSEs and SSEs. There was less distortion in the capital market and SSEs fared better. Korea needs to encourage the growth of SSEs, because of their greater potential for employment creation and egalitarian income distribution. However, Taiwan may need to expand its capital-intensive industrial structure, due to growing competition in labor—intensive products from other LDCs. This di This dissertation is dedicated to my mother. - NAMU MYRUK JONBUL GUWANSEUM BOSAL - I like encouragement education wit brothers , neg I an hor he was my advz‘ He gave me 5 one. I than] insight guide< I earn helpful in Sh; The sou] revising Procl Finally during th e on HEY, : ACKNOWLEDGMENT I like to give most credits to my mother. Her encouragement and financial support made me finish my education without agony. Also my lovely sisters , brothers , nephews and niece gave a lot of moral support. I am honored to know Professor Kannappan in person; he was my advisor and has a generous and warm personality. He gave me an exciting human experience beyond academic one. I thank to Professor Liedholm whose intelligent insight guided me safely through the empirical study. I appreciate to Professor Dinopoulos who was very helpful in shaping the model of my thesis. The sound advices by Professor Koo were helpful in revising process of my thesis. Finally I like to thank to everyone who helped me during the ordeal. Hey, Mom. I did it! List of Table: List of Figure Definitions CHAPTER I. IN'. A. Each B. Majo: 6. Obje. 1. T K 2. E- T D- The 1. T 2- E E~ The CHAPTER II. 1 1" Int: B. DUal TABLE OF CONTENTS List of Tables . . . . . . . . . . . . . . . . . List of Figures . . . . . . . . . . . . . . . . Definitions . . . . . . . . . . . . . . . . CHAPTER I. INTRODUCTION A. Background . . . . . . . . . . . . . . B. Major Theme & Hypothesis of Thesis . . . . . 0 Objective of Thesis . . . . . . . . . . . . 1. Theoretical Macro Analysis of . . . . Korean SSEs 2. Empirical Micro Study of Korean and . . . Taiwan SSEs D. The Theoretical and Empirical Approach 1. Theoretical Model . . . . . . . . . . . . 2. Empirical study . . . . . . . . . . . . . E. The Organization of the Dissertation . CHAPTER II. LITERATURE REVIEW FOR THREE SECTORS (ULS, USS, R SECTORS) MODEL A. Introduction . . . . . . . . . . . . . . . . B. Dual Sector . . . . . . . . . . . . . . 1. Introduction. . . . . . . . . . . 2-a. Classical Dual Sector Model 2-b. Extension of Classical Model. 3-a. Todaro’s Dual Sector Model. . . . 3-b. Modification of Todaro’s Model. 4. Conclusion. . . . . . . . . . . . . Informal Sector . . . . . . . . . 1. Introduction. . . . 2. Definition. . . . . . . . . 3. Interaction between the Informal. and Formal Sectors 4. Dynamic Future of the Informal Sector 5. Policy for the Informal Sector. . . 6. Conclusion. . . . . . . . . . Rural Non-farming Activities in LDCs 1. Introduction. . . . . 2. The Pessimistic View of Non—farming Activity 3. Skepticism about H—R Study. 4. The Optimistic View of Non-farming. Activity 5. A Possible Modification of H-R’s. Model 6. Conclusion. . . . . . . . . . Small Scale Enterprises . . 1. Introduction. . . . . . . . 2. Definition. . . . . . . 10 12 13 15 17 18 l8 18 21 24 26 27 27 27 28 28 3O 31 32 33 33 33 N“ M" luv—“4...”... ; CHAPTER III. 1 A. Intr 3. Mode 3. Characteristics of SSEs . . . . a. Location . . . . . . . . . . b. Demand for Outputs of SSEs 4. Labor in SSEs . . . . . . . 5. Capital in SSEs . . . . . . . . . . m Efficiency of SSEs. . . . . . a. Partial Efficiency Measurement b. Comprehensive measurement. 7. Conclusion. . . . . . . . . . CHAPTER III. MODEL DEVELOPMENT WITH THREE SECTOR 0 LABOR MARKET & THREE DIMENSIONAL PRODUCTION POSSIBILITY FRONTIER Introduction . . . . . Model with Labor Market. . . 1. Assumptions . . . . . . . . . 2. Mathematical Framework. . . . . . . . 3. Geometric Framework . . . . . . . . 4. Limitations of the Model. . . Model with Three Dimensional PPFs. 1. Assumptions . . . 2. Mathematical Framework. . . . 3. Geometric Framework . . . . . Policy Implication . . . . . . . . . . 1. Tariff Policy in the ULS Sector 2- Capital Subsidy Policy in the ULS Sector. vi 34 34 34 36 36 37 37 38 39 41 41 41 42 43 45 45 45 46 46 48 48 57 59%“; . . .- 3. Taril 4. Capi1 5. Mini] 6. Tari: the : CHAPTER IV. A RE‘ IN' TH SE A. Introdu B. Three 3 1. Assu 2. Math 3} Geom 4. Rati in t 3. Tariff Policy in the USS Sector . . . 4. Capital Subsidy Policy in the USS Sector. 71 5. Minimum Wage Policy in the ULS Sector . . 76 6. Tariff Policy in the ULS Sector in. . . . 80 the Situation Where Homogeneous Products Are Produced by the ULS and USS Sectors CHAPTER IV. A REVIEW OF CHANGES IN THE KOREAN A. B. INDUSTRIAL STRUCTURE IN TERMS OF A THREE SECTOR MODEL & AN EMPIRICAL TIME- SERIES STUDY DURING 1966-1984 Introduction . . . . . ; . . . . . . . . . . _88 Three Sector Model . . . . . . . . . . . . . 89 1. Assumptions . . . . . . . . . . . . . . . 89 2. Mathematical Framework . . . . . . . . . 89 3; Geometric Framework . . . . . . . . . . . 9O 4. Rationale of the Three Sector Model . . . 92 in the Korean Case The Main Character of Korean Industrial. . . 94 Policy during 1966-1984 A Comparative Static Analysis for Capital. . 96 Inflow into the ULS Sector A Time-Series Analysis of the Korean . . . . 99 Industrial Structure & Change over 1966-1984 1. The Establishment & Employment in the . . 99 LSE vs. the SSE vii 2. Remu: in t2 & Ag: 3. Real 4. Valu of L F. Some Em of Taiw G. Conclus CHAPTER V. SMALL A. Introdu 3. Methods C. The Sta Korean 1. The 2. Stru in u 3° Spat Mann 4' Time SSE: 5- Effj 6. Char 7' The int 2. Remuneration & the Average Nominal Wage . 103 in the LSE vs. the SSE & Agricultural Household incomes 3. Real Wage in the Three Sectors. . . . . . 107 .5 Value Added, Rent and Capital Intensity . 113 of LSE and SSE r F. Some Empirical Findings in the Case . . . . 119 of Taiwan G. Conclusion . . . . . . . . . . . . . . . . . 120 CHAPTER V. SMALL SCALE ENTERPRISE IN KOREA A. Introduction . . . . . . . . . . . . . . . . 122 B. Methodology of Research . . . . . . . . . . 123 C. The Status, Location, and Efficiency of. . . 124 Korean SSE l. The Share of VSSEs in Manufacturing . . . 124 2. Structural Characteristics of SSEs . . . 126 in Manufacturing Industry 3. Spatial Distribution of SSEs in . . . . . 128 Manufacturing Industry 4. Time-series Relative Position of the . . 135 SSEs in Manufacturing Industry 5. Efficient Plant Size . . . . . . . . . . 137 6. Change in_the Competitive Bases of SSEs . 143 7. The Factors Influencing Competitiveness 145 in the SSEs viii 8. Effi Inte D. Conclus CHAPTER VI. THE SSE A. Introdu B.Iarge s 1. SSES 2. Kore D. 8. Efficiency Relative to Capital. Intensity in SSEs Conclusion . . . . . . . . . . . . . . CHAPTER VI. THE ISSUES & POLICY IMPLICATIONS FOR V SSES IN KOREA Introduction . . . . . . . . . . . . Large Scale Oriented Industrial Structure. 1. SSES in Korean Industry . . . . 2. Korean Industrial Policy. . . . . . a Capital Market . . . . . . . b. Labor Market . . . . . . . . . c. Product Market . . . . . . Urban Concentration of SSES. . . . . Relative Inefficiency of SSES in Small Scale Industry 1. HO’s Method . . . . . . . 2. Social Benefit-Cost Analysis. 3. Technical & Pecuniary Economies of Scale. General Policy toward SSES for Entry. Expansion 1. Entrepreneurship Promotion. 2 Fair Access to Capital . . 3. Information Service . . . . . .p Competitive Relations with LSE U1 Complementary Relations with LSE. . . ix 150 154 158 158 161 164 165 168 172 173 181 187 189 189 190 191 193 “a’}' ,, a.E b.€ F. Conclus CHAPTER VII. A I A. Introdt B. A Comp: Korea a C. Early 1 D. Taiwan 1. Feel a. Scitovsky’s External Economy b. Subcontracting . . . . . . . . . F. Conclusion . . . . . . . . . . . . . . . CHAPTER VII. A REVIEW OF TAIWAN’S SSES A. Introduction . . . . . . . . . . . . . B. A Comparison of Social Indicators for. Korea and Taiwan C. Early Economic Background for Taiwan’s SSE . 199 D. Taiwan’s Industrial Policy . . . . . . . . . 202 1. Factor Market . . . . . . . . . . . . . . 202 a. Capital Market . . . . . . . . . . . . 203 b. Labor Market . . . . . . . . . . . . . 203 2. Product Market . . . . . . . . . . . . . 205 a. Export-Import Policy . . . . . . . . . 205 b. Agricultural Policy . . . . . . . . . 206 c. Capital Formation Policy . . . . . . . 207 d. Industrial Strategy and Innovations. . 208 E. SSEs in Taiwan's Industry . . . . . . . . . 208 1. Small Scale Dominating Industrial . . . . 208 Structure 2. Decentralization of SSEs . . . . . . . . 214 F. Efficiency of SSES in Taiwan . . . . . . . . 219 1. Characteristics of SSEs in Taiwan . . . . 219 2. Partial Efficiency Measurement of SSES. . 221 G. TFP'of SSEs in Taiwan . . . . . . . . . . ._ 226 x 1. HO‘ 2. 80¢ . H. Conch CHAPTER VIII. I A. Gener; B. Summa C. The M APPENDICES Appendix Appendix BIBLIOGRAPHY 1. Ho’s Method . . . . . . . . . 2. Social Benefit-Cost Analysis. . . .H. Conclusion . . . . . . . . . . . . . . CHAPTER VIII. CONCLUSION A. General Summary . . . . B. Summary of Each Chapter . C. The Major Lessons from the Study . . APPENDICES Appendix A.Korean Capital Market Distortion Appendix B.Taiwan’s Capital Market Distortion BIBLIOGRAPHY . . . . . . . . . . . . . xi 226 233 238 240 241 246 249 253 257 Table Number L1 L2 43 L4 L5 L6 L7 L8 t9 A10 All A12 A13 i1 12 i3 i4 i5.1 i5.2 is Number Korear Employ Nomina Major Whole: Real 1 Real . Non-f Compa Value Rent Capit Labor The p Mfg, ssnls SSE'S LIST OF TABLES Table Number Page 4.1 Numbers of Establishments in Korean Mfg. 4.2 Korean Employment Survey . . . . . . . 4.3 Employee’s Nominal Remuneration . . 4.4 Nominal Average Wage per Worker 4.5 Major Indicators of Farm Household 4.6 Wholesale Price Indexes . . 4.7 Real Wage based on Non-food Price. . 4.8 Real Ag. Income per capita based on. Non-food Price 4.9 Comparison of Rural & Urban Income 4.10 Value Added Survey in Korean Mfg. 4.11 Rent in Korean Mfg. . . . 4.12 Capital Intensity in Korean Mfg. 4.13 Labor Productivity & Real Wage in LSE & SSE. 5.1 The Relative Position of SSEs in Mfg.. 5.2 Mfg. Employment & its Percentage in Industry 5.3 SSE’s Employment Distributed by Industry 5.4 SSE’s Value Added Distributed by Industry. 5-5-1 Mfg. Employment Distribution by Location 5~5r2 Mfg. Value Added Distributed by Location 5.6 Distribution of Mfg. Emploment by. Establishment Size in Percentage xii No. 100 101 104 105 106 108 109 111 112 114 115 117 118 125 127 129 130 133 134 136 %.e%, ’ H8 Efficil ’ Survi‘ 'Survi' Size f Share and Va Relati in whi Korean Establ Worker 5‘10'1 Output of 4-c 5'10‘2 Output Indust Predor 6‘1 Korea] 6'2 Polio} in La: Hz‘l Mfg. j of Es‘ 6'2‘2 Mfg, j by s 6'3‘1 Ten 0 in Wh . 6.3.2 Relat 5.10-1 5.10-2 Efficient Plant Size according to. ’Survivor Criterion’ ’Survivor’ Estimates of Efficient Plant. Size for Korean Mfg. Industries: 1979-84 Share of Small Establishments in Employment. and Value Added Relative Importance of Types of Industry in which Small Establishments Predominate Korean Industries in which Small Establishments Employed 50% or more of the Workers in the Particular Industry in 1984 Output/Capital & Capital/Labor Ratios. . of 4-digit Industry Output/Capital & Capital/Labor Ratio for Industries in which Small Establishments Predominate, by Industry Korean Employment in Mfg. Industry Policy-induced Capital Cost Distortion in Large and Small Scale Enterprises(Korea) Mfg. Establishments by Location & by Size. of Establishment, Korea. 1979 & 1984 Mfg. Employment by Location and. by Size of Establishment TFP of Establishments for Industries in which Small Establishments Predominate Relative Efficiency of Establishments. xiii 139 140 144 147 148 151 155 160 163 169 176 178 7.1-3 12 7.3.1 7.3.2 14 7.5 7.7 for In! Predom Number Create Benefi for In Predom Number is Gre Averag GINI ] Income Polio} in IS] The R4 Share Emploj Share Value Numbe Non~m Emplo NUmbe Distr Numbe . arui c for Industries in which Small Establishments Predominate Number of Industries where TFP is. Greatest in the Size Category Indicated Benefit-Cost Ratio of Establishments for Industries in which Small Establishments Predominate Number of Industries where Benefit-Cost. is Greatest in the Size Category Indicated Average Annual Growth Rates in Real Terms GINI Index of Inequality of. . Income Distribution Policy-induced Capital Cost Distortion in LSE and SSE (Korea and Taiwan) The Relative Position of SSES in Mfg. Share of SSE with 1-99 Workers in. Employment by Four-digit Industries Share of SSE with 1—99 Workers in. Value added by Four-digit Industries Number of Enterprises in Selected. Non-manufacturing Industries , by Size Employment of Mfg., distributed by Location. Number of Establishments in Mfg. Distributed by Location Number of Enterprises Unit of Industry and Commerce by Locality & Industry 180 183 186 200 204 211 213 213 215 217 217 218 "peerage... .. ,- .. 18 L9 110 L11 L12 113 114 L15 116 L17 Number of Mfg Number Rural. Employ Small Distri Output in Sma Output for In (TFP of 'in whi Number in the Benefj for Ir Number Ratio Number of Enterprises Units. . . . of Mfg. by Locality Number of Major Enterprises of Mfg. in Rural Area Employment in Mfg. by Industry . . . . .'. Small Scale Mfg.: Employment & Value Added Distributed by Industry - Taiwan Output/Capital & Capital/Labor Ratios. in Small Establishments of Major Industry Output/Capital & Capital/Labor Ratios. for Industries in which SSEs Predominate ‘TFP of Establishments for Industries 'in which SSEs Predominate Number of Industries where TFP is Greatest in the Size Category Indicated Benefit-Cost Ratio of Establishments for Industries in which SSES Predominate Number of Industries where Benefit- Cost Ratio is Greatest in the Size Category XV 218 220 222 223 225 227 230 232 235 237 Wk"; 1'" .. . M Figure Number L1 L1 ll 12 L3 L4 L5 L6 L7 L8 L9 llo Large Capit Consu betwe Labor in Th Three Labor Polic Effec in TI Effe< by Ta Labor Capil Effe< Sect. Effel Capri Lab0: Tari Effe LIST OF FIGURES Figure Number 1.1 Large Firm Favoring Policy: 0 v 0 Capital Subsidy Case Consumption Possibilities . . . between Z and M Labor Market Equilibrium. . . . . . . in Three Sectors(ULS,USS, R) Three Dimension PPF at Equilibrium . Labor Market Adjustment after Tariff. Policy on the ULS Sector Effects of Tariff on ULS Sector Product in Three Dimensional PPFs Model Effects on Sectoral Output . . . . by Tariff on ULS Sector Labor Market Adjustment after . . . . Capital Subsidy Policy on ULS Sector Effects of Capital Inflow on ULS. Sector in Three Dimensional PPFs Model Effects on Sectoral Output by . Capital Inflow into ULS Sector Labor Market Adjustment after Tariff Policy on the USS Sector Effects of Tariff Policy on USS Sector. xvi Page No. 31 44 47 50 53 55 59 62 64 67 7O Product in Three Dimensional PPFs Model Labor Market Adjustment after Capital Subsidy Policy on the USS Sector Effects of Capital Inflow on USS. . Sector in Three Dimensional PPFs Model Labor Market Adjustment after Minimum Wage Policy on the ULS Sector Labor Market Adjustment after Tariff. Policy on the ULS Sector in the Homogeneous Products Case of ULS and USS Sector Effects of Tariff on the ULS Sector . . . . Product in Three Dimensional PPFs Model Labor Market Equilibrium in Korean. . Three Sectors Comparative Static Process with Capital Inflow into Urban Large Scale Sector Area Map by Province; Korea . . . . . . . xvii 73 75 78 82 84 91 97 132 —M‘s_:...._. Kore The One Tail The One ‘rwfl ' DEFINITIONS ULS Sector: Urban Large Scale Sector USS Sector: Urban Small Scale Sector LSE : Large Scale Enterprise SSE : Small Scale Enterprise TFP : Total Factor Productivity Korean Currency: Won The prevailing official exchange rate was One US $ = Won 827.40 in 1984. Taiwan Currency: NT $ The prevailing official exchange rate was One US $ = NT $ 39.47 in 1984. xviii Over th achieved relat other NICs(New Korea and Taiw endowment and small in te Taiwan:36,000 their land a: million peoplej opted for expl their cheap la' started by textiles and electronics an Though differences i small scale er of the nations manufacturing average size c INTRODUCTION A. BACKGROUND Over the last two decades, Korea and Taiwan have achieved relatively higher rates of economic growth than other NICs(Newly Industrialized Countries). Interestingly, Korea and Taiwan have some similarity in terms of resource endowment and economic growth patterns. Both countries are small in terms of land area (South Korea: 99,000km2, Taiwan:36,000km2) but have a large population relative to their land area (South Korea:40 million, Taiwanz20 million people). Since the early 19605, both nations have OPted for export-promotion policies in order to exploit their cheap labor and compete in the world market. They started by exporting labor-intensive goods such as textiles and shoes and shifted to capital-intensive electronics and automobile industries later. Though their trade policies are alike, there are differences in their industrial structure. In Taiwan, small scale enterprises (SSES) contribute a major portion 0f the national GDP. Between 1966 and 1976, the number of manufacturing firms in Taiwan increased by 150%, but the ‘average size of the individual enterprise, as measured by the number 0 Korea,on the 0 have come to d During the earn in Korea inc employees per Currentl change their competition products from capital-intens .uhereas, fa inequality, K These have a income distribl Our com industrial st: Policy. We 1 SSES in Korea of economic g helpful in ev as well as 0th B. Maj or Underlyi the number of employees, increased by only 29%. In Korea,on the other hand, large scale enterprises (LSEs) have come to dominate important sections of the economy. During the same period the number of manufacturing firms in Korea increased by a mere 10%, while the number of employees per enterprise increased by 176%. Currently, both countries are facing challenges to change their industrial structure. Due to tougher competition in the world market in labor-intensive products from other LDCs, Taiwan needs to expand its capital-intensive large scale industrial structure; ‘ whereas, faced with growing concerns about income inequality, Korea needs to encourage the growth of SSEs. These have a potential for employment creation and better income distribution. Our concern is with why and how the prevailing industrial structure in Korea was shaped by its economic P°liCY. We need further to compare the current state of SSES in Korea with SSEs in Taiwan with its better record 0f economic growth and income distribution. This may be helpful in evolving a better industrial policy for Korea as well as other LDCs. B. Major Theme and Hypothesis of Thesis Underlying the study of SSEs in Taiwan and Korea, ““4... . ; _ _7. there is the related market allocation. Neo-clas competition th minimize cos and outputs. optimum resou pricing. In an externalities prices of inp between privat make their i1 their decisior allocation. 1 such instances However, distortion in favor one sect will only can As example, created by a p government at analysed in a the Social Mar there is the old issue that government intervention and related market failure may lead to inefficient resource allocation. Neo-classical economists believe that under perfect competition the entrepreneurs will maximize profits and minimize costs , based on the correct prices of inputs and outputs. The market mechanism will thus lead to optimum resource allocation, or can be made to by shadow pricing. In an imperfect market, for instance, due to externalities and increasing returns to scale, the wrong prices of inputs and outputs will cause a divergence between private and social benefits(or costs). If firms make their investment plans based on distorted prices, their decision will not lead to efficient resource allocation. Government intervention can be justified in such instances of market failure. However, the government may itself create a distortion in the factor or output market in order to favor one sector against another sector. Such intervention will only contribute to inefficient resource allocation. AS example, I look at the case of a market distortion created by a policy favoring LSEs. How this policy of the Government affects the resource allocation can be analysed in a partial equilibrium framework; I assume that the Social Marginal Benefit(SMB) and Social Marginal Cost schedule (SMC) Figure 1 In the market system,a large - Capital Subsidy case? firm producer fixes output at qE because the equality f of social marginal benefit and social marginal cost at the point A, would maximize profits. If government favors a large firm by l providing a capital subsidy I (2 ,the large firm will face a lower private marginal cost(PMC) schedule. Then production will expand to ql , where social marginal cost exceeds social marginal benefit by the amount of the subsidy (BC). The cost of a policy favoring LSE is the triangle, ABC, the sum of amounts by which social marginal cost exceeds social marginal benefit for each unit produced beyond the optimal quantity, qE . Therefore, this market distortion created by a policy favoring LSE would lead to inoptimal resource allocation. The government then can be an independent source of market failure. I give below the order of my investigation. 1. I will determine the extent to which the Korean Policy of promoting LSEs and discriminating against SSEs has led to distortions in the factor (capital) market. 2. I will distortit SSES. I ‘ producti* industri' 3. Fina and perf SSES. A compr impact of Ker the Purposive is clear eVid enterPrise par markets with F m0de1 is mo: °°mparative iInDucal-ions LSEs. However. discussiOns it C. Obj e< 1‘ Theo] Korea 1 market. 2. I will further examine the effect of this market distortion on the efficiency and performance of SSES. I will do so by looking at the total factor productivity (TFP) of Korean SSEs in small scale industries. 3. Finally, I will compare the industrial policies and performance of Taiwan and Korea with regard to SSES. A comprehensive analysis should also include the impact of Korean and Taiwanese economic organization and the purposive uses of capital subsidies. In Korea, there is clear evidence that the government actively stimulated enterprise participation and expansion in profitable world markets with potential dynamic gains for the economy. My model is more limited in scope and rests within a comparative static framework in analyzing the implications of capital subsidies to privately owned LSEs. However, I take note of these dynamic aspects in my discussions in the text and conclusion. C. Objective of Thesis 1. Theoretical Macro Analysis of Korean SSEs Korea has favored the allocation of capital to w__.——-——_—. ;_____———— LSEs. This t industrial domi I base th labor market ; (USS), and n prevailing pol the ULS sectc sectors. This empirical data 2. Empi Taiwan During 1 relative decl taming LSEs Interest rates m the °Vera11 that SSES ham severa1 measure for Purpoee ( SSES. D' The T] This tn. LSEs. This has encouraged their rapid expansion and industrial dominance. This merits theoretical analysis. I base this analysis on a three-sector model of the labor market ; urban large scale (ULS), urban small scale (USS), and rural (R) sectors. I will demonstrate how the prevailing policy increases employment and real wages in the ULS sector but decreases them in the other two sectors. This prediction will be tested against Korean empirical data for the period 1966-1984. 2. Empirical Micro Study of SSES in Korea and Taiwan During the last two decades, there has been a relative decline of SSEs in Korea. The policy package favoring LSEs included an overvalued Won, and lower interest rates (Table 6.2). The composition of Korean SSEs in the overall industrial structure (Table 6.1) confirms that SSEs have been losing ground. We will develop several measures of the relative efficiency of Korean SSEs f°r Purpose of testing and comparison with Taiwanese SSEs. D. The Theoretical and Empirical Approach This thesis will develop a theoretical model and a complementary 1. Theorl The theo a three-secto mobile across assume full em calPital market alternative po analysis. H0(1980) Korea and Tai is 'The Repor 19845 For Tai and Wilmer-oi Republic of C thus “tends a E‘ The I will chapters. complementary framework for empirical analysis of SSES. 1. Theoretical Model The theoretical analysis involves the development of a three-sector labor market model , in which labor is mobile across sectors, but capital is sector-specific. We assume full employment and a competitive labor market; the capital market is regulated. We assess the effects of alternative policies on each sector by comparative static analysis. 2. Empirical Study Ho(1980) has done an empirical study of SSEs in Korea and Taiwan. The main data source for Korean SSEs is ’The Report on Mining and Manufacturing Survey ; 1966— 1984’. For Taiwanese SSEs it is ’The Report of Industrial and Commercial Census' of Taiwan-Fukien District of Republic of China ; 1976, 1981’. My empirical analysis thus extends and updates Ho’s earlier study. E. The Organization of the Dissertation I Will present below a synopsis of the different chapters. Chapter ' will deal wit‘ sectors, and w and 5513's. Chapter labor market to analysis of th Chapter previous chap policies favor on the growth Chapter in terms of 51 chapter and relative e implications f Chapter the sea eXpe Chapter preViou 1' ECOmmendatiOU Chapter Two will present a literature review. This will deal with the concepts of the dual and informal sectors, and with the role of rural non-farming activities and SSEs. Chapter Three will elaborate on the three- sector labor market model to be followed by a comparative static analysis of the effects of different policies. Chapter Four will extend the analysis of the previous chapter by focusing on the effects of Korean policies favoring LSEs. This empirical study will focus on the growth of employment and real wages. Chapter Five will review the state of Korean SSEs in terms of size, location and survivor technique. Chapter Six will elaborate on industrial policy and relative efficiency issues in Korea, and explore their implications for policy as regards SSES in Korea. Chapter Seven will conclude with a comparison of the SSE experience in Korea and Taiwan in the context of their general industrial policies. Chapter Eight will summarize general findings from the previous chapters. This will include policy recommendations for industrial policies toward SSEs. A. INTRO In buil scale, and dealing vi rellllisite: tr, and Harris-Tod field. The 1 the importance the urban area in the rural a source of i1 Finally, t1 .evaluated in 1 B. DUAL I 1' Intrr Current mutt: CHAPTER 2 LITERATURE REVIEW FOR THREE SECTORS (ULS, USS, R SECTORS) MODEL A. INTRODUCTION In building a theoretical model for large,small scale, and rural sectors, a review of the literature dealing with a sectoral model of the economy is a requisite: the classical dual sector literatures by Lewis and Harris—Todaro et al. provide the best insights in this field. The literature on the informal sector establishes the importance of informal or small scale activities in the urban area. The literature on non-farming activities in the rural area, which are usually small scale, is the source of information on-potential rural SSEs in LDCs. Finally, the literature about SSES in LDCs is re- evaluated in the light of this study. B. DUAL SECTOR 1. Introduction Current literature on economic development resorts to multi-sector models in order to analyze the interactions b« these multi-se simple dual se to review the . multi-sector as considered tho effects of rur 2-a. Cla Lewis(19 sectors: the MthWQh he Silbdivisioh’ asSumed that non-capitalist Perfectly ela subsistence 1e Preduct, becau is Zero, Bu constant Wag e suhsistence Is 0““ mi bag 1113 ’ 9001 supply from expands’ and c 10 interactions between major economic sectors. In fact, these multi-sector models have evolved over time, from a simple dual sector model. Therefore it seems meaningful to review the dual sector model, which is the root of the multi—sector model. Among the dual sector models are considered those by Lewis and H—T , which consider the effects of rural-urban migration. Z-a. Classical Dual Sector Model Lewis(1954) divided a developing economy into two sectors: the capitalist and non-capitalist sectors. Although he recognized the possibility of further subdivision, he treated each as a single sector. He assumed that there is an unlimited labor supply in the non—capitalist sector and so this rural labor supply is perfectly elastic. The rural sector wage is at a subsistence level, which means that it equates the average product, because the rural marginal productivity of labor is zero. But the urban capitalist sector pays ahigher constant wage (about 30% higher) than the rural sector’s subsistence level. Once migration from the rural to the urban sector beQins, economic growth proceeds. With unlimited labor suPply from the rural sector, the capitalist sector exPands,and generates profits, which are reinvested into the capitalist capitalist sec Lewis a supply in the study by Godfr found that i labor deficit agriculture pu Price . This 1 the Ricardian that the Lewis ' d”91091119 ecc Mavis a favorable for than P°Plllatic Slipply in the future When based on We intensin’ the in the page a sector. Lewis n as only a Slip] deemphhsize t] batman the SthOr 11 the capitalist sector itself, so economic expansion by the capitalist sector continues. Lewis assumed that there is an unlimited labor supply in the rural sector; but according to the empirical study by Godfrey, this is not always true. Godfrey (1969) found that in West Africa agriculture suffered from a labor deficit,not from a surplus. The labor shortage in agriculture pushed up the price of agricultural products’ price . This led to delayed industrialization, in terms of the Ricardian food bottle neck phenomenon. Godfrey showed that the Lewis model does not necessarily hold in all developing economies. Lewis also recognized that if conditions are favorable for the capitalist surplus to grow more rapidly than population, there will be an exhaustion of labor supply in the non-capitalist sector at some point in the future when capital catches up with labor supply. And based on whether it is capital-intensive or labor- intensive, the choice of technology can make a difference in the pace at which labor is absorbed in the capitalist .sector. Lewis regarded the role of the non-capitalist sector as only a supplier of unlimited labor. This view tends to deemphasize the role of agriculture and the interaction between the capitalist sector and the non-capitalist sector. Lewis led the industry-oriented theories for LDC’s developm agriculture t without impair Lewis as rural-urban unemployment i Therefore his urban sectors. Z'b. Ext Ranis-Fe frillework and expansion of i labor from the labor Supply c after exhaust: This is cal: 12 LDC’s development, because a rapid labor transfer from agriculture to industry can expand the industrial sector without impairing the growth of the agricultural sector Lewis assumed a full employment potential for the rural—urban migrants. But, in reality , urban unemployment is a prominent and common phenomenon in LDCs. Therefore his model failed to explain unemployment in the urban sectors. Z—b. Extension of Classical Model Ranis-Fei(1961) started with Lewis’ theoretical framework and extended it for clarifying the optimistic expansion of the capitalist sector by exploiting unlimited labor from the non-capitalist sector. Lewis shows that the labor supply curve in the industrial sector turns upwards after exhausting excess labor supply in the rural sector. This is called Lewis' turning point. At this turning point, the rural wage starts to be determined by the labor market mechanism of supply and demand. But Ranis—Fei argued that the turning point in the industrial labor supply curve starts earlier than Lewis thought. If the MPL of rural sector becomes positive due to labor migration into the industrial sector, the declining total output of the rural sector can worsen the industrial terms of trade. The relatively cheap price of industrial go industrial sec Therefore R-F industrial la}: point. To come] the labor marl of full employ the study of t 3-a. Toc' Todarou °f unlimited 3 full employmel ObserVEd that the urban Sect In order to migrate has“ the urban s« than on actua; Implici' traditional s for migrants. lahOr mOVes then into the 13 industrial goods can push up the real wage of the industrial sector. This is called the shortage point. Therefore R—F interpreted Lewis’ turning point in the industrial labor supply curve as starting at the shortage point. To conclude, R-F’s concept of wage determination in the labor market seems to be valid under the assumptions of full employment and perfect competition, specially for the study of the Korean and Taiwanese case. 3-a. Todaro’s Dual Sector Model Todaro(1969) raised doubts about Lewis’ assumption of unlimited labor supply in the non-capitalist sector and full employment in the capitalist sector,because Todaro observed that in many LDCs there was open unemployment in the urban sector, and also soaring wages in this sector. In order to solve these issues he suggested that people migrate based on the differentials of expected income of the urban sector and income of the rural sector, rather than on actual income differences. Implicitly, he recognized the existence of the urban traditional sector which is a temporary informal sector for migrants, because he pointed out that, rural unskilled labor moves first to the urban traditional sector ,and then into the permanent modern sector. Based c model of migra the urban se finding urban expected urba income, peopl sector. In the 5 market tends t further Inigra negatively the long run, 1 difference bet migration Wil: MOrEOVe] farmers Will 1 in°°me over Since the over tinle. t1 market, or . Opportunities overflow into labor absorpt The (sea requirements rural sectors 14 Based on this observation, he developed his own model of migration. He assumed that the expected income in the urban sector is the product of the probability of finding urban jobs and the actual urban wage. If the expected urban income is larger than the rural actual income, people will migrate from the rural to the urban sector. In the short run the rigid wage of the urban labor market tends to raise the expected urban income and cause further migration. But massive unemployment may affect negatively the probability of finding urban jobs . In the long run, the adjusted probability may result in no difference between expected urban and rural incomes, when migration will stop. Moreover, in the long term decision to Imigrate, farmers will compare the present value of expected urban income over time and the present value of rural income. Since the probability of getting a job tends to improve over time, there is an incentive to enter the urban labor market, or join the unemployed while waiting for job opportunities to open up. So, in order to prevent an overflow into the city, the rural sector must improve its labor absorption. The weak point in this model is that it ignores the requirements of skill differentials between urban and rural sectors. Also, some LDCs, including Korea do not have an offici Departir (1983) sugge incorporating in the labom established s< other tradit: important fat the labor mar] 3'1). Mod. Harris-' two sector (; They assumea . the actual r Also they ass Production of bEtween the t Further- limited Wage order to imp the urban Sec wage by Eqm product 0f la of the Produ 15 have an official minimum wage system . Departing from Todaro’s dual sector model, Kannappan (1983) suggested a more realistic analytical framework incorporating supply factors and institutional practices in the labor market. His study stressed that the established social order such as family,caste,religion and other traditional networks should be considered as an important factor in shaping the structure and process of the labor market in LDCs. 3-b. Modification of Todaro’s Model Harris-Todaro(H-T) (1970) supported the basic Todaro two sector (rural- urban) model and extended it further. They assumed that as long as expected urban income exceeds the actual rural income, migration continues to occur. Also they assumed that the urban sector specialized in production of manufactured goods. The goods can be traded between the two sectors. Further, H-T suggested policy prescriptions of a limited wage subsidy and restriction on migration in order to improve the welfare of sectors. Wage subsidy in the urban sector can reduce unemployment under the minimum wage by equalizing the shadow wage and the marginal product of labor,because this subsidy can reduce the costs of the producers, therefore increasing the demand for labor. It give urban income e Therefor can prevent tr unemployment Otherwise, tt actually inc: Paradox). TI subsidy and mj sector in term sector faces a low rural wag. Bhagwat; reCommended bj The best pos: 5mm}! :or, Production suggest a dir they Prefer-r and urban see a mixed pac subsidy) Wor ( Such as mechanism. B~s ’s ellInitiation 16 labor. It gives further incentives to migrate as long as urban income exceeds rural income. Therefore, a -restriction on rural-urban migration can prevent the minimum wage from having its effects on, unemployment in terms of increasing the supply of labor. Otherwise, the accelerated urban employment creation may actually increase the level of unemployment(Todaro’s Paradox). The combination of these two policies (wage subsidy and migration restriction) favors only the urban sector in terms of income distribution, because the urban sector faces a high minimum wage, and the rural sector has low rural wages. Bhagwati-Srinivasan(B-S) (1974) studied the policies recommended by H—T and graded them in the following way. The best possible optimal policy may be a uniform wage subsidy ;or, a wage subsidy in manufacturing plus a production subsidy in agriculture. But they did not suggest a direct migration restriction like H—T , because they preferred optimal allocation of labor in the rural and urban sectors through the price mechanism, rather than a mixed package , in which one policy (such as wage subsidy) works through the price mechanism and the other ( such as migration restriction) is a physical quota mechanism. B-S’s policy prescription is based on the idea of elimination of the labor market distortion by a minimum wage through : Todaro(l I find an empir: under which a ,designed to level and rat. His con. absolute levu result of an the elasti Probabilities - °f jOb creati His emDirica Conside Taiwan are model S e 91118 4° Conc were those weaknesSes actiVities Powerful lY tt the Urban a] 17 wage through a wage subsidy in both sectors. Todaro(l976) ,based on his original model, tried to ' find an empirically testable formula to explain conditions under which autonomous increase in urban job creation ,designed to get rid of urban unemployment, can cause the level and rate of unemployment to rise (Todaro’s Paradox). His condition which is np> g*Eu/M indicated that the absolute level of urban unemployment will increase as a result of an autonomous expansion of urban labor demand if the elasticity of migration with respect to job probabilities(np ) exceeds the ratio of the normal amount of job creation(g*Eu) to the normal level of migration(M). His empirical study in Tanzania confirmed this condition. Considering that the labor markets in Korea and Taiwan are free of such distortions,the Harris-Todaro model seems to be less valid there. 4. Conclusion The basic dual rural-urban migration models reviewed were those of Lewis and Harris-Todaro. These. showed weaknesses in explaining the role of small scale sectors as informal activities in urban area or non-farming activities in rural area; but they have contributed powerfully to the explanation of the interactions between the urban and rural sectors. C. INFO! 1. Intrc It has 1 LDCs proceede< continued. Om capacity to a] P°Verty in urI for unemploym. identified as AS the deve10pment 1 for disregard LDCS has 3012).] growth. Bei 2« Defj General mean the ] emPloyrment' to mean the self~empl°m 18 C. INFORMAL SECTOR 1. Introduction It has been observed that as economic development in LDCs proceeded, urbanization due to sectoral migration continued. Once the degree of urban migration exceeded the capacity to absorb labor, the problems of unemployment and poverty in urban areas noticeably emerged. One solution for unemployment and poverty in urban areas came to be identified as employment in the informal sector. As the potential of this sector was articulated, development literature of the dual model became criticized for disregarding its importance. The informal sector in LDCs has actually contributed to employment and economic growth. Being small scale, the informal sector concept can now be incorporated into the small scale sector. 2. Definition Generally, the formal sector has been understood to mean the protected urban sector which offers wage employment, while the informal sector has been understood to mean the sector that is not protected and that offers self-employment _- this includes petty traders and street hawkers. In f sector was an Hart(19 study of Ghan self employme job opportun without any 9 a job in t recognize tha The IL informal sec specific char report, the opportunities family owners labor-intensi outside the competitive informal se< formal sectw discriminati< Mazumd terms of th urban enter unprotected cannot absc 19 hawkers. In fact in the dualistic literature, the informal sector was ambiguously defined. Hart(1973) introduced the informal sector in his study of Ghana. He distinguished between wage earning and self employment activities. The informal sector provided job opportunities for new migrants who came to towns without any special training, and those who could not find a job in the formal sector. Hart was the first to recognize that the informal sector can generate an income. The ILO(1972) report on Kenya identified the informal sector as a target sector in analyzing the specific characteristics of the economy. According to this report, the informal sector can provide easy access to opportunities, using indigeneous resources allowing for family ownership of enterprises, small scale operation, labor-intensive and adapted technology skills acquired outside ,the formal school system, and unregulated and competitive markets. The report concluded that the informal sector can create jobs more quickly than the formal sector. Therefore, ILO recommended lessening discrimination in policy towards the informal sector. Mazumdar(l976) looked at the informal sector in terms of the urban labor market rather than in terms of urban enterprises. He saw the informal sector as an unprotected one. Because the industrialized formal sector cannot absorb all the urban labor force, the informal sector absor}: Mazumda relationships turnover of 1 into this sec regular emplc in the info unnecessary overurbanizat found that en Weeks(1 sector in ter exchange acti a-vis the St sector tends types, wherea and regulatec The in: operations , 1a and indigenl 4 local input He poi sectors dyna 0f the two 5 positively, sector in 20 sector absorbs most of those left out. Mazumdar observed that the lack of long contractual relationships in the informal sector stimulates a high turnover of labor, which, in turn, favors an easy entry into this sector. Some job seekers who are unable to find regular employment in the formal sector participate easily in the informal sector. This easy entry may cause unnecessary overmigration, which then leads to overurbanization. Breman(1976) unlike Mazumdar, however, found that entry into the informal sector is not easy. Weeks(1975) like Mazumdar defined the informal sector in terms of the .organizational characteristics of exchange activity ,and position of economic activity vis— a-vis the State. The insecure operation of the informal sector tends to limit its access to resources of all types, whereas the formal sector is recognized, nurtured and regulated by the State. The informal sector is distinguished by small scale operations,labor-intensive techniques,low income levels and indigeneous ownership, with production using largely local inputs and catering for a low income market.’ He pointed out that population growth makes both sectors dynamic regardless of the inherent characteristics of the two sectors. Weeks looked at the informal sector positively, in that as an evolving and dynamic low wage sector in LDCs, "it can provide a large quantity of consumer go: capital goods using labor- the long rt distribution informal good between the 1 Sethura of the inform of productior activities: can be desc organized vs. scale activit simple technc The informal simple tecl constraints a 3. I: Sectors Hart(1 on demand c the formal 5 Weeks ( 21 consumer goods for lowest income groups, indigeneous capital goods, and can utilize labor more efficiently by using labor-intensive techniques. But he argued that in the long run the high wage formal sector makes income distribution more unequal , reduces the demand for informal goods and results in widening wage differentials between the formal and informal sectors. Sethuraman(1981) reviewed many different definitions of the informal sector, specifically, those based on mode of production, on mode of organization and on scale of activities; he suggested that an urban dichotomy in LDCs can be described in terms of technological dualism ,or organized vs. unorganized sectors, and large vs. small scale activities. The informal sector is characterized by simple technology, unorganized and small scale activities. The informal sector is relatively labor—intensive and use ~ simple technology in production because of capital constraints and limited access to technical knowhow. 3. Interaction between the Informal and Formal Sectors Hart(1973) believed that the informal sector depends on demand created by the current levels of activities in the formal sector. Weeks(1975) pointed out that high wages in the sector, latter. Al increases c sector. capital-inte the long relationship Also, informal 5: because the consumer (in: output, exce] boosting Ag- predominantl} scale and rm Moser(2 between the : Why of as by promoting boosting th Poverty and 22 formal sector tend to stimulate production in the informal sector because costs increase in the former. Then the formal sector begins to subcontract with the informal sector, making it possible for more production in the latter. Also, high wage incomes in the formal sector increases consumption of goods produced in the informal sector. Therefore, in the short run, it can increase employment in informal sector by shifting production from capital—intensive to labor-intensive processes. But in the long run he projected paradoxically negative relationships. Also, Weeks thought that the relation between the informal sector and the Ag-sector is complementary, because the informal sector can provide simple Ag—tools, consumer durables, ,processing and transportation of output, except fertilizers and seeds; and he believed that boosting Ag-development in LDCs where the Ag—sector is predominantly a small holder business, can stimulate small scale and rural non-farming activities. Moser(1978) recognized the dependent relationship between the formal and informal sectors. In his view, a policy of assisting the informal sector will likely end up by promoting the formal sector. Therefore, a policy of boosting the informal sector is not effective in reducing poverty and unemployment unless there is a fundamental change in ove Gerry(1 relationship but emphasize in favor of c indigenous p dependent up large urban l: Tokman( between the ' choosing a should he s an autonomo links with tt it also has a Bromle} the informal although a entrepreneur socio-econom. suppliers, 04 the supplier: range of pub Sethur sector may subordinate 23 change in overall political and economic structure. Gerry(1978) also pointed out the dependent relationship between the formal and informal sectors, but emphasized that the relationship is heavily weighted in favor of capitalist hegemony. He agreed with Amin that indigenous petty production and distribution is largely dependent upon materials made available by or through large urban Dakar-based capitalist enterprises. Tokman(1978) tried to explain the interaction between the informal sector and the rest of the sectors by choosing a more neutral position . The informal sector should be seen neither as a completely integrated nor as an autonomous sector, but rather as one with significant links with the rest of the economy, while simultaneously, it also has a considerable degree of independence. Bromley(1978) also confirmed the interaction between the informal sector and the other sectors. In his View, although a street trader is an independent small entrepreneur, he is ,however, tied into a very complex socio-economic network, relating not only to his Suppliers, competitors, customers, but also money lenders, the suppliers of equipment, the authorities,and a wide range of public and private institutions. Sethuraman(1981) pointed out that the informal sector may depend on the formal sector for capital. This subordinate relationship tends to make the informal sector surren predetermined dependent uni 4. Dyna Weeks ( 1 sector in t because the f to the forma more labor. incremental of the eco utilization variables a] relative prit Mazumda informal 561 the formal s of earnings , average inc01 eupirical s growth of ea that in the that the search model 24 sector surrender all 'or part of its output at prices predetermined by the formal sector lender, or operate as a dependent unit serving the interests of the formal sector. 4. Dynamic Future of the Informal Sector Weeks(1975) projected the future of the informal sector in terms of positive employment opportunities, because the future growth of the informal sector relative to the formal sector means a shift towards utilization of more labor. Theoretically, this trend should add to the incremental employment/output ratio in the non-Ag sector of the economy as‘ well as ensure a more efficient utilization of capital. He pointed out that the crucial variables are changes in the aggregate income, the relative price levels, technology and consumer habits. Mazumdar(1976) argued that the future of the informal sector is dependent on the growth of income in the formal sector with some lag, and that the growth rate of earnings in the informal sector lags behind that of the average income in the rural traditional sector. But in his empirical study of Peru, this was not the case, because growth of earnings in.the informal sector was higher than that in the rural sector. He explained, by conjecture, that the migration function in his probabilistic .job search model may have exaggerated the migration rate, and that the migr protect earni Gerry(1 a massive co job in urban petty produce cannot esca inability oi fully emerge. 0f different such as establishmeni These extrel the Capacitj the e"0111mm Order to tr: 39thur about the m pessilnistic. informal Se to absorb market impel: the infOrmal between the The 9933111115 25 that the migration cost and job search in the urban sector protect earnings from falling too far. Gerry(1978) in his study of Senegal found that with a massive continuous migration, the burden of finding a job in urban areas tends to fall on the shoulders of the petty producers, and that the majority of the population cannot escape poverty. This result reinforces the inability of an embryonic Senegalese capitalist class to fully emerge. He recognized the emergence of the process of differentiation, as two extreme transition movements, such as proletarianization and the progressive establishment of a capitalistic environment of production. These extreme transitions require an enormous increase in the capacity of capitalist industry to absorb labor, and the evolution of a vital indigenous capitalist class, in order t0“ transform the mass of petty producers into wage workers. Sethuraman(1981) analysed two contrasting views about the future of the informal sector: optimistic and pessimistic. The optimistic view seems to be that the informal sector can accumulate capital & grow fast enough to absorb labor flow if government policy is favorable, market imperfections are eliminated ,and the market for the informal sector’s goods' is expanded through linkage between the informal sector and the rest of the economy. The pessimistic View holds that the informal sector cannot accumulate c since it is s 5. Poli Gerry ( 1 for the info: formal sect: keeping the big to be influence 01 favoring tn. Producers W01 raw materi; impoverished Bromlej training and in the info achieve up“ in the fOr linkage, finanCQ. Nihan, majority of nominterver suggeSted t1 26 accumulate capital and generate an evolutionary growth since it is subordinate to the formal sector. 5. Policy for the Informal Sector Gerry(1978) cited Dannis' comments (1974) on policy for the informal sector. " The optimal strategy for the formal sector is not one of elimination but rather of keeping the informal sector at an optimal size, not too big to be competitive, not so small as to have no influence on wages." Otherwise because of the policies favoring the formal sector, a large number of petty producers would be driven towards the miserable fringe of raw materials production, unskilled _ labor, and impoverished clients, generally. Bromley(1978) recommended easy credit, technical training and technical assistance for selected activities in the informal sector so that a minority of households achieve upward socia-economic mobility. This needs support in the form of fundamental changes in the vertical linkage, governmental regulation, and institutional finance. Nihan, Demol and Jondoh (1979) found that the majority of entrepreneurs in the informal sector prefer non-intervention by the State to any assistance. They suggested that action programs should avoid any systematic recourse to granting wide supposed to 1: 6. Conc With ox tends to cor further migi oPPortunitie: activities; 1 With reCommendati. informal sec growth in th seCtOl‘ can n D. RUE 1- Int The lj “Wines . theoret i Ca 1 27 recourse to the solution of injecting fresh capital and granting widespread access to credit facilities, which are supposed to promote rapid business expansion. 6. Conclusion With over—urbanization in LDCs, the informal sector tends to contribute to employment generation. To avoid further migration, LDCs need to boost employment opportunities in rural areas through non-farming activities; this should form part of state policy. With slight differences in definition and policy recommendation, the literature seems to agree that the informal sector contributes to job creation and economic growth in the urban areas of LDCs. The concept of informal sector can now be incorporated into the study of the small scale sector. D. RURAL NON-FARMING ACTIVITIES in LDCs 1. Introduction The literature on economic develOpment has explored and debated the importance of the rural non-farming activities in LDCs over the past 20 years , using various theoretical models and empirical studies. Currently, the increase in employment c urban areas. scale. 2.The 1 In ths Liedholm-Chm their analys: and consume. Possibility : They stipula' as economic demand less They t non-farming the Ag‘Secto pmd‘lction. GEVelopmem: , 3 ' Ske Byerle H ~ I R s predi‘c 28 increase in these activities seems to be generating employment opportunities and controlling migration into urban areas. They tend to be labor intensive and small scale. 2.The Pessimistic View of Non-farming Activity In the classical debate between Hymer-Resnick and Liedholm-Chuta, H-R(1969) introduced a Z goods concept in their analysis. They assumed Z goods can only be produced and consumed within the rural sector, excluding a possibility of Z goods trade with urban sector outlets. They stipulated that Z goods are inferior in quality; and as economic growth proceeds, Ag-households would tend to demand less Z goods. They took a pessimistic view of the future of rural non-farming activities, because the increasing income of the Ag-sector would bring about a decline 'in Z goods production. Thus, they considered these activities as temporary phenomena, prior to full fledged economic development. 3. Skepticism about H-R Study Byerlee and Eicher(B-E) (1972) raised a doubt about H-R’s predibtion regarding rural non-farming activities in LDCs. They c important, increase in migration int their analysi a small agri< sector. Liedhol their study ; the importam and forward I The increas. Various manu the same t in a further pr In f and Processi LDCS' Lied aCtivities expanded to of the 1”Ural Gibbh confirm that 1. Des‘ He the“? to t1 29 LDCs. They observed that these were becoming increasingly important, because they appeared to contribute to an increase in rural employment and to the control of migration into the urban areas. Within the framework of their analysis, the rural sector became disaggregated into a small agricultural sector and small scale rural non-farm sector. Liedholm(1973) supported B-E’s views and extended their study further. Like B-E, Liedholm also emphasized the importance of non-farming activities through backward and forward linkages in the agricultural sector itself. The increase of Ag-production will create a demand for various manufactured farm inputs(backward linkage), and at the same time a higher agricultural production would need a further processing of these Ag-outputs(forward linkage). In fact, these activities, producing farm inputs and processing foods, do occur in the rural areas of many LDCs. Liedholm suggested that research into non-farming activities must recognize that the Z goods concept be expanded to include intermediate goods beyond the consumer ones. Therefore he took an optimistic view of the future of the rural non-farming sector. Gibb(1974), in his Phillippine study, tried to confirm that the rural non-farming sector was important in LDCs. He applied the growth theory and the locational theory to the rural non—farming sector in LDCs. The growth~ theory sugge productivity dependent re industries . of transport: production a1 should be 1. 4. The Liedho the H~R's v Leone, whic f°r a large °f Sierra Le Also t t“mild out 1 rural incom PrOductS was increases, increases, Their Strong dem. baCkWard 1i Furthermore 3O theory suggested that an increase in agricultural productivity and export from this sector can stimulate dependent regional industries such as the non-farming industries. The locational theory stressed the importance of transportation costs in determining the location of production and suggested that a non-agricultural industry should be located near a developing agricultural sector. 4. The Optimistic View of Non-farming Activity Liedholm and Chuta(L-C) (1976) directly challenged the H—R’s view through their empirical study of Sierra Leone, which showed that small scale industries accounted : for a large segment of total employment, in the rural area of Sierra Leone. Also the Z goods concept of the non-farming sector turned out to be irrelevant in Sierra Leone, because the rural income elasticity of demand for most small scale products was strongly positive; therefore, as rural income increases, the demand for the non-farm output also increases. Their study also revealed that Ag-sectorv showed a strong demand for non—farming sector products through a backward linkage( e.g. the small farm tool industry), and a forward linkage(such as the food processing industry). Furthermore, the foreign export sector showed a strong demand for t1 Theref< R’s pessimisi sector would 5. A PI Liedho farming outp If we 31 manufactured Pessimistic sector can t Figure (ems‘mptior bet p: > 31 demand for the non-farming output(gara dying industry). Therefore their empirical study clearly rejected H- R’s pessimistic view, predicting that the non-farming sector would not decline as rural income increases. 5. A Possible Modification of H-R’s Model Liedholm(1973) demonstrated that the demand for non— farming output by the urban sector also tend to be strong. If we allow for Z goods being traded for urban manufactured goods, even in the original H-R model, their pessimistic outlook on the future of the non-farming sector can be modified. Figure 2.1 In the original H-R model, ‘agriculture,the new equilibrium Z of production and consumption b will occur at C2. The movement from C1 to A is due to the C1 substitution effect and the a ___¥ A movement from A to B is due to \\ \\1\ the income effect. The final C I II d M=P*F movement from B to C is due to 2 the curvature effect. Now we may assume, 2 goods can be traded for urban manufactured goods with the other assumptions unchanged. Under this C] in agricultv occur at C3 occur at B. of F goods 1 because b-a and consum consumption curvature ef Finall after a t Production 11 a contradict 5. Cox 32 Under this changed assumption, after technological change in agriculture the new equilibrium of production will occur at C3 and the new equilibrium of consumption will occur at B. So Ag-sector can produce ob of Z goods and oc of F goods but consume 53 of Z goods and 53 of M goods, because 53 of Z goods can be traded with Ed of M goods, and consumption at point B offers higher utility than consumption at point C2. Under the relaxed assumption the curvature effect is not necessary. Finally although Z goods are relatively inferior, after a technological change in agriculture, their production may increase or at least not decline, which is a contradiction of H-R’s original prediction. 6. Conclusion The controversial aspects of rural- non-farming activities in LDCs ,in the classical economic development literature needs analysis. First H-R’s model suggested a pessimistic View about the future of the non-farming activities. But L-C’s empirical study proved that is not the case, at least in Sierra Leone for several reasons. Based on L-C’s studies on the epositive future of this sector in LDCs, there is some rationale for a policy of encouraging this activity for rural development in LDCs. 1. Int: As SSE and economi industry in seems to be SSE can disaggregatj scale sector 2. Dei 33 E. SMALL SCALE ENTERPRISE 1. Introduction As SSES have become important in terms of employment and economic growth in LDCs, our study disaggregating industry in urban areas into large and small scale sectors seems to be valid. Therefore a review of the literature on SSE can give us_ background information before disaggregating the urban sector into large and small scale sectors in the three sectors model. 2. Definition The SSE is not a very well-defined unit. There are various definitions of SSEs. Some are quantitative, involving assets ( maximum ranging from $25,000 to $2 million ), number of employees (maximum ranging from 15 to 500), and sales turnover. Others are qualitative. The definition by the World Bank ,in 1976,includes firms with Up to a maximum of $250,000 of fixed assets excluding land (Liedholm 1986). Based on these definitions, a firm can be classified into a LSE or a SSE. Therefore, any definition of an SSE seems arbitrary. Eiedholm(1986) defined a SSE as one with less than fifty workers. H with less th off point consistency, SSEs seems a 3.Char a. Loc Accord 1979), the areas. Some household 0] difficult Official ce: and underes Ho ( deve1°Dueut rural areas in urban ar general 12 e b~ De 34 workers. Ho(1980),on the other hand, defined it as one with less than 100 workers in Korea and Taiwan. The cut— off point in SSEs still remains ambiguous. For consistency, Ho’s cut-off design(100 workers) in Korean SSEs seems a good starting point. 3.Characteristic of SSEs a. Location According to various studies(Liedholm 1986,Page 1979), the majority of SSEs in LDCs are located in rural areas. Sometimes the small scale activities in a rural household or a farm compound are invisible, and rather difficult to ascertain in a census. Consequently the official census often overlooks the smallest rural firms and underestimate the existence of SSEs in a rural area. Ho (1980) found that rural infrastructure development contributed to decentralization of SSEs in the rural areas of Taiwan. Some LDCs have their SSEs located in urban areas, as in Korea. Therefore one cannot generalize about their location. b. Demand for Outputs of SSEs. Surprisingly, the studies show the primary market for products of SSEs in LDCs tend to come from local domestic consumers. Even in LDCs with strong .export- oriented po not from the The so consumer go order to mee population. The l sector, or factors in t linkage( 51 (such as i agriculturai areas, thou: The comPP-titive dented. In Could Coupe West Afric response to they Could scale rice ProceSSQS ‘ Agaix to be impo] demand 80] by SSES is 35 oriented policies, the major demand for SSE’s goods was not from the export market. The major small scale industries represent simple consumer goods, such as clothing, furniture, shoes, in order to meet the needs of a low income rural and urban population. The linkages between SSEs and the agricultural sector, or between SSEs and LSEs have become important factors in the developmental process in LDCs. The forward linkage( such as food processing) and backward linkage (such as agricultural tools) between SSEs and the agricultural sector emerged as important for SSEs in rural areas, though their role is still disputed. The relationship between SSEs and LSEs, the competitive one vs. the complementary one was intensively debated. In some LDCs, the former role prevailed, and SSEs could compete effectively against LSEs. For example, in West Africa small scale rice mills expanded rapidly in response to an increase in rice output. They proved that they could compete directly and effectively against large scale rice processing facilities and existing modern processes (Page 1979). Again, the backward linkage from LSE to SSE seemed to be important in LDCs; this backward linkage where LSEs demand some intermediate goods or capital goods produced by SSEs is known as sub-contracting. This where 54 9s 0 large firms established against lar government d goods (Liedhc 4. Lat Varior that there 1979), beca] Wages of th. institution; is true in But distortion 333 still 9 with littie difference the Wage differentia Var-1‘ QXist wit] 36 This subcontracting was very prevalent in Japan, where 54 % of all small scale units were subcontractors to large firms in 1966 (Liedholm 1986). Empirical studies established that SSEs in LDCs could compete effectively against large ones, as well as complement them. The government does not seem to have a major demand for SSEs’ goods(Liedholm 1986). 4. Labor in SSEs Various studies of the labor market in LDCs show that there exists a wage gap between LSEs and SSEs(Page 1979), because the labor market is imperfect, and usually wages of the formal sector are rigid downward ,at an institutionally or legally established minimum wage. This is true in Africa (Page 1979). But in some cases, although there is little distortion in the labor market, a wage gap between LSE and SSE still exists. Ho(1980) found that in Korea and Taiwan, with little distortion in the labor market, the quality difference between the two sectors’ workers may explain the wage gap. Whatever may be the reasons , the wage differential still continues in LDCs. 5. Capital in SSEs Various studies show that in many LDCs the SSEs exist within a segmented capital market and they are excluded fro Page 1979, r borrowing f lending rate and loans fr Usual] lending to 5 higher admir to be able credit markt there exist Gener Economic collprehensi a. Pa The F to the uso labor rat capital/lal capital/lal the“ may 1 Coilsiderim 37 excluded from access to institutionalized credit(Ho 1980, Page 1979, Liedholm 1986). Hence, the SSEs tend to rely on borrowing from the informal credit market ,where the lending rates are much higher; or use their own savings, and loans from friends or relatives. Usually public agencies in LDCs tend to avoid lending to SSEs because of the higher default risk and higher administrative costs. On the other hand LSEs tend to be able to get institutional finance from the formal credit market at a much lower interest rate. In most LDCs there exist fragmented capital markets. 6. Efficiency of SSEs Generally representative methods for measuring the economic efficiency of firms are the partial and comprehensive measurements. a. Partial Efficiency Measurement The partial efficiency relates output or value added to the use of one input only. Output-capital or output- labor ratios are used for partial measurements. Capital/labor ratio is used as an alternative. This capital/labor ratio can provide some idea as to whether there may be a trade-off between output and employment. Considering the abundant labor and scarce capital in LDCs, a productivity measurement is often used (Liedholm 1986). Empiri intensive. high capit limitations measure of resources . b. Co: The c< general can and econom; _ .Productivitj weighted a. reflect the social bene '10 cost of The added and functiml ar into tech; TechniCal maximizp, t1 other ham ability t. fact” ~ Pr 38 Empirical study shows SSEs in most LDCs are labor intensive. There are, however, cases of SSEs in LDCs with high capital intensity. In order to overcome the limitations of partial measurements, the comprehensive measure of economic efficiency should include all scarce resources . b. Comprehensive Measurement The comprehensive measure can be classified into two general categories, total factor productivity measurement and economic return measurement. The total factor productivity measurement relates _the value added to a_ weighted average of all scarce inputs, where the weights reflect the shadow prices of inputs. For this measurement social benefit-cost ratios mean the ratios of value added to cost of capital and labor at the shadow price. The production frontier method relates the value added and all scarce inputs by means of a frontier function analysis. This method decomposes total efficiency into technical efficiency and allocative efficiency. Technical efficiency indicates the firm’s ability to maximize the output from a given set of inputs. On the other hand allocative efficiency indicates the firm’s ability to select the optimal input given the existing factor price. This breakdown of total efficiency was suggested by Farrel(1957). This to outliers. of extreme Forsund, Cc numerous ap] function, function am Liedhl enterprises enterprises de‘il‘ee of e LDCs. The efficiency 7. Cc The “Ch as a relationsh: the agrim LDCs. Th. eXiSt be The SSES f ASa ratio is c to be la 39 This method has been criticized as being sensitive to outliers. Ho suggested discarding a small percentage of extreme observations until the frontier stabilized. Forsund, Covell and Schmidt(1980), also, have provided a numerous approaches to estimating the frontier production function, such as deterministic frontier production function and stochastic frontier production function. Liedholm(1986) suggested that in LDCs small scale enterprises continue to be more efficient than large scale enterprises in a sizable number of industries. But the degree of efficiency of SSEs can be different in different LDCs. Therefore, we must be careful in judging the efficiency of SSEs in LDCs. 7. Conclusion The definition of SSE is ambiguous in terms of size such as a cut-off point of 50 or 100 workers etc. The relationships between SSEs and LSEs or between SSEs and the agricultural sector have become important issues in LDCs. The wage gap between LSEs and SSEs continues to exist because of labor market imperfections and differences in the quality of workers in the two sectors. The SSEs face a fragmented capital market. As a partial efficiency measure, the capital/labor ratio is often used. In many cases SSEs in LDCs turn out to be labor intensive. Also, the benefit—cost or production f efficiency II 88135 in LI room for cla its importar 40 production frontier methods are used as comprehensive efficiency measurements for SSEs in LDCs. In many cases, SSEs in LDCs turn out to be more efficient. There is room for clarification about the role of the SSE, though its importance has been definitely established. MODEL In th oriented po 0f its effe Some have argued ’La necessary t lab” marks Which can , industrial three sec+ Rural sec' begin with 3.14 1.1a The 41 CHAPTER 3 MODEL DEVELOPMENT WITH THREE SECTOR LABOR MARKET & THREE DIMENSION PRODUCTION POSSIBILITY FRONTIER A. INTRODUCTION In the past the issue of large vs. small scale oriented policy in LDCs has caused controversy, in terms of its effects on economic growth and income distribution. Some have said ’Small is beautiful’,while others have argued ’Large is efficient’. Therefore it seems to be necessary to develop a simplified version of models with a labor market and production possibility frontier(PPF), which can clarify some of the issues about the impact of industrial structure. The following models consist of three sectors(Urban large scale, Urban small scale, and Rural sectors) and two inputs(Capital,Labor). First I begin with the factor market model(Labor market). B. Model with Labor Market 1. Assumptions The model consists of three sectors(Urban large scale sector, Urban small scale sector, Rural sector) and two factor: scale secto: the rural product). three sect Finally f assumed in sector lab specific-fa 2. Ma Maths follows ; Urbar WE Urba: Rura differem: 42 two factors(Labor, Capital). The large scale and small scale sectors produce different manufactured products and the rural sector produces an agricultural product(Ag— product). Labor is homogeneous and mobile between the three sectors, while capital is fixed and specific. Finally full employment and a perfect competition are assumed in the labor market model. This framework of three sector labor market model evolved from the two country specific-factor model by Ethier (1983). 2. Mathematical Framework Mathematically the model can be expressed as follows; I Urban large scale sector (ULS sector) production : X1 = f(N1,§1) fN> o,fK> 0,fNN< o,fKK< o,fNK> 0, wage : PlafN = W1 rent : Pl-fK = R1, where Nland K1 are the labor and capital in ULS sector. Urban small scale sector (USS sector) production : X2= g(N2,E;) gN> 0,gK> 0,gNN< 0,gKK< 0,gNK> 0, wage : PZ-gN = W2 rent : PzegK = R2,where P2 is the output price in USS sector. Rural sector has no underemployment and hires different capital such as land, technology etc. production : X3= h(N3,K3) Total Total 3. Ge Figur sector 131 output, e expressed 2 ULS : USS Rura (Not the emplc)y line a( S the ULS se line of t} the Wage The curv, 43 hN> 0,hK> o,hNN< o’hKK< o,hNK> 0, wage : P3*hN= W3 rent : P3*hK= R3 Labor market equilibrium condition : Wl=W2=W3 , or P1*fN=P2*gN=P3*hN, or fN=P2/ 131* 9N=P3/ P1 * hN' Total labor : N. =N1+N2+N3 Total capital : E IEI+ET+ES 3. Geometric Framework Figure 3.1 shows the equilibrium point of three sector labor market graphically. At equilibrium , the output, employment, wage and rent of each sector can be expressed as follows: ULS sector: employment; 55 output;E]OCFK wage share 7DOHFC rent share; kHFK USS sector: employment; A5 output;DABEI wage share ;C3ABEG rent share; b.EGI Rural sector:employment; OE output;ClOBEJ wage share;E]OBEH rent share; A EHJ (Note: In Figure 3.1, the horizontal line AB shows the employment movement among three sectors. The vertical line 5% shows the real wage in terms of output price in the ULS sector, which is W/Pl. The curve RE shows the wage line of the ULS sector, which means fN. The curve IE shows the wage line of the USS sector, which means PZ/P1* 9N The curve E3 shows the wage line of the rural sector, Labor M (ULS 44 Figure 3.1 Labor Market Equilibrium in Three Sectors (ULS, USS, and Rural sector) W/P,(real wage) r which mean; defined in can be deri E and F. because 53 4. Li a. i assumed 1 Common pher applied 01 b. a assumed p. market wll in many L .because 0 asSulllption a Simuliti A th to Clarit poliCieS C The 45 which means P3/P1* hN . If the production function is defined in a Cobb-Douglas form, a demand curve of labor can be derived. The equilibrium points of labor market are E and F. Total labor is given as 5A + SE = CC + CD = OD because 5A is equal to CB.) 4. Limitations of the Model a. full employment assumption: In the model we assumed full employment. But high unemployment is a common phenomenon in many LDCs. Therefore the model can be applied only to the fast growing economies. b. an equilibrium in real wage : In the model we assumed perfect competition. Any disturbance in the labor market wll spontaneously clear the market. But it is true in many LDCs that a wage gap exists among the sectors ,because of the productivity differentials. Therefore this assumption is a little more difficult to sustain. In such a simplified model,however, it remains valid. A three dimension PPFs model gives us an opportunity to clarify the effects of the large scale oriented policies on the output production of each sector. C. Model with Three Dimensional PPFs 1. Assumptions The model consists of three sectors (Urban large scale secto and two fac produce di sector pro homogeneous and specifi competition the three c‘ model of (1974). 2. Ma Math: fellows: ULS . USS Rura Tota Tota With different VS ~ rural , dimension: Fig, 46 scale sector,Urban small scale sector, and Rural sector) and two factors (Labor, capital). The ULS and USS sectors produce different manufactured products and the rural sector produces a homogeneous Ag-product. Labor is homogeneous and mobile among sectors but capital is fixed and specific to each sector. Full employment and perfect competition are assumed in the model. This framework of the three dimensional PPFs model originated from the PPFs model of three commodities & two factors by Sheikh (1974). 2. Mathematical Framework Mathematically the model can be expressed as follows: ULS sector production : Xl= f(Nl,K1) USS sector production : X2= g(N2,K2) Rural sector production : X3= h(N3,K3) Total labor : N. =N +N +N Total capital: E. =§-+E—+E— With the above assumptions we can draw PPFs in three different ways between two sectors (Large vs. small, large vs.rural, small vs. rural). Now we can call it a three dimensional PPFs model. 3. Geometric Framework Figure 3.2 shows the equilibrium point of the three lhrer Fron‘ 47 Figure 3.2 Three Dimension Production Possibility Frontier '. 3 _PA/ 1 M C P /.',I 3 A“! l l ‘ ‘ l a O I ( . i i , \ \‘ \ XL ‘ - 33 dimensions can be 1 line. 1) I Fig) frontier 7 tariff i: output p. price(1>2 / price of a releas curve exp TEIatiVe is p. Rchzyns]. ' inCr938e5 declines 2) Fie frontier tariff °“tput ‘ Price“:3 54 dimensional PPF model. The output changes of each sector can be further analyzed with the aid of the Rybczynski line. 1) Description of the Geometric Form in Figure 3.5—1 Figure 3.5-1 shows the production possibility frontier curve between the ULS and USS sectors. Before the tariff is imposed on the ULS sector, the equilibrium output point of PPF(RS) is at N with given relative price(P2/P After the tariff is imposed, the relative 1 )‘ price of P3 /p1 will decline to P3/Pl‘, which will lead to a release of labor from the rural sector. Now the PPF curve expands to the new PPF curve (6T). With a declining relative price(P2 /P£ ), the new output equilibrium point is P. The line going through P and N is called the Rybczynski line(R line). The output of the ULS sector increases from 5? to 5W. But the output of the USS sector declines from OE to 5A. 2) Description of the Geometric Form in Figure 3.5-2 Figure 3.5-2 Shows the production possibility frontier curve between the ULS and rural sectors. Before a tariff is imposed on the ULS sector, the equilibrium OUtPUt point of PPF(RU) is at L with given relative price(P3/Pl). After the tariff is imposed, the relative I . Price of P2, /P1 declines to P2 /P , which leads to a .19 To 7.7 \AA_ U 13:14 nqfiv 190v J Monumv. rf...‘~\~ ~ firsanUUWn n~Do o P. pivlfin~flw ...~ n~no Tu duvmvmums mum . m 0.2% .2 m iJfiX . . 4» AV . _ :— c we _ _ . _ _ «in L _ a a / _ a, L .3 \ A _ x 5 \ _ fir.H|i-iI< _\ a 3.11.... is. \\ H of? _xmcxxoczx : ax “Cu_;sc_p kn. Leouom mg: :0 LDQDDO scLOuoom :0 muocwum m . m chow E Him. . m Or~5w w; release of expands to price(P 3 / line going line). Oui output 0: 3) ' Fig frontier tariff 1 output. pc price(P3 Price 01 reSpectii U53 and to is. output and F is If °f the 5 t° the FiSure ! t° 5A. ‘ 0C. 56 release of labor from the USS sector. Now the PPF curve expands to the new PPF curve(QV). With declining relative price(PZ3/qf), the new output equilibrium point is M. The line going through L and M is called the Rybczynski line(R line). Output of ULS sector increases from OR to OW. But output of the rural sector declines from OD to 5C. 3) Description of the Geometric Form in Figure 3.5-3 Figure 3.5-3 shows the production possibility frontier curve between the USS and rural sectors. Before a tariff is imposed on the ULS sector, the equilibrium output point of PPF(fik) is at G with a given relative price(P3 /P2 ). After the tariff is imposed, relative price of P2 /P1 & P 3 /P1 declines to R2 /Pil& P3 /%- respectively, which leads to a release of labor from the USS and rural sectors. Now the PPF curve shrinks from ER to £3. With relative price (P3/P2) unchanged, the new output equilibrium point is F. The line going through G and F is called the Rybczynski line(R line). If the R line is above the 450 line(OE), the output of the small scale sector declines more than in proportion to the decline in the output of the rural sector. In Figure 5.3, the output of the USS sector declines from SE to 5A. The output of the rural sector declines from 5D to 07:. But if the R line is below the 45° line(cTs), the- output . proportio: The outco the rural d.T Wit competiti the out lead to a sector, r. and the decline c So °n the U: dimensim , Scale or ‘in the PPF mode 57 output of the rural sector declines more than in proportion to the decline in the output of the USS sector. The outcome depends on the character of the USS sector and the rural sector of each LDC. d.The Result of Tariff Policy on ULS sector With the assumption of full employment , perfect competition, and heterogeneous products, the increase in the output price of the ULS sector due to a tariff will lead to an expansion of employment and output in the ULS sector, while the output and employment in the USS sector and the rural sector will decline. The real wage will decline due to the decline of relative price. So far we have analyzed the impact of tariff policy on the ULS sector in the labor market model and the three dimensional PPFs model. Next we analyze another large scale oriented policy, which is that of a capital subsidy in the ULS sector, in terms of the labor market model and the three dimensional PPFs model. 2. Capital Subsidy Policy in ULS Sector A capital subsidy policy can be analysed in two ways -in the labor market model and in the three dimensional PPF model. a. If policy fa' on loans ULS secto more cap sector p Because labor wi] to the U] adjustmei 1) Be injectio =p3 /P: hiSher n U15 sect U15 sec “33 and the 5hr) the Wagl Finally 58 a. Labor Market Mechanism If the government practices a capital allocation policy favoring the ULS sector through low interest rates on loans etc., the marginal productivity of labor in the ULS sector will rise, because labor in the ULS sector has more capital to work with. This higher MPL in the ULS sector push the real wage in the~ ULS sector upward. Because of this disturbance of the wage differential, labor will start to migrate from the USS and rural sectors to the ULS sector in order to restore equilibrium. The adjustment process is as follows: 1) Adjustment Process ( Figure 3.6) Before a disturbance is created by the capital injection, equilibrium points are I and V where PZ/P1 *g =P3 /l>:1 * hN= fiu' Now with an easy capital policy the higher MPL in the ULS sector shifts the wage curve of the ULS sector up to the right(f1;). A higher real wage in the ULS sector can attract AB (=6S) amount of labor from the USS and the rural sector into the ULS sector. Considering the shrunken labor force in the U88 and the rural sectors, the wage curve of the USS sector shifts to the right(5W). Finally the new equilibrium points are J and U. 2) Description of the Geometric Form at Equilibrium At a new equilibrium point after the adjustment Lab 59 Figure 3.6 Labor Market Adjustment after Capital Subsidy Policy on ULS Sector full employment ’\> at l_\ u: I ._.,_—— —. “gun—u- _. process in ULS emp] out} wag« ren‘ USS emp out wag rer Ru] em) on wa re No PPFs M the U: the m the th capital 60 process in Figure 3.6, ULS sector employment : 5R which is higher than previous 56. output: DORUX which is larger than previousDOQVP. wage share: DORUL which is larger than DOQVM. rent share: [SLUX (previous BMVP) USS sector employment: ED which is lower than previous AC or 52. output:EjBDJG which is smaller than previoustjACIF. wage share:[j BDJY (previous QACIE) rent share: ijJG which is smaller thantSEIF. Rural sector ‘ employment: 55 which is lower than previous' 5C. output:C]ODJK which is smaller than previouslfiOCIK. wage share:EJODJL (previous E3OCIM) rent share: AJLK which is smaller than AIMK. Now we turn to analysis in the three dimensional PPFs model. b. Three Dimensional PPFs Mechanism With capital subsidy on the ULS sector, the PPF of the ULS sector expands but the PPF of the USS sector and the rural sector does not change. The relative prices of the three sectors do not change in this process. The capital inflow into the ULS sector will lead to expansion of ULS productio: 1) Bet eguilibri into the from B t< expands : rural 5. relative e(Illilibr Ontput: Sect-'0r 61 of ULS production but reduction of U88 and rural production. 1) Adjustment Process ( Figure 3.7 ) Before a distortion by a capital subsidy policy, the equilibrium is L, given P1,P2,P3. With the capital inflow into the ULS sector, the PPF for the ULS sector expands from B to C. The PPF curve for the ULS and the USS sectors expands from AB to AC and PPF curve for the ULS and the rural sectors expands from BR to CE. With no change of relative prices among the three sectors, the new equilibrium production point is M. 2) Description of the Geometric Form at Equilibrium At the new equilibrium point after the adjustment process in Figure 3.7, the output of the ULS sector rises from 5N to OS, the output of the USS sector declines from OR to EC and the output of the rural sector declines from NR to SE. C. Analysis of Sectoral Output Change with Rybczynski Line. We have seen in the three dimensional PPFs model that a capital subsidy on the ULS sector will affect the outputs of all three sectors. The output changes in each sector can be further analysed with the Rybczynski line- Figure 3.7 62 Effects of Capital Inflow on ULS Sector ional PPFs Model X \ in Three Dimens 1) 1 Fig‘ frontier capital i point on ). After curve exp price(Pz, going t) line). . 5}, whi to 56, 2) Fi frontie] the ca) Output Price sector, Change POint Rybczy, increa: declim 63 1) Description of the Geometric Form in Figure 3.8—1 Figure 3.8-1 shows the production possibility frontier curve between the ULS and USS sectors. Before the capital inflow into the ULS sector, an equilibrium output point on PPF(§I) was at A with given relative price(PZ/Pl ). After the capital inflow into the ULS sector, the PPF curve expands from ET to SI. With no changes in relative price(Pz/P the new equilibrium point is B. The line 1). going through A and B is called the Rybczynski line(R line). The output of the ULS sector increases from OE to SF, while the output of the USS sector declines from 5D to DC. 2) Description for Geometric Form in Figure 3.8-2 Figure 3.8-2 shows the production possibility frontier curve between the ULS and rural sectors. Before the capital inflow into the ULS sector, the equilibrium output point of PPF (EU) was at L with given relative price (P /P1). After a capital inflow into the ULS 3 sector, the PPF curve expands from EU to 63. With no change in the relative price (PB/Pl)’ the new equilibrium point is K. The line going through L and K is the Rybczynski line (R line). The output of the ULS sector increases from 5D to 5F,while the output of rural sector declines from 56 to 5D. FCUUGW Tip: 00:... 30_WCH _fi..q.n:\.uv \AE DSCUZC pawl—OUUUWJ. CO WUUODNH r. - m, ULSM a: 64 mlw.m osswwb NA. . m 9:3 2 roboch me: coca ' 30—wCH alm.m osjwmm A i i x H c u c _;\As . //. ’ ~ // a Q 4L I \ x/ ./ <, a . , , , s , .. a ,, /,. cccs s , .x _cu41mo >2 ussusc FcLODOom co unbound x.m assure 3 ) Fig frontier the cap output p price(P3 the labc ULS sect no char equilibi the RYDt decline: also de B Output PrOport Sector, 0f the the a. occurs 38¢.th °°mpet increa 65 3) Description of the Geometric Form in Figure 3.8-3 Figure 3.8-3 shows the production possibility frontier curve between the USS and rural sectors. Before the capital inflow into the ULS sector, the equilibrium output point of the PPF(MU) was at R with given relative price(PB/PZ). After a capital inflow into the ULS sector, the labor migration from USS and rural sectors into the ULS sector makes the PPF curve shrink from MD to UT. With no change in the relative price (P3/P2), the new equilibrium point is S. The line going through S and R is the Rybczynski line(R line). The output of the USS sector declines from 5D to OC and the output.of the rural sector also declines from OD to 5P. ' But if the R line is above the 450 line(OT), the output of the small scale sector declines more than proportionately to the decline in the output of rural sector. If the R line is below the 45°line(5§), the output of the rural sector declines more than proportionately to the decline in the output of the USS sector. Which one occurs depends on the character of the USS and rural sectors of each country. d. The Result of a Capital Subsidy on the ULS Sector With the assumption of full employment , perfect competition, and heterogeneous products, a capital increase in the ULS sector will lead to an expansion of output a employmer sector l the capi‘ So sector dimensio small 5: capital model at an anal} Produoi the us: Wage c this Start sector adihst 66 output and employment in the ULS sector, while the employment and output in the USS sector and the rural sector will decline. The real wage will increase due to the capital increase. So far we have analyzed a capital subsidy on the ULS sector in the labor market model and the three dimensional PPFs model. Now we will analyse the impact of small scale oriented policies, which are a tariff and a capital subsidy on the USS sector, in the labor market model and the three dimensional PPFs model. We begin with an analysis of tariff policy. 3. Tariff Policy in the USS Sector Tariff policy can be analysed in two ways: in the labor market model and in the three dimensional PPF model. a. Labor market mechanism If the government imposes a tariff on the USS product in order to protect the USS sector, the price of the USS product ( Pz/Pl) will rise. Therefore the real wage of the small scale sector becomes higher. Because of this disturbance of the wage differential, labor will- start to migrate from the ULS and rural sectors to the USS sector in order to attain a new equilibrium. The adjustment process is shown in Figure 3.9. 1) Adjustment Process in Figure 3.9. 67 Figure 3.9 Labor Market Adjustment after Tariff Policy on the USS Sector [w/P, J \ Y , K ‘ 3' .. /Pg Oh) i ’/PI 7'. t we. Z P T— ”x— ‘ ‘ I t i . i‘ ,’ k 133/PI "brill l ‘r‘ I . y B CLP IE 0 F jb 1.h ié‘““‘-“‘-—-—“N L155 ---———9(—-N NI'M- ——-NuL§ , .n—. —— _ -— _—.. g- 1 I 0 e5, 5 i . Be: points with tar the USS equilibr higher. 6%, F71) sector. sector, lettu‘fi; V. 2 A Process C 68 Before a distortion by tariff policy, equilibrium points are L and W where P2 /P1 *g =P3 /P1*h =f . Now N N N with tariff policy, P2 rises to P; and the wage curve of the USS sector shifts up to the right(§N), where equilibrium wage between the USS and rural sectors becomes higher. Because of the wage differential between OD and OR, F§(=AD) labor migrates from the ULS sector to the USS sector. Because of the expanded labor force in the USS sector, the wage curve of the USS sector shifts to the V left(JM). Finally, the new equilibrium points are M and V. 2) Description of the Geometric Form at Equilibrium At the new equilibrium point after adjustment process in Figure 3.9, ULS sector employment: 5C which is lower than previous OE. output:[]OGVT which is smaller than previousCJOHWT. wage share: DOGVQ (previous E3 OHWP) rent share: B QVT which is smaller than previous h PWT. USS sector employment: AD which is higher than previous DC. outputztjADMJ which is larger than previous CJBCLK. wage share: l3 ADMZ which is larger than previous D BCLX . rer Ru] em) on wa re the USS /P1&P higher to an j the 1< output rural equilf on thr Will- not c new e 69 rent share: is ZMJ (previous QXLK) Rural sector employment: OD which is smaller than previous 5C. output:ClDOSM which is smaller than previousE3COSL. wage share: E3 DOQM (previous DCOPL) rent share :leQS which is smaller than previous ALPS. b. Three Dimensional PPFs Mechanism With a tariff on the USS sector, the output price of the USS sector will rise and the relative prices such as P2 /P1 & P2 /P3 will rise, while P3/P1 does not change. The higher price for the products of the USS sector will lead to an increase in the production of the USS Sector, while the lower relative price of the ULS and rural sector output will lead to a decrease of output in the ULS and rural sectors. 1) Adjustment Process in Figure 3.10 Before a distortion by tariff policy, the' equilibrium point is B, given P1,P2,P3 . Now with a tariff ' 'dPP&PP on the USS sector, P2 rises to P2 , an 2 / 1 2 / 3 will rise to P; /P1 & P2‘/P3respectively. But P3/P1 does not change. Because of the change in relative prices, the new equilibrium production point is C. 2) Description of the Geometric Form at Equilibrium "‘Q" 70 Figure 3.10 Effects of Tariff Policy on USS Sector Product in Three Dimensional PPFS Model X) At process declines rises f1 decline: o W competi output to an sector, and the due to U33 5. margir riSe' work ‘ Wage : °f tn. the ' 71 At the new equilibrium point after adjustment process in Figure 3.10, the output of the ULS sector declines from EU to E5, and the output of the USS sector rises from 5} to 5%. The output of the rural sector declines from $6 to EE. c. The Effect of the Tariff Policy on the USS Sector With the assumptions of full employment, perfect competition, and heterogeneous products, the increase of output price in the USS sector due to the tariff will lead to an expansion of output and employment in the USS sector, while the output and employment in the ULS sector and the rural sector will decline. Real wage will increase due to the increase in the relative price. 4. a Capital Subsidy Policy in the USS Sector a. Labor Market Mechanism If the government practices a policy favoring the USS sector through low interest rates on loans etc., the marginal productivity of labor in the USS sector will rise, because labor in the USS sector has more capital to work with. This higher MPL in the USS sector pushes real wage in the USS sector upward. Because of this disturbance of the wage differential, labor will start to migrate from the ULS and the rural sectors to the USS sector in order to reach 1} Bq equilib: = f'N' in the Up to t in the sector force j shifts and U. Figure 72 to reach an equilibrium point. 1) Adjustment Process ( Figure 3.11 ) Before the distortion by a cheap capital policy, equilibrium points are K and M where P2 /P1 *gN =P3 /P1 *hN = fli' Now with the cheap capital policy, the higher MPL in the USS sector shifts the wage curve of the USS sector up to the right(§E) or (Pia/P].*gbi)° The higher real wage in the USS sector attracts fi(=§A) labor from the ULS sector to the USS sector. Considering the expanded labor force in the USS sector, the wage curve of the USS sector shifts to the left(ST). The new equilibrium points are T and U. 2) Description of the Geometric Form at Equilibrium At new equilibrium point after adjustment process in Figure 3.11 ULS sector employment: 5? which is lower than previous Si. output:E30WUE which is smaller than previousCIOLME. wage share: [1 OWUV (previous [:1 OLMP) rent share: BVUE which is smaller thankPME. USS sector employment: 62 which is higher than previous AN. output: E] QXTS which is larger than previous D ANKJ. 73 Figure 3.11 Labor Market Adjustment after Capital Subsidy Policy on USS Sector w/R full .i employment . . fl) ha ,/ / / / / / / / PPF moc the US the m output Proces to an and n the e inflo incre Sectc 74 wage share: [3 QXTR which is larger than previous 0 ANKZ rent share :BRTS (previous (3sz) Rural sector employment : 5? which is lower than previous 5N. output:fi}OXTD which is smaller thanffiONKD. wage share: DOXTV (previous DONKP) rent sharezATVD which is smaller than previouszflKPD. Now we turn to an analysis of the three dimensional PPF model. b. Three Dimensional PPFs Mechanism With a capital subsidy on the USS sector, the PPF of the USS sector expands but the PPFs of the ULS sector and the rural sector do not change. The relative prices of the outputs of the three sectors do not change during this process. The capital inflow into the USS sector will lead to an expansion of USS production but a reduction of ULS and rural production. 1) Adjustment Process ( Figure 3.12) Before the distortion by a capital subsidy policy, the equilibrium is K, given P1,P2,P3 . Now with a capital inflow into the USS sector, the PPF for the USS sector increases from R to B. The PPF curve for the ULS and USS A A sectors expands from AR to AB and the PPF curve for the 75 Figure 3.12 Effects of Capital Inflow on USS Sector in Three Dimensional PPFS Model X\ USS and in the the new 2 A process decline rises : declin. compet increa and ot in the wage 1 marks the 1 76 USS and rural sector expands from RE to BE. With no change in the relative prices of outputs of the three sectors, the new equilibrium production point is J. 2) Description of the Geometric Form at Equilibrium At the new equilibrium point after the adjustment process in Figure 3.12, the output of the ULS sector declines from Ml to fit, the output of the USS sector rises from 5E to 5B,and the output of the rural sector declines from TS to BF. c. The Result of a Capital Subsidy on the USS Sector With the assumptions of full employment, perfect competition, and heterogeneous products, the capital increase in the USS sector will lead to expansion of labor and output in the USS sector, while the output and labor in the ULS sector and the rural sector will decline. Real wage will increase due to the capital increase. 5. Minimum Wage Policy in the ULS Sector Minimum wage policy can be analysed in the labor market model. a. Labor Market Mechanism If the government imposes a minimum wage policy on the ULS sector, the wage level in the ULS sector becomes indepen sector. equilil: this c the fol l I 1 the UL rural proces USS at in Fig equil in th worke sect: r“re: thej 77 independent of wage determination in the USS and rural sector. The migration process determines the new equilibrium wage level in the USS and rural sectors. In this case, a labor market equilibrium condition becomes the following. W1=Wm>W2=W3or P1*fN>P2*gN=P3*hN or fN>P2/P1*gN=P3/Pl*hN Under the minimum wage system, the left-out labor in the ULS sector starts to migrate to the USS sector and rural sector in order to find jobs there. This migration process determines the new equilibrium wage level in the USS and rural sectors. The adjustment process can be shown in Figure 3.13. b. Adjustment Process ( Figure 3.13 ) Before a distortion by minimum wage policy, the equilibrium points are L and V where P2 /P1*gN= P3/P1*hN = f Now with a minimum wage policy , the real wage level N O in the ULS sector is determined at 53 or 5Wm. The left—out workers (EF) start to migrate into the USS and rural sector. Because of the expanded labor force in the USS and rural sectors, the wage curve of the USS sector shifts to the left(fik). The new equilibrium points are-K and U. c. Description of the Geometric Form at Equilibrium At new equilibrium point after adjustment process in 78 Figure 3.13 Labor Market Adjustment after Minimum Wage Policy on the ULS Sector m /P, I _L\ U] I l Figure 071W] hete the the and 79 Figure 3.13, ULS sector employment: 5E which is lower than previous 6F. output :EJOEUT which is smaller than previous[]OFVT. wage share: U OEUS (previous BOFVQ) rent share:b,SUT which is smaller than previous 5 QVT . USS sector employment: XE which is higher than previous ED. outputztlACKH which is larger than previousEJBDLI. wage share:ClACKG (previoustBDLJ) rent share:fi>GKH which is larger than previouskaLI. Rural sector employment: 65 which is larger than previous 65. output:CiCOPK which is larger than previous[jDOPL. wage sharezd CORK (previousEJDOQL) rent sharezzfiKRP which is larger than previous ALQP. Equilibrium wage in the USS and rural sector: 5% or 6R which is lower than previous Si or 56. d. The Effect of a Minimum Wage in the ULS Sector Given full employment, perfect competition, and heterogeneous products,the imposition of a minimum wage in the ULS sector will lead to decline of output and labor in the ULS sector, but a corresponding increase in the USS and rural sectors. The rent in the ULS sector will will de will in demand a maj01 Where USS Se assume produc diffe1 capita marke' anal} threc Prod the 80 will decrease, while the rent in the USS and rural sectors will increase. Given this loss of rent the capitalist may demand a compensatory rent subsidy. As yet this is not a major issue. 6. Tariff Policy in the ULS Sector in the Situation Where Homogeneous Products Are Produced by the ULS and USS Sectors Changing one of the previous assumptions , we now assume that the ULS and USS sectors produce the same products at the same price. The ULS and USS sectors use different production technologies and have different capital endowments. Under these assumptions, the labor market equilibrium condition becomes the following: fN =9N = P2 /P1 *hN (Note: P - the price of the USS and ULS sector 1 products, Pz- the price of the rural sector product) Now the tariff policy on the ULS sector can be analysed in two ways: in the labor market model and the three dimensional PPFs model. a. Labor Market Mechanism If the government imposes a tariff on the ULS product in order to protect the ULS sector, the price of the output of the ULS sector will rise and relative prices such a: USS a: distur migrat in ord proces shit and Fig); 81 such as Pz/Pl will decline. Therefore the real wage of the USS and rural sectors becomes lower. Because of this disturbance of the wage differential, labor will start to migrate from the USS and rural sectors to the ULS sector in order to reach an equilibrium point. The adjustment process can be shown in Figure 3.14. 1) Adjustment Process ( Figure 3.14 ) Before a distortion by tariff policy, equilibrium points are E and D where fN =gN =P2 /P1 *hN . Now with tariff policy, P1 rises to Pfi and the wage curve of the rural sector shifts down to the EM curve, where the equilibrium wage in the USS and rural sectors falls to the 5% level. Because of wage differential between 53 and Si, AP(=5N) labor migrate from USS and rural sectors to the ULS sector. Considering the shrunken labor force in the USS and rural sectors, the wage curve of the USS sector V O O O . shifts to the right(RS). The new equilibrium pOints are S and U. 2) Description of the Geometric form at Equilibrium At new equilibrium point after adjustment process in Figure 3.14, ‘ULS sector employment: EU which is higher than previous 53. output:EJOVUH which is larger than previousEJOJDH. 82 Figure 3.14 Labor Market Adjustment after Tariff Policy on the ULS Sector in the Homogeneous Products Case of ULS and USS Sector 1 /P‘ full employment I I I Lit, mode] Pric rela Pric int lows doc] 83 wage share:C30VUT (previousCJOJDI) rent share: bTUH which is larger than previousbIDH. USS sector employment: PW which is higher than previous AB. outputzfijPWSR which is larger than previousEJABEF. wage share:E) PWSQ (previousDABEC) rent share: BQSR which is larger than previous bCEF. Rural sector employment: 5W which is lower than previous 6B. outputcEJOWSL which is smaller than previousEiOBEG. wage share:CJOWST which is smaller than previous D OBEI. rent share: A STL (previous AEIG) Now we turn to analysis of three dimension PPFs model. b. Three Dimensional PPFs Mechanism With a tariff on the output of the ULS sector, the price of the output of the ULS sector will rise and the relative price ( P2 /P1) will decline. The higher output price of the ULS and USS sectors will lead to an increase in the production of the ULS and USS sectors, while the lower relative price of the rural sector will lead to a decrease in the production of the rural sector. 1) Adjustment Process ( Figure 3.15 ) rr 1 fl. «I. 84 Figure 3.15 Effects of Tariff on the ULS Sector Product in Three Dimensional PPFS Model Xx .0) 1 equili] of hom derive the ru three USS 5 0P. 1 dimen: urban ULS s rises Chang new 6 prOCI rise rise decl out; eQUE 85 Before the distortion by tariff policy, the equilibrium point is L, given P P . Under the assumption 1’ 2 of homogeneous products in the ULS and USS sector, we can derive the PPF curve of the urban sector(ULS and USS) and the rural sector(6B). At relative price Pz/P1 from the three dimensional PPF, the ULS sector produces PM and the USS sector produces PN, while the rural sector produces OP. If a 45° line passes through point L of the three dimensional PPF, 5N is equal to MR. Therefore the total urban output is PR. Now with a tariff on the output of the ULS sector (as well as the output of the USS sector), P 1 rises to P1' and P 2/P1 declines to P_2 /P1' . Because of the change of relative price on the three dimension PPF, the new equilibrium production point is H. 2) Description of the Geometric Form at Equilibrium At the new equilibrium point after adjustment process in Figure 3.15, the output of the ULS sector rises from PM to FE, and the output of the USS sector rises from PM to FT, while the output of the rural sector declines from 63 to SF. Using the 45° line technique, the urban sector output(ULS and USS) can be derived as SF because ET is equal to SG. c. The Result of a Tariff on the ULS Sector compet price will ULS se in the decli: model goods in t1 modei as t are Poli and EXP; rura tar exp rUI to 86 With the assumptions of full employment, perfect competition and homogeneous products, the increase in the price of the output in the ULS sector due to the tariff will lead to an expansion of the output and labor in the ULS sector and the USS sector, while the output and labor in the rural sector will decline. And the real wage will decline due to the decline in the relative price. E. Conclusion So far we have developed a simple three sector model(ULS, USS, and Rural sector), which is based on three goods and two factors(capital, labor). GeOmetric exercises in the labor market model and ,the three dimensional PPF model gave us an opportunity to clarify some issues such as the change in output,employment, and real wage , which are caused by large vs. small scale oriented policy. The analysis of the three sector model shows that a policy of boosting the large scale sector such as tariff and capital subsidy on the ULS sector leads to an expansion of the ULS sector but reduction of the USS and rural sectors in terms of employment and output. A policy of boosting the small scale sector such as tariff and capital subsidy on the USS sector leads to an expansion of the USS sector but reduction of the ULS and rural sectors in terms of employment and output. Also,a minimum wage policy in the ULS sector leads to a reduction ”of the ULS but expansion of the USS and urn-"r- “ha. rural homage sector produc but tc emploj emplo can b many analy tests 87 rural sectors in terms of employment and output. Finally,in the case where the ULS and USS produce homogeneous products, a policy of boosting the large scale sector such as tariff on the ULS(as well as USS) sector product leads to an expansion of the ULS and USS sectors but to a contraction, of the rural sector in terms of employment and output. Our analysis is based on the assumptions of full employment and perfect competition. Therefore our model can be applied to only fast growing economies, not to the many LDCs with high unemployment. Whether the theoretical analysis of the model is empirically true remains to be tested. 81 opted its c POPUI Open for 1 Of St comp. and dire of disc fac1 ihdl fav. 01’ 88 CHAPTER 4 .A REVIEW OF CHANGES IN THE KOREAN INDUSTRIAL STRUCTURE IN TERMS OF A THREE SECTOR MODEL & AN EMPIRICAL TIME-SERIES STUDY DURING 1966-1984 A. INTRODUCTION Since the early 1960’s, the Korean government has opted for an export-oriented developmental strategy,given its constraints of poor natural resources,high density of population, and a relatively small domestic market. .This open door policy resulted in an expansion of the markets for Korean products. With this widening market, economies of scale and the large size of firms assumed importance in competing in the world market; thus large firms emerged and small firms expanded into larger ones. In this process, the Korean government encouraged , directly or indirectly, the growth of LSEs at the expense of SSEs through various policy instruments. This discriminatory policy resulted in a- distortion of the factor market,particularly the capital market. The main industrial policy during the last two decades seemed to favor LSEs,to the detriment of SSEs, in allocating capital or credit for their expansion. This policy has become controversial because it has an im meanin struct (ULS: compel the l reviei with wage, SECtI inpu' PrOd sect and to and litan- 89 an impact on economic growth and equity. Therefore it is meaningful to analyse the change in the Korean industrial structure during 1966-1984, based on a three sector model (ULS, USS, and Rural sectors). In building up this three sector model, I did a comparative static analysis based on a capital inflow in the ULS sector. Secondly, based on this analysis , I reviewed the change in the Korean industrial structure with time series data (1966-1984) on employment, real wage, capital intensity , etc. B. Three Sector Model 1. Assumptions The model consists of three sectors ( ULS, USS,rural sector) and two factors (labor,capital) that are the inputs for the three ’sectors. The ULS and USS sectors produce homogeneous manufactured products and the rural sector produces agricultural goods. Labor is homogeneous and mobile between the sectors, but capital is specific to each sector. In this model we assume full employment and perfect competition. 2. Mathematical Framework If the ULS and USS sectors produce the same manufactured product, the output price of the two sectors has to Labor W1 =V :Agr; sect capi i th 3) i MI for mar hor 90 has to be equal. ULS sector Production X1=f(N1,K1) fN >0,fK >O’fNN0,gK >O’gNNO,hK >0,hNN<0,hKK 134 Table 5.5-2 Manufacturing Value Added distributed by Location Unit: Million Won Year Total Mfg. Urban Rural 1984 24,654,407 17,397,352 7,257,055 (Z) (100) (71) (29) Small Est. (5-99workers) 4,279,773 3,529,928 749,845 (Z) (100) (82) (18) Year Total Mfg. South4west South—east 1984 24,654,407 1,802,355 9,925,532 (%) (100) (7) (40) Small Est. » (5-99workers) 4,279,773 248,719 1,418,555 (%) (100) (6). (33) Note: Urban; Seoul,Pusan,Inchon,Kyonggi—Do,Kyongsangnam-Do Rural ; The rest of area South-west; Chollabuk-Do, Chollanam-Do South-east; Pusan,Taegu,Kyongsangbuk-Do, Kyongsangnam-Do Source: 'Report on Mining and Manufacturing Survey 1984' Economic Planning Board, The Republic of Korea. It between 15 correlated urban ma: labor mar} Whel region Do,Kyongs location share 0 Southwest %- As Ta manufactt Southeasi Th1 regional themselv. haVe b consider under-Sta 4. Manufac) T: positio: industr 135 It appears that because of 'a possible linkage between LSEs and SSEs, their location is positively correlated in the urban area. The SSEs seem to rely on the urban market with its purchasing power, and an urban labor market with its skilled or semi-skilled labor pool. When the Southwest (Chollabuk-Do,Chollanam-Do) region is compared to the southeast(Kyongsangnam- Do,Kyongsangbuk-Do) region, the regional differences in location of manufacture are distinct. In Table 5.5-1, the share of manufacturing employment , in 1984,in the Southwest is only 5 %, while that in the Southeast is 40 %. As Table 5.5-2 shows, the bulk of the value added in manufacturing by SSEs is from those located in the Southeast rather than the South-west. The distinct differences in location can lead to regional inequity ,and agitations in those areas which see themselves as unfairly treated; but this situation might have been due to optimal locational and economic considerations which the concerned population may not understand. 4. Time-series Relative Position of the SSEs in Manufacturing Industry Table 5.6 shows the dramatic change in the relative position of the SSEs within the overall manufacturing industry. During the last two decades, SSEs have been Ilimfi Distrib1 Es tabl: Year 5 1958 1 1963 1 19-75 1984 Sources ; For in Sta F01 3111 136 Table 5.6 Distribution of Manufacturing Employment by Establishment Size in Percentage Year 1958 1963 1975 1984 Sources: 5—9 10-19 20—49 50—99 100-499 500+ Total 17 16 21 13 21 12 100 15 14 16 12 21 22 100 4 5 8 9 30 44 100 4 6 12 12 28 38 100 For 1958, 1963, 1975, ' Small Scale Enterprises in Korea and Taiwan' by Ho, 1980, World Bank Staff Working Paper, No. 384, PP 27, Table 3.1. For 1984, 'Report on Mining and Manufacturing Survey 1984' issued at April, 1986 by Economic Planning Board, The Republic of Korea. sustaining employmen1 workers) %, alth01 trend ind are ECOHO‘ 5. The depends important be establ 0f scale SSEs. techniqu. calculat industry the est in their De [SurViVC cenSus w°rkers Plants1 °utput unable 137 sustaining a loss in the share of manufacturing employment. In 1958 , the employment share of SSEs(5-99 workers) was 67 % but, by 1984 , this dropped to only 34 %, although it reached a bottom of 26 % in 1975. This trend indicates that the SSEs became larger and that there are economies of scale. 5. Efficient Plant Size The relative position of SSEs in a given industry depends on whether or not economies of scale are important. The minimum efficient size of a plant needs to be established on the available evidence about economies of scale in order to understand the dynamic change in SSEs. For this purpose I have applied the ’Survivor technique’. By definition, the Survivor technique calculates for two points in time the share of an industry’s output(or the value of shipment) by the size of the establishment. Those groups which produce increases in their shares are presumed to be efficient. Despite the convenience in calculation, this ’Survivor technique’ has the following limits: first, the census bases its.establishment data on the number of workers in each plant. If innovation is labor saving, all plants may appear to be shrinking faster than their true output levels are. But the actual survivor tests have been unable to discern coherent shift patterns in most industries size , as as the pl: long run 2 estimated Thirdly, and benef efficienc ’Social E In four-dig: 1984. Th( the Surv industri and 1984 categori a indust Value ir Size, 1 identif: is cont Change, C939 of T t° eff 5.7-1 138 industries(Shepherd 1979). Secondly, the efficient plant size , as determined by a survivor test , is not the same as the plant size corresponding to the lowest point on the long run average cost curve, either outside or inside the estimated range of optimal sca1e(Ho 1980, Shepherd 1979). Thirdly, because the test considers only the private cost and benefit, the Survivor test does not measure social efficiency. For social efficiency, we need to apply a ’Social Benefit and Cost analysis’ ( Chapter 6). In conducting this ’Survivor technique’, I selected four-digit breakdowns in Korean manufacturing, in 1979 and 1984. Then I identified those similar industries for which the Survivor technique is applicable. As a result, 72 industries were selected for two time periods between 1979 and 1984. Next, I compared each industry’s shares by size categories of the value of shipment in 1979 and 1984. When a industry in a given size category shows a positive(+) value in Table 5.7—2, this is identified as the efficient size, but when it shows a negative(-) value, this is identified as the inefficient size. A size class , which is contiguous to an efficient size but registering no change, is identified as efficient. This occurred in the case of jewelery production. Table 5.7-1 shows the number of industries according to efficient sizes during 1968-75 and 1979-84. As Table 5.7-1 shows, of the 72' industries, 14 are in the food Efficie Industry group(00) F00d(31) 1979-84 (1968-75) Iextile(32' 1979-84 (1968-75) Wood(33,34 1979-84 (1968—75) Chemicals( 1979-84 (1968-75) Non-metals 1979-84 (1968-75) 1979-84 (1968-75) Fabricate< 979-84 (1968-75) Other Mfg 1979-84 (1968-75) Total 1979‘84 139 Table 5.7-1 Efficient Plant Size according to 'Survivor Criterion': 1968-1975 and 1979-1984 Industry # of Industry # of Industry with efficient group(00) in group , plants in size category 5—49 50-99 100-199 200—499 500+ Food(31) 1 1979-84 14 5 10 9 10 2 (1968-75) (11) (2) (4) (6) (6) (3) Textile(32) 1979-84 11 8 9 6 4 4 (1968-75) (15) (2) (3) (8) (11) (11) Wood(33,34) 1979-84 8 4 5 4 5 2 (1968-75) (8) (1) (2) (4) (6) (4) Chemicals(35) 1979-84 9 8 5 6 5 3 (1968—75) (5) (1) (1) (3) (3) (2) Non-metals(36) 1979-84 5 3 4 2 1 2 (1968-75) (7) (0) (3) (4) (3) (3) Basic metals(37) 1979-84 2 1 1 1 1 2 (1968-75) (1) (0) (0) (O) (0) (l) Fabricated metal(38) 1979-84 19 10 7 7 7 13 (1968-75) (12) (1) (3) (7) (8) (5) Other Mfg.(39) 1979-84 ' 4 3 3 2 1 3 (1968—75) (4) (0) (1) (2) (2) (3) Total 1979-84 '72 42 43 37 34 31 (1968—75) (63) (7) (17) (34) (39) (32) Sources: For 1968-75 data,'Small Scale Enterprises in Korea and Taiwan' by Ho.1980,World Bank Staff Working Paper, No384, Table 3.5. For 1979-84 data,'Report on Mining and Manufacturing Survey 1979,1984' by Economic Planning Board, The Republic of Korea. 'Surviy Korear 1984 Indus Code Descr Food(31) 3111 Land a1 3112 Dairy 1 3113 Fruits 3114 Fish,C' 3115 Vege.a 3116 Grain 1 3117 Bakery 3l18 Sugar 3122 Food p 3123 Animal 3131 Distil 3132 Wine 3133 Malt l 3140 Tobacc Textile(32) 3211 Sllk&) 3214 Textid 3215 Knitti 3216 Carpet 3212 Weavir 3213 Bleac) 3219 Texti? 322 Wear,) 3231 Tanne- 3233 Leath 3240 Footw “00mm 3311 Sani 3312 Woode 140 Table 5.7—2 'Survivor' Estimates of Efficient Plant Size for Korean Manufacturing Industries; 1979—84 1984 Industry Changes in the Percentage Share of Value of Code Description Shipment by Size of Establishmentgl979-84 5-49 50-99 100-199 200-499 500+ Food(31) 3111 Land animal meat +.3 +4.5 +7.0 —3.5 -8.2 3112 Dairy products +2.3 +2.5 + .1 +29.6 —34.6 3113 Fruits,vegetable+7.3 -4.3 -10.4 +6.0 +1.5 3114 Fish,crustacea +5.6 +2.9 + .3 —6.2 —2.6 3115 Vege.animal oil -9.9 —1.0 —8.0 +67.4 -48.6 3116 Grain mill -11.6 + .6 +5.1 +5.8 3117 Bakery products —2.0 + .4 +2.7 + .9 —2.1 3118 Sugar refinery - .6 — .3 +16.5 —15.5 3122 Food prod.N.E.C.—1.9 +5.6 +3.8 -1.5 -6.0 3123 Animal feeds —2.5 +1.8 -2.5 +5.6 -2.3 3131 Distilling&etc. +'.2 +4.8 +3.0 +2.4 -10.3 3132 Wine —17.7 +1.0 +5.9 +10.8 3133 Malt liquors —36.1 +36.1 3140 Tobacco + .3 + .2 +1.2 -1.6 Textile(32) . 3211 Silk&yarn spin - .2 +2.5 — .9 — .6 - .8 3214 Textile,wearing +6.7 +3.5 +4.4 —4.4 -10.3 3215 Knitting mills +7.6 +2.0 -3.1 -6.8 + .2 3216 Carpet & rugs -15.0 —5.9 +20.8 3212 Weaving textile +2.7 + .2 + .6 +2.3 —5.8 3213 Bleaching,dye +3.1 +1.1 -5.7 -4.9 +6.4 3219 Textiles N.E.C.+16 +11 +13.1 +8.4 —48.6 322 Wear,exc.footwear—1 +5.1 +6.7 +5.3 -16.2 3231 Tannerie&leather+4.1 +14.3 - .6 + .4 -18.3 3233 Leather +16.5 +4.2 +1.9 —26.7 +4.0 3240 Footwear +4.7 - .2 —6.3 - .4 +2.4 Wood(33,34) 3311 Sawmill,plan +21 +5.4 +3.7 + .1 ~30.l 3312 Wooden,cane -4.3 +3.2 -23.5 +24.6 3319 Wood&cork,N.E.C.+18.4 -14.9 —3.5 3320 Furniture&fix. -7.7 +1.7 —2.7 -1.9 +10.7 3411 Pulp,paper -11.2 + .2 +1.2 +13.1 —12.3 3412 Cont. of paper +7.2 + .1 -2.1 -5.3 3419 Pulp,paper,N.E.C-6.7 -10.3 +8.6 +6.5 +1.9 3423 Print allied +17.4 -2.6 +2.5 +2.3 —19.6 Chemicals(35) 3521 Paint&lacquare + .4 —10.2 —7.2 +4.5 +12.5 3522 Drug&medicine +1.4 +3.9 +2.2 +4.5 -12.1 3523 Soap&cleaning +1.9 +1.4 +4.4 -1.4 -6.4 1984 Indus) Code Descri 3529 Chemica 3530 Petrol) 3540 Misc. ] 3551 Tyre & 3559 Rubber 3560 Plasti Non-metalli 3610 Potter 3620 Glass 3691 Clay 3692 Cement 3699 Non—me minera Basic Metal 371 Ir0n&s 372 Non—fe Fabricated 3811 Cutley 3812 Fur. c 3813 Struc. 3821 Engine 3822 Ag. me 3823 Metal 3824 Spec.r 3825 Office 3829 Machir Elect 3831 Eleet 3832 Sound 3833 Elect 3841 Shi Pb 3342 Railr 3873 Motor .141 Table 5.7-2(cont'd.) 1984 Industry Changes in the Percentage Share of Value of Code Description Shipment by Size of Establishment31979—84 5—49 50-99 100—199 200—499 500+ 3529 Chemical,N.E.C +1 -1.7 —18.8 +24.5 -5 3530 Petroleum ref. - .5 - .3 + .5 —34.2 +34.6 3540 Misc. petroleum+3.4 +7.3 +2.4 —3.6 —9.4 3551 Tyre & tube + .3 —1.3 - .4 + .1 +1.2 3559 Rubber N.E.C. +2.3 +1.2 +2.4 +2 —7.9 3560 Plastic N.E.C.+10.7 +8.9 +6.9 -9 -17.5 Non-metallic Mineral(36) 3610 Pottery&china —1.3 —2.2 +3.7 —10.3 +10.1 3620 Glass +1.2 +2 -7 —22.6 +26.3 3691 Clay +4.5 +4.0 -7.8 +5 -5.9 3692 Cement,lime —1.6 + .8 +13.3 -2.0 —10.6 3699 Non-metallic mineral +16.1 +14.3 —2.1 —12.6 -15.7 Basic Metal(37) 371 Iron&steel + .5 + .1 -2.2 +1.0 + .6 372 Non—ferrous met.-1.0 -1.6 +3.1 -12.4 +12 Fabricated Metal(38) 3811 Cutley,tool -3.4 +6.7 - 5 -4.2 +1.4 3812 Fur. of metal -4.7 +4.7 +1.9 -1.8 3813 Struc. metal + .2 —11.7 , -2.6 —6.5 +20.5 3821 Engine&turbine -7.0 —15.4 +12.4 +10.2 3822 Ag. machine + .8 +6.3 - 9 —8.1 +2.0 3823 Metal machinery+25.4 +9.7 +8.2 +8.6 -51.8 3824 Spec.machin. +6.9 -2 -5.3 -14.7 +15.0 3825 Office machin. +1.9 —12.6 + .2 -2.1 +12.7 3829 Machine exc. . Electric N.E.C.+ .3 —1.2 + .9 + .8 — .8 3831 Elect.machine - .9 -2.9 -8.8 -10.4 +23 3832 Sound,Image eq.+ .9 - 5 -1.3 —4.3 +5.1 3833 Elect.appl. +1.3 +2.5 +3.2 +3.3 -10.4 3841 Shipbuilding —4.2 -2.2 ~ .1 -1.6 +8.2 3842 Railroad eq. -2.2 -4.1 +2.9 +1.5 +1.9 3843 Motor vehicle +1.5 + .5 + .9 +2.9 -5.8 3844 Mot.&bicycle —1.1 —9.8 —5.1 — .2 +16.3 3849 Transport eq. N.E.C. —56 —33.6 +89.6 3852 Photogra.goods + .9 +4.0 —3.8 -5.3 +4.1 3853 Watch&clock -1.6 —3.4 -1.1 -26.9 +32.9 Other Manufacturing 3901 Jewellery +2.3 +13.1 0 +3.5 -18.7 3902 Musical Ins. + .2 -1.2 —2.6 —10.2 +13.9 3903 Sporting goods +4.9 +3.5 —4.3 +15.1 +12.2 3909 Mfg.N.E.C. -11.8 -1.5 +1.4 -18.2 +7.7 Sources: Korea. , 'Report on Mining and Manufacturing Survey 1979, 1984' by Economic Planning Board,The Republic of group , a industrie 1979-84. had effic in the category workers; the Surv sizes of Th with in during 1 more eff 0f 63 ir 5'49 we] the 512. zoo-499 T future h°WeVer their e innovat 5.7-1 slight: the ss: indust 142 group , and the efficient plant size in 5 of the 14 food industries include the size category 5-49 workers during 1979-84. Between 1979 and 1984, 42 of the 72 industries had efficient plants in the size category 5-49 workers; 43 in the size category 50-99 workers; 37 in the size category loo-199 workers; 34 in the size category ZOO—499 workers; and 31 in the size category 500+ workers. Thus the Survivor criterion indicates that during 1979-84, all sizes of plant were fairly efficient. The situation in the period 1979 to 1984 contrasts with in the period 1968 to 1975. As Table 5.7-1 shows, during 1968-75, the statistics indicate that the LSE was more efficient than the SSE. Between 1968 and 1975, 7 out of 63 industries had efficient plants in the size category 5-49 workers; 17 in the size category 50-99 workers; 34 in the size category 100-199 workers; 39 in the size category 200-499 workers; and 32 in the size category 500+ workers. This data about an efficient size suggests that the future of the SSEs is not so pessimistic. It is true, however, that the LSEs have more opportunities to improve their efficiency through capital accumulation, technology innovation, and scale of economies than the SSEs. Table 5.7-1 shows that the LSEs with 100+ workers show a slightly higher frequency of being the efficient size than the SSEs with 5-99 workers. The LSEs are efficient in 102 industries, but the SSEs are efficient in only 85. 6. If or more c be predon predomina Following 1984 S: more of digit ma workers small sc Tl Sizable added i] the in] scale overall emPloye 8 3 of C small I manufat Value ; the t indust 143 6. Changes in the Competitive Bases of SSEs If SSEs with fewer than 100 workers account for half or more of the industry’s employment, they may be said to be predominant in an industry. An industry in which SSEs predominate is called a small scale industry(Ho 1980). Following this classification, Table 5.8 shows that in 1984 SSEs(those with 5-99 workers) accounted for half or more of the industry’s employment in 37 of the 104 four— digit manufacturing industries. SSEs employed 61% of the workers and produced 47 % of the value added in these 37 small scale industries. The establishments with 5-49 workers employed a sizable 44 % of the workers and produced 31 % of the value added in the same 37 industries. The statistics indicate the importance of SSEs with 5-49 workers in the small scale industry. As for their relative importance in overall manufacturing, SSEs in the small scale industries employed 13.4 % of all manufacturing workers and produced 8 % of the total manufacturing value added in 1984. On the other hand, SSEs in the 67 industries where small plants do not predominate employed 21.7 % of all manufacturing workers and produced 10.9 % of the total value added in the manufacturing industries, greater than the total volume of the small firm activity in the 37 industries, in which they were predominant. In those Share of S1 by Five—d1, 7 of Indus workers 511 Small Est. 75—100 50-74 25-49 0 -24 Total 7 of Indu: Workers i; Small Est 75-100 50-74 25-49 0 -24 Total SOUICQS : For 19 T331973] No.38) For 1 SUTVe 0f K0 144 Table 5.8 Share of Small Establishments in Employment and Value Added by Five—digit Industries:1975& by Four—digit Industrieszl984 Z of Industry Year # of Share of Total Mfg.Employment(%) workers in Industry Size of Industries Small Est.(5-99) 5-99 10—49 50-99 100-499 500+ Total 75-100 1984 10 .3 .9 2 3 O 1.7 1) (2.2) (1) (.4) (0) (4.6) 6.6 3.6 5.8 2.4 20.2 ) (3.1) (1.6) (3) (1) (10) 7.9 5.8 14.5 11.7 41.2 (1975) (86) ( ) (5.0) (3.6)(10.2) (6.6)(27) 8 3 3 .6 0 -24 1984 28 .5 3 8 2.4 7.0 23.2 36.9 5 9 4 (1975) (114) ( 50-74 1984 27 1. (1975) (61) (1. 25-49 1984 39 1 1 (1975)(112) (. ) (2:6) (2.6)(16.6)(36.1)(58.4) Total 1984 104 3. 19.2 12 27.6 37.3 100 (1975)(373) (4 )(13) (8.7)(30.2)(43.7)(100) Z of Industry Year # of Share of Total Mfg.Value Added(%) workers in Industry Size of Industries Small Est.(5-99) 5-99 10-49 50499 100-499 500+ Total 75—100 1984 10 .2 .5 .1 .4 0 1.2 (1975) (114) (.5) (1.4) (.9) (.3) (0) (3.0) 50-74 1984 27 .7 3.8 2.7 5.4 3.1 15.7 (1975) (61) (.5) (1.6) (1.1) (2.7) (1.6) (7.4) 25-49 1984 39 .5 3.8 3.5 13.0 12.4 33.2 (1975) (86) (.6) (2.4) (1.8) (7.4) (4.5)(16.6) 0 -24 1984 28 .1 1.5 1.5 7.2 39.6 49.9 (1975) (112) (.2) (1.4) (2.9)(18.4)(50.0)(72.9) Total 1984 104 1.5 9.6 7.8 26 55.1 100 (1975) (373)(1.7) (6.7) (6.8)(28.7)(56.1)(100) Sources: For 1975 data, 'Small Scale Enterprises in Korea and Taiwan' by Ho, 1980, World Bank Staff Working Paper, No.384, PP40, Table 3.7. For 1984 data, 'Report on Mining and Manufacturing Survey 1984' by Economic Planning Board, The Republic of Korea. industrie exist by (Staley a SSES Fo appraise (small 50 a study favoring ,future < follows Costs b and Gus economi S the a Predomj 145 industries where they are not predominant, SSEs seem to exist by "filling the cracks’ not occupied by the LSEs (Staley & Morse 1965). 7. The Factors Influencing Competitiveness in the SSEs - For the small entrepreneurs who would like to appraise particular product groups for feasibility of small scale manufacture in their own economic environment, a study of the economic and technical circumstances favoring SSEs can give some practical guidance for their future operation. The main factors favoring SSEs are as follows: a) locational incidence factors of high transport costs b) the need for personal contact between producers and customers c) a simple technology not based on scale economies , and d) market influences. Staley and Morse(1965) classified into 8 categories the above (factors that help determine the SSEs predominance in manufacturing industries. 1)Locational Influence 1A. Dispersed resources processors 18. Market oriented industries 1C. Service industries 2) Process Influence 2A. Separable manufacturing operations ZB. Craft handwork 2C 3) 3] Th: scale 1 5.9-2, 5 as 1A cl ; 3 as 2 Ta faciliti importax 55.3 1 industr (30.3%) to be urban SSEs matEIia 23.4% 26.4% ; P°licy SimPle separa 146 2C. Simple assembly,mixing and finishing operation 3) Market Influence 3A. Differentiated products having low scale economies BB. Industries serving small total markets This categorization was applied to 37 Korean small scale industries, where SSEs are predominant. In Table 5.9-2, 5 out of 37 small scale industries were identified as 1A class in 1984; 9 as 1B; 4 as 1C 7 6 as 2A ; 2 as 2B 7 3 as 2C; 8 as 3B. Table 5.9-1 shows that with better transportation facilities, the locational influence seems to be less important.(1A,1B,1C group employment; 59.5% in 1975 but 55.3 % in 1984) Within this category market oriented industry was growing larger in terms of employment (30.3%) and value added (32.3%) in 1984. The reason seems to be partly that the SSEs are concentrated around an urban area for the urban market. On the other hand, the SSEs in the industries which process dispersed raw materials were contracting in terms of employment (from 23.4% in 1975 to 19.6% in 1984) and value added (from 26.4% in 1975 to 16.6 % in 1984). Again, with a wider market due to an export‘oriented policy, the pattern of production in SSEs changes from a simple assembly and mixing method, to more advanced separable manufacturing stages. (2A. group employment Relative In Establishme Industry T} 1A DisperS) resourc 10 Market . industr 10 Service 2A Seperab operati 2B Craft h 20 Simple 33 Small I Market Total Industry 1 1A Dispen resour< 13 Market indust 1C Servic. Indus t- A Sepera' °Perat 3 Craft 33 Small Total SourceS: 147 Table 5.9—1 Relative Importance of Types of Industry in which Small Establishments(5—99) predominate: 1975 & 1984 Industry Type Year Employment in Small Establish. Person Percentage 1A Dispersed 1984 63,894 19.6 resource proces.(1975) (38,845) (23.4) 1B Market oriented 1984 98,426 30,3 industry (1975) (38,020) (26.3) 1C Service industry1984 17,676 5.4 (1975) (14,202) (9.8) 2A Seperable Mfg. 1984 59,725 18.4 operation (1975) (25,022) (17.3) 2B Craft handwork 1984 6,038 1.9 (1975) (1,666) (1.2) 2C Simple Assembly 1984 32,492 10,0 (1975) (25,462) (17.6) 3B Small Total 1984 46,770 14.4 Market (1975) (6,474) (4.5) Total 1984 325,021 100 (1975) (144,691) (100) Industry Type Year Value Added in Small Est. Million Won Percentage 1A Dispersed 1984 324,732 16.6 resource proces.(1975) (44,095) (26.4) 1B Market oriented1984 632,605 32.3 industry (1975) (41,790) '(25.1) 1C Service 1984 107,702 5.5 _ Industry (1975) (12,370) (7.4) 2A Seperable Mfg. 1984 373,943 19.1 operation (1975) (26,577) (15.9) 2B Craft handwork 1984 26,160 1.3 (1975) (800) (.5) 2C Simple Assembly1984 200,073 10.2 (1975) (35,997) (21.6) 3B Small Total 1984 294,638 15.0 (1975) (5,037) (3.0) Total 1984 1,959,853 100.0 (1975) ( 166,667) (100.0) Sources: For 1975 data,'Small Scale Enterprises in Korea and Taiwan‘ by Ho. 1980, World Bank Staff Working Paper, No.384, PP42,Table 3.8. For 1984 data,'Report on Mining and Manufacturing Survey 1984' by Economic Planning Board, The Republic of Korea. Korean Ind employed 5 Industry ] Industry(( 1A Disper: Proces 3114 Pres: 3119 Proc. 3132 Wine 3231 Tan . 3311 Sawm Tota lB Market Indust 3122 Food 3312 Wood 3412 Pape 3419 Pulp 3516 Ag, 3560 Plag 3693 Conc 3699 Non- Tote 1C Servi) 3723 Fri: 3711 SCEI 3724 Non. 3814 Met Tot 2A Seper Opera 3722 She 3822 A _ 3819 Fab 148 Table 5.9—2 Korean Industries in which Small Establishments(5—99) employed 50% or more of the Workers in the Particular Industry in 1984 Industry(0000) # of Small Workers in Small V.A in Small Est. Est. Est. person %share of V.A. %share of Ind. (million) Ind. 1A Dispersed Resources Processors 3114 Preserving Fish 994 24,162 60 81,766 52 3119 Processing Food 510 7,575 92 36,052 83 3132 Wine 927 8,088 88 57,535 75 3231 Tan & leather 184 5,687 52 39,029 39 3311 Sawmill,plan. 1,288 18,382 55 110,350 53 Total 3,903 63,894 69 324,732 60 1B Market oriented Industries 3122 Food,N.E.C. 549 8,165 57 50,697 29 3312 Wooden contain 124 1,844 74 8,621 67 3412 Paper contain 639 14,387 74 84,301 65 3419 Pulp,paper board257 5,375 53 33,774 30 3516 Ag. medicine 14 2,056 83 14,400 20 3560 Plastic N.E.C.1,705 34,824 64 223,881 43 3693 Concrete 1,457 17,293 55 127,161 43 3699 Non—metal min. 532 9,234 71 64,188 59 3812 Metal furniture 276 5,248 71 25,582 58 Total 5,553 98,426 60 632,605 41 1C Service Industries 3423 Printing allied 280 4,655 78 23,379 84 3711 Steel works 35 1,343 52 19,820 51 3724 Non-ferrous fou. 84 1,303 86 6,932 82 3814 Metal heating 545 10,375 62 57,571 47 Total 944 17,676 70 107,702 66 2A Seperable Manufacture . Operation 3722 Smelting Metal 26 493 63 5,472 62 3822 Ag. machinery 238 5,505 52 28,276 34 3819 Fabricated metal 1052 21,714 50 135,002 38 3823 Wood machine 749 14,434 69 92,750 61 3824 Special equip. 844 15,687 62 103,855 40 3851 Medical Instru. 102 1,892. 53 8,588 44 Total 3,011 59,725 58 373,943 47 Industry(000 2B Craft Hat 3232 Dressed 3901 Jewlly Total 20 Simple As 3215 Knittir 3512 Inorgar 3513 Pigment Total 33 Small T01 3214 Madeup 3216 Carpet 3219 Textil) 3221 Custom 3233 Leathe 3319 Wood, ‘ 3922 Commer 3540 Misc.o Total Source; 'Rep by E 149 Table 5.9-2(cont'd.) Industry(0000) Est. ZB Craft Handwork 3232 Dressed furskin 35 3901 Jewlly 228 Total 263 2C Simple Assembly 3215 Knitting millsl,328 3512 Inorganic chem. 201 3513 Pigment 55 Total 1,584 3B Small Total Market 3214 Madeup textile 489 3216 Carpet & rugs 46 3219 Textile N.E.C. 222 3221 Custom tailor 731 3233 Leather prod. 457 3319 Wood, N.E.C. 225 3422 3540 Commercial print159 Misc.of Petrol. 270 Total 2,599 Source: Est. 1,156 4,882 6,038 25,892 5,246 1,354 32,492 11,272 1,087 4,759 6,290 10,056 3,683 2,156 7,467 46,770 Est. person %share of V.A. %share of Ind. (million) Ind. 79 4,883 73 77 21,277 76 78 26,160 75 50 108,527 37 71 75,868 60 51 15,678 40 57 200,073 46 53 56,542 49 75 3,839 61 72 27,166 61 54 26,509 55 63 44,753 53 88 11,573 87 80 12,141 82 62 112,115 53 68 294,638 63 # of Small Workers in Small V.A in Small 'Report on Mining and Manufacturing Survey 1984' by Economic Planning Board, The Republic of Korea. 150 17.3% in 1975, 18.4% in 1984, 2C. group employment: 17.6% in 1975, 10.0% in 1984) The SSEs in an industry serving a small total market, are expanding in terms of employment(14.4% in 1984). As Table 5.9-1 shows, in 1975 the locational and process influences were most important in determining the SSEs’ predominance - in manufacturing industry in Korea.(1A,B,C and 2A,B,C group employment took a major share of 95.5% in 1975, 97% of the total value added in the same year) 8. Efficiency Relative to Capital Intensity in SSEs Empirical studies in LDCs have often shown that SSEs are labor intensive. Now let us see the relative efficiency of Korean SSEs in terms of capital productivity. In Table 5.10-1, the capital productivity (value added /fixed assets) of SSEs in the textile industry is relatively high (1.19 in 1984), while that of the food industry is relatively low (0.67 in 1984). The capital productivities in basic metal and fabricated metal industries are 1.12 and 1.26 in 1984 respectively. The SSE in the textile industry is less capital - intensive (3,593,000 won/worker). Within the textile industry, the SSEs in the silk reeling section have a high capital-intensity (7,097,000 won/worker), while the SSEs in the ready made apparel have a low capital- 151 Table 5.10-1 Output/Capital & Capital/Labor ratios of 4-digit Industry Year; 1984 % V.A % emp. V.A per Fixed Assets in Small.in Small. Fixed Assets per worker Manufacturing (000) Foodsbeveraae ( 31) 13 _ ___3.9_,___-,,_ 0. .6_7,____,___,.-_._9.000, Proce. of meat(3111) 27 34 0.67 13022 Dairy product(3112) 9 16 0.80 14906 Can of fruit(3113) 40 43 0.84 6370 Proce. of fish(3114) 52 61 0.90 3744 Vegetable oil(3115) 14 34 0.63 12590 Grain mill(3116) 21 49 0.30 23357 Bakery(3117) 7 19 0.58 6301 Sugar(3118) O 1 0.10 38125 Food for trade(3119)83 93 0.33 314303 Condiment(3121) 15 33 0.77 9488 Food N.E.C.(3122) 29 58 0.48 12919 Animal feed(3123) 34 49 0.97 16891 Distilling(3131) 7 13 1.72 10786 Wine(3132) 76 88 1.09 6502 Non-alchol(3134) 4 7 0.46 27255 Tobacco(3140) 0 1 5.06 2500 Textile(32) 24 34 1.19 3593 Silk reel(3211) 7 13 0.69 7097 Weaving(3212) 27 38 0.81 5465 Bleaching(3213) 36 45 0.92 5854 Madeup textile(3214)49 53 1.12 4492 Knitting mill(3215) 37 51 1.53 2742 Carpet(3216) 61 76 0.83 4249 Cordage(3217) 23 27 0.81 5223 Textile N.E.C(3219) 61 73 0.91 6303 Custom tailor(3221) 55 54 1.80 2338 Ready apparel(3222) 21 29 2.94 1178 Tanneries(3231) 39 52 1.39 4932 Furskin(3232) 73 80 1.43 2963 Leather(3233) 53 64 2.50 1778 Footwear(3240) 19 28 1.93 2031 Wood(33) .541111154- 0.91 5325 Sawmill(3311) 53 55 0.85 7091 Cane contain(3312) 67 75 1.08 4325 Cork N.E.C.(3319) 87 89 0.97 3236 Furniture(3320) 27 46 1.03 3478 nger(34) _ --m“_12911.11531 ,_“110711- .5660 Pulp paper(3411) 15 36 1 0.93 7012 Paper contain(3412) 65 75 1.22 4821 Pulp N.E.C.(3419) 30 53 1.04 6052 Table 5.10—1(cont'd.) Output/Capital & Capital/Labor ratios of 4-digit Industry Year; 1984 Manufacturing % V.A % emp. V.A per Fixed Assets in Small.in Small.Fixed Assets per worker (000) Newspaper(3421) 24 45 1.06 6067 Comm. print(3422) 82 81 0.98 5744 Print allied(3423)76 78 1.18 3831 Chemical(35) 17 30“,, ”1:03l-“___1;§929 Organic chem(3511)10 39 0.69 13741 Inorga. chem(3512)60 71 0.75 19405 Tanning mat.(3513)40 52 1.33 8697 Chem.fibre(3514) 3 13 0.53 10734 Chem.fertil.(3515) 4 17 0.34 21656 Ag-medicine(3516) 20 24 1.64 14851 Paint(3521) 21 34 1.33 8007 Drugs(3522) 15 29 1.12 10164 Soap(3523) 8 16 1.00 11872 Chem.N.E.C.(3529) 33 45 1.20 8217 Petrol.refi.(3539) 1 8 1.54 13182 Petrol.coal(3540) 53 62 1.78 8436 Tyre & tube(3551) 2 7 1.00 4921 Rubber prod.(3559)12 10 1.24 3824 Pastic prod.(3560)43 65 0.97 6653 Nonmetal min;(36) 29 40 0.86 9192 Pottery(3610) 14 21 0.77 4396 Glass(3620) 11 23 1.19 4511 Clay(369l) 32 44 0.62 7785 Cement(3692) 23 15 1.90 26050 Concrete(3693) 43 55 0.60 12340 Refractory(3694) 19 41 0.74 6498 Nonmetal min(3699)59 72 1.17 5937 Basic metal(37) 11” 26 __ 1.12 2324 Blast fur.(3711) 51 53 2.12 6952 Steel roll(3712) 5 12 0.91 11268 Iron(3713) 31 44 1.29 4294 Iron N.E.C.(3719) 17 40 1.14 8169 Primary sme.(3721)17 20 1.90 9757 'Secon.smelt(3722) 62 63 1.40 7903 Rolling(3723) 29 39 1.00 6612 Hammer(3724) 82 86 1.17 4563 Nonfer.N.E.C(3729)38 46 0.89 7837 Fabricated me.(38)17,, ~“30“,,“ 1.26 4618 Handtool(3811) 37 41 1.17 4680 1.52 3212 Furniture(3812) 58 71 153 Table 5.10—1(cont'd.) Manufacturing % V.A Z emp. V.A per Fixed Assets in Small.in Small.Fixed Assets per worker ) (000 Struc.metal(3813) 32 42 1.35 5208 Metal stam.(3814) 70 62 1.27 4385 Fabr.metal(3819) 38 51 1.26 4922 Engine(3821) 2 7 0.47 9088 Ag-machine(3822) 34 53 0.93 5502 Metal mach.(3823) 61 70 1.26 5086 Specia.mac.(3824) 40 63 1.33 498 Office mac.(3825) 15 20 1.32 5854 Service ma.(3826) 27 39 1.84 4076 Mach.N.E.C.(3829) 23 36 1.15 5749 Elect.mach.(3831) 25 38 1.62 3995 Sound mach.(3832) 7 18 1.38 3514 Elec.appli.(3833) 15 31 1.22 3757 Elect.mach.(3834) 8 13 1.16 3694 El.mac.N.E.C(3839)20 36 1.27 4300 Shipbuild(3841) 5 10 1.32 4564 Rail. equi.(3842) 7 20 1.31 6048 Motor veh.(3843) 10 22 1.00 5860 Bicycle(3844) 22 37 1.08 4182 Trans.equi.(3849) 7 34 0.73 6218 Med.instru.(3851) 44 53 1.12 4036 Optical good(3852)24 33 1.02 5112 Watches(3853) 13 19 1.87 2527 Science eq.(3854) 43 46 1.60 3876 chgr_manu.(39) 35 41 1.57 2671 Jewellery(3901) 76 77 1.70 2562 Musical ins.(3902) 8 12 0.98 4336 Sport good(3903) 32 40 1.47 2985 Dolls(3904) 34 39 2.57 1372 Costume jew.(3905)50 47 2.22 1922 Manu.N.E.C.(3909) 39 47 1.13 4316 Source; 'Report on Mining and Manufacturing Survey 1984' issued at April, 1986 by Economic Planning Board, The Republic of Korea. 154 intensity(1,178,000won/worker). 0n the other hand, the SSEs in the food industry are more capital- intensive (9,000,000 won/worker). This shows that food industry has become larger and more capital- intensive. Within food industry, the SSEs in the sugar section have a high capital intensity (38,125,000won /worker),while the SSEs in the tobacco section have a low capital intensity (2,500,000won/worker). Because of its potential for employment generation, the Korean small scale textile industry has a bright future. As Table 5.10-1 shows, depending on the specific industry, SSEs tend to have varying degrees of capital intensity. When we examine the capital intensity of SSEs in the small scale dominant industries, within the overall picture of the manufacturing industry, they still tend to be labor intensive, compared to the LSEs. Supporting this idea, Table 5.10-2 shows that their capital/labor ratio lies between 4,000,000-6,000,000 won/worker but the capital/labor ratio of all manufacturing industry (including the large and small scale) is far above 10,000,000 won/worker. Therefore in labor abundant LDCs ,like Korea, SSEs have a growth potential. D. CONCLUSION So far I have reviewed the structure of Korean SSEs in terms of micro prospects. The analysis covered the ' “ii" Table 5.10-2 Output/Capital and Capital/Labor Ratio for Industries in which Small Establishments predominate,by Industry 1984 a) Value Added per Fixed Assets per Unit of Fixed Workers Assets (000won/person) Total Manufacturing 1.00 10,534 Small Scale Industry Accross-the-board Mfg. 1.06 5,496 1A Resources processors 0.79 5 6,413 1B Market oriented ' 1.19 5,359 1C Service Industry 1.37 4,447 2A Seperable Mfg.operation 1.25 . 5,028 ZB Craft handwork 1.64 2,639 2C Simle Assembly 1.08 . 5,680 3B Small Total Market 1.44 4,386 Source: 'Report on Mining and Manufacturing Survey 1984' by Economic Planning Board, The Republic of Korea. Note: a) This data covers only small establishments(with 5—99 workers) in small scale industries. b) This data covers only small establishments(with 5-99 workers) in small scale industries. b) specific industrial, regional and dynamic prospects, efficient size, and capital intensity. The findings from this analysis follow: 5 First, with economic development the role of very small scale activities(with 1-4 workers) in Korea is declining. The share of very small scale activities declined from 36% in 1975 to 30 % of total manufacturing employment in 1984. Second, as industrialization progresses, SSEs are emerging in capital-intensive and heavy industries such as chemicals and the metal machinery industry; in 1984, the SSEs in the chemical industry took a sizable share ( 15.3% - those with 5-49 workers and 17.4% - those with 50— 99 workers) of the total small scale manufacturing value added. Third, the SSEs are generally located in the urban area: in 1984, 73% of the total labor force employed by SSE was in the urban area. Fourth, during the last couple of decades, the SSEs have been losing in the share of manufacturing employment. In 1958 the employment share of SSEs (5-99 workers) was 67%, but by 1984 this figure dropped to 34%, although it reached a low of 26% in 1975. Fifth, the Survivor criterion shows that during 1979-84 all plants, .SSEs and LSEs were fairly efficient. On the other hand, during 1968-75 the Survivor test shows 157' that the LSEs were much more efficient than the SSEs; but it must be noted that the Survivor test is a measurement of private efficiency, not social efficiency. This Survivor test implies that the private efficiency of the SSEs improved during 1979-84. Sixth, with better transportation , the locational influence in the competitiveness of SSEs became less important. The market-oriented industry (by the Staley & Morse classification) was getting larger in terms of employment (30.3%) and the value added (32.3%), in 1984. Seventh, depending on the specific industry, the SSEs in the manufacturing industries tend to have various ranges of capital intensity. However, compared with the overall average for manufacturing industry, SSEs still tend to be labor intensive. The study shows that there exists a bright future for Korean SSEs, if they can keep improving their efficiency and correct the urban concentration problem, and if government policies are not biased against them. 158 CHAPTER 6 THE ISSUES AND POLICY IMPLICATIONS FOR SSES IN. KOREA A. INTRODUCTION During the last two decades of economic development, the Korean industrial structure has been transformed into one dominated by. LSEs. The importance of SSEs has been intensively debated; with the expansion of LSEs, problems of income inequality have surfaced. The Korean government has been accused of needlessly accelerating the growth of LSEs at the expense of SSEs; so it is meaningful to review the problems besetting Korean SSEs and to suggest a feasible solution. First, I begin with the past industrial policies in terms of overvaluation of the exchange rate, subsidized interest rate and taxes. Second , I review the location of SSEs. Third, I continue to check the relative efficiency of the SSEs, using Ho’s method, and ’Social Benefit and Cost(SBC)’. Finally, I review the general policy implications for SSEs. B. Large Scale Oriented Industrial Structure 159 1. SSEs in Korean Industry From the early 19605, the government chose an export-oriented policy as its economic development strategy. This was an open door policy that led to the expansion of markets. With this enlarged market, economies of scale and a large size of the firm have become more important. As a result LSEs emerged and SSEs expanded into LSEs. Table 6.1 shows that in 1979 SSEs with 5-99 workers employed 28% of the labor force in the manufacturing industry, while LSEs with 100+ workers employed 72%. But by 1984 ,the LSEs with 100+ workers employed 66% of this labor force. Today LSEs dominate in manufacturing employment. The large scale oriented industrial structure is a controversial issue because of its potential for monopolies and restrictive behavior, and its potential for reducing social welfare. LSEs, also, tend to be capital intensive. The Korean case is not exceptional. This situation can adversely affect employment and income equality objectives. For these reasons, I favor a greater role for the small scale industry. 2. Korean Industrial Policy In Korea, technological necessity, possible economies of scale, and government policy had all combined to increase the size of the plant. Here I would like to 160 Table 6.1 Korean Employment in Manufacturing Industry (Table 6.1—1) Mfg. employment by size of establishment Korea 1975, 1979, 1984 (unitz000person) Year 5—9 10-19 20-49 50-99 100-199 ZOO—499 500+ Total 1975 63 69 115 124 163 266 619 1,420 (2) (4) (5) (8) (9) (11) (19) (44) (100) 1979 76 94. 205 205 256 361 920 2,117 (%) (4) (4) (10) (10) (12) (17) (43) (100) 1984 93 143 289 277 294 357 891 2,344 (%) (4) (6) (12) (12) (13) (15) (38) (100) (Table 6.1—2) Employment Growth of Manufacturing Korea 1975—1979 and 1979-1984(in percentage) Year 5-9 10-19 20-49 50—99 100—199 200—499 500+ Total 1975 -1979 21 36 78 65 57 36 48 49 1979 —1984 23 52 41 , 35 15 -1 —3 11 Source: 'Korean Statistical Yearbook 1975' 'Report on Mining and Manufacturing Survey 1979, 1984' by Economic Planning Board, The Republic of Korea. ' 161 focus on government policy, which has favored LSEs. It is reflected in the lower relative cost of capital for LSEs. This discriminatory policy resulted in creating a distortion in the factor (capital) market. Tariffs favor the importation of machinery and equipment. Import licensing is linked to minimum export requirements ,and the government provided repayment guarantees to foreign suppliers(Kusnets 1977). Also, the overvaluation of the Korean currency the ’won’ and the high domestic interest rate policy, after the fall of 1965, made imports and foreign loans (primarily supplies credit) cheaper relative to domestic goods and loans. The Korea Development Banks provided funds for industrial enterprises (government invested corporations) at rates well below those charged in the unorganized money market. Firms with access to these institutions, mainly LSEs, obtained domestic credit at subsidized rates. Clearly the government favored LSEs in the allocation of credit; for example, in 1963, SSEs accounted for 60 % of manufacturing output, but received less than 27% of domestic loans (here SSEs are defined as firms with less than 30 million won in assets, or with fewer than 200 workers)(Kusnets 1977). a. capital Market To see the policy-induced capital cost distortions 162 over the last decade , we need to analyse the quantitative data on policy in terms of the trade regime, interest rate,and tax rate. Table 6.2 shows the impact of these policy-induced distortions on the LSEs and the SSEs. The trade regime favored LSEs by overvalued exchange rates , making capital costs cheaper for them relative to the capital cost for SSEs; again, the interest policy favored the LSEs by a lower official rate, further accentuating the condition. During the last decade( except 1975 , 1981), the overvalued exchange rates tended to favor LSEs despite .some negative effects on balance of payment. On the other :hand, the official lending rate for LSEs was lower than the black or private market rate for the SSEs, although this differential has narrowed since 1981. The higher interest rate on the black market could be argued as reflecting the higher risk and transaction costs where the borrowers are small, such as SSEs. But Liedholm(1986) found in some carefully designed programs that only the small portion of interest gap could be traced to administrative and risk transaction costs differential. Specially when I compare the Korean situation with situation in Taiwan with a similiar risk condition, the interest gap still gives some rationale for measuring the capital cost differential for LSEs & SSEs. Over time, however, the tax policy favored the 163 Table 6.2 Policy—induced Capital Cost Distortion in Large and Small Scale Enterprises (Korea) (Expressed as the percent difference in large firms' cost to small firms) Percentage Difference in Capital Cost owing to: Year Trade regime Interest Taxes Total Capital (overvaluation) 1973a) —5 —35 +10 -30 1976 —2.1 —30.4 +10 -22.5 1977 -5.3 —28.5 +10 -23.8 1978 -7.7 -28 +10 -25.7 1979 —15 —33 +10 -38 1980 . —1.6 -32.5 +10 -21.1 1981 —2.2 —23 +10 -15.2. 1982 -7.5 -21.7 +10 -19.2 1983 -5.8 ~15.5 +10 -11.3 1984 —0.1 -16.4 +10 -6.5 Source:a) For 1973 data,'The Effect of Policy & Policy Reforms on Non-Agricultural Enterprises & Employment in DevelOping Countries: A Review of Past Experiences ' by Steve Haggblade, Carl Liedholm, & Donald C Mead. MSU IDP Working Paper, No.27, 1986, P32, Table 7. Note: The calculation for Korean data is shown in Appendix A. 164 small firms with a 10% lower corporate tax rate. Overall, the government policy favored LSEs, though, since 1981, the difference has been declining. In 1973,the total capital cost for the LSEs was 30 % cheaper than that for the SSEs, but ,in 1984, this figure fell to 6.5 % . Considering that the payoff from capital investment occurs over a relatively long period of time, the impact of the declining capital cost gap on the industrial structure may appear with lag. b. Labor Market The studies(Hong,1981,Ho,1980,Lindauer,1984) agreed- that the labor market in Korea is relatively free from distortions. The reason for this conclusion is the fact that the union activities are suppressed by the government, which limits the scope of bargaining agents and the resort to strikes; and the minimum wage scheme is not generally practiced. These factors contributed to preventing wage determination above market clearing levels. But as shown in Ch.4, there is a real wage gap between LSEs and SSEs; this gap is explained in different ways. According to a competitive theorist, training and hiring costs are believed to produce this differential; a bargaining theorist says this is the consequence of some bilateral monopoly; and the dual labor 165 market theorist points to the size of the firm as a critical boundary segmenting the primary and secondary labor markets(Lindauer 1984). In Korea this gap seems to be relatively small, if the labor productivity differential (quality, skill or training) is considered. c. Product Market Levy(1986) observed that the Korean duty drawback scheme tends to discriminate against SSEs. The original purpose of the drawback scheme, which repays exporters their taxes on imports, was to ensure that exporters did not face high input costs in competing on equal terms with foreign competitors. At the same time, it ensured‘ that local suppliers of inputs needed by exporters could compete against foreign imported inputs. But this scheme seems to be available only for firms above a minimum size threshold because of transaction costs that firms must bear in dealing with the government. These transaction costs imply that the minimum scale at which firms may enter export markets, is likely to be relatively large. Levy(1986) further indicated that the Export Processing Zone (EPZ) tended to discriminate against SSEs because of stiff licensing Orequirements for entry. Originally, an EPZ was conceived for giving firms, within a delineated area, the opportunity for tariff-free trade in order to ensure competitiveness in the world market. In 166 1971, the first zone was opened in Masan, the second zone was opened, in 1975, in Iri city; this system provided a favorable environment for foreign investors. Levy(1986) found that the Korean targeted cross- subsidization policy was likely to discriminate against small and medium enterprises; from the government’s point of view, the time and effort required to negotiate a subsidy, in return for a specific export response, tends to be high in the SSEs compared with that in the LSEs. Therefore, the government rationally tends to allocate scarce time and effort in negotiating cross-subsidization arrangements only with LSEs. From these experiences we may conclude that the government should alter the bias against the SSEs in order to sustain fair competition between the SSEs and the LSEs.‘ Levy suggested that in the duty drawback scheme the government should set up information channels, such as trading institutions, to be a source of information cost or transaction cost for SSEs. According to Levy, a "laissez-faire" outward-looking policy is preferable to a policy involving protectionism plus a drawback scheme for the SSEs. Liedholm(1986) pointed out that there exists a minimum export value for firms , qualifying for subsidies under the transaction law of 1957; exporters have to exceed an annual $ 20,000 and importers have to exceed 167 $100,000 in order to maintain this privileged status; this pre-condition worked against the SSE. Since late 19605, the Korean government has selectively exercised the incentive policies and instruments of protection in favor of large scale 'heavy industries where there remained opportunities for substantial import substitution. Further this selective import substitution strategy has allowed the concentration of scarce investment resources in a few key sectors at a time and exploited the effects of economies of scale and linkage among closely allied activities(Westphal 1977). 'For example, during 1977-79, to assist in accelerating the process of industrialization, the government focused on the heavy and chemicals industries pumping in financial assistance of 2,806 billion won. This investment amounted 'to approximately four times that in light industries. But the majority of large plants built during this period faced a serious shortage of demand in the domestic and foreign markets. This massive injection of capital proved to be too ambitious(Lau 1986). Westphal (1981) suspected that the difficulty of this strategy stemmed from the Korean government’s decision to promote too many infant industries at once. The sudden change in the industrial strategy contributed to the domination of the LSEs, heavy industry belonging to this category. 168 Table 6.2.a) Percentage composition of manufactured output 1960 1971 1979 all light industry 70% - 54.7% 44.7% all heavy industry 30% 45.3% 55.3% source:Tibor Scitovsky, 1985 ’Economic Development in Taiwan and South Korea; 1965-1981’ Table 6.2.a shows that as industrialization progressed, the heavy industry sector output dominated the scene. In 1960, light industry output was 70 % of the total industrial output; but by 1979, heavy industry produced a major portion ( 55.3 %) of the total industrial output, further. establishing the large scale dominated ‘industrial structure. C. Urban Concentration of SSEs As industry expanded, labor migration from the Ag sector accelerated because of job opportunities in the urban area. This continuous migration resulted in overurbanization, city life becoming intolerable with a shortage of housing, public facilities etc. The problem of overurbanization lies partly in the fact that many manufacturing industries are located in the urban area; the Korean case is not exceptional; the Korean SSEs, also, are concentrated in the urban area of Seoul and Pusan. Table 6.2-1 shows that the majority of manufacturing 169 Table 6.2-1 Manufacturing Establishments by Location and by Size of Establishment, Korea. 1979 and 1984 Size of Establishments(workers) Year Total 5-9 10-19 20-49 50-99 100-199 200-499 500+ Est. 1984 41549 14009 10521 9128 3942 2109 1213 627 1979 31804 11915 6851 6482 2895 1829 1200 632 Seoul 1984 71 64 74 73 74 75 74 75 Pusan ‘ area (1979)(62) (50) (65) (68) (72) (71) - (74) (77) Seoul 1984 31 35 35 27 24 20 19 16 (1979)(26) (22) (31) (28) (26) (25) (22) (25) Pusan 1984 12 9 13 13 11 14 14 16 (1979)(11) (7) (12) (14) (13) (13) (16) (14) Kyong 1984 23 15 22 28 33 34 31 25 gi-Do(1979)(18) (13) (16) (20) (27) (27) (27) (25) Kyong 1984 5 5 4 5 6 7 10 18 sang namDo(1979) (7) (8) (6) (6) (6) (6) (9) (13) Rest 1984 29 36 26 27 26 25 26 25 of Korea(1979)(38) (50) (35) (32) (28) (29) (26) (23) Total(%) 100 100 100 100 100 100 100 100 Source:'Report on Mining and Manufacturing Survey, 1979 & 1984' by Economic Planning Board, The Republic of Korea. 170 establishments are urban-based. In 1984, 71 % of these were located in and around Seoul and Pusan, this status having developed between 1979 and 1984. In 1984, 64% of the firms, with 5-9 workers and 74 % of the firms with 10-19 workers , were urban-based. Table 6.2-2 indicates that employment for each category by size of firms was similar in terms of urban concentration. The urban concentration of firms is directly related to the issues of overurbanization. This locational pattern acted as an inducement to this momentum of overurbanization. In the short run, it deepens the regional income inequality, .unless there is a drastic improvement in the rural quality of life. Given this status quo, there seems to be a great need to expand SSEs in the rural areas. The Liedholm study(1986) pointed out that rural non- farming activitiy within a small scale may be promising in the LDCs. This may be true for Korea also; as the rural income rises, the Ag household tends to demand more; consumer or non-Ag goods produced by the local SSEs. Liedholm, also, found the income elasticity of demand for goods produced by the rural non-farming activities tends to be ,very often, positive in many LDCs; therefore there seems to be a potential role for SSEs in the rural areas in terms of forward linkage (food processing etc.) and backward linkage (ag-tool, fertilizers etc.) industries. This potential for rural non-farming activity in Korea 171 Table 6.2-2 Manufacturing Employment by Location and by Size of Establishment, Korea. 1979 and 1984 890 919 77 (80) 14 (22) 18 (20) 22 (22) 23 (16) 23 (20) Size of Establishments(workers) (000) Year Total 5-9 10-19 20—49 50-99 100—199 200-499 500+ Est. 1984 2843 93 143 288 276 293 356 1979 2116 75 94 204 204 256 361 Seoul 1984 75 66 74 74 74 75 75 Pusan area (1979) (74) (52) (65) (68) (72) (71) (74) Seoul 1984 20 36 35 27 24 20 19 (1979) (24) (24) (31) (28) (27) (25) (22) Pusan 1984 15 10 13 13 11 14 14 (1979) (16) (7) (12) (14) (13) (13) (16) Kyong 1984 27 16 22 28 33 34 31 giDo (1979) (23) (13) (16) (26) (26) (27) (27) Kyong 1984 13 4 4 6 6 7 11 sang namDo(1979) (11) (8) (6) (6) (6) (6) (9) Rest 1984 25 45 26 26 26 25 25 of Korea(1979) (26) (48) (35) (32) (28) (29) (26) Total(%) 100 100 100 100 100 100 100 Source:‘Report on Mining and Manufacturing Survey 1979 & 1984' by Economic Planning Board, The Republic of Korea. 100 172 suggests that the government’s policy should encourage SSEs to move into the rural area through tax concessions, loans etc; for example, the Local Industry Development Law was enacted in 1970 to promote the decentralization of industrial development. This law provided a reduction in acquisition , property , registration and corporation taxes within 5 years ,for those industries located in the designated local industrial estate, ( where the size is 2 not less than 100,000m for the inland estates and 200,000m2 for the coastal ones). The government should consider an uni-modal agricultural strategy. rather than a bi-modal policy, because the former will have more demand effects on the labor intensive and small scale oriented industrial activities(Liedholm 1986). At the same time the policy of developing the infrastructure, such as transportation, power, and water availability, should precede the above aggressive agricultural policy; as a good example, the Korean ’New Village Movement’, since the early 1970s , should continue and expand in order to improve the environment for agricultural productivity. D. Relative Inefficiency of SSEs in Small Scale Industry Before advocating the cause of the SSEs in Korea, we need to review whether SSEs are ,in fact,efficient. In the I___—E _ rah-=- 173 measurement of efficiency, the concept of the total factor productivity is gaining popularity in economic development literature. A firm’s efficiency depends not only on its capital productivity , but also on other factors. Technically there are various modes of measuring the total factor productivity. Here, Ho’s(1980) relative efficiency measurement and ’Social Benefit-Cost’ ratios are adopted for a more objective assessment, each method,though, having its own limitations. 1. Ho’s Method Ho reformulated the Christensen-Jorgensen index. Ho’s method appears to be crude, but it offers a check on the results of more sophisticated methods; it has the advantage of being relatively easy to calculate and depends on fewer assumptions about the production relationship. But this method should be adjusted for differences between the actual factor price and the opportunity costs as well as any differential impact of trade protectionism on the value added between the LSEs and SSEs. There is .a tendency towards biased efficiency measures in different sized categories of firms, because the SSEs tend to underreport their value added, use a lower quality of labor, have a lower rate of capacity utilization and operate in far more competitive 174 conditions, which make for a lower efficiency rating. there is, also, an element for favoring SSEs in terms of efficiency ,because the capital usage may be understated to a greater extent(Tyler Biggs 1986). Let us begin by applying Ho’s method. For convenience we assume two factors of production , capital and labor, and we apply this measurement to the Korean 4- digit manufacturing small scale-dominated industry, where the SSEs have over 50% of employment. The reason I chose ’the small scale industry’ for an efficiency analysis is that the small scale.industry is a favorable environment for employment creation /generation and tends to face competitive market conditions. According to Ho, we assume a 20% shadow price of capital and no distortion in the labor market. Technically the total factor productivity(A) becomes ( value added )(‘é-fii-LEQ‘ * (value added 712m (.2)*K w ) where K is fixed assets and W is wage bill,capital factor share is (.2)K/(.2)K+W & labor factor share is W/(.2)K+W. Total factor productivity of j- th category(A(j)) becomes . (.2)K(1’) . wm (Value added(3))(.2)1<(j)+wj) *(Xalue added(3))(.2)1<(j)+wrfi (o2)K(j) WI) The relative efficiency measurement then becomes the ratio of the total factor productivity of the establishment in a given size group to that of all .4?” 175 establishments in the industry, such as A(j)/A. Table 6.3-1 shows A and A(j) of the industries dominated by SSEs in 1984. Table 6.3—2 shows the relative efficiency measurement(A(j)/A) in the 4-digit industries dominated by SSEs in 1984. Table 6.3-3 shows that in only a few instances is the total factor productivity by the relative efficiency measurement highest in SSEs of small scale industries. In 1984 , the total factor productivity was highest in only 9 small establishments, with 5-99 workers, but the total factor productivity was highest in 21 large establishments with over 100 workers. This 1984 result contrasts with the 1968 result. In 1968 total factor productivity was highest in 38 SSEs with 5-99 workers of small scale industries but the total factor. productivity was highest in 33 LSEs with over 100 workers of small scale industries. In 1968 the SSEs were far more efficient than LSEs in small scale industries(Ho 1980). In 1984, the LSEs in the small scale industry became relatively more efficient than the SSEs and the numbers of small scale industries also declined. This data implies that by 1984 the social efficiency of the SSEs in the small scale industry declined, compared to the 1968 level. According to Ho’s study ,however, the relative efficiency measurements tend to be biased against the SSEs. Although he recognized that understating the capital 176 Table 6.3-1 Total Factor Productivity of Establishments for Industries in which Small Establishments predominate,by Type of Industry & by Size of Establishments(Worker), Korea. 1984. A A(j) 4-digit average 5-9 10—19 20-49 50-99 100-199 200-499 500+ industry of industry 1A 3114 3.7 2.69 3.35 3.36 3.94 3.64 _g,;; 2.71 3119 2.16 1.75 1.82 2.12 3.40 2 82 4.95 3132 4.61 3.2 3.61 6.17 6.78 1,91 5 0 3231 4.12 3.52 2.94 3.02 3.81 4 59 5,;2 3.86 3311 2.67 2.25 2.94 2.85 3,22 2.37 1 5 2.61 1B 3122 3.88 1.67 2.02 2.43 6.57 1.93 4.28 ,7,09 3312 2.76 2.82 2.64 2,98 2.58 2.27 2.54 3412 3.37 3.16 3.27 3.22 3.3 3.69 3.4 3419 4.47 3.23 3.51 3.32 2.95 3.33 6,5] 6.22 3516 9.08 .86 2.59 9.88 6.23 8.73 13,59 3560 4.04 2.91 2.66 3.2 3.95 3.57 4 04 .21 3693 3.53 2.09 2.64 3.35 3.59 9,22 3.81 3 57 3699 3.79 2.64 3.66 3.48 3.57 4.14 4 6 4.71 3812 3.28 2.43 3.18 2.91 2.8 2.79 5.14 " "' C . 3423 2.65 3.05 2.72 2.6 2.3 3,1, 2.13 3711 5.1 4.00 2.59 2.73 7.75 3.36 6.55 3724 2.82 1112 2.58 2.75 2.06 2.81 3814 3.66 2.82 2.73 2.69 3.75 6.43 3.27 3.58 A ‘““ 3722 4.07 1.71 4.22 3.49 2.1 4.22 3819 3.64 2.79 2.97 3.31 3.71 4.21 3.8 3.2 3822 3.53 2.56 2.42 2.69 3.02 2.61 3.25 4.48 3823 3.09 2.65 2.83 2.9 2.97 3.06 3.66 3824 3.66 2.87 2.95 2.98 3.38 3.68 3.27 4,92 3351 3.01 2.54 2.33 2.99 2.67 3192 3.13 3232 2.75 2.22 1.89 2.51 2.85 g,g1 3901 2.76 2.97 2,9; 2.9 2.40 2 8 2.48 c 3215 3.58 2.59 2.45 2.77 3.06 3.12 3.3 5,33 3512 4.12 2.76 3.29 4.22 3.85 5129 3.76 2.97 3513 4.54 4.67 2.84 3.84 4.14 9199 4.6 B 3214 2.86 3.13 2.85 2.99 3.19 3.15 1.99 2.34 3216 3.09 2.15 2.26 2.50 2.44 4.78 3219 3.49 2.46 2.69 2.92 3.62 3.47. 5.68 _ 3221 2.73 3.08 3.84 2.44 1.75 2.60 2.87 2.00 3233 3.34 2.75 2.58 3.05 2.80 3.87 4.40 3.74 17.7 Table 6.3-1 (cont'd.) 4-digit average 5-9 10—19 20—49 50—99 100—199 200-499 500+ industry of industry 3319 2.39 2.54 2.53 2.16 3.06 3.13 3422 2.74 3.20 2.43 2.82 3.77 2.16 2.06 3540 6.39 3.98 4.56 8.18 4.84 6.14 8.15 Note: A(total factor productivity in given industry) measurement was obtainted by (V.A/.2K)a%(V.A/W)a‘ where V.A is value added, K is fixed asets, W is wage bill. We assume shadow price of capital is 20% and there is no distortion in labor market. And a, is capital factor share which is .2K/(.2K+W), while a, is labor factor share which is W/(.2K+W). A(j) is total factor productivity in.j—th size category of given indUstry. This measurement can be obtained by same process by only using data of j-th category. Source: The calculation is based on data in 'Report on Mining and Manufacturing, 1984' 178 Table 6.3-2 Relative Efficiency of Establishments for Industries in which Small Establishments predominate, by Type of Industry and by Size of Establishment(Worker), Korea,1984. A(j)/A 5-9 10—19 20-49 50-99 100-199 200-499 500+ 1A Resource processors 3114 fish .73 .91 .91 1.06 .98 1:1] .73 3119 food proces. .81 .84 .98 1.57 1.31 4122 3132 wine .69 .78 1.34 1.47 1,61 1.08 3231 leather .85 .71 .73 .92 1.14 1.27 .93 3311 saw mill .84 1.1 1.07 1,49 .89 .56 .98 1B Market-oriented industries 3122 food N.E.C. .43 .52 .63 1.69 .50 1.10 1.83 3312 cane contain 1.02 .96 1,08 .93 .82 .92 3412 paper board .94 .97 .96 .98 1,09 1.01 3419 paper board ' N.E.C. .72 .79 .74 .66 .74 1,47 1.39 3516 Ag.medicine .09 .29 1.09. .69 .96 1&6 3560 Plastic NBC. .72 .66 .79 .98 .88 1 1.29 3693 concrete .52 .75 .95 1.02 1.2“ 1.08 1 01 3699 mineral NBC, .7 .97 .92 .94 1.09 1.21 1,24 3812 furniture) ' .74 .97 .89 .85 .85 £121 10 Service industries ’ 3423 printing all.1.15 1.03 .98 - .87 1.55 .8 3711 blast furnace .78 .51 .54 1.52 .66 1.28 3724 foundaries 1.11 .91 .98 .73 1 3814 metal stamp. .77 .75 .73 1.02 ‘1476 .89' 2A Seperable manufacturing operation 3722 smelt&refine .42 1.04 1.1 .52 1.04 3819 fab.metal .77 .82 .91 1.02 1.16 1.04 88 3822 Ag.machine .73 .69 .76 .86 .74 .92 1:27 3823 metal machine .86 .92 .94 .96 .99 1.18 3824 special mach. .78 .81 .81 .92 1.01 .89 1,; 3851 medical ins. .84 .77 .99 ' .89 1;1§_ 1.04 2B Craft handwork 3232 dressed fursk..81 .68 .91 1.04 1.06 3901 Jewelley 1.08 1.09 1.05. .87 1.01 .9 20 Simple assembly,mixing and finishing 3215 knitting mill .72 .68 .77 .85 .87 .92 1.49 3512 inorga.che. .67 .8 1.02 .93 1.28 .91 .72 3513 dyestuff 1.03 .63 .85 .91 1432 1.01 3B Small total market 3214 textile 1.09 1.00 1.05 1,12 1.1 .7 .82 3216 carpets .7 .73 .81 .79 1L55 3219 textile EEC, .7 .77 .84 1.04 .99 1m63 3221 custom tailor1.13 1L41 .89 .64 .95 1 05 .73 3233 leather sub. .82 .77 .91 .84 1.16 _l;§Z 1.12 ~IHE.' 4 179 Table 6.3-2 (cont'd.) 5—9 10—19 20—49 50-99 100-199 200-499 500+ Small total market continued 3319 cork NEC. 1.06 1.06 .9 1.28 1.3 3422 com.print 1.17 .89 1.03 1.38 .7 12 \OH .75 3540 mis.petroleum 5 & coal .62 .71 1 _8 .76 .96 1.28 Note: Small establishments(those with 5-99 workers) are considered to predominate in an industry if they account for half or more of the industry's employment. The relative efficiency measure is the ratio of the total factor productivity of establishments in a given size group to that of all establishments in the industry. This relative efficiency measures were obtained by assuming capital share to be .2K/(.2K+W), where K is the industry's fixed assets and W is the industry's actual wage bill. All industries which contain the words 'miscellaneous' or 'not elsewhere classified' in their titles are classified as 'excluded' category of size group in Table 6.3-3. Source: The calculation is based on data in 'Report on H mining and Manufacturing,1984'. .ngu: 180 Table 6.3-3 Number of Industries where Total Productivity(Capital Shadow Priced at 20%) is greatest in the Size Category Indicated, Korea 1968 and 1984 —Small Scale Industries(lndustries where SSEs have 50% + of employment) - Year 5-9 10-19 20—49 50—99 100-199 200-499 500+ excl. 1A 1984 0 0 0 1 1 3 0 (1968) (0) (0) (1) (3) (2) (4) (0) 1B 1984 0 0 1 0 2 2 0 (1968) (0) (1) (1) (5) (4) (3) (3) 1c 1984 1 0 0 1 2 o 0 (1968) (0) (1) (0) (0) (0) (1) (1) 24 1984 0 0 1 0 2 1 2 (1968) (0) (0) (1) (3) (3) (4) (0) 2B 1984 0 1 0 0 1 0 5 0 (1968) (0) (0) (0) (0) .(0) (0) (0) 20 1984 0 0 0 0 2 0 1 (1968) (2) <1) (5) (7) (1) (2) (2) Sub total 1984 1 1 2 2 10 6 3 (1968) (2) (3) (8) (18) (10) (14) (6) 3B 1984 0 1 0 2 1 1 0 (1968) (1) (0) (2) (4) (2) (1) (0) Total 1984 1 2 2 4 11 7 3 (1968) (3) (3) (10) (22) (12) (15) (6) Source: For 1968 data,'Small Scale Enterprises in Korea Taiwan' by Ho, 1980. P65, Table 4.5. 0 (3) 4 (5) 0 (0) 0 (6) For 1984 data,'Report on Mining and Manufacturing 1984' by Economic Planning Board, The Republic of Korea. 181 may have a positive effect on these figures, he insisted that a negative bias is still strong in relative efficiency measurements. According to the above data, SSEs with 5-99 workers do not tend to generate large amount of employment efficiently because SSEs were efficient in only a few industries. Therefore-we may expect further invasions and take-over bids by modern ,efficient, and capital intensive LSEs, unless the efficiency of SSEs increases so that they can effectively compete against the LSEs. The possible domination by LSEs may weaken the competitive market conditions in the small scale industries ,reduce the employment effects, and endanger the status of the small scale industry itself. 2. Social Benefit - Cost Analysis To complement Ho’s measurement of total factor productivity, I will use another technique, the ’Social Benefit and ost’ ratio; this involves calculating the social opportunity cost in terms of capital and labor. The SBC method, in a some sense, only measures technical efficiency in production possibility frontier measurement, because the SBC using social input cost takes care of the allocational efficiency issues. In practice, this method uses a single estimate of the social opportunity cost of capital applied to all 182 sources, as well as a sectoral average wage for each skilled category as the cost to all firms of labor in that category. If the SBC does not use a shadow price to evaluate outputs or raw material inputs, it can only be used to compare Benefit and Costs of firms in the same sector, having a similar input and output mix. When the SBC ratio is used to measure the relative performance of enterprise , a ratio greater than one implies that the firm has a positive effect on the total output of the economy, while a ratio of less than one implies a negative effect on the economy(Tyler Biggs 1986). I apply this measurement to the 4-digit manufacturing small scale dominated industry, where the SSEs have over 50 % of employment. Following Ho, I assumed a 20 % shadow price of capital and no distortion in the labor market. Keeping in mind that union activities are suppressed and that the minimum wage system is not generally practiced, I can justify the actual wage expenditures as the social labor costs. Technically, a Benefit and Cost ratio of j-th size category in the given industry becomes Value Added(j) (-2)*K(j)+W(J') where K(j) is the fixed assets of j - th size category, and W(j) is the wage bill of j -th size category. Table 6.4-1 shows the Benefit and Cost ratio of .183 Table 6.4-1 Benefit-Cost ratio of Establishments for Industries in which Small Establishments predominate,by Type of Industry & by Size of Establishments(workers), Korea. 5-9 10—19 20—49 50-99 1A Resource processors 3114 fish 1.38 1.68 1.75 2.08 3119 food proce. .94 .95 1.06 1.76 3132 wine 1.63 1.85 3.32 3.39 3231 leather 1.87 1.59 1.76 2.07 3311 saw mill 1.13 1.55 1.58 2.19 IR Market-oriented industries 3122 food NBC. .87 1.02 1.22 3.32 3312 cane EEfita.1.52 1.62 1.67 1.33 3412 paper board1.70 1.86 1.80 1.75 3419 paper board NLELQ, 1.78 1.84 1.81 1.32 3516 Ag.machine .43 1.97 5.14 3.11 3560 plastic EECI.49 .85 1.66 2.04 3693 concrete 1.05 1.33 1.67 1.80 3699 mineral EE§1.35 1.89 1.88 2.00 3812 furniture 1.32 1.81 1.76 1.70 1C Service i ndustries 3423 printing all. 72 1.47 1.52 1 30 3711 blast furn. 2. 39 1.44 1.55 4,14 3724 foundaries 1_Z5 1.41 1.54 1. 64 3824 metal stamp1.54 1.52 1.53 2. 06 2A Seperable manufacturing 3722 smelt&refin .87 2. 77 3819 fab. metal 1.47 1.67 3822 Ag. machine1.28 1.24 3823 metal mach.1.51 1.63 3824 spec.mach. 1.57 1.60 3851 medi.ins. 1.50 1.38 ZB Craft handwork 3232 dres.fur 1.25 1.06 3901 jewelley 1.72 1.75 20 Simple assembly A 3215 knit mill 1. 39 1.42 3512 inor. chem. 1.41 1.69 3513 dyestuff 2.33 1.43 3B Small total market 3214 textile 1.68 1.55 3216 carpets 1.08 1.16 operations 2.34 1 1.85 1.98 1.46 1.67 1.59 1.75 1.72 1.95 1.9.2 1-38 1.59 1:99 1.81 1.48 1.68 1.85 2.12 2.00 2.19 2.22 1.57 1385 1.30 1.51 1984 100-199 200—499 500+ 1.98 1. 53 3 82 2 40 1. 21 1.03 1.64 1:22 1.67 4.37 1.72 1.87 1. 78 2. 93 —_ -_-._._ 2.47 2. 53 2. 50 2.66 .75 .16 .69 .87 .35 .32 2.02 1.97 2.31 3198 -. |\mb H++ha 1.50 3.48 1. 40 1. 87 1. 88 2.15 2. 30 1.01 2.04 (A) :0 HI g) am» 1.18 —184 Table 6.4—1 (cont'd.) 5—9 10-19 20—49 50—99 100—199 200-499 500+ Small total market continued 3219 textile NgE. C. 1.24 1.37 1.52 1.99 1.84 _2,20 '3221 custom ta. 1.79 .ZLQQ 1.47 1.40 1.80 1. 92 1.15 3233 leather su 1.68 1.73 2.03 1.79 2.20 892. 2.49 3319 cork NEC, 1.34 1.37 1.22 1.32 1.88 3422 com.print 1.79 1.3 1.52 1494 1.14 1.09 3540 mi§92 petroleum& coal 2.00 2.35 4.22 2.65 3.41 4.26 Note: Small establishments (those with 5-99 workers) are considered to predominate in an industry if they account for half or more of the industry's employment. The benefit-cost ratio is a way of measuring total factor productivity and is the ratio of the value added to the input cost in given size group of industry. The input cost is assumed to be r * K + W, where r is the shadow price of capital (here 20%), K is fixed assets and W is wage bill. Here we assume no distortion in labor market. Source: The calculation is based on data in 'Report on Mining and Manufacturing Survey, 1984'. 185 the 4-digit small scale industry based on the size category; for example, the Social Benefit-Cost ratio of fish industry(3114), 'with a size category of 5-9 workers, is 1.38. The most efficient firm size in the fish industry becomes the size category of 200-499 workers, which has a highest SBC ratio(2.47). Based on this process,Table 6.4- 2, shows the number of industries in which the Benefit- Cost ratio is the highest in the size category indicated. Table 6.4-2 shows that in only a few of small scale industries, SSEs have the highest total factor productivity. These figures confirmed the conclusions of Ho’s, relative ,efficiency measurements. In 1984, the Benefit-Cost ratio was highest in only 9 small establishments with 5—99 workers of small scale industries but the Social Benefit and Cost ratio was highest in 21 large establishments with over 100 workers of the small scale industries. The optimal size among the most efficient SSEs was 50-99 workers. The industries which do not fare very well on the efficiency-criterion by Ho’s method and the SBC ratio are as follows; Cane container(3312), Smelting and refining (3722), Medical instrument(3851), Dressed fur skin(3232), Jewellery(3901), Mis. petroleum and coal(3540). In these cases, the two methods do not reveal the consistent bias in efficiency measures for either the LSEs or the SSEs. From the analysis of results based on these two 186 Table 6.4—2 Number of Industries where Benefit-Cost(Capital Shadow Priced at 20%) is greatest in the Size Category Indicated —Small Scale Industries(Industries where SSEs have 50% + of Employment) - Korea 1984 5—9 10-19 20—49 50-99 100-199 200-499 500+ exc. 1A 0 0 0 1 l 3 0 0 IB 0 0 0 0 2 3 0 4 1C 1 0 0 l 2 0 0 0 2A 0 1 1 0 1 1 2 0 2B 0 0 0 1 l 0 0 0 2C 0 0 0 0 2 0 1 0 Subtotal 1 1 1 3 9 7 3 4 3B 0 1 0 2 1 1 0 3 Total 1 2 1 5 10 8 3 7 Source:'Report on Mining and Manufacturing Survey,1984' Note: by Economic Planning Board, The Republic of Korea. All industries which contain the words'miscellaneous' or 'Not Elsewhere Classified(N.E.C.)' in their titles are classified as 'excluded' category of size group in Table 6.4—2. 187 measurements, I conclude that policies favoring the LSEs including the cheap capital policy might partly contribute to the increasing efficiency displayed by the LSEs as! compared to the SSEs in the small scale industry, by utilizing the economies of scale in the LSE. The lower capital cost resulted in accelerating the modernization of the LSEs in the small scale industries. On the other hand, the higher capital cost resulted in delaying modernization and endangering the survival of SSEs in the small scale industries. The selective manner in which government intervention in Korea was carried out is not fully covered and could not be within the framework of my simple model. But it is important because Korean policy resembles that of Japan which directed investment to the profitable export markets. This may be a significant explanation of the efficiency of Korean LSEs. 3. Technical and Pecuniary Economies of Scale In the literature on industrial organization , the term ’economies of scale’ is defined as including the technical as well as pecuniary economies of scale. The former refers to the actual physical organization of production activities by reducing the ratio of inputs to output; thus achieving an increase in social economic efficiency. 0n the other hand, a pecuniary economy of 188 scale stems from a lower input price and the ability to exert market power by exploiting the advantage of a larger sized firm , which provides private economic gains. The policy favoring LSEs, by providing cheaper financial assistance to LSEs, might enable them to accelerate their technical and pecuniary economy of scale at the expense of SSEs; this,again, is partly reflected in the higher efficiency of LSEs in small scale industries. But this result was obtained at the expense of weakening the employment increases created by SSEs, which partly contributed to worsening income inequality. Therefore the government should try to establish a fair and equal investment climate for both SSEs as well as LSEs. In order to do this , first ,the factor market for capital and labor should have no distortion. Then the rent and the wage will reflect the correct values of capital and labor. The correct price signal through the market mechanism will lead to an optimal resource allocation and improve the total factor productivity of SSEs in the small scale industry. But as the study of Korean SSEs shows, a distortion in the capital market, created by a policy favoring LSEs , led to less efficient resource allocation and a worsening TFP in SSEs in the small scale industry. By the same token, another distortion in the capital market, created by a policy favoring SSEs, also l 189 may make SSEs more capital intensive, reduce their employment potential_ and let inefficient SSEs survive longer in the market. E. General Policy toward SSEs for Entry & Expansion 1. Entrepreneurship Promotion Clearly, the future of the SSEs lies in the hands of entrepreneurs ,who are willing to take risks in order to make profits. These entrepreneurs are called innovative organizers. In order to promote SSEs, first the cultivation -of entrepreneurs in terms of numbers and quality is more important. The way to promote entrepreneurs is through education or Social campaign encouraging potential entrepreneurs to enter small scale industries. The government should remove legal and policy constraints for easy entry of potential entrepreneurs. Complicated licensing procedures usually tend to discourage the development of a small scale business. As long as the legal regulations require minimum qualifications such as preventing overcrowding, or avoiding an excessive demand for imported equipment etc., the government should simplify these procedures. 2. Fair ACcess to Capital 190 Usually, the SSEs rely on personal saving and the unorganized money market for their capital funding, because the low credit rating and high risks of SSEs tend to ruin the chance of borrowing from institutional banking organization. Also the failure to meet needs of working capital can draw a' cloud over‘ the future of SSEs. Hence, the existing banking institutions for assisting SSEs should be fully utilized. Liedholm’s study(1986) suggested that the interest rate charged by lending agencies should be high enough to cover the operating expenses including the cost of funds; then these agencies would be more successful in negotiating loans to the SSEs. 3. Information Service In order to enter and expand into the market, entrepreneurs require information about the market ,f managerial skills and technical training. To promote efficiency in SSEs , the government‘ or lending agency should assist with specific information regarding the product market, utilization of capital, expertise and other relevant data, through the medium of loan screening. This non—financial assistance can significantly alter their outlook. The government, or trade and industry associations such as the small and medium industry promotion corporation and Korea production technology service 191 corporation can set up or utilize some agencies specialized in such information services. Kilby's study(1986) found that this non-financial assistance was especially successful in situations where there seemed to be only 'one single missing ingredient' that needed to be supplied to the firms in LDCs. Also providing market informations to small firms is one way of strengthening the market condition. 4. Competitive Relations with LSEs When the SSEs aim at the domestic as well as the export market as an outlet for their products, they face tough competition from the LSEs. When the economy grows fast enough to cause labor shortages in the industrial sector, and reduces the wage disparity between the LSEs and the SSEs, it is more difficult for SSEs to compete against LSEs by utilizing cheap wages as a weapon. The only way to overcome this challenge is by SSEs maximizing their efforts at increasing labor productivity through mechanization. Further, the SSEs should improve their total factor productivity , by advances in technology, resulting in technical efficiency. Since the early 1960's Japan has experienced a labor shortage.(Japanese economic survey 1963-64) In the future Korea may face a similar situation. Table 6.5 Korean average wage per worker in manufacturing ( real unit; million won/worker) 1975 1977 1979 small firms(5-99workers):A .76 1.14 1.71 large firms(100+workers):B 1.07 1.50 3.09 B/A (%) 140 131 122 (Source:report on mining and manufacturing 1975, 1977, and 1979, Economic Planning Board,Korea note:real wage is calculated by nominal average wage/ non-food prices) Although the wage gap between the LSEs and the SSEs still exists in Korea as shown in Table 6.5, the future labor shortage may further reduce the wage gap. In Japan, when the SSEs relied on loans to mechanize their facilities, it caused a deterioration in their financial composition, and an unsettled management due to the sudden change in size. This situation resulted in a vicious cycle because it weakened the SSEs' ability to secure funds. The SSEs were being forced out of business until the government rescued them. From the Japanese experience, we may conclude that the government should assist in the SSE's modernization through various schemes, such as advice, guaranteed loans, and tax concessions for environmental improvement for SSEs. Clearly ,this modernization scheme for SSEs should be distinguished from 193 an excessively biased capital subsidy for SSEs, which only helps to delay the death of inefficient ones. In the long run, the survival of the SSEs can contribute to strengthening the existence of a free and competitive market. Also, the SSEs can aim at a different market( the low income market) by utilizing product differentiation in quality and package. 5. Complementary Relations with LSEs In order to survive, the SSEs should adapt themselves to the new environment Fby diversifying, by utilizing their unique characteristics, and by mutual cooperation; a potential exists for the SSEs in certain industries for complementarity and interaction between LSEs and SSEs. This includes the ancillary production of component parts by the small factories for the large factories(backward linkage), and also by the large for the small(forward linkage). This backward linkage is called subcontracting. . a. Scitovsky's External Economy Scitovsky(1964) showed that the integration between two industries by using each others'output can create a peculiar external economy through a market mechanism; applying Scitovsky's externality concept to the complementary relationship between the LSEs and - the SSEs,we can arrive at the rationale for the important role 194 of SSEs, in terms of an industrial policy. As the LSEs use the SSEs' output as their inputs, this backward integration creates an external economy; if the LSEs decide to invest more for their expansion and hence demand more inputs(such as small firms' output),this incremental demand can push the factor price higher and lead to higher profits in the SSEs. In turn,these profits can lead to further investment and can cheapen the SSEs' output price, establishing the mechanism of the external economy. The LSEs and the SSEs can each constrain the other's investment plans; with backward integration, the limited capacity of the SSEs will constrain the demand by the LSEs and, again, the limited capacity of LSEs will constrain supply by the SSEs. This limitation can be removed only by the simultaneous expansion of both LSEs and 'SSEs. Therefore with vertical integration (backward linkage) ,the role of SSEs becomes important for the LSEs' survival. b. Subcontracting Subcontracting is found to be of as much benefit to the smaller units, as it is to the larger ones. The system of contracting to small units enables LSEs to operate at relatively lower costs, since the cost of manufacture by SSEs is often comparatively less because of lower overheads and lower wages. Larger exporters tend to turn 195 to the foreign suppliers for intermediate inputs, a situation which negatively affects the balance of payments. Considering the advantages of subcontracting, I think the government should assist the growth of SSEs by permitting, encouraging, and requiring its agencies to purchase from them; large government contractors should be encouraged to subcontract to smaller suppliers; finally the government may indirectly persuade large private industries to give a fair chance to SSEs by subcontracting to small domestic suppliers. A good example of the subcontracting system with LSEs is in Japan. .Japanese multi-national LSEs have their own subcontractors, sub-sub contractors, sub-sub-sub contractors etc., which makes a vertical grouping connection with the large parent firms. Japanese parent companies gave excellent technological ,managerial, .and financial support to small sub- contractors,. improving their efficiency (Watanabe 1978); to encourage subcontracting, the Japanese government provided financial assistance to smaller producers through the parent firms(The Law of Cooperative Association of Medium and Small Enterprises, August 1949). Subcontracting can be used to further the goals of self-sufficiency in industrialization and decentralization of industrial location in Korea. The Korean government can encourage subcontractors to locate their firms in 196 secondary cities or rural areas, by providing financial assistance and friendly persuasion through the large parent firms; the managerial and technological guidance from the central or local government and from the large parent firms to small subcontractors is valuable in improving the efficiency of SSEs in Korea. F. CONCLUSION So far I have reviewed the problems of the SSEs in Korea and tried to suggest possible solutions for them. The findings about these issues are: First, the past industrial policies in terms of overvaluation, interest etc. have consistently favored the LSEs. This policy favoring the LSEs reduced the capital cost for the LSEs relative to the SSEs. In 1973, the capital cost for LSEs was 30% cheaper than the capital cost for SSEs. Although this advantage has lessened since 1981, even in 1984 the capital cost for LSEs was . 6.5% cheaper. Second, the SSEs tend to be concentrated around urban areas. The SSEs in urban areas employed about 70% of the total labor force in SSEs in 1984. Third, Ho's and the SBC method confirmed that the SSEs in ' small industry became less efficient than the LSEs in such industry during 1968-84. The suggetions are mainly aimed at creating a fair environment based on the 197 market mechanism where the LSEs and the SSEs can compete on even terms and complement each other in order to survive and co-exist. Therefore the balanced growth and composition of SSEs and LSEs in the overall industrial structure may contribute to fast economic growth, strong growth of employment, and a more equitable income distribution in the economies of other LDCs as well as in Korea . 198 A REVIEW OF TAIWAN’S SSES A. INTRODUCTION In the previous chapter, I found that in the Korean capital market the SSEs were discriminated against; they were concentrated around urban areas (specially Seoul) and were not efficient compared to LSEs in small scale industry. As conditions in Taiwan are similar to those in Korea ,in terms of a high density of population with a poor resource base though the record is better in terms of economic growth. and income distribution, it is meaningful for us to compare SSEs in Korea and Taiwan. First I will review the relative share of SSEs in the industrial structure of Taiwan, then their location and finally the efficiency of SSEs in small scale industry. From these studies, we will be able to get a perspective on the SSEs in Taiwan and derive a course of action for SSEs in Korea and other LDCs. B. A Comparison of Social Indicators for Korea and Taiwan Korea and Taiwan are known as NICs (Newly Industrializing Countries) that have achieved a high economic growth as well as an equitable income 199 distribution,re1ative to other LDCs. But a careful look at these two countries will reveal the superior record of Taiwan in terms of growth and income distribution. As Table 7.1-1 shows, during the period 1965-81, Korea achieved a growth rate of GNP of 8.7% , and Taiwan 9.4% . Also, during 1965-76,the GINI index worsened in Korea from .344 to .381. On the other hand, Taiwan recorded an improvement of the GINI index from .322 to .289. This indicates that Taiwan registered a better economic growth and income distribution. The better income distribution in Taiwan seems to result from various policies including one of land reform. In Taiwan, land reform between 1949-53 reduced tenant rents,and public lands .were sold to farmers; land holdings were limited to about 1.7 acres of paddy and 7.2 acres of dry land, and the rest was redistributed to other rural households. This resulted in an effective improvement in income distribution, better management of land and higher productivity. The existence of a large number of small business enterprises also contributed to the equitable income pattern in Taiwan, because these small businesses made the competitive market system work in an efficient fashion. C. Early Economic Background for Taiwan’s SSEs 200 Table 7.1-1 Average Annual Growth Rates in Real Terms 1965—1981 (%) Korea Taiwan Population 1.9 2.3 Employment 3.4 3.7 Gross National Product 8.7 9.4 Manufacturing Output 20.6 15.5 Total factor productivity a) b) in manufacturing 2.1 4 Exports 26.0 18.9 GNP per capita 6.7 “ 6.9 Labor productivity 5.2 5.4 7.9 7.3 Real wage in manufacture Table 7.1-2 GINI Index of Inequality of Income Distribution 1965 1970 1976 Korea .344 .332 .381 Taiwan .322 .293 .289 Note: a) covers the period of 1966—1975, while b) covers the period of 1952-1980. ‘ Sources: Tibor Scitovsky, 1985 ' Economic Developmet in Taiwan and South Korea; 1965—81‘ Table1,P216& Table3, P 218 Food Research Institute Studies No.3. For a) & b), Harry Oshima, 1986 'The Transition from an Agricultural to an Industrial Economy in East Asia ‘Economic Development & Cultural Change. 201 Before 1950, Taiwan was a agriculture—oriented economy with relatively equal income distribution. The cultivable area was 25 % of total land, and the tropical weather was warm enough to grow crops all the year around. The major crops were rice and sugar cane. Since early 1950, the land reform program has made a dramatic change in the agricultural sector. Taiwan’s land reform went through three stages; 1) rent reduction in 1949, 2) sale of public land in 1951, 3) a land-to-tiller program in 1953(Koo 1968). This land reform contributed effectively to income redistribution. During 1950-60, rent reduction and the higher land productivity due to land reform increased rural tenants’income, consumption, and saving. The increased rural income made it possible for the farmers 'to afford education for their children and demand more agricultural inputs as well as consumer goods. Consequently the migration outwards by well educated young people to seek off-farm employment reduced Ag-unemployment. Also the higher demand for agricultural inputs and consumer goods made possible the development of industries with backward and forward linkages to agricultural sector. Therefore the land reform program in the agricultural sector contributed to promoting the industrial sector indirectly. The opportunity cost of the land reform program turned out to be minimal, because there was no competition 202 for skilled labor between the Ag and industrial sectors, and the demand for land input for industrial projects was not high during that period(Koo 1968). During the early 19505 , the migrants from main land China began to establish many Small businesses in the non- Ag sector. Taiwan’s government took over LSEs from the Japanese so that the revenues created from these public enterprises went to the treasury. At the same time, all efforts for industrial development focused on the recovery of the then-existing infrastructure, the Ag-based industries, and the import-substituting industries. Since early 19505, Taiwan’s high interest rate policy has accelerated saving and capital accumulations by curbing hyperinflation, and resulted in keeping the deposit & loan interest rate close to the equilibrium rate, so as to maximize the return on the investment. Also the high loan interest rate played a role of limiting the profits of the enterprises, which slowed down the growth of the firm size(Scitovsky 1985). D. Taiwan’s Industrial Policy A review of Taiwan’s industrial policy in terms of the factor market and the output market. 1. Factor Market 203 a. Capital Market In Taiwan the. capital market is less distorted; Table 7.1-3 shows that the Korean capital market conditions favored LSEs against SSEs clearly, while the Taiwanese capital market did not have any such bias. In 1978 the capital cost for Korean LSEs was 25.7 % less than that of SSEs, in terms of overvaluation, interest and taxes. On the other hand, the relative capital cost for Taiwan’s LSEs was 0.4 % less than that of SSEs. Therefore Table 7.1-3 proved clearly that Taiwan’s capital market is less discriminating(more neutral) against SSEs‘ than the Korean' market. Although some distortions in Taiwan’s capital. market still exist, the degree of distortion is much less. b. Labor Market The studies(Myers 1986, H0 1980) agreed that Taiwan’s labor market was relatively distortion- free and highly competitive and open. The rationale for this statement lies in the facts that unions are weak and disorganized,while the ’minimum wage' is not generally enforced . During the last two decades labor productivity has more than kept pace with the increasing real wage. But the real wage gap between the LSEs and the SSEs still exists, though this is smaller than in Korea; 204 Table 7.1-3 Policy—induced Capital Cost Distortion in Large and Small Scale Enterprises (Korea and Taiwan) (Expressed as the percent difference in large firms' cost to small firms) Percentage Difference in Capital Cost owing to: Year Nation Trade regime Interest Taxes Total Capital (overvaluation) 1976 Korea —2.1 -30.4 +10 —22.5 Taiwan -4.6 —16.27 +22 + 1.13 1977 Korea -5.3 -28.5 +10 -23.8 Taiwan —2.5 -17.05 +22 + 2.45 1978 Korea -7.7 -28 +10 -25.7 Taiwan -5.9 —16.5 +22 — 0.4 1979 Korea -15 -33 +10 -38 Taiwan -10.5 -l6.99 +22 - 5.49 1980 Korea —1.6 -32.5 +10 -21.1 Taiwan -5.1 -17.31 +22 - 0.41 1981 Korea -2.2 -23 +10 -15.2 Taiwan —4.3 —18.18 +22 - 0.48 1982 Korea —7.5 —21.7 +10 -19.2 Taiwan -3.5 —19.64 +22 - 1.14 1983 Korea —5.8 -15.5 +10 -11.3 Taiwan -4.0 -18.26 +22 - 0.26 1984 Korea -0.1 —16.4 +10 — 6.5 Taiwan —5.9 -17.92 +22 — 1.82 Source: The Korean data are from Table 6.2 in Ch.6. Notes: The calculation for Korean data is shown in Appendix A. The calculation for Taiwan data is shown in Appendix B. 205 in 1971 the annual average wage paid by LSEs with 500+ workers was about 60 % higher than that paid by SSEs with 1-9 workers(H0 1980). The main reason for wage gap between LSEs and SSEs was the difference in quality, training and skills of the work force. 2. Product Market a. Export/Import Policy In the early 19605, the government chose an export oriented policy in order to dispose of the surplus local products and to explore the foreign market. Since then, Taiwan’s government has adopted a more realistic foreign exchange rate; liberalized import controls and generally reduced the factor and product market distortions. At the same time, it offered tax rebates, and five year tax holidays for newly established enterprise. In the early phase of the export promotion policy, SSEs were not much affected because they produced for the domestic market. But as this gave an advantage to economies of scale, the size of firm assumed importance; SSEs in Taiwan realized that they were in competition with. many new LSEs for the same .unskilled workers. These factors in the export oriented policy made conditions less favorable for SSEs. But because the government policy has been less discriminating, the SSEs in Taiwan have played a 206 much larger role in' industrial development as well as exports. b. Agricultural Policy To improve living standards in the rural areas, the Taiwanese government relied on land reforms (1949-53), improving the infrastructure and raising the ag-output; to raise Ag-output and income, government agencies(the Joint Commission on Rural Reconstruction and the Taiwan Provincial Department of Agriculture and Forest) increased the supply of fertilizer,improved irrigation facilities, provided better seeds, undertook pest control and developed high-yielding crop varieties. The increase in rural incomes also led to a strong demand for non—food consumer goods and services, and for material and equipment inputs in Ag production; the attempt' to develop a new export-oriented food processing industry in the rural area was successful. This adaptive Ag sector combined with a more outward -looking strategy contributed to the strong growth of new Agro-industries, which have been a major source of employment and non-farming income in rural Taiwan. The improving infrastructure (transportation system, roads, etc) and the small land area helped to extend industrial growth from the major urban centers to smaller towns and the surrounding country side. 207 c. Capital Formation Policy In the 19505, US aid played a significant role in financing Taiwan’s economic growth and national defence. But as economic conditions improved in the 19605, other foreign loans and investment replaced US aid. To attract foreign investment, in 1959-60 the Taiwan government amended the investment law, providing tax concessions (five year income tax holidays), allowing repatriation of profits and interest two years after completion of investment, and giving extensive protection against expropriation. But the bulk of 5 investment was financed by high domestic savings; thus household saving supplied from 36 % - 47% of total capital formation in 1977, levelling off to 36 % in 1981. ‘The average savings propensity of households was high, ranging from 11 % to 24 % during 1965—1981. The policy that enabled this high rate of savings was one of insuring a positive real rate of return on savings deposits and a stable price level and offering attractive tax incentives. With the aid of high domestic savings, Taiwan successfully financed its growth without relying heaVily on foreign capital. This is in contrast with the Korean case. At the end of 1984, the Korean external debt was 43 billion $ or 53 '% of its GNP, but the Taiwan debt was 208 under 12 billion $ or 21 % of its GNP. Between 1960 and 1983, Taiwan saved, in gross terms, 25% of its GNP while Korea saved 17 % (Myers 1986). d. Industrial Strategy and Innovations Levy(1987b) characterized the Korean industrial strategy as an ’Assembly Strategy’ which requires substantial initial investments and may absorb substantial initial loss; this strategy can help to acquire complex technology, specially the process technology permitting productivity gain. On the other hand, Levy saw Taiwan’s industrial strategy as a ’Bootstrap Strategy’ which requires small size at entry, and allows low cost of failure under high uncertainty; this strategy can focus on products with short life cycles, can help to acquire technology facilitating the design of new products & can be more supportive of innovation. As a result, the entry barrier to Korean SSEs limited the advance of innovation and development of technology to a few LSEs, but the easy entry for Taiwan’s SSEs made it possible for innovation and development of technology to be driven evenly by many SSEs. E. SSEs in Taiwan’s Industry 1. Small Scale Dominating Industrial Structure 209 The development and prosperity of Taiwan’s SSEs can be attributed to many factors, including a cultural preference for economic independence through owning business concerns. As Scitovsky(1986) pointed out, the migrants from the mainland, as well as from abroad, brought in their own capital ,and commenced establishing their own enterprises in the manufacturing and service sector; a high domestic saving rate made it easy to fund new business establishments, and a large field of small sized enterprises offered better conditions for new comers entering the market. Political stability and the greater stock of human capital of high quality combined to provide an favorable environment for the development of SSEs(Levy 1987c). Levy(1987b) regarded the emergence of Korean large conglomerates as an efficient institutional response to greater .underdevelopment of the nation & a host of associated market failures; the lack of subcontracting opportunity and the fact that few indigeneous traders willing to eXplore the export market for SSEs’ products seemed to make the initial investment cost and the size at entry substantially larger. On the other hand, Levy found that Taiwan had a favorable environment for SSEs, where the market functioned more efficiently and SSEs did not face an entry barrier; the easy entry to subcontracting and the presence of Taiwanese traders willing to explore 210 the export market for SSEs’ products seemed to make it easy for SSEs to start production at a small scale with a little investment and specialized market information. Taiwan’s policies did not favor one industrial sector against another. Consequently, aggressive small scale entrepreneurs and a neutral government policy ensured easy access to funds and prosperity for the SSEs. Government policies nurtured private enterprises which were market-oriented and had a comparative advantage, using abundant resources efficiently and only gradually switched to the use of more scarce resources like capital goods. These conformed to the criteria of efficiency as noted by Myer(1986), who argued that if the markets are highly competitive and allocate resources efficiently, then technical diffusion, productivity gain and employment opportunity occurs in all sectors. So there is no case for a discriminatory policy . As the Levy(1986) study shows, some of Taiwan’s policies such as cross-subsidization of domestic sales in ~export and import competing industries did negatively affect SSEs but the extent of intervention by the Taiwan government was less extensive than in Korea. A5 a result, - the position of SSEs in Taiwan is better. Table 7.2 shows that in manufacturing , SSEs with below 100 workers are significant in terms of employment. 211 Table 7.2 The Relative Position of SSEs in Manufacturing Taiwan (1966) workers % of total Total employment 589,660 100 Small scale enterprise 251,879 43 1—4workers 23,447 4 5-49workers 177,256 30 50-99workers 51,176 9 Large scale enterprise 100+ workers 337,781 57 (1971) Total employment 1,201,539 100 Small scale enterprise 427,988 36 1-4workers 31,360 3 5—49workers 285,843 24 50-99workers 110,785 9 Large scale enterprise 100+workers 773,551 64 (1981) Total employment 2,178,191 100 Small scale enterprise 957,591 43.97 1-4 workers 112,649. 5.17 5—49 workers 571,168 26.23 50—99 workers 273,774 12.57 Large scale enterprise 100+ workers 1,220,600 56.03 Source:1966 & 1971 data are based on Ho's World Bank Staff Working Paper , No.384,'Small Scale Enterprises in Korea and Taiwan' P5. ' 1981 data are based on 'The Report of 1981 Industrial and Commercial Censuses of Taiwan-Fukien District of the Republic of China,‘ Volume 3. 212 In 1981, Taiwan’s SSEs with below 100 workers employed 43.97% of the total labor force in the manufacturing industry. In 1966 it was 43%. On the other hand, Korean SSEs in 1984 employed 34% of the manufacturing labor force. Taiwan’s SSEs are significant enough to employ a considerable section of labor in the manufacturing industry, particularly , the SSEs with 5-49 workers. They dominate in terms of employment among the SSEs (1981, 25.23%). If SSEs account for half or more of the industry’s labor force, SSEs (those with fewer than 100 workers) may be said to be predominant in an industry. An industry where SSEs predominate is called a small scale industry (Ho 1980). Following this classification, Table 7.3—1 shows that in-1976 the SSEs with 1-99 workers accounted for half or more of the country’s employment in 67 out of 134 four digit manufacturing industries; the SSEs employed 70% of the workers and produced 60% of the value-added in these 67 small scale industries. As for their relative importance in overall manufacturing, SSEs in small scale industries employed 22.4 % of all manufacturing workers and produced 13.7% of the total manufacturing value added in 1976. SSEs in 67 industries, where small plants do not predominate, employed_16% of all manufacturing labor and produced 9.2% of the total. value added in the Table 7.3—1 Share of Small Enterprises with 1-99 workers in Employment by Four-digit Industries, Taiwan. 1976 # of workers(000) share of employment(%) % of workers # of small large all small large all in small est. est. est. est. est. est. est. est. 75—100 28 192 39 231 10.1 2.0 12.1 50-74 39 234 142 376 12.3 7.5 19.8 25-49 38 190 339 529 10.0 17.8 27.8 0 -24 29 115 654 769 6.0 34.3 40.3 Total 134 731 1,174 1,905 38.4_ 61.6 100.0 Table 7.3—2 Share of Small Enterprises with 1—99 Workers in Value Added by Four-digit Industries, Taiwan. 1976 Value Added share of manufacturing (NT $ million) value added(%) Z of workers # of small large all small large all in small est. est. est. est. est. est. est. est. 75-100 28 11,865 3,413 15,278 6.1 1.8 7.9 50—74 39 14,661 14,204 28,865 7.6 7.3 14.9 25—49 38 11,144 32,570 43,714 5.8 16.9 22.7 0 —24 29 6,587 98,866 105,453 3.4 51.1 54.5 Total 134 44,257 149,053 193,310 22.9 77.1 100.0 Note: Employee is based on data of persons engaged as of Dec.31,1976, and value added is based on data of annual gross value added. Source: The calculation is based on data in ' The Report of 1976 Industrial and Commercial Censuses of Taiwan-Fukien District of the Republic of China, Volume 3, book 1' 214 manufacturing industry. Table 7.4 shows indirectly through the number of enterprises that in the other industries apart from the manufacturing sector SSEs with 1-4 workers employed a majority of the work force: in the electricity, gas and water sectors there are 40 % of the total numbers employed; in the construction sector, 38.2%; and especially, in whole sale or retail sale trade, 87.7%. Therefore, we may conclude indirectly that in Taiwan the SSEs are absorbing a significant portion of the labor force in the across-the-aboard industries including the manufacturing and the service industries. 2. Decentralization of SSEs In Taiwan, the rural areas have not been excluded from the benefits of economic development ,because rural households can add to their incomes from non-farming activities. The variety of non-farming activities in Taiwan’s rural areas is impressive. With good infrastructure like roads and transportation, farmers can commute to the cities to seek part time employment; the rural locations of SSEs give the opportunity to farmers to engage in non-farming activities. In the early phase of development, the government stressed the production of consumer goods using products from the agricultural sector as inputs; this resulted in the development of Number of Enterprises in Selected Non-manufacturing 215 Table 7.4 Industries, by Size Industry Total Electricity gas&water 40 (Z) (100) Mining 944 (Z) (100) Construction 12,541 (Z) (100) Trade 315,382 (Z) (100) Other Industry 92,944 (Z) (100) Source: Taiwan - 1981 1—4 5—9 10-19 20—49 16 1 2 2 (40) (2.5) (5) (5) 210 232 203 162 (22.2) (24.6)(21.5)(17.2) 4,792 2,787 2,083 1,635 (38.2) (22.2)(16.6)(l3.1) 276,679 26,032 8,455 3,276 (87.7) (8.3) (2.7) (1.0) 74,264 9,265 4,637 3,347 (79.9) (10) (5) (3.6) 50—99 6 (15) 57 (6) 593 (4.7) 620 (0.2) 874 (0.9) 100—499 500+ 9 4 (22.5) (10) 67 13 (7.1) (1.4) 515 136 (4.1) (1.1) 290 30 (0.1) (0) 449 108 (0.5) (0.1) 'The Report on 1981 Industrial and Commercial Census Taiwan-Fukien Area, the Republic of China,‘ Volume 1 General Report published by Directorate—general of Budget, Accounting and Statistics, Executive Yuan. June, 1983. 216 food processing industries. a rising income from non- farming activities made farmers less dependent on agriculture as their sole source of income. Table 7.5 shows that in 1981 rural employment in manufacturing took a considerable portion( 43%) of the total manufacturing labor force. In 1966, the rural employment in manufacturing was 32 %. This data confirms the continuous increase in rural manufacturing employment through non-farming activities. This table assumes that urban areas in Taiwan cover Taipei, Kaohsiung, Taipei Hsien, Keelung, Taichung and Tainan(Ho 1980). In Table 7.6 the majority of enterprises regardless of size including the large and the small scale are located around rural areas; the number of very small establishments with 1-4 workers located in the rural areas was 61 % in 1981. In across-the-board industries, a rural location is also very common. Table 7.7 shows that in 1981 the number of mining establishments located in the rural areas was 67%; manufacturing ,57%; electricity,70% etc. This data implies that many SSEs are located in the rural areas of Taiwan. Further Table 7.8 shows the manufacturing locations in more detail; a majority of the food , wood and minerals industries, are located in rural areas. 82 % of food based industries, 78 % of wood based industries, and 71 % of mineral based industries are to be found in rural areas. LhA-IH' "-A 217 Table 7.5 Employment of Manufacturing,distributed by Location Taiwan (unit:000) Year Total Urban Rural 1966 590 402 188 (Z) (100) (68) (32) 1971 1,202 613 589 (%) (100) (51) (49) 1976 1,906 1,114 792 (%) (100) (58) (42) 1981 2,196 . 1,255 941 (%) (100) (57) (43) Note: Urban areas cover'Taipei,Keelung,Taichung,Tainan, Kaohsiung and Taipei prefecture'. Table 7.6 Number of Establishments in Manufacturing,distributed by Location Taiwan - 1981 Location Total 1-4 5-99 100+ Urban 40,827 17,892 21,350 1,585 (%) 43 . 39 47 40 Rural 53,719 27,519 23,786 2,414 (%) 57 61 53 60 Sources: 1966,1971 data are based on Ho's World Bank Staff Working Paper No.384"Small Scale Enterprises in Korea and Taiwan' P 21 Table 2.7. 1976 data are based on 'The Report of 1976 Industrial and Commercial Censuses of Taiwan-Fukien District of the Republic of China,‘ Volume 3 Book 1. 1981 data are based on 'The Report of 1981 Industrial and Commercial Censuses of Taiwan-Fukien District of the Republic of China,‘ volume 3. 218 Table 7.7 Number of Enterprises Units of Industry and Commerce by Locality and Industry Taiwan — 1981 Industry Total Urban Rural Mining 944 311 633 (%) (100) (33) (67) Manufacturing 91,562 39,543 52,019 (Z) (100) (43) (57) Electricity gas & water 40 12 28 (Z) (100) (30) (70) Construction 12,541 6,832 5,709 (Z) (100) (54) (46) Wholesale trade retail trade 315,382 152,456 162,926 (%) (100) (48) (52) Other Industry 92,944 44,762 48,182 (%) (100) (48) (52) Table 7.8 Number of Enterprises Units of Manufacturing by Locality Taiwan — 1981 Manufacturing Total Urban Rural Food,beverage,tobacco 8,748 _1,620(18%) 7,128(82%) Textile, leather 9,996 5,041(50%) 4,955(50%) Wood, its product 8,662 1,986(22%) 6,676(78%) Paper, publishing 6,426 4,061(63%) 2,365(37%) Chemical products 11,294 4,894(43%) 6,400(57%) Non—metal mineral 3,716 1,114(29%) 2,602(7l%) Basic metals 2,007 1,057<48°/.) ' 950(48‘7.) Metal products 36,497 17,756(48%) 18,741(52%) Other manufacture 4,216 2,014(47%) 2,202(53%) Note: Urban areas cover 'Taipei, Kaohsiung,Taipei Hsien, Keelung, Taichung and Tainan' . Source:'The Report on 1981 Industrial and Commercnal Censuses of Taiwan—Fukien District of the Republic of China'. Table 7.9 shows in further detail the aspects of these three rural based industries; in the food industry ,70% of the total number of food processing enterprises, 93% of the oil pressing and milling enterprises, and 79% of the condiment producing enterprises are located in rural areas. In the wood industry, 79% of the total number of wood and bamboo processing enterprises and 74% of non- metal furniture producing enterprises are located in rural areas. In the non-metallic mineral industry, 84% of glass product enterprises and 85% of cement enterprises are located in rural areas. This strong trend towards rural location of manufacturing industries in Taiwan is in contrast with the situation in Korea, where the majority of the LSEs and SSEs are located in urban areas. This pattern in Taiwan suggests that expansion of Korean SSEs and those of other LDCs in rural areas may improve. rural incomes. Considering the lecal conditions, the non-farming activities based on a linkage with LSEs and the agricultural sector could be ideally developed in SSEs in the rural areas. F. Efficiency of SSEs 1. Characteristics of SSEs in Taiwan The Taiwanese policy of reliance on the market 220 Table 7.9 Number of Major Enterprises of Manufacturing in Rural Area Taiwan - 1981 Industry Total Urban Rural Food,beverage,tobacco 8,748 1,620(18%) 7,128(82%) Food manufacture 8,605 1,570(18%) 7,035(82%) food processing 1,827 551(302) 1,276(70%) oil pressing & mill industry 4,091 318(71) 3,773(93%) sugar factory & refineries 36 14(38%) 2(62Z) tea manufacture 213 72(33%) 141(67%) condiments 370 81(21%) 289(79%) other food 2,068 534(25Z) l, 534(75Z) Beverage& Tobacco 143 50(34%) 3(661) Wood & Wood product 8,662 1,986(22%) 6,676(78%) Wood & bamboo 6,389 1,389(21%) 5,000(79%) Non-metal furniture 2,273 597(26%) 1,676(74%) Non-metallic mineral 3,716 1,114(29%) 2,602(71%) Pottery, China 735 426(577.) ’ 309(432) Glass & its product 510 83(16%) 427(84%) Cement & Oment 870 133(15%) 737(85%) Other non—metallic mineral 1,601 472(291) 1,129(71%) Note: Urban areas cover 'Taipei, Kaohsiung, Taipei Hsien, Keelung, Taichung and Tainan'. Source:'The Report on 1981 Industrial and Commercial Census Taiwan—Fukien Area, The Republic of China' Volume 1 General Report published by Directorate—general of Budget, Accounting and Statistics, Executive Yuan June, 1983. % 221 system has improved the efficiency of SSEs; to see this in greater detail, we need to review the SSEs in terms of partial efficiency. First let us observe the change in the manufacturing sector. Table 7.10 shows the change in the manufacturing sector over time in terms of employment; the fast expansion of heavy industries, chemicals and metals is especially notable. The metal products industry increased its share of employment from 19% in 1966 to 32% in 1981 , within the total manufacturing labor force, and the chemical products industry also increased employment by 18.1% during the period 1966-71 and by 24% during 1976-81. On the other hand, the light industries, food and textiles suffered a decline or slowing down in share of employment; the food industry figure decreased from 21% in 1966 to 7 % in 1981 within the total manufacturing employment; this trend is similar to that in Korea. Heavy or capital intensive industry is expanding faster than the simple light industry, with capital deepening taking place. 2. Partial Efficiency Measurement of SSEs Table 7.11 shows that regardless of the size of the SSEs, heavy industry took a significant portion of the employment and the value added; in 1976, a VSSE with 1-4 workers in the metal product industry employed 39% of the 222 Table 7.10 Employment in Manufacturing by Industry Taiwan (unit: 000 persons) Industry 1966 1971 Net 1976 1981 Net increase increase (1966—71) (1976—81) Total Mfg. 566 1,170 604 1,908 2,197 289 Z of total Mfg. 100 100 100 100 100 100 Food,beverage 21.2 10.9 1.3 8 7 -4 Textile,leat. 20.6 26.4 31.8 25 22 Wood, paper 11.6 11.1 10.7 10 9 5 Product of chemical 14.3 16.2 18.1 16 17 24 Non—metal mineral 8.1 .7 4 5 6 Basic metal 3.1 . .1 3 3 6 Metal product 19.0 22.9 26.7 30 32 51 Other Mfg. 2.2 4.4 6.5 4 5 8 Sources: 1966 and 1971 data are based on Ho's World Bank Staff Working Paper, No.384 'Small Scale Enterprises in Korea and Taiwan' P 28. 1976 data are based on 'The Report of 1976 Industrial and Commercial Censuses of Taiwan—Fukien District of the Republic of China,‘ Volume3,book1. 1981 data are based on 'The Report of 1981 Industrial and Commercial Censuses of Taiwan—Fukien District of the Republic of China,’ Volume 3. 223 Table 7.11 Small Scale Manufacturing : Employment and Value Added distributed by Industry -Taiwan Industry Est. with Est. with Est. with 1-4workers 5—49workers 50—99workers (1971) employ. V.A. employ. V.A. employ. V.A. (NT$ mil.) (NT$ mil.) (NT$ mil.) Manufacturing 31,360 725 285,843 5,840 110,785 2,359 Z distribution 100 100 . 100 100 100 100 Food,beverage 56 58 11 10 7 2 Textile,leather 3 3 12 ll 20 18 Wood,its product 8 9 9 10 7 7 Paper, print 3 3 8 8 4 5 Product of chemical 7 6 15 14 18 25 Non-metal mineral 3 3 10 9 10 8 Basic metal, Metal prod.&equip. 17 15 29 32 27 29 Other Mfg. 4 2 4 4 5 6 (1976) Manufacturing 78,702 4,828 4446,973 26,299 209,702 13,138 Z distribution 100 100 100 100 100 100 Food,beverage 22 24 6 7 6 8 Textile,leather 6 5 14 13 21 19 Wood, its product 10 10 10 9 8 6 Paper, print 6 5 7 7 4 5 Product of chemical 10 9 16 17 18 18 Non-metal mineral 3 4 7 7 7 7 Basic metal 1 l 3 3 4 4 Metal prod.&equip. 39 39 33 33 28 28 Other Mfg. 3 3 4 4 4 5 Sources: 1971 data are based on Ho's World Bank Working Paper No.384,'Small scale Enterprise in Korea and Taiwan' 1976 data are based on 'The report of 1976 Industrial and Commercial Censuses of Taiwan—Fukien District of the Republic of China,’ Volume 3, book 1. 224 total labor force, produced 39% of the value added; an SSE with 5—49 workers in the same industry employed 33% of the labor and produced 33 % of the value added; an SSE with 50-99 workers in the metal product industry employed 28% of labor and produced 28% of value added. But light industries such as food, did not experience much of a change in terms of number of workers employed. In the 1-4 workers category, the food industry in 1971 employed 17,561 workers and produced 420 million NT$ of value added, but in 1976 employed 17,314 workers and increased the production to 1,158 million NT$ of value added. From this data we can conjecture that food industry has become a more capital-intensive industry. This can be confirmed in Table 7.12; in 1976 the capital/labor ratio of SSEs in the food industry was 474,000 NT$/worker. This ratio is high, compared to the figure in the other industries except for the basic metal industry(529,000 NT$/worker). SSEs in the metal products industry have a relatively low capital/labor ratio(205,000 NT$/worker) and produced an insignificant 27.6% of the value added; but in 1976 the SSEs in the basic metal industry had the highest capital/labor ratio(529,000 NT$/worker) and produced 98.3% of the value added. From these above data, the Taiwan case study has established' that SSEs can not necessarily be categorized as being labor-intensive; Capital/labor ratios 225 Table 7.12 Output/Capital and Capital/Labor ratios in Small Establishments(1—99workers) of Major Industry Groups. Taiwan - 1976 Z of V.A. Z emp. Value added Assets per in small in small per unit of worker(000NT$) est. est. assets in in small est. small est. 1) Food,beverage tobacco 11.2 37.6 .14 474 2) Textile,leat. 17.7 23.7 .19 281 3) Wood,its prod.49.3 59.2 .24 227 4) Paper,its product publishing & printing 40.7 59.8 .26 241 5) Products of , ' chemical 17.8 36.4 .18 348 6) Non-metallic . mineral 25.3 59.8 .25 227 7) Basic metals 98.3 43.8 .12 529 8) Metal products machine&equip.27.6 41.6 .29 205 9) Other Mfg. 36.9 36.9 .29 178 Note: Small establishments employ 1—99 workers, value added is based on data of annual gross value added, assets are based on data of value of assets in operation as of Dec. 31, 1976, employee is based on data of persons engaged as of Dec. 31, 1976. Source: The calculation is based on data in 'The Report of 1976 Industrial and Commercial Censuses of Taiwan—Fukien District of the Republic of China,‘ Volume 3, book1. 226 vary, depending on the nature of the industry irrespective of size. Table 7.13 shows in more detail, the capital/labor ratio in the manufacturing industry. In Table 12, I found SSEs in the basic metal industry have a high capital/labor ratio(529,000 NT$/worker). Within the basic metal industry ,SSEs of the ship scrapping industry have an impressively high capital/labor ratio (1,446,000 NT$/worker); a relatively high capital/labor ratio was also found in the SSEs in the food industry (474,000 NT$/worker); within the food industry, SSEs of the flour mill industry have a high capital/labor ratio(1,306,000 NT$/worker). G. The Total Factor Productivity of SSEs in Taiwan In the previous chapter, the SSEs in the Korean small scale industry proved inefficient, compared to LSEs.‘ Scitovsky(1986) pointed out that the high interest rate policy and free market forces made a significant contribution to the prosperity of SSEs in Taiwan. For a comparative study, we need to review the efficiency of SSEs. Once again , using Ho’s(1980) total factor productivity measurement and the Benefit and Cost ratio, I arrive at a more objective assessment. 1. Ho’s Method 227 Table 7.13 Output/Capital and Capital/Labor ratios for Industries in which Small Enterprises Predominate, Taiwan, 1976 4—digit small Value scale industry Added Assets (000NT$) per unit of asset per employee (all est.)(small est.(all est.)(small est. 1—99workers) 1) Food,beverage,tobacco dehydrated food .17 pickled food .22 bakery& cofectionery .21 edible oil fats .08 flour mills .10 husking of rice .18 tea processing .11 other flavoring product .18 misc.food preparation .24 2) Textile, leather spinning of silk .26 rope,cable & rugs .28 other textile . .13 3) Wood products sawmill & planning mill .17 timber drying .30 wooden container .36 bamboo products .51 other wooden products .34 wooden furniture .33 4) Paper & paper products paper container .24 other paper .23 publishing .19 printing .25 book binding .31 engraving & etching .31 5) Products of chemicals other chemical material .18 paint & lacquer .17 medicine & drugs .23 agricultural insecticide .11 other chemical product .23 industrial rubber prod. .31 other rubber product .29 plastic bags .22 plastic articles .24 other plastic products .27 209.79 176.15 392.75 1,475.06 1,611.23 393.56 361.00 409.79 348.91 260.71 193.32 483.50 353.03 301.23 147.09 91.00 160.01 183.15 341.07 372.81 313.99 309.06 149.77 181.74 364.58 634.49 472.45 815.11 594.23 245.94 234.47 228.89 273.22 215.27 l—99workers) 248.04 185.05 252.34 819.12 1,306.21 393.56 262.69 387.28 331.46 274.85 177.27 226.14 353.35 314.76 156.24 99.26 150.55 178.67 191.67 333.70 311.88 210.26 167.45 181.74 409.64 557.21 381.48 885.35 438.33 245.72 188.30 217.66 181.11 234.74 228 Table 7.13(cont'd.) 4—digit small Value added Asset5(000NT$) scale industry per unit of assets per employee (all est. )(small est. )(all est. )(small est.) 6) Non- metalli.c products glass & glass prod. .24 .33 451.13 191.33 cement products .25 .03 384.01 7,091.05 cons.clav.products .22 .23 232.12 205.85 abrasive material .21.21 357.82 357.82 marble products .30 .31 239.63 233.34 misc. mineral .27 .24 310.94 260.05 7) Basic metal Iron & steel found. .14 .12 560.39 498.42 ship scrapping .04 .04 1,273.43 1,446.38 copper refining .17 .17 . 324.36 389.02 other non—ferrous .23 .24 337.36 358.08 8) Metal product, machine and equipment metallic hand tools ' .34 .32 206.75 192.35 metallic mould & dies .30 .35 228.01 186.94 metallic cons.material .29 .29 184.41 191.06 aluminum products .13 .24 577.75 239.69 copper products .26 .35 271.15 185.35 metallic surface fin. .35 .39 166.68 151.85 other metal product .28 .32 235.57 195.34 prime mover .27 .28 292.49 229.63 farm machine .09 .28 416.09 226.54 metals & mat. process. .25 .28 310.18 232.30 textile machine .21 .27 351.01 246.84 food&chem.eng.machine .19 .31 537.56 210.42 mine & cons. machine .17 .27 480.01 219.29 misc. industrial mach. .30 .29 235.85 238.02 other machine .27 .31 261.79 197.38 railway rOIling stocks .46 .46 56.46 56.46 bicycle .20 .18 300.05 318.59 other trans. equipment .27 .29 170.91 213.49 scientific controlling .42 .36 237.87 311.96 medical apparatus .25 .25 211.06 211.06 other precision machine.22 .22 327.38 327.38 9) Other manufacture ‘ jewellery .38 .30 158.91 173.61 ice making .13 .13 , 484.10 500.96 Note: Value added is based on data of annual gross value added, assets are based on data of Value of assets in operation as of Dec. 31,1976, employee is based on data of persons engaged as of Dec. 31, 1976. Source: The calculation is based on data in 'The Report of 1976 Industrial and Commercial Censuses of Taiwan-Fukien District of the Republic of China,‘ volume 3 book 1. 229 For convenience we assume two factors in production namely capital and labor and apply this measurement to Taiwan’s 4-digit manufacturing small scale industry where the SSEs have over 50% of employment. Following Ho’s technique, I assume a 20% shadow price of capital and no distortion in the labor market. Then the total factor productivity of j th category - A(j) becomes ( value added(j))(.ii§(§§:3a(j*) (value added(j)} (V2?K(j)+W(j) (.2)K(J') W(j) where K(j) is assets in operation of j th category and W(j) is the wage bill of j th category and the Value added(j) is the net value added of j th category of the given industry. Table 7.14 shows A(j) of the small scale industries in 1976; the total factor productivity of the dehydrated food industry in the 1-4 workers category is 1.09; this result in Table 7.15 shows that in a large number of small scale industries, the SSEs have the highest total factor productivity. In 1976, the total . factor productivity was highest in 46 SSEs with 1-99 workers but the total factor productivity was highest in 21 LSEs with over 100 workers. This result for Taiwan contrasts with the Korean situation. In the Korean case, SSEs have the highest total factor productivity in a fewer number of small scale industries in 1984 than LSEs. In 1976 Taiwan SSEs in 230 Table 7.14 Total Factor Productivity of Establishments for Industries in which Small Establishments predominate, by Type of Industry & by Size of Establishments(Worker),Taiwan, 1976: A(j) 4-digit Industry 1-4 5—9 10-19 20-49 50-99 100—299 300—499 500+ 1) Food,beverage,tobacco dehydrated food 119? 1.04 .68 .94 .68 .90 pickled food 1.11 1.36 1.75_ 1.28 1.16 .92 .54 bakery&confect. 1.23 1.16 1.05 .84 .97 1.55 1.09 1.02 edible oil fat .93 .63 .18 .001 1.82 .38 flour mill 1,12 .94 .56 .33 .35 .31 .69 husking of rice 1.11 1.01 1.16 .43 1,24 tea processing .65 1.03 1.10 1121 .09 .09 1.03 other flavor.pro. .92 1.09 .62 .19 .87 .84 1,51 mis.food prepar. 1.11 1.02 .69 1.03 .98 2,88 2) Textile, leather Spinning of silk 1.42 .55 1:52 .9 1.20 1.47 rope,cable& rugs 1.15 1,45 .98 .11 1.29 1.08 1.34 1.30 other textile .83 1.26 1.18 .87 1.15 .99 1.04 .27 3) Wood & wood product saw&plan. mill .84 .96 .97 1.07 .67 1.23 2.22 .52 timber drying .70 2.17 .99. .70 1.29 1.62 wooden contain. 1.23 1.36 1.37 1.53 1.44 1.50 bamboo products 1.38 1.52 1.64 1.59 1.64 1.68 1.17 other wooden 1.34 1.51 1.59 1 48 1.37 1.28 .66 .84 wooden furniture 1.32 1.42 1.35 1.66 1.19 1.41 1.33 1.28 4) Paper& its products paper container 1.19 1.34 1.39 1.47 1.01 .94 .53 1.22 other paper 1.17 1.39 1.22 1.08 .62 1.52 1.55 publishing .85 1.22 1.08 1.05 1.38 .60 printing 1.15 1.33 1.40 1.41 1.29 1.04 .73 .93 book binding 1.16 1.41 1.43 1.36 1.58 engrave&etching 1.11 1.41 1.29 1.53 1.05 5) Products of chemicals other chemical .85 .44 .86 1.03 .86 .59 1_25 paint& lacquer 1.02 1.01 .87 1.07 .82 .81 67 medicine&drugs .59 .88 .87 1.02 .95 1.56 1153 1.17 ag.insecticide .79 .71 .70 .63 .10 ,_96 .75 .70 chemical products .34 .80 .96 1.11 1.13 1 44 1_68 .85 industrial rubber2.00 1.16 1.54 1.27 .62 1.73 1 14 other rubber 1.17 1.19 1.15 1.26 1.23 1.16 4,32 1.51 plastic bag 1.17 1.07 1.13 1483 .74 .86 1 15 plastic articles 1.33 1.33 1.23 1.28 1.35 .83 1.25 .51 other plastics 1.18 1.34 1.36 1.22 1.17 1.23 1.29 1.28 6) Non-metallic products glass & its prod.1.41 1486 1.43 1.38 1.15 .89 .03 .90 cement products 1.16 1.23 1.33 .92 1.40 1.21 .74 1.78 con. clav. 1.05 .97 .98 1.21 .92 .96 _,58 .72 abrasive mater. .85 1.10 1.12 .60 1,53 marble products 14;] 1.53 1.38 1.34 .83 1.11 mis.minerals .88 1.12 .98 1.27 1.01 1.21 .2419 231 Table 7.14(cont'd.) 4-digit Industry 7) Basic metal 1-4 5-9 10-19 20—49 50-99 100—299 300—499 500+ Iron&steel found. 1.16 1,38 .88 .47 .51 .70 .92 1.02 ship scrapping: .03 .02 .08 .07 .25 .26 73 c0pper refine 1,51 .89 .73 .83 .95 1.01 “_— other non—ferr. 1,38 1.26 .72 1.34 1.18 .56 8) Metal products, machine and equipment metal,handtools 1,43 1.35 1.33 1.28 1.35 1.41 1.39 metallic mould 1.48 1.53 1.42 1.36 1.09 1.33 .18 metal.con.mat. 1.41 1,48 1.47 1.19 1.07 1 00 1.36 1.47 aluminum prod. 1.32 1.21 1.28 1.18 .87 1151 .37 copper products 1.35 1.40 1.41 1.42 1.54 81 metallic surface 1.48 1.48 1.48 1.38 1.58 .57 other metal prod. 1.41 1.42 1.46 1.29 _,21 1.11 .95 1.13 prime mover manu. 1.06 1.40 1.07 1,42 2 1.31 1.10 farm machine 1.31 1.76 .99 1 34 1.18 .28 .68 metal&mat.proc. 1.39 1,48 1 31 1.19 1.18 1.11 1.13 .94 textile machine .33 1.28 1,42 1.33 .95 .22 1.25 food&chem.mat. 1.23 1.33 1 34 1.32 1.24 1.36 1.04 .76 mine&cons.machine 1.39 1.09 1.40 1,41 .89 .53 1.36 misc.ind.machine 1.29 1.32 1.25 1.09 1.36 1.46 1 45 other machine 1.29 1.41 1.19 1.36 1.21 1.42 1,83 .91 railway rolling 1,10 .98 bicycle 1.36 .23 2,98 1.27 .98 1.12 1.60 other trans.equip.1.51 1.39 1.14 1,62 1.03 1.58 .83 .15 scie.controlling 1.31 1.31 1.42 1.83 1,29 1.61 1.79 medical appara. 1.16 1.42 1.05 .98 1.59 other preci.mach. 1,36 .83 1.31 .78 1.09 9) Ohther manufacture jewellery 2,08 1.60 1.36 .96 1.22 1.58 ice making .64 .71 .64 .61 .96 1.13 Note: A(total factor p roductivity in given industry) measurement was obtained by (V.A./.2KY“ "‘(V.A/W)"z where V.A. is net value added, K is operating assets value, W is wage bill. We asume shadow price of capital is 20Z and there is no distortion in labor market. And a, is capital factor share which is .2K/(.2K+W) and a, is labor factor share which is W/(.2K + W). A(j) is total factor productivity in j-th size category of given industry. This measurement can be obtained by same process, by using data of j—th category. Source: The calculation is based on data in 'The Report of 1976 Industrial and Commercial Censuses of Taiwan-Fukien District of the Republic of China,’ volume 3 book 1. 232 Table 7.15 Number of Industries where Total Factor Productivity (Capital shadow priced at ZOZ) is greatest in the Size Category indicated Taiwan — 1976 Industry 1—4 5—9 10—19 20—49 50—99 100—299 300-499 500+ Food 1 0 l l 2 3 1 0 Textile 0 2 1 O 0 O O 0 Wood 0 l 1 2 0 1 l 0 Paper 0 0 0 3 1 1 1 0 Chemical 1 0 1 2 1 1 4 0 Mineral 1 1 O 0 1 O 3 0 Basic metal 2 1 0 O O O 1 0 Metal prod. 2 5 2 3 5 3 1 0 Other manu. 1 0 0 0 0 1 0 0 Total 8 10 6 11 10 10 12 0 Note: The surveyed industries are small scale industries where small scale enterprises have over 50% of employment. Source: 'The Report of 1976-Industrial and Commercial Censuses of Taiwan—Fukien District of the Republic of China,‘ volume 3, book 1, published by the committe on industrial and commercial censuses of Taiwan—Fukien district of the Republic of China, Executive Yuan. 233 small scale industries were more efficient than the LSEs in small scale industries. From this data we may conjecture that the Taiwanese SSEs with 1-99 workers tend to generate a large amount of employment efficiently because SSEs were more efficient than LSEs in many small scale industries. The rationale in comparing Korean data in 1984 with 1976 Taiwan data is that Taiwan was about six years ahead of Korea in terms of per capita income; Korean per capita income in 1981(US$ 1,697) was about the same as Taiwan’s per capita income in 1975(Scitovsky 1986). 2. Social Benefit - Cost Analysis To complement Ho’s measurement for a total factor productivity, I perform a Benefit - Cost ratio analysis. In the Benefit and Cost ratio measurement, I assume two factors , capital and labor. I apply this measurement to 4—digit manufacturing small scale industries where SSEs have over 50 % of employment. Following Ho, I assume a 20% shadow price of capital and no distortion in the labor market. Technically Benefit and Cost ratio of the j th size category in a given industry becomes Value added(j)/(.2)K(j)+W(j) where K(j) is assets in operation of j th size category and W(j) is the wage bill of j th size category and value added(j) is the net value added of j th size category. 234 Table 7.16 shows Benefit and Cost ratio of 4—digit small scale industries based on size category; the Benefit and Cost ratio of dehydrated food industry in 1-4 workers sized category is .56. As a result, Table 7.17 shows that the majority of industries have the. highest TFP in SSEs of the small scale industries. These results confirmed those by Ho’s measurement. In 1976, the Social Benefit and Cost ratio was highest in 43 SSEs with 1-99 workers, but the SBC ratio was highest in 24 LSEs with over 100 workers. From these two sets of efficiency measurements, we may conclude that SSEs in the small scale industry in Taiwan are more efficient compared to LSEs in small scale industries. On the other hand, Ho(1980, see Table D12 p141) showed in the relative efficiency in terms of TFP that in 1971, SSEs in 21 out of 40 small scale industries were more efficient than LSEs. In 1971, Taiwan’s SSEs and LSEs were fairly efficient in the small scale industries. Therefore, over time Taiwan’s SSEs improved their efficiency in small scale industries, and the number of small scale industries also increased . I can conjecture that the more neutral market condition in Taiwan contributed to the continuous improvement of the SSE’s efficiency in the small scale industry, by providing an equitable investment climate for both LSEs and SSEs. The efficient SSE in Taiwan’s small scale industry shows 235 Table 7.16 Benefit-Cost ratio of Establishments for Industries in which Small Establishments predominate with over 50Z of Employment by Type of Industry and by Size of Establishment(worker) Taiwan - 1976 4—digit 1-4 5-9 10—19 20—49 50-99 100-299 300—499 500+ 1) Food,beverage, tobacco dehydrat.food ,66 .54 .42 .48 .34 .46 pickled food .61 .69 .89 .66 .59 .50 .27 bakery&cofect..63 .59 .53 .44 .50 .80 .61 .57 edible oil fat.51 .36 .12 —0.001 .93 .29 flour mills is .50 .35 .23 7276 .24 .43 husk of rice .59 .54 .62 .25 ,6; tea process .38 .53 .57 ,63 .07 .08 .51 other flavor. .48 .56 .33 .11 .45 .44 :77 mis.food pre. .57 .53 .37 .54 .55 1,38 2) Textile, leather spin of silk. .82 .32 .77 .46 .62 .75 rope,cable,rug.59 .73 .50 .06 .66 .54 .71 .66 other textile .44 ,63 .60 .44 .58 .50 .53 .17 3) Wood products saw&plan.mill .45 .51 .51 .55 .35 .62 1.11 .27 timber drying .38 1,09 .50 .37 .65 .82 wooden contain.63 .69 .69 .79 .80, .77‘ bamboo product.69 .77 .84 .83 ' .87 1.08 .88 other wooden .68 .76 .82 .76 :70 .65 .38 .44 wooden furni. .67 .71 .68 .84 .61 .74 .69 .65 4) Paper & paper products paper contain..61 .67 .70 75 .51 .49 .29 .62 other paper .59 .70 .62 5 publishing .47 .63 .55 .5 printing .59 .67 .707 .7 book binding .59 .72 .74 6 engrave&etch. .57 .71 .66 _2, 5) Products of chemicals other chem. .46 .30 .45 .53 . paint&lacquer .662 .53 .45 .552 .45 .47 .34 medicine&drug T34 .46 .46 .52 .49 .80 .84 .63 ag.insecticide.43 .37 .39 .35 —.06 .54 .48 .82 chemical prod..21 .42 .50 .58 .59 .82 ,90 .48 indust.rubber1.00, .59 .78 .64 .32 .92 .58 other rubber .59 .60 .58 .64 .63 .60 2.32 ..81 plastic bag .60 .55 .57 .67 .38 .43 .58 plastic arti. .67 .67 .62 .65 .69 .43 .67 .26 other plastic .60 .67 .69 .61 .59 .61 .65 .65 6) Non—metallic products glass&its pro..71 ,86 .73 .71 .58 .48 .02 .50 cement produ. .59 .63 .67 .47 .72 .62 .88 .90 cons.clav. .54 .51 .50 .62 .47 .49 .73 .39 abrasive mat. .44 .55 .57 .33 ‘77 marble prod. ,79 .78 .70 .68 .43 .56 misc.minerals .46 .57 .51 .64 .51 .61 9 .66 .52 .39 .51 H H 236 Table 7. 16(cont' d. ) 4—digit Industry 1-4 5— 9 10— 19 20— 49 50— 99 100- 299 300— 499 500+ 7) Basic metal iron&steel foun. .59 .69 .44 .30 .28 .39 .46 .53 ship scrapping .03 .02 .05 .05 .14 —0.18 L24 copper refining ,18_ .46 .37 .44 .53 .52 other non-ferr. ,69 .64 .39 .68 .59 .29 8) Metal products, machi.ne ,&equipment metal. handtools .71 .68 .67 .65 .68 .71 ,72 metallic mould .75 ,89 .74 .70 .54 .67 .11 metal.cons.mat. .71 .75 .74 .61 .53 .51 .72 80 aluminum prod. .66 .62 .64 .60 .45 ,70 .25 copper products .68 .71 .71 .72 180 .45 metallic surface .75 .75 .76 .71 .87 .30 other metal prod..71 .72 .74 .65 ,76 .56 .50 .58 prime mover manu..54 .706 .54 .708 .49 .67 .56 farm machine .66 .64 .50 .68, .60 .16 -0.40 metal&mat.process.70 ,72 .66 .60 .59 .56 .57 .50 textile machine .22 .65 129 .67 .49 .19 .66 food&chem.machine.62 .67 .68 .67 .63 .69 .53 .43 mine&cons.machine.69 .55 ,712 .710 .45 .28 .69 misc.ind.machine .65 .66 .63 .55 .73 ,81 .73 other machine .65 .71 .65 .69 .61 .72 ,83 .46 railway rolling ,92 .52 _ bicycle .69 .16 1,02 .64 .49 .56 .80 other trans.equi..76 .71 .57 ,84 .52 .81 .43 —.08 scientific cont. .66 .66 .72 .95 .96 .97 1,17 medical apparatus.59 .72 .54 .50 ,82_ other precis.mach.79 .45 .68 .42 .59 9) Other manufacture jewellery 1.05 .82 .69 .49 .63 .90 icemaking .36 .38 .35 .33 .49 i_5__8_ Note: Small establishments(those with 5— 99workers) are considered to predomi.nate in an industry i.f they account for half or more of the industry‘ 5 employment. The benefit— cost ratio is a way of measuring total factor productivity and is the ratio of the net value added to the input cost in given size group of i.ndustry. The input cost is assumed to be r *K + W, where r i_s shadow price of capi.tal(here 20Z), K is operating assets and W is wage bill. Here we assume no distortion in labor market. The operating current asset value which I use for K is far larger than capital services which the capital contributes to production. This tends to exaggerate the capital cost. As a result, the SBC ratio tends to be underestimated. Source: The calculation is based on data in 'The Report of 1976 Industrial and Commercial Censuses of Taiwan—Fukien District of the Republic of China,’ volume 3,book 1. 237 Table 7.17 Number of Industries where Benefit —Cost Ratio(Capital Shadow priced at 20%) is greatest in the Size Category indicated Taiwan - 1976 Industry 1—4 5-9 10—19 20—49 50-99 100—299 300—499 500+ Food 2 0 1 l 2 2 1 0 Textile 1 2 0 0 0 0 O 0 Wood 0 l 1 l 1 l 1 0 Paper 0 O 0 3 1 l 1 0 Chemicals 2 0 l 1 l 0 4 1 Mineral 1 1 0 0 1 0 2 1 Basic metal 2 1 O 0 0 O 1 0 metal prod. 1 3 3 3 4 3 2 2 other manu. 1 0 O 0 0 l '0 0 Total 10 8 6 9 10 8 12 4 NotezThe surveyed industries are small scale industries where small scale enterprises have over 50% of employment. Source: 'The report of 1976 Industrial and Commercial Censuses of Taiwan—Fukien District of the Republic of China,’ volume 3, book 1, published by the committe on industrial and commercial censuses of Taiwan-Fukien district of the Republic of China, Executive Yuan. 238 bright future for more competitive market , a strong growth in employment and improvements in the income distribution. It is possible to hypothesize that the Korean large scale oriented industrial structure can provide an advantage in the development of technology—intensive industries, compared to Taiwan; highly technology- intensive industries involving long lead time, high fixed cost and uncertain outcomes require massive capital investment. But Levy(1987a) found that the small size at entry of the firms did not result in poorer performance at least by Taiwanese computer-related industries as compared to their Korean counterparts. Also the role Of public enterprises in Taiwan and Korea is not an important part of my analysis, because public enterprises are relatively insignificant in Korea. But this would certainly qualify any judgement we can make in comparing Korea and Taiwan because it introduces a variable not covered in my model or in the rest of analysis. In respect of the quality of the work force I believe Taiwan and Korea are similar. But public sector management of LSEs in Taiwan certainly introduces the possibility of recurring deficits. H. CONCLUSION From the above study we can derive the following 239 findings; First, Taiwanese SSEs are significant in the overall industrial structure. In 1981, the SSEs with 1-99 workers employed 44 % of total manufacturing employment. Second, Taiwanese SSEs are also decentralized in location. In 1981, 47% of SSEs were rural-based. Third, Taiwanese SSEs are efficient in small scale industries. Ho’s and the SBC ratio methods confirmed that Taiwan’s SSEs in small scale industry became more efficient than LSEs in small scale industry by 1976. This result contrasts with the Korean case in 1984. From the previous chapter, we know that Korean SSEs are not a significant force in the overall industrial structure; they are urban based; and inefficient in small scale industry, compared to LSEs. Considering these comparative characteristics and the better record of growth and income distribution in Taiwan, policies to‘ improve their situation are suggested for Korean SSEs and those of other LDCs. 240 CONCLUSION A. GENERAL SUMMARY I started with raising the questions of why and how the large scale industrial structure in Korea was shaped by its economic policy. The findings on the issue are that considering the competitive edge based on economies of scale in industry, for the world export market,the government chose a policy favoring the LSEs. This large scale oriented policy in terms of low interest rates and overvaluation of the Won resulted in reducing the capital cost of LSEs relative to SSEs. This discriminatory policy created a distortion in the capital market. As Korean SSEs were unfairly treated in allocation of capital, their status declined in terms of employment and value added. Then I tried to compare the current state of Korean and Taiwanese SSEs.‘ Korean SSEs are largely urban-based and inefficient compared to LSEs in small scale industries. On the other hand, Taiwan’s SSEs have been strong in the overall industrial structure. Taiwanese SSEs are decentralized in location and efficient compared to LSEs in small scale industries. These contrasting results are due to the different policy priorities of the 241 governments of the two countries; the Korean policy favored the LSE against the SSE, and manufacturing industry against the agricultural sector, while Taiwan's policy emphasized the progressive balance between the LSE and the SSE, & between manufacturing industry and the agricultural sector. From these comparisons we can envisage what would be a better policy for SSEs in Korea and other LDCs; no discrimination between LSEs and SSEs,the decentralization of SSEs, and an improvement in the infrastructure and information services. B. SUMMARY OF EACH CHAPTER In Chapter Two, I began with reviewing the literature by Lewis & Harris-Todaro. Lewis & H-T models were the basic dual rural-urban migration models. Although these models have contributed to the explanation of the interaction between urban and rural sectors, they were weak in their explanation of the small scale sectors such as the informal activities in urban area or non-farming activities in rural area. I continued to review the informal sector as small scale activities in the urban area. With over—urbanization in LDCs,the informal sector contributes to employment creation. To prevent further migration, there should be a potential for employment generation through rural non- 242 farming activities. Then I reviewed the controversial future of rural non-farming activities in LDCs through the economic development literature. First Hymer—Resnick’s model adopted a pessimistic view about the future of non- farming activities; but Liedholm-Chuta’s empirical study proved that is not the case, at least in Sierra Leone for several reasons. Finally I reviewed the SSEs. The definition of SSEs is ambiguous in terms of size such as a cut-off point of 50-100 workers. The relationship between SSEs and LSEs and between SSEs and the agricultural sector has become important in LDCs. The wage gap between the LSEs and SSEs continues to exist because of labor market imperfections and differences in the quality of the work force. SSEs exist in a fragmented capital market in LDCs. ‘As a partial efficiency measure, the capital/labor ratio is often used. In many cases, SSEs in LDCs proved labor intensive. The Benefit-Cost or production frontier methods are used as comprehensive efficiency measurements for SSEs in LDCs. In many cases the SSEs in LDCs turned out to be more efficient. In Chapter Three, I developed the three sector (ULS,USS and Rural) labor market and PPFs models, which are based on three goods and two factors. The analysis of the three sector model shows that a policy favoring the 243 LSEs such as a tariff and capital subsidy on the ULS sector leads to an expansion of the ULS sector but contraction of the USS and rural sectors in terms of employment and output. A policy favoring the SSEs such as a tariff and capital subsidy on the USS sector leads to expansion of the USS sector but reduction of the ULS and Rural sector in terms of employment and output. Also, a minimum wage policy in the ULS sector leads to contraction of the ULS and expansion of the USS and Rural sectors in terms of employment and output. Finally in the case of homogeneous products of the ULS and USS sectors, Ithe large scale boosting policy such as tariff on the ULS(as well as USS) sector product leads to expansion of the ULS & USS sectors and reduction of the rural sector in terms of employment and output. In Chapter Four, I reviewed the theoretical three sector model based on Korean time series data (1966-84) about industrial structure change. According to the comparative static analysis with a capital inflow in ULS sector, the employment of the ULS sector was supposed to improve. Empirically it was so. Employment in rural areas was supposed to decline. Empirically it was ' so. Employment in the USS sector was supposed to decline. But empirically it was not so, because we ignored the investment actiVities by the USS sector in our model. The real wage in the three sectors was supposed to increase in the model. Empirically it was so, but contrary to the prediction, the difference in real wage between the three sectors persisted because of productivity differences and a different pace of expansion in the three sectors. The most distinctive characteristic of the Korean industrial structure change was that the ULS sector expanded faster than the USS sector during the years 1966— 84. As a result the position of SSEs in the overall manufacturing structure worsened. This result proved that the capital allocation clearly favored the LSE against the SSE. In Chapter Five, I reviewed the specific industrial, regional, and dynamic prospects; efficient size,and capital intensity of Korean SSEs. First, with economic development the role of very small scale enterprises(1-4workers) in Korea is declining. Second, as industrialization proceeds, the SSE is emerging as a big player in the capital-intensive and heavy industries such as chemical and metal machinery industries. Third, 'SSEs are heavily located in the urban area. Fourth, during the last few decades, the SSES have been losing in the share of manufacturing employment. Fifth, the survivor criterion shows that during 1979-84, both SSEs and LSEs were fairly efficient. Sixth, with infrastructural improvements, the locational influence in SSE’s competitiveness became less important. Seventh,depending on the specific industries, the SSEs in manufacturing industries tend to have various ranges of capital intensity. Within the overall picture of the manufacturing industrty, the SSEs still tend to be labor intensive. In Chapter Six, I reviewed the problems of Korean SSEs and tried to suggest possible policy measures to alleviate them. Firstly, past industrial policies in terms of overvaluation,interest etc. have consistently favored LSEs. This policies reduced the capital cost for LSEs relative to SSEs. Second, the location of SSEs tend to be urban-based.- Third, Ho’s and SBC methods confirmed that Korean SSEs in small scale industry have become less efficient than LSEs in small scale industry. The main suggestions are aimed at creating a fair environment based on the market mechanisms where LSEs and SSEs can compete on fair terms and complement each other in order to survive and co-exist. In Chapter Seven, I reviewed Taiwan’s SSEs. First, Taiwan’s SSEs are still significant in the country’s overall industrial .structure, with help from less 246 discriminatory policies. Second, Taiwan’s SSEs are also decentralized in location. Ho’s and the SBC ratio methods confirmed that Taiwan’s SSEs in small scale industry became more efficient than LSEs in small scale industry. Considering these comparative characteristics and better records of growth and income distribution in Taiwan, Korean SSEs as well as other LDCs' SSEs should diagnose their weakness and determine their future direction for themselves. C. THE MAJOR LESSONS FROM THE STUDY The thesis found that the Korean policy of promoting LSEs indeed discriminated against SSEs, favoring LSEs in terms of low interest rates ,overvaluation of exchange rates,taxes, etc. during the last two decades. This agressive government intervention resulted in creating a distortion in the capital market. But because of the effects of economies of scale and efficient managements, the capital market distortions favoring LSEs partly contributed to improving the social efficiency of the LSEs in the small scale industry. This result is not inconsistent with Neo—classical ideas, because the Nee-classical economists ruled out the effects of economies of scale and selective managements in the market system. The favorable condition for capital 247 funding in LSEs contributed to accelerating the improvement in efficiency in LSEs at the expense of employment creation by SSEs. On the other hand, the Taiwan’s government allowed a freer market mechanism with less government intervention. As a result, the less distorted market contributed to the more efficient resource allocation,compared to the Korean case. This was reflected in the consistent improvement of the TFP of Taiwan' SSEs in small scale industry. The more neutral condition for capital funding in both LSEs and SSEs contributed to' continuous improvements in SSE’s efficiency in the small scale industry. The Taiwan’s case proved that Nee-classical ideas are still valid. This comparative study confirmed the importance of a market mechanism in achieving the optimal resource allocation and improving the efficiency of SSEs. Even if the intentions are good, the government interventions tend to distort and worsen the imperfect market in LDCs. Therefore government should try to provide a more fair environment for both LSEs and SSEs in terms of trade regime and credit allocatiion etc. At the same time, the study suggests that-the role of government should be directed toward eliminating the market imperfection, particularly through providing market information in a selective manner. Finally this study confirmed that the neoclassical 248 axiom such as "Get the price right" and " Let the market do it" can still play an important role in economic development in Korea as well as other LDCs. At the same time, it is important to keep in mind that Korean intervention in capital markets had two offsetting characteristics: a) capital was directed towards profitable markets; and b) subsidies were not of an indefinite duration. These aspects of the discretionary lending system in Korea were dealt with in various studies(Kwack Sung Yeung 1986, Scitovsky 1985). This sets the Korean economic regime on a different footing from Taiwan, or import substitution regimes such as that of India, and closer to Japan. The details of these aspects of Korean policy and implementation are obviously important and have been noted in the discussion, particularly chapters 6 and 7. The general implications of such a regime must be taken account of in any consideration of policy reform. APPENDIX A KOREAN CAPITAL MARKET DISTORTION 249- Appendix A : Korean Capital Market Distortion Table 6.2 shows how the distortion in the capital market affects on the capital cost of the LSEs and SSEs. The distortions in the capital market are largely due to trade regime, interest rate and taxes. The quantative calculation about these distortions are followed; 1) Trade regime: The tariff rate, overvaluation of ’won’ or licensing system in trade regime can make distortions in the capital market. In Korean case, the tariff rates or licensing system seem not to be serious factors of distortions. It is important to see how the overvaluation of ’won' affects the LSEs and SSEs in terms of the capital cost. Usually, the government tends to allow the LSEs to import the capital goods more easily at the low official rate than the SSEs. Therefore the degree of overvaluation can show that the capital cost of the LSEs is cheaper than that of the SSEs. The Official Exchange Rate(A) and the Black Market Exchange Rate(B) (exchange unit; won/$) .. —._o_ ——_.—_.M——_h—4 -...-...._... _ -_-.. 250 ’76 '77 ’78 '79 ’80 ’81 A: 484 484 484 484 659.90 700.504 B: 494.25 ‘ 510 521.41 556.83 671 716.5 (A-B)/A*100(%): -2.1 -5.3 -7.7 ~15 -1.6 -2.2 ’82 ’83 ’84 A: 748.8 795.5 827.4 B: 805.25 841.75 828.38 (A-B)/A*100(%): -7.5 -5.8 -o.1 Source:’ World Currency Yearbook 1985’ edited by Philip.P Cowitt, 1986. International Currency Analysis INC. Note: The black market exchange rate is calculated as the average monthly rate. The data ShOWS'that the Korean currency ’won’ was overvalued over time. Indirectly, it implies that the capital cost of the LSEs is cheaper than that of the SSEs in terms of overvaluation. 2) Interest rate: The formal and informal interest can make a distortion in capital market. Usually, the SSEs rely on funds from a private capital market with an informal interest, while the LSEs rely on funds from regular banking institution with a formal interest. So we need to see the formal interest and informal interest in order to calculate the capital cost of the LSEs and the SSES. 251 Interest Rate on Official Market(A) and Black Market(B) ’75 '76 ’77 ’78 ’79 ’80 "81 ’82 ’83 ’84 A: 15.2 17.8 16.5 18.5 18.5 22.5 18 13 13 10.6 B: 49.5 48.2 45 46.5 51.5 55 41 34.7 28.5 27 A-B:-34.3 -30.4 -28.5 -2 -33 -32.5 -23 -21.7 -15.5 -16.4 ’85 ’86 A: 10.7 10.7 B: 26.2 25 A-B: -15.5 -14.3 Source: For official interest,Economic Statistics Yearbook,various issues the Bank of Korea For black interest,’A study of the underground economy in Korea: Theoretical Approaches, Empirical Evidence and Policy Implication’ by Chae Gwang,April 1987. NotezThe official interest is based on the discounts on bills. The data shows that the formal interest is lower than the informal interest. Indirectly, it implies that the capital cost of the LSEs is cheaper than that of the SSEs in terms of interest rate. 3) Tax: The tax rate can make a distortion in the capital market. According to the corporate tax law, the corporate tax rate on SSEs is 20% on below 50 million won 252. profits, but the tax rate on the LSEs is 30% on above 50 million won profits and 10 million won, which is is 50 million* 20%. So the approximate tax rate differences favoring SSEs became that 30% -20% = +10%. This shows that tax rate on LSEs is higher than tax rate on SSEs. Indirectly, it implies that the capital cost of the SSEs is cheaper than that of the LSEs in terms of tax. Finally we can combine the above three differences from distortions in order to calculate the capital costs of the LSEs relative to the capital cost of the SSEs. Total percentage difference in LSEs’ cost relative to SSEs ’76 ’77 ’78 ’79 ’80 ’81 ’82 l)(A-B)/A*100(%): -2.1 -5.3 -7.7 -15 -1.6 -2.2 —7.5 2)A-B: -30.4 -28.5 -28 -33 -32.5 -23 ~21.7 3)A-B: +10 +10 +10 +10 +10 +10 +10 Total: -22.5 -23.8 -25.7 -38 *21.1 -15.2 -l9.2 ’83 ’84 1)(A-B)/A*100(%): —5.8 -o.1 2)A-B: -15.5 -16.4 3)A-B: +10 +10 Total: -ll.3 -6.5 APPENDIX B TAIWAN’S CAPITAL MARKET DISTORTION 253 APPENDIX B : Taiwan’s Capital Market Distortion Table 7.1—3 in Ch.7 shows how the distortion in the capital market affects on the capital cost of the LSEs and the SSEs in Taiwan and Korea. The distortions in the capital market are largely due to trade regime, interest rate and taxes. The quantitative calculation for Taiwan’s capital market distortions are followed: 1) Trade regime: The tariff rate, overvaluation or licensing system in trade regime can make distortions in the capital market. In Taiwan case, the tariff rate or licensing system seem not to be serious factors of distortions. It is important to see how the overvaluation of 'NT $’ affects the LSEs and the SSEs in terms of the capital cost. Usually, the government tends to allow the LSEs to import the capital goods more easily at a low official rate than the SSEs. The degree of overvaluation can show that the capital cost of the LSEs is cheaper than that of the SSEs. The Official Exchange Rate (A) and the Black Market Exchange Rate (B) (exchange unit; NT $/ $) 254 ’76 '77 ’78 '79 ’80 ’81 A: 38 38 36 36.03 36.01 37.84 B: 39.77 38.96 38.15 39.82 37.86 39.49 (A-B)/A*100(%): -4.6 -2.5 -5.9 -10.5 -5.1 -4.3 '82 ’83 '84 A: 39.91 40.27 39.47 B: 41.33 41.9 41.82 (A-B)/A*100(%): -3.5 -4.0 -5.9 Source: 'World Currency Yearbook 1985’ edited by Philip.P Cowitt, 1986. International Currency Analysis INC. Note: The black market exchange rate is calculated as average monthly rate. The data shows that Taiwan’s currency ’NT S’ was overvalued over time. Indirectly, it implies that the capital cost for the LSEs was cheaper than that for the SSEs in terms of overvaluation. 2) Interest rate: The formal and informal interest can make a distortion in capital market. Usually, the SSEs rely on funds from a private capital market with an informal interest, while the LSEs rely on funds from a regular banking institution with a formal interest. So we need to see the formal interest and informal interest in order to calculate the capital~cost of the LSEs and the SSES. 255 Interest Rate on Official Market(A) and Black Market(B) ’76 ’77 ’78 ’79 ’80 ’81 ’82 A: 10.5 9.25 9.25 12 13.5 13 9 B: 26.77 26.30 25.75 28.99 30.81 31.18 28.64 A-B:-l6.27 -17.05 -l6.5 ~16.99 -17.31 -18.18 -19.64 ’83 ’84 A: 8.5 8 B: 26.76 25.92 A-B:-18.26 -17.92 Source: Domestic economic index report, 1986 Taiwan. The data shows that the formal interest is lower than the informal interest. Indirectly it implies that the capital cost of the LSEs is cheaper than that of the SSES. 3) Tax: The tax rate can make a distortion in the capital market. Firms’income tax rates are ranging from 22 to 25 % (The Statue for Encouragement of Investment 1960). But SSEs with below 50,000 NT$ exempt from income tax. The tax rate on profit-seeking income rises from 15 % on NT$ 50,001-100,000 to 25 % on NT$ 100,001-500,000 and 35% on the excess over NT$ 500,000 , which make average 25 % as overall tax rate.( Source: CJ P1att., 1982,'Tax System of Africa, Asia and the Middle East’) So the approximate tax rate differences favoring SSEs became 22%. This shows that tax rate on the LSEs is 256 higher than that on the SSEs. Indirectly, it implies that the capital cost of the SSEs is cheaper than that of the LSEs in terms of tax. Finally we can combine the above three differences from distortions in order to calculate the capital costs of the LSEs relative to the capital cost of the SSEs. Total percent difference in LSEs’ cost relative to SSEs ’76 '77 ’78 '79 ’80 1)(A-B)/A*100(%): -4.6 -2.5 -5.9 -1o.5 . —5.1 2) A—B: -16.27 -17.05 -16.5 -16.99 —17.31 3) A-B: +22 +22 +22 +22 +22 l+2+3)Total: +1.13 +2.45 —o.4 -5.49 —o.41 ’81 ’82 ’83 ’84 1)(A-B)/A*100(%): -4.3 —3.5 -4.0 -5.9 1) A-B: -18.18 -19.64 -l8.26 -17.92 2) A-B: +22 +22 +22 +22 1+2+3)Total: -0348 -1.14 -0.26 -1.82 BIBLIOGRAPHY 257‘ BIBLIOGRAPHY Amsden, Alice ,1985,'The State and Taiwan's Economic Development' in " Bring the State Back in " Cambridge University Basu,K.,1984,"The Less Developed Economy-A Critique of Contemporary Theory" New York, Basil Blackwell. Bhagwati,J and Srinivasan, T.N.,1974," On Reanalysing the Harris—Todaro Model: Policy Rankings in the Case of Sector Specific Sticky Wages" American Economic Review 64, 502-8. Biggs, Tyler , 1986," On Measuring Relative Efficiency in a Size Distribution of Firms" E.E.P.A Discussion Paper No.2. Blomqvist,A.G.,1978,"Urban Job Creation and Unemployment in LDCs" Journal of Development Econopmics 5,3-18. Breman,J.,1976,"A Dualistic Labor System ? A Critique of the 'Informal Sector' Concept :2:A Fragmented Labor Market" Economic and Political Weekly 11, 49 (December 4), 1905-08. Bromley,R., 1978," Organization, Regulation and Exploitation in the so-called ' Urban Informal Sector': The Street Traders of Cali, Columbia" World Development 6, 1161-72. Byerlee,D. and Eicher, C.K., 1972,"Rural Employment, 258 Migration and Economic Development: Theoretical Issues and Empirical Evidence from Africa" African Rural Employment Paper No.1, Michigan State University. Collier,P.,1975," Labor Mobility and Labor Utilization in Developing Countries" Oxford Bulletin of Economics and Statistics 37, 169-89. Dinopoulos,E.,1983(May),"Import Competition, International Factor Mobility and Lobbying Response: The Schumpeterian Industry Case" JIE. 'Economic Survey of Japan' published by Economic Planning Agency Japanese Government, 1959-65 years. 'Economic Survey of Korea' published by Economic Planning Board Government Republic of Korea, 1968-77 years. Economic Statistics Yearbook, various issues, Bank of Korea, Republic of Korea. Ethier, W., 1983,"Modern International Economics“ New York, W.W. Norton & Company. Fields,G.S., 1975," Rural-Urban Migration, Urban Unemployment and Underdevelopment, and Job—Search Activity in LDCs" Journal of Development Economics 2, 165-87. Gerry,C., 1978," Petty Production and Capitalist Production in Dakar: The Crisis of the Self— Employed" World Development 6, 1147-60. Gibb, Arthur.jr.,1974,"Agricultural Modernization ,Non- 259 farm Employment and Low-level Urbanization" Ph.D Dissertation in Economics, University of Michigan. Godfrey,E.M.,1969,"Labor-Surplus Models and Labor—Deficits Economics: The West African Case" Economic Development and Cultural Change 17, 382—91. Haggblade,Steven.,1984,"The Shebeen Queen or Sorghum Beer in Botswana" MSU Ph.D Dissertation in Economics. Hagglblade, Steve, Liedholm, Carl and Mead, Donald.,l986,"The Effects of Policy and Policy Reforms on Non-agricultural Enterprises & Employment in Developing Countries: A Review of Past Experience" MSU International Development Papers No.27. Hammermesh, D and Rees, A., 1984, "The Economics of Work and Pay" Harper and Row publishers,New York. Harris,J.R. and Todaro, M.P.,1970,"Migration, Unemployment and Development: A Two-sector Analysis" American Economic Review 60,126-42. Hart,J.K.,1973,"Informa1 Income Opportunities and Structure of Urban Employment in Ghana" The Journal of Modern African Studies 11. Ho,Samuel ,1980, "Small Scale Enterprises in Korea & Taiwan" World Bank Staff Working Paper No.384, Washington. Ho,Samuel ,1979," Decentralized Industrialization and Rural Development; Evidence from Taiwan" Economic 260 Development and Cultural Change, The University of Chicago press. Ho,Samuel ,1978,"Economic Development of Taiwan 1860—1970" New Haven, Yale University press. Hsiung,James(ed.),1981,"Contemporary Republic of China : The Taiwan Experience 1950-1980" Praeger Publishers. Hymer,S. and Resnick,S.,1969,"A Model of an Agrarian Economy with Non-agricultural Activities" American Economic Review 59, 493-506. ‘ Kannappan,S.,1983,"Employment Problems and the Urban Labor Market in Developing Nations" University of Michigan. Kilby,Peter and Liedholm,Carl ,1986,"The Role of Non-farm Activities in the Rural Economy," Employment and Enterprise Development Policy Analysis Discussion Paper # 7, Cambridge, Harvard Institute for International Development Koo,Anthony,Y.c.,1968,"The Role of Land Reform in Economic Development; A Case Study of Taiwan" Frederick A. Praeger, Publishers. Korean Statistical Handbook, 1984, published by Economic Planning Board, the Republic of Korea. Kuzunets,Pau1 ,1977,"Economic Growth and Structure in the Republic of Korea" Yale University Press. Kwack Sung Yeung , 1986,'The Economic Development of the Republic of Korea, 1965-1981' in "Models of 261 Development: Comparative Study of Economic Growth inr South Korea and Taiwan" Institute for Contemporary Studies. Lardy, Nicholas , 1983, "Agriculture in China's Modern Economic Development" Cambridge University Press. Lau,Lawrence(ed.) ,1986,"Models of Development: A Comparative Study of Economic Growth in South Korea and Taiwan" Institute for Contemporary Studies. Levy,Brian ,1986,"Prospects and Perils for Small & Medium Enterprises in Outward-oriented Industrial Expansion; Lessons from South Korea and Taiwan"(mimeo.). Levy,Brian & Kuo, Wen-jeng ,1987a,"Investment Requirements & The Participation of Korean & Taiwanese Firms in Technology-intensive Industries" E.E.P.A. Discussion Paper , No.11. Levy, Brian and Kuo, Wen-jeng , 1987b, "The Strategic Orientations of Firms and The Performance of Korea & Taiwan in Frontier Industries: Lessons from Comparative Case Studies of Keyboard & Personal Computer Assembly" E.E.P.A. Discussion Paper,No.12. Levy,Brian , 1987c, "Export Intermediation & The Structure of Industry in Korea & Taiwan" E.E.P.A. Discussion Paper, No. 13. Lewis,W.A.,1954,"Economic Development with Unlimited Surplus of Labor" The Manchester School of Economic 262 and Social Studies. Liedholm,C.,1973,"Research on Employment in the Rural Nonfarm Sector in Africa" African Rural Employment Paper No.5, Michigan State University. Liedholm,C. and Chuta,E.,1976,"The Economics of Rural and urban Small-scale Industries in Sierra Leone" African Rural Economy paper No.14, michigan State University. Liedholm, Carl , 1986,"The Effects of Policy and Policy Reform on Non-agricultural Enterprises and Emoployment in Developing Countries" MSU IDP Working paper, No.27. Liedholm, C and Mead, Donald., 1986,"Small Scale Industries in Developing Countries: Empirical Evidence and Policy Implication" Lindauer, David ,1984,"Labor Market Behavior in the Republic of Korea" World Bank Staff Working Paper No.363. Little, I., 1987, "Small Manufacturing Enterprises in Developing Countries" The World Bank Economic Review, Volume 1, Number 2. Lu, Min-jen , 1981," Conference on the History of the Republic of China,Taiwan's Economic Development Retrospect and Prospect" Taipei, Taiwan. Mazumdar, D., 1976, "The Urban Informal Sector"' World Development 4,655—79. 263 Meir,G.M.,(4th ed.),1984," Leading Issues in Economic Development" Oxford University Press. Moser,C.,1978,"Informal Sector or Petty Commodity Production: Dualism or Dependence in Urban Development?" World Development 6,1041-64. Myrdal, Gunnar , 1968, "Asian Drama ; An Inquiry into the Poverty of Nations" Volume 3. The Twentieth Century Fund. Nihan,G.,Demol,E., and Jonndoh,C.,1979,"The Modern Informal Sector in Lome" International Labor Review, 118,5(September-October),630-44. Oshima, Harry , 1986, "The Transition from an Agricultural to an Industrial Economy in East Asia," Economic Development and Cultural Change, July. Page,John ,1979,"Small Enterprise in African Development: A Survey" World Bank Staff Working Paper No.363. Patrick,Hugh and Rosovsky,Henry, ed.1976,"Asia's New Giant: How the Japanese Economy Works" The Brookings Institution, Washington,D.C. Ranis,G and Fei,J., 1961," -A Theory of Economic Development" AER, vol.51, pp.533-565. Rees, A and Schultz.,1970,FWorkers and Wages in an Urban Labor Market" The University N of Chicago Press Chicago. Sabot,R.H.,1977,"The Meaning and Measurement of Urban Surplus Labor" Oxfbrd Economic Papers 264 29(November)389-411. Scitovsky,Tibor,1985,"Economic Development in Taiwan and South Korea: 1965-1981" Food Research Institute Studies, No.3. Scitovsky,Tibor,l964,'Two Concepts of External Economies' in "Papers on Welfare & Growth" Stanford University Press, Stanford,California. Seturaman,S.,1981,"The Urban Informal Sector in Developing Countries: Employment, Poverty and Environment" I.L.O.Geneva. Sheikh,M.,1974,"Smuggling, Production and Welfare" Ch.16 in "International Trade: Selected Readings" by Jagdish,Bhagwati,eds.MIT Press. Shepherd, William , 1979, "The Economics of Industrial Organization" Prentice-Hall,INc,N.J. Shetty,M.C.,1963,"Small Scale and Household Industries in a Developing Economy" Asia Publishing House,New York. 'Statistical Yearbook of the Republic of China, 1986' . Directorate-general of Budget, Accounting and Statistics, Executive Yuan. The Republic of China. Staley,Eugene and Morse, Richard , 1965,"Modern Small Industry for Developing Countries" Mcgraw Hill Book Company. Stee1,W. and Takagi,Y.,1983," Small Enterprise Development and the Employment Output Trade -off " Oxford Paper, 2'6 5 423-44 The Report on Mining and Manufacturing Survey: 1979, published by Economic Planning Board, the Republic of Korea, issued at May 1981. The Report on Mining and Manufacturing Survey; 1984,published by Economic Planning Board, the Republic of Korea, issued at April 1986. The Report of 1976 Industrial and Commercial Censuses of Taiwan-Fukien District of the Republic of China, General Report Volume 1, Volume 3 Book 1, published by the Committee on Industrial and Commercial Censuses of Taiwan-Fukien District of the Republic of China, Executive Yuan, issued at Dec. 1978. The Report of 1981 Industrial and Commercial Censuses of Taiwan-Fukien District of the Republic of China, General Report Volumel Volume 3, published by the Committee on Industrial and Commercial Censuses of Taiwan-Fukien District of the Republic of China, Executive Yuan,issued at June 1983. Swapna,Mukhopadhay and Chee,Peng Lim,1985,"Development and Diversification of Rural Industries in Asia" Asian and Pacific Development Center, Kuala Lumpur, Malaysia. Tidrick,G.M.,1975,"Wage Spillover and Unemployment in a Wage Gap Economy: The Jamaican Case" Economic Development and Cultural Change 23,306-24. 266 Todaro,M.P.,1976a,"Internal Migration in Developing Countries" International Labor Office, Geneva. Todaro,M.P.,1976b,"Urban Job Expansion, Induced Migration and Rising Unemployment" Journal of Development Economics 3,211-25. Todaro,M.P.,1969,"A Model of Labor Migration and Urban Unemployment in Less Developed Countries" American Economic Review 57. Tokman,V.E.,1978," An Exploration into the Nature of Informal-Formal Sector Interrelationships" World Development 6, 1065-76. Watanabe,Susumu, 1978,"Technological Linkage Between Formal and Informal Sectors of Manufacturing Industries" WEP 2-22, ILO,Geneva. Weeks,J.,1975,"Policies for Expanding Employment in the Informal Urban Sector of Developing Economies" International Labor Review 3, 1-13. Westphal, Larry and Kim , Kwang Suk , 1977, " Industrial Policy & Development in Korea " World Bank Staff Working Paper No. 263, Washington. Westphal, Larry ,1981 ,"Empirical Justification for Infant Industry Protection" World Bank Staff Working Paper No.445. 'Yearbook of Financial Statistics of the Republic of China 1985' Department of Statistics ,Ministry of Finance. " * ”I11111111111111I