RURAL-URBAN MIGRATION AND ITS ‘ RELATION TO UNEMPLOYMENT IN THE URBAN AREA OF . VALLE DEL CAUCA. COLOMBIA Dissertatim for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY ALFREDO ROA MEIIA. 1974 LIBRARY Michigan State University This is to certify that the thesis entitled RURAL-URBAN MIGRATION AND ITS RELATION TO UNEMPLOYMENT IN THE URBAN AREA OF VALLE DEL CAUCA, COLOMBIA presented by Alfredo Roa Mejia Department of Agricultural Economics has been accepted towards fulfillment of the requirements for ‘ Ph . D. n Agricultural Economics degree i Date 8/28/74 4!“! “1639.1 ION men It“ {3,17 Q’ n,“ . ' twlmwmm m “2.1!; uh: «km. r, V. VALL! L (ZAUPA, v,;‘,, u.. l_~l ,A ' '” rates of urban ug‘..:t.i- :. we, r w M l tipcct of (deny rd It" ’ . ' . study. The prim -‘ ‘ ., 4“ ~ -__1.1. of rural—urn» ’.-.-.-. ‘- ;- a mt in the urban ates- - - .~ ;. - an the (arias 0t ‘35,, 4 H? n- 32‘; H‘.‘ The specific: o‘.‘_;e-r:'.‘ . ‘.~ I. » :¢;.:m‘.;u~ VI. jg”! on rural-urban manner. t:.-.v.'.-.-_- :z-'.f‘.‘.tnrn JOMC Ind inperfem unearthing ‘;.‘l the part at _" V. migrants. (Z) to touwlace a week for t " labor force of urns: mute and particularly lap a theoretical framework that napkins r Mrptum of Yang‘s Mm-iq- teeter a . (4;) to ecu-u: the nub-r. ref 71m -mloyme tu- M 0-! ~-.- . nut in man. m m £04.45) :cu Wu!” n3“, ' was». and new. ABSTRACT RURAL-URBAN MIGRATION AND ITS RELATION TO UNEMPLOYMENT IN THE URBAN AREA OF VALLE DEL CAUCA, COLOMBIA By Alfredo Roa Mejia High rates of urban unemployment, considered by many an important aspect of many related problems of Colombia, motivated this study. The primary objective of this research was the analysis of rural-urban migration and its relation to unemployment in the urban areas of Valle, especially Cali, and to determine the causes of the high rates of urban unemployment. The specific objectives were: (1) to formulate a decision model on rural-urban migration under conditions of urban unemployment and imperfect knowledge on the part of the prospective migrants, (2) to formulate a model for estimating the labor force of urban Valle and particularly Cali, (3) to develop a theoretical framework that explains the limited labor absorption of Valle's manufacturing sector for unskilled labor, (4) to estimate the number of jobs needed to (a) prevent unemployment from increasing and (b) lower unemployment rates in urban Valle during the period 1974 to 1980, and (5) to identify and examine policy alternatives with respect to labor employment. Ito. ml‘ ‘.‘ ~‘no - Alfredo Roa Mejia Urban unemployment rates have risen every year during the period 1960 to 1969 and will continue to be high if basic changes are not affected in the economic organization and activity of Valle. The unemployment rate in Cali was estimated at 18.3 percent in 1969. Migration to the urban centers, which continued during the 19608 despite high unemployment rates, takes place under imperfect knowledge on the part of the prospective migrants. Investment/disinvestment theory helps us to understand (1) migration under imperfect knowledge, (2) high urban unemploy- ment rates, (3) the high proportion of young people and females among migrants and (4) people living in both rural or urban areas under very difficult economic conditions without being able to seek better opportunities in other places. The lack of educational opportunities causes a low rate of conversion of unskilled to skilled labor. This fact along with migration to urban areas, high fertility rates, and the increasing participation of women in the labor force, results in large numbers of unskilled laborers in the urban zones of Valle, especially in Cali. The demand for unskilled labor grows very slowly in the urban zone because the industrial sector has a low demand for additional unskilled labor. Manufactured goods are largely produced in big firms with high capital-labor - ratios while labor-intensive small and medium-sized firms are relatively few and absorb little labor. Ia _o‘vA ‘ Alfredo Roa Mejia The total number of jobs that need to be created during the period 1964 to 1980 to give everyone of the projected labor force the opportunity to find a job and to eliminate low paid jobs in Cali are estimated at 114,000 to 162,000 under three alternative assumptions about migration to Cali. In the past, policies oriented toward production growth and productivity have produced (1) high unemployment and underemployment rates, and (2) uneven income distribution. Agricultural and manufacturing sectors of Valle are charac- terized by an acute dualistic system. Under dualism policies oriented to increase agricultural and manufacturing produc- tion in the modern sectors can hurt the traditional sectors. In this study policies aimed at (1) lowering urban unemploy- ment rates, (2) obtaining more even distribution of income, and (3) increasing production were identified and examined. The specific policies considered were: (1) revision of labor code especially as it affects unskilled laborers and small and medium-sized enterprises, (2) reduction in growth of total labor force, (3) changing composition of labor force and upgrading of skills, (4) slow down the rate of migration to the large urban centers such as Cali, (5) creation of jobs in urban areas for large pool of unskilled workers, and (6) improvement in the performance of the labor market. The first four policies are aimed at affecting the amounts and kinds of labor at the large urban centers of Alfredo Roa Mejia Valle. Policies to create additional jobs in urban Valle are focused in (a) relative price of different kinds of labor and different kinds of capital, (b) factor proportions used by size of enterprises, (c) use of installed capacity, and (d) amounts and kinds of labor absorbed by different products. The lack of information on urban wages and employment opportunities on the part of the prospective rural migrants and the existence of urban unemployment and rural underemployment despite high migration rates to the urban areas of Valle suggested the need for improving the performance of the labor market. Research on recent migratory currents and the effect of the present National Development Plan on employment in the last two years should be carried out as data becomes available. Also, future research is needed on (1) the kinds of products being produced by small and medium-sized enter- prises, the prices they received for their products and on its costs of production; (2) estimates of elasticities of substitution between different kinds of labor and capital and between imported and domestically produced inputs; and (3) performance evaluation of public institutions created to help small farmers and manufacturing firms. 1‘” Iv , "r I -URBAN MIGRATION AND ITS RELATION ‘moUNEm’LOYMENIINIHEURBANAREAOF VALLE DEL CAUCA, COLOMBIA By Alfredo Roa Mejia A DISSERTATION Submitted to Michigan State university “.13 partial fulfillment of the requirements . , for the degree of DOCTOR OF PHILOSOPHY 1924 I.” 3“...th for 'tp 92.: :7, .'~ “-mc‘ln‘ly gratni ;; . -. \ - ' m the sum-.2- ma .; I“ Dr. "scam I. Inn- , m; .- “m “if! of this sum w. u flatly ”preciated. ' H’u' . . .' of m Wfip‘atefith 3l’e‘iirizr and Zoila, A a. magma marten PX Nita. Luc11a . it}. vabCKBfeIieYF .MJ.’ .1} V I _‘.‘. '; ‘ '., :1 m for the melanin... . ~ fl “3 th- propane ion of Lh '- -‘.‘ ..,. .~ » .... .~ ,__,.. ‘- .- m “mm a 7’ h GotpOtaci-nn Aurore :a New}. 1:21 dH fen-:9. (NZ, _. 'Olifiod the author T’Wr e. 'm ’necessary for the ,Irey‘ili" “ m .tfly while an ample}- ~ vi the I‘oxporarim‘n and hi author by its interest in the subject of ACKNOWLEDGMENTS The author wishes to express his gratitude to his major professor, Dr. Glenn L. Johnson who offered several useful suggestions for improving this thesis. The author is particularly grateful to him for the patience he displayed throughout the author's graduate program. Dr. Gerald I. Trant suggested the topic and read the first draft of this study. His advise and friendly attitude is highly appreciated. Thanks are also given to the other members of the thesis committee, Drs. Harold Riley, Lester V. Manderscheid and Kelly Harrison. The Rockefeller Foundation gave the financial assistance necessary for the undertaking of my post-graduate studies, and for the preparation of this study, for which the author is very grateful. The Corporacién Auténoma Regional del Cauca, CVC, kindly allowed the author the time necessary for the prepara- tion of the study while an employee of the corporation, and encouraged the author by its interest in the subject of rural-urban migration and unemployment. iii Chapter I. II. III. IV. TABLE OF CONTENTS INTRODUCTION . Need for the Study . Objectives . Theoretical Framework for the Study Organization of the Study . . . . DEMOGRAPHIC AND ECONOMIC CHARACTERISTICS FOR CALI AND VALLE . . . Population of Valle Population of Cali . Migration to Valle . . . Migration Within Valle . Migration to Cali Production in Valle Income Distribution DECISION MODEL FOR RURAL-URBAN MIGRATION . Factors Affecting Rural-Urban Migration . Decision Model for Rural- Urban Migration With Urban Unemployment . Decision Model for Rural- Urban Migration Under Imperfect Knowledge THE URBAN LABOR FORCE AND SUPPLY . Skilled Labor Supply . . Unskilled Labor Supply . Union's Effect on the Supply of LabOr The Labor Force in the Urban Zone of Valle . . The Labor Force in the City of Cali URBAN LABOR ABSORPTION . Demand for Labor . . Modifications in the Demand for Labor Elasticities of Substitution . . iv 41 45 53 65 66 78 79 84 94 100 111 v‘fl Chapter Page Labor Input- Output Analysis . 121 Labor Employment and Size of Firms. 138 Used of Industrial Installed Capacity . 143 Other Factors Affecting the Manufacturing Industry Growth Rates . 149 VI. URBAN UNEMPLOYMENT RATES 151 Unemployment Rates, 1963-1971 . 152 Additional Jobs Needed in Cali 165 VII. ALTERNATIVE POLICIES TO REDUCE RURAL-URBAN MIGRATION AND INCREASE PRODUCTIVE EMPLOYMENT . 169 Revision of the Labor Code Especially as it Affects Unskilled Laborers and Small and Medium-Sized Enterprises . . 172 Reduction in Growth of Total Labor Force. 176 Changing Composition of Labor Force and Upgrading Skills 177 Slow Down the Rate of Migration to the Large Urban Centers Such as Cali . 179 Creation of Jobs in Urban Areas for Large Pools of Unskilled Workers 186 Improve the Performance of the Labor Market . . . . . . . . 201 VIII. SUMMARY AND CONCLUSIONS . 204 Future Research . 216 APPENDICES . 224 A. Some Characteristics of the Population of Cali and Valle del Cauca . 219 B. Data on Valle' 8 Industrial and Agricultural Sectors . . . 224 C. ILO' 5 Model for Projection of Employment in Colombia . . 231 D. Questionnaire on Migration Applied to Rural Inhabitants. Munici io of Palmira, Corregimientos R320, Acequia, Latorre . . . 238 BIBLIOGRAPHY . . 240 v I Aha--- 2.1 Table LIST OF TABLES Comparison of the Rate of Growth of Total Population of Colombia, Valle and Cali, 1951 and 1964 . . . . . . . . . . Relation Between Economic Active and Total Population of Valle del Cauca According to Residence and Age, 1951 and 1964 . . . Comparison Between the Total Population Living in Valle and the Population Born Outside of Valle del Cauca and its Growth Rates, 1951 and 1964 . . . . . Place of Origin of Valle' 3 Non- Native Population, 1951 and 1964 . . . . Place of Origin and Present Residence of the Valle s Non-Native Population, By Sex, 1964 . . . . . . . . . . . . Place of Origin and Present Residence of the Non-Native Born Population of Cali by Sex, 1964 . . . . . . . Distribution of the Population of Cali by Place of Origin, Sex and Age Groups, 1969 Total Value of Manufactured and Agrarian Production of Valle Measured in 1958 Constant Prices for the 1960 to 1970 Period Distribution of Family Income in Valle, 1965 . . . . . . . . . . . . . . . . . Distribution of Family Income in the Rural Sector of Valle, According to Sample July 1972 . . . Distribution of Family Income in Cali, August 1969 . . vi Page 13 15 20 21 23 27 28 31 36 36 37 Knowledge of Unemployment in the City; Sample from Municipio of Palmira, Corregimientos Rozo, Acequia, Latorre, 1973 Knowledge of Unemployment and Relative Urban Wages of Prospective Migrants from the Countryside to the City; Sample from Municipio of Palmira, Corregimientos Rozo, Acequ a, Latorre, 197 Population of the Flat Zone of Rural Valle by Sex and Age, Migrating to Urban Valle During 1964 to 1973 . . Population of the Mountainous Zone of Rural Valle by Sex and Age, Migrating to Urban Valle During 1964 to 1973 . . . Levels of Education of People from the Rural Flat Zone of Valle Migrating to Urban Valle during 1964 to 1973 . . Levels of Education of People from the Mountainous Zone of Rural Valle, Migrating 4 . . to Urban Valle During 1964 to 197 Urban Labor Force in Valle With and Without Migration During the Period 1964 to 1980 . Urban Labor Force in Cali With and Without Migration During the Period 1964 to 1980 . Total Manpower Employment in the Urban Sector of Valle, 1962 to 1969 . . . Total Manpower Employment in the Consump- tion, Intermediate, and Capital Producing Goods of Valle' s Manufacturing Industry, 1956 to 1968 . . . Employment Growth Rates in the Industrial Sector of Valle, 1956 to 1968 . Percentage of Total Employment Absorbed by the Principal Industrial Centers of Valle, 1965 to 1969 . . . . vii Page 50 52 75 76 77 77 85 89 95 97 99 . 100 Table 5.5 Regression Coefficients of the CES Production Function When the Efficiency Parameter Changes Over Time for Valle' 8 Industrial Branches . . . . Regression Coefficients of the CES Production Function When the Efficiency Parameter is Constant for Valle's Industrial Branches . . . . . Industrial Input— Output Matrix (1- A) of Colombia . . Matrix of Technical Coefficients of Workers' Employment in the Industrial Sector of Valle . . . . Final Demand from Other Sectors for Industrial Products: Present and Modified Aggregate Demands, Valle, 1969 . Industrial Employment in Valle With Present Industrial Aggregate Demand, 1969 . Industrial Employment in Valle With Modified Industrial Aggregate Demand, 0’ 1 69 . . . . . . . . . . . . . . . . . . Industrial Employment in Valle With Mbdéfied. Industrial Aggregate Demand, d1, 196 . . . Industrial Employment in Valle With Modified Industrial Aggregate Demand, d2, 1969 . . Industrial Employment in Valle with Modified Industrial Aggregate Demand, .d3, 1969 . . Size of Valle's Craftmen and Manufacturing Enterprises According to the Number of Employees, 1963, 1966 and 1969 Rates of Utilization of Colombian Industrial Equipment Based on the Maxi mum Number of Shifts Worked, by Industry, 1971 . . . . viii Page 117 119 125 125 128 131 134 135 137 148 144 Table Page 6.1 Economically Active Population of Valle According to Types of Activity, Sex, and Time Worked, 1964 . . . . . . 156 6.2 Active and Non-Active, Employed and Unemployed Population of Cali in Different Years . . . . . . . . . 158 6.3 Percentage Classification of Population Employed and Unemployed in the Major Cities of Colombia, 1964 . . . . . 160 6.4 Distribution of the Unemployed in Cali According to Origins and Characteristics, 1969 . . . . . . . . . . . . . . . . . . 161 6.5 Number and Rate of Unemployed in the Urban Area of Valle .During the Period 1964 to 1969 . . . . . 164 6.6 Estimated Number of Jobs Needed in Cali From 1974 to 1980 . . . . 167 A.1 Number of People of Valle and Rate of Growth Accordin to Residence, Age and Sex, 1951 and l 64 . . . . . . . 219 A.2 Economic Active Population of Valle, 1951 and 1964 . . . . . . . . . . . . 220 A.3 Relation Between Migrant Population and Total Economic Active Population of Valle del Cauca, by Age and Economic Sectors, 1964 . . . . 221 A.4 Population of Cali Classified by Residence, Age and Sex, 1951 and 1964 . . . 222 A.5 Economic Active Population of Cali, 1951 and 1964 . . . . . . . . . . . . 223 8.1 Value Added of 17 Industrial Branches of Valle del Cauca, 1956 to 1968 . . . . 224 B.2 Number of Workers Employed in 17 Industrial Branches of Valle del Cauca, 1956 to 1968 . 225 3.3 Annual Wage of Laborers Employed in 17 Industrial Branches of Valle del Cauca, 1956 to 1968 . . . . . . . . . . . . 226 ix -eacturing Sector Price Index and icit GNP Price Index, 1956 to 1968 . and Area of Farms Classified rang Area of Farms Classified by , eatiOn Between the Amount of Manpower Ht lized.by Each Industry of Valle and the Respective Value of Production in Constant 1958 Prices . 491?“? Page 228 229 230 Figure LIST OF FIGURES Page Segments of iso-MVP lines when acquisition price is greater than salvage value for both inputs. . . . . . . 55 The prices of labor for unionized and non-unionized laborers and the marginal value product curve of labor . . . . . . . 104 Demand for labor when capital is variable. 106 Expansion paths of firms for non-unionized and unionized labor . . . . . . . . . . . 108 Change in technology and expansion paths for a firm . . . 110 Function of supply and demand in the unskilled labor market . . . . . . 151 Labor employment when the minimum number of workers legally required to form unions is increased . . . . . 175 LIST OF MAPS Page Colombia 2 xi CHAPTER I INTRODUCTION The purpose of this study was to analyze rural-urban migration andils'relation to unemployment in the urban areas of the Department Valle del Cauca1 and particularly in Cali, the department's largest and capital city. Thus the area of study is the department of Valle but frequent references are made to other areas of Colombia when considering migration. The location of Valle, in the southwest area of Colombia and the location of other departments can be seen in the map included on the following page. Need for the Study High unemployment rates have been found in Colombia, especially in the big cities like Cali, where the rate has exceeded 18 percent. Analyzing the simultaneous existence of such high urban unemployment and rural underemployment despite high migration rates to the urban areas of Valle is the central focus of this research. Studies relating rural—urban migration and unemployment 1Hereafter Valle del Cauca will be referred to in a short form--Valle . Barranquilla Atlantico Caribean Sea Venezuela Arauca Risarald '3 a Vichada c u !|||||l: « 'H a o a a Narifio Vaupes Caqueta . ‘ Brazil Ecuador Amazonas Peru Map 1. Colombia have not been carried out for Valle. Data on migration are also quite limited. Approximations on migratory currents in Valle were possible, however, as a result of the Universidad del Valle's detailed analysis of the 1964 population census data wherein information was made available about the Valle's non-native population. These figures were utilized as proxy variables for migration. Unemployment rates in Cali have only been estimated for four different years beginning in 1965.2 There are recent studies of unemployment for Colombia as a whole. The Colombian government requested and received an economic mission of the International Labor Organization (ILO) which undertook the study of unemployment and employ- nent of the Colombian labor force. The mission presented its report in 1970.3 This effort was a general analysis of the Colombian labor force and the economic situation of the country. The study recommended several policies to lower the unemployment rate to 5 percent by 1985. The Colombian government adopted an Economic Development Plan4 in 1972 to achieve a similar goal but the policies and strategies differ from those suggested by ILO. There was, in fact, a significant 2 3Oficina International del Trabajo, Hacia e1 Pleno Eymlgg. Bogota: Imprenta del Banco Popular, See Chapter VI, Table 6.2. 4Dep artamento Nacional de Planeacién, DNP, Las Cuatro Mags. Bogota DNP 3 " h ,1- no. .u.. I... a a I h . u ‘6 . I. I .I . I a difference in the policy formulated in one of these two approaches. The Economic Development Plan gives the con- struction sector (houses and urban growth) the leading role in absorbing the expanding labor force while ILO's proposal stresses agricultural and manufacturing growth using labor intensive methods. Hence, it was felt important to examine the difference in policies suggested by ILO and the ones adopted by the Colombian government. More importantly it was felt necessary to examine the specific applicability of these general recommendations for Valle. All efforts which enhance understanding of the magnitude of unemployment in Valle, the identification of possible means of reducing it, and the evaluation of the effectiveness of policies designed to lower the unemployment rates, are of prime importance. Objectives The present study has the following specific objectives: 1. To formulate a decision model on rural-urban migration under conditions of urban unemployment and imperfect knowledge on thepart of the prospective migrants. 2. To formulate a model for estimating the labor force of urban Valle and particularly of Cali. 3. To develop a theoretical framework that explains the .limited labor absorption of Valle's manufacturing sector for unskilled labor. To estimate the number of jobs needed to (a) prevent a rise in unemployment and (b) lower unemployment rates ('1 in urban Valle during the period 1974 to 1980. 5. To identify and examine policy alternatives with respect to labor employment. Thgoretical Framgwork for the Study Initially we will consider the basic theory to be used in formulating decision models for use in understanding rural-urban migration under conditions of (1) differentials between the cost of adding laborers to the farm economy and the returns obtainable from off-farm migration and, (2) imperfect knowledge. The flow of rural-urban migration which is quite heavy in Valle despite the high rate of urban unemployment has attracted the attention of many economists. The migration decision is viewed in this study as an investment/disinvest- tent decision where potential migrants choose that alter— native whichmaximizesthe present value of the expected income streams. An early explanation to rural-urban migra- tion when there is unemployment in the recipient areas was 5 used by Glenn L. Johnson which was later redeveloped by Michael P. Todaro6 and John R. Harris and Todaro.7 The 5Glenn L. Johnson, "Allocative Efficiency of Agricul- tumal Prices as Affected by Chan es in the General Level of Ehmloyment," unpublished Ph.D. t esis, Department of Economics, University of Chicago, 1949. 6Michael P. Todaro, "A Model of Labor Migration and Urban Unemployment in Less Developed Countries," American Economic Review, Vol. 59, No. 1 (1969). 7John R. Harris and Michael P. Todaro, "Migration, basic idea is that when the expeCted real urban income, that is, the real urban income adjusted by theprobability of getting employment in a given period of time is greater than the rural real income E(Yu) > Yr’ there will be rural- urban migration. Therefore there is a level of unemployment consistent with equilibrium in the labor market, when E(Yu) = Yr' Glenn L. Johnson's more recent investment— disinvestment theory8 more adequately explains rural-urban migration when there is imperfect knowledge in the urban areas. If the expected salvage value of rural workers, that is earnings available to them in urban industries, times the probability of finding a job in the urban areas in a given period of time is greater than wages or marginal value products (MVPs) of rural laborers, then migration will take plaCe. Since the decision to stay on the farms or to migrate is taken in an uncertain situation, the risk taking attitude of the migrants has to be considered. Under uncertain situations right and wrong decisions will be taken with respect to staying or moving out of agriculture and the possibility of reversing wrong decisions has to be examined. This means that many individuals and groups of people can be unemployment and Development, A Two-Sector Analysis," American Economic Review, Vol. 60, No. 1, (1970). 8See, for instance, Glenn L. Johnson and C. Leroy Quance (editors), The Over roduction Tra in U.S. A ricul— ture. Baltimore: JoEfis HopEins UnIversIty Press, I972. simultaneously overcommitted to both agrarian and urban pursuits. Thus, we see how the simultaneous existence of high urban unemployment and rural underemployment despite high rural-urban migration, the central focus of this thesis as we said before, can be understood. We turn now to the second objective which is to formulate a model for estimating the labor force of urban Valle and particularly of Cali. The labor force can be imperfectly divided into two broad categories: skilled and unskilled workers. The supply of skilled labor reflects investment in human capital during the period of training and preferences of workers between greater income and leisure time. The slope of the supply function should be positive for low income levels and, according to some theories, has a backward bending segment for high salaries. In the case of the unskilled labor supply human investment is much less relevant. In Colombia the unskilled labor supply function appears to be a stepped supply function. Colombian labor legislation established 25 as the minimum number of laborers working for a given firm to form a union. Thus, for the first 24 workers the relevant salary is the minimum legal wage; for 25 workers and over a higher salary is generally established on the basis of collective bargaining agreements between the firm owner and the union. Unemployment of unskilled workers in Valle and Cali is very serious and consequently special attention will be concentrated on them in this study. , . K. \ . '1.‘ I“ It is postulated that for a given wage the additional number of unskilled workers entering the labor market depends on the following factors: The rate of growth of the native population of working age living in the urban areas, the flow of rural-urban migration, the rate of the participa- tion of women in the labor market, and the rate of conversion of workers from the unskilled to the skilled labor market. The rate of growth of the native population depends on birth and mortality rates. The rate of feminine parti- cipation is expected to increase both as a consequence of technological change which makes domestic tasks easier and allows the substitution of manufactured for home-made goods, and the need to contribute to family income. The flow of rural-urban migration is quite heavy despite the high rate of urban unemployment. The conversion rate of workers from the unskilled to the skilled labor market is a function of the difference between the wages of skilled and unskilled workers, the unemployment rates for both kinds of labor, the proportion of the income which can be invested in training, educational and training facilities available, and the distribution of income. In what follows, the basic ideas and hypothesis for developing a theoretical framework for the third objective, that is, to explain the limited labor absorption of Valle's nanufacturing sector for unskilled labor will be presented. Under perfect competition the demand for labor when V‘ the entrepreneurs are profit-maximizers is given by L a f(VMP) where L is the quantity of labor and VMP is the value of marginal physical product of laborers. When firms have monopolistic power, the demand for labor is L = f(MRP) where MRP is marginal revenue product, that is the marginal physical product multiplied by the correspond- ing marginal revenue. Under uncertainty, entrepeneurs maximize non-monetary as well as monetary values. The kinds of capital, the capital-labor ratio, and cost structure are related to size of firms. For handcraft and very small manufacturing industries, labor is the most important factor input. For medium size enterprises, the capital-labor ratio is lower than in large firms and labor cost is relatively high. Increases in the cost of labor have different effects on medium size firms. Those firms with enough funds can acquire labor-saving capital and do not hire over 25 workers to avoid unions and labor cost increases. Other firms which cannot afford changing technology may decide to hold the number of laborers below 25. For these firms laborers,when unionized,make labor more expensive than capital, that is, the ratio between marginal value product of labor and marginal factor cost of labor is less than the ratio between the marginal value product of capital and the marginal factor cost of labor-saving capital. The largest firms in Valle were inaugurated with big plants and modern equipment and therefore need a number of workers 10 greater than the legal minimum number for unionization. Most of their capital saves unskilled labor but is complementary with skilled labor, especially in the intermediate, durable and capital goods producing industry. We will study factor proportions in Valle's industry examining elasticities of substitution for industrial branches. Also we will examine the effects on employment of (1) an increase in the aggregate demand for consumption goods and (2) an equal decrease in the aggregate demand for durable, intermediate, and capital goods. To examine this, an input-output table of the Colombian manufacturing sector will be utilized, transformed in such a way that it only shows the labor input of each of the industrial sectors. We will also analyze the relation between cost of labor and the level of utilization of installed capacity. With respect to the fourth objective, that is to estimate the number of additional jobs that need to be created in urban Valle (1) to prevent a rise in unemployment and (b) to lower unemployment rates during the period 1974 to 1980, we will investigate alternative assumptions with respect to rate of migration. The difference between the number of people of working age estimated under alternative assumptions with respect to migration (using the model for estimating the labor force of Valle developed in this study) and the number of jobs that will be created in the manu- facturing and service sector if the same trends observed 11 in the past continue until 1980, will give us the additional jobs that need to be created to prevent further increases in unemployment. Additionally we will estimate the total unemployed and low paid jobs at the end of 1973 using some parameters estimated by ILO for Colombia. Finally, we will identify policy alternatives with respect to labor employment and to estimate their conse- quences, the fifth objective of this study. The conclusions on employment policies of this study are compared with the present Colombian economic and social development plan as well as the suggested policies by ILO made in 1970. Organization of the Study The study consists of eight chapters. This intro- duction is Chapter I. Chapter II presents selected aspects of population growth and the economic activity of Valle and Cali. These aspects are important for the analysis carried out in the following chapters. Chapter III contains a model that is used to explain rural-urban migration when there is urban unemployment and imperfect knowledge on the part of the prospective migrants. Chapters IV and V contain the analysis of the number of laborers looking for jobs in the urban areas of Valle and labor absorption in those areas, respectively. In Chapter VI the unemployment rates of the urban areas of Valle and particularly in Cali are quantified and analyzed. In Chapter VII employment policies are identi- fied and examined. Finally, Chapter VIII contains a summary of our findings and the conclusions to be drawn from these. CHAPTER II DEMOGRAPHIC AND ECONOMIC CHARACTERISTICS FOR CALI AND VALLE As a basis for a sound comprehension of the extent of unemployment in the region, some of the economic and demo- graphic data for Cali and Valle are presented in this chapter. Special attention is focused on the population growth rate, geographic distribution, migratory patterns, regional production, and income distribution. We will deal first with information for Valle as a whole, then for Cali. Population of Valle Figures from the census of 1951 and 1964 show that the population of Valle has grown at a rate of 3.5 percent a year. This implies Valle's population will double in the next twenty years if the same rate continues. The growth rate for this department is higher than the national level of 3.2 percent according to the same census. Table 2.1 shows the population of Cali, Valle, and Colombia in the census years 1951 and 1964. It will be observed that the rate increase in the population of Cali is twice the rate of increase in the population of Colombia, and almost doubles that of Valle. 12 13 Table 2.1. Comparison of the Rates of Growth of Total Population of Colombia, Valle and Cali, 1951 and 1964. Item Year Annual 1951 1964 Growth Rate b ------- Number; --------- --Percent--- City of Cali 241,357 618,215 7.5 Department of Valle 1,106,927 1,733,053 3.5 Colombia 11,548,209 17,484,508 3.2 Source: DANE, Censos de Poblacion de Colombia, 1951 and 1964. The total population of Valle in 1973 is estimated to be 2,361,972, and the population of Cali to be 1,115,042. Cali thus has 47 percent of the total population of Valle. The urban population, those living in administrative centers called cabeceras municipales,1 has grown during the years 1951 and 1964 at the rate of 6.3 percent, while the rural population decreased at the rate of 0.6 percent. In 1964, 70.4 percent of the Valle population was classified as urban, and 29.6 percent as rural. The cabecera municipal of Cali had 52.3 percent of the Valle's urban population in 1964 while the cabeceras_municipglg§ of Buenaventura, Buga, Cartago, Palmira and Tulua had 29.2 percent of the Valle's urban population in the same year. The difference between the two rates is largely explained by the phenomenon of migration from the country 1The definitions of rural and urban areas are discussed in Chapter III. 14 to the city. As the distribution of the population by age and sex shows, a high proportion of young persons and women is concentrated in urban areas. It is a known fact that the population which migrates mostly from the country to the city are the young and the women.2 The urban population between 15 and 64 years of age doubled in the period from 1951 to 1964, while the same age group in the rural sector decreased at an annual rate of 1.5 percent during the same period, as can be computed from data given in Table 2.2.3 In the urban sector the female population is larger than the male in both census years, whereas in the rural areas the situation is reversed, there are more men than women (see Appendix A.1). The concept of economically-active population is useful in understanding the participation of the population in the labor force and the rate of economic dependence. We use here the definition of economically active population adopted by the National Administrative Department of Statistics (DANE) for application in the 1964 census, which states: 2The reasons for higher rates of migration for young people and females are discussed in Chapter III. 3The age group less than 15 years of the rural popu- lation is the only age group which has an increase in the number of eople, showing high birth rates in the rural area of Va 1e. 15 .qomH can Han .mHQEOHoo on GDHomHnom on mousse .m2¢a .aOHumHomom 0>Huo< hHHMUHaouoom "oouuom H .. me.m www.mH u- oNH.o ONN.MH Ho>0 use no q.nm omo.n¢H nmo.om~ m.oo omm.wwH Hwo.mom «oan -u ooH.m mmm.¢¢~ I- qu.m moo.¢m~ uHo mummy nH noun mmoH m.Hm qu.ooH oH~.mHm N.om mmo.Hom oom.omm Hanna .. nmo.w an.mm .- owo.m www.5H uo>o can no m.mm owm.~mm nua.wmo m.om oo~.me w¢~.n~m «ounH .. mam.w mne.m~m .. mum.o ¢N¢.nom uHo memo» mH coup mmoH m.om omo.mom mmw.mH~.H w.nm NHH.an wnn.onm noun: .. mum.mH nHm.n¢ u- NHN.HH oOH.Hm uo>o was no o.¢m oHo.mm¢ «mo.mHm m.mm oom.mnm mam.¢mo «ounH us mnm.oH oo~.omn u- mo GOHuoHuaom Houoa can o>Huo< oflaoooom nooapom GOHuoHuM .«omH can HmmH .ow< use oooovaom ou waHuuooo< .N.N «Hana. l6 "Economically active population includes all those of age twelve (12) or more, who received income from a job held during the census year, in the production of goods and services, or who worked without pay as family helpers in the enterprise of the respective head of family, for at least a third of normal working hours. During the census year, the census establishes a limit of continuous or discontinuous activity for nine months in order 20 discriminate between employed and unemployed." Upon comparing the economically active population with both the total population and with the population between the ages of 15 and 64, one notices that in the period under consideration the proportion of the population which is economically active diminished in both rural and urban areas as shown in Table 2.2. Moreover, in 1964, only 53.5 percent of the urban population between the ages of 15 and 64 was classified as economically active. This percentage of participation in the urban labor market of the working-age population is lower than the participation rate of rural Valle which was 57.4 percent while for the department as a whole was 54.6 percent during the same year. The percentage of the economically active population with respect to the total population was 30.6 percent in 1964, which illustrates a high rate of economic dependency in the Valle's population. The low rate of participation in the economically active part of the population of working-age is due to various 4DANE, "XIII Censo Nacional de Poblacion. Resumen del Valle del Cauca," Bogota: Dane 1967. "I. o.- o ‘0 17 causes: (a) the high proportion of women dedicated to home- making, which was 78 percent in 1964, (b) students and incapacitated, who constituted 7.5 percent of the male working age population, and 5.4 percent of the females in the same age group, and (c) unemployed, retired and beggars. Population of Cali In 1951, 25.7 percent of the population of Valle lived in Cali, but in 1964 a third of the department's population was living in the capital.5 This concentration shows the attraction of Cali for inhabitants of other departments, and of other municipalities of Valle. In 1973 it is estimated that Cali includes 47 percent of the total population of the Department of Valle. The urban population of Cali grew at the rate of 7.5 percent annually between 1951 to 1964, which implies Cali's population will double in ten years, only half the time which it will take the Department of Valle, if the same rates continue. This high growth rate reflects the fact that Cali is a greater absorber of populationthan Valle. The population of the city of Cali is young, with a high proportion of women, especially between the ages of 15 to 64.6 As was said before, this is characteristic of centers which receive a great number of migrants. 5These proportions were computed from data given in Table 2.1. 6The distribution of the population by age, sex and place of residence may be consulted in Appendix A. 18 The economically active population of Cali shows the same decreasing tendency observed in Valle. The rate of economic dependency in Cali was 32.2 percent in 1964.7 The rate of participation in the labor market in 1964 is similar to the rate of the department and the same reasons explain this situation. As we have seen, Valle and the capital city, Cali, receive large numbers of migrants. Many migrants come from outside the department of Valle, many change residence within the department, and the majority of these come to Cali. Statistics on these different types of migration are scarce; the only reliable data is that from the census of 1951 and particularly for the census of 1964.8 During the preparation of the present research, primary information on the causes of rural-urban migration in some zones of Valle was obtained; the analysis of those results is presented in Chapter III. Migration to Valle The high level of migration to Valle from other depart- ments is shown by the high proportion of the population of 7In Table 6.2 of Chapter VI on unemployment, it is seen that the economically active population rose from 32.2 percent in 1969 to 47.7 percent in 1971; this is due to the Panamerican Games. 8The Universidad del Valle with collaboration from DANE published more detailed data on migration for the census year of 1964.- See Universidad del Valle, Comité Universitario de Investigaciones sobre poblacién, CUIP, Censo de Poblaci6n ara e1 Departamento Del Valle, Vol. II, Cali, 1968. 19 Valle born in other departments of Colombia. If the total population of Valle is compared with the number of those born in other departments, in 1951, 37.5 percent were born outside Valle, and in 1964, 33.4 percent. The annual increase of the migrant population in the period between 1951 and 1964 was 2.6 percent, while the total increase of the population each year was 3.5 percent. In Table 2.3, the non-native born population of Valle is distributed by age groups. This shows that more than 75 percent of this group is constituted by persons of working age, for both census years, confirming the fact that the young are the most mobile. In Table 2.4 the previous residence of the migrants to Valle is identified. The categories include the original department of Caldas (subsequently subdivided into three departments), Antioquia, Cauca, and two groups which include several departments. The first of the two groups includes Cundinamarca, NariHo, and Tolima. The second group contains Atlantico, Bolivar, Boyaca, Choc6, Huila, Magdalena, Norte de Santander, Santander, and the Intendencias and Comisarias. Caldas, Antioquia, and Cauca account for 67 percent of the migration to Valle in 1951, and 63.1 percent in 1964. In 1951, 38 percent of the total migratory group arriving in valle came from Caldas, and 36 percent came from Caldas in 1964. The different kinds of migratory currents among non-native born population of Valle and their relative importance shows h .momH .HHmo .oHHo> Hon cuooauuommn Hm mumm sOHomHnom on commm .mHDo .oHHm> Hon umvauo>Hoa "ooHoom ¢.m w.~ mHm.n¢ m.~ 00H.Hm Ho>o new mummy mm m.N w.~n «mo.mHm m.nm mwm.¢mo «you» acumH q.¢ ¢.¢¢ oow.omn o.o¢ N¢¢.Hq¢ uHo mummy nH umnu mde 0 n.m o.o0H mmo.mmn.H o.OOH num.oOH.H ooHuMHoaom Hmuoa 2 is as 08.3 5m Sine 88 as $3» 3 q.~ «.05 Hmo.N¢¢ n.wn m¢~.omm memo» soumH H.m m.wH onm.m0H w.NH NHw.mn uHo mums» mH coup mmoq o.~ c.00H nmm.mmm o.OOH aom.mH¢ mHHm> mo GOHumHomom m>Hum2uaoz Hmuoa uuuomouomn- uuumoummuu -uuonaoz- uuooouomuu uuquBSZI nusouu mo oOHuHOQOHm HWuoa GOHuuonoum Hmuoa spam Hmono< «omH and» HmmH and» EouH .«omH was HmaH .moudm £u3ouo nuH can dunno Hon oHHm> mo ouHmuoo anon cOHuaHnmom onu can oHHu> aH waH>HH GOHuuHouom Hmuoy on» soosuom uOmHunmaoo .m.~ OHHHH 21 Table 2.4. Place of Origin of Valle's Non-Native Population, 1951 and 1964. Place of Birth Year 1951 Year 1964 and Group Ages Total Proportion Total Proportion Number Percent Number Percent Caldas 159,300 38.3 209,586 36.2 Less than 15 Years Old 36,744 49.8 50,456 46.1 15-64 Years 117,869 36.1 150,809 34.1 65 Years and Over 4,687 30.3 8,321 30.9 Antioquia 65,710 15.8 73,391 12.7 Less than 15 Years Old 7,329 9.9 9,877 9.0 15—64 Years 54,052 16.6 57,555 13.0 65 Years and Over 4,329 28.0 5,959 22.1 Cauca 56,488 13.6 82,724 14.3 Less than 15 Years Old 9,317 12.6 14,095 12.9 15-64 Years 44,879 13.8 64,567 14.6 65 Years and Over 2,292 14.8 4,062 15.1 Group 1* 87,761 21.1 140,181 24.2 Less than 15 Years Old 14,259 19.3 24,168 22.1 15—64 Years 71,102 21.8 110,679 25.0 65 Years and Over 2,400 15.5 5,334 19.8 Group II** 39,394 9.5 63,612 1.0 Less than 15 Years Old 5,457 7.4 9,405 8.6 15-64 Years 32,528 10.0 51,559 11.7 65 Years and Over 1,409 9.1 2,648 9.8 Foreigners 6,856 1.7 9,733 1.7 Less than 15 Years Old 706 1.0 1,569 1.4 15-64 Years 5,815 1.8 7,522 1.7 65 Years and Over 335 2.2 642 2.4 Total 415,509 100.0 579,227 100.0 Less than 15 Years Old 73,812 100.0 109,570 100.0 15-64 Years 326,245 100.0 442,691 100.0 65 Years and Over 15,452 100.0 26,966 100.0 *Cundinamarca—NariHo-Tolima **At1§ntico-Bolivar-Boyaca-Chocé-Huila-Magdalena—Norte de Santander— Intendenias y Conisarias. Source: 1970 (mimeo). Universidad Del Valle, CUIP "Estudios de Poblacién," Cali, CA. on. 5.. ‘4 ‘. 1 s, u 22 in Table 2.5. The cabeceras municipales are considered urban areas, but this is not really a valid picture of reality, for there are many cabeceras in which the inhabitants primary activity is agricultural. For this reason, the figures under- estimate the rural-urban migration.9 Table 2.5 shows four kinds of migratory currents: (l) urban-urban, (2) urban-rural, (3) rural-urban, and (4) rural-rural migration. Urban-urban migration involves the highest proportion of the Valle non— native born population, 56.5 percent followed by rural-urban migrants 16.9 percent, rural-rural migrants 13.8 percent and urban-rural migrants, 12.8 percent. A higher proportion of men than women migrate to rural areas while a higher pro- portion of females than males migrate to urban areas. If Valle's native and non—native migrants are considered, the relative importance of the four kinds of migratory currents do not change substantially: 54.8 percent for urban—urban migrants, 17.5 percent for rural-urban migrants, 13.9 percent for urban-rural migrants and 13.8 percent for rural-rural 10 However, we can see that Valle native population migrants. is more mobile from the country to the city and between rural areas. Out migration of natives for Valle, measured as the change in the number of Vallecaucanos living in other departments 9A discussion of this definition is found in Chapter III. 10Rural-rural migration was also found by CVC in the flat and mountainous rural areas of Valle, as pointedcnn:in Chapter IV of this study. v- 6'.. .womH .HHoo .HH .Ho> .oHHm> Hon ouchMuummon Ho mumm uOHUMHnom on omooo .mHDo .oHHm> Hon vmuHmuo>HuD "oouoom n-.mmm mnm.HaN qnw.nw~ wa.an mmo.ow mnH.Hm mmm.Ho¢ ooh.HquIaoz m.oHHm> onu mo wooouHmum uaomoum use GHMHHO mo oouHm .n.~ oHnoH. ~ 24 of Colombia between 1951 and 1964 was estimated at 272,000 people, while the number of immigrants was 400,000 people during the same period.11 Migration Within Valle It was said above that Valle's municipios had in 1964 over half of their population born in other municipios of Valle and of other departments, indicating Valle's high population mobility. L6pez Toro found the following charac- teristics with respect to migration within Valle: (1) rural- urban migration within a municipio was not very important in the period 1951 to 1964, while there are high potential rural-urban migration between municipios, especially in municipios with less than 45,000 inhabitants in 1964;12 (2) there is little evidence of stepwise migration in Valle;13 (3) there was urban-rural migration among natives of Valle from 1951 to 1964;14 (4) Valle's municipios with less than 45,000 inhabitants had considerable immigration and outmigration during the period 1951 to 1964;15 (5) when the immigration 11Alvaro L6pez Toro, "Nota sobre los Fendmenos migra- torios del Valle del Cauca." Bogota (Mimeo). 12Ibid., p. 4. 13Ibid.. pp. 11-12. 14Ibid., p. 18. 151bid., p. 14. 25 rates to the Valle's municipios for the periods 1952 to 1958 and 1958 to 1963 are compared, Cali's migration rate was stable while the other municipios, classified in three groups according to their population16 had increasing immigration rates;17 (6) outmigration rates from Valle's municipios increases as the municipios' population decreases and, thus, Cali had the lowest outmigration rate; Cali did not have evidence of outmigration.l8 Migration to Cali 0f the persons who migrate to Valle, a high percentage, 72 percent, settle in Cali. In 1964, 37.5 percent of the total population of Cali came from departments other than Valle, the natives of Valle not born in Cali constituted 18.6 percent, foreigners 1.2 percent. Thus 57.3 percent of the population of Cali was born elsewhere. Alvaro L6pez Toro showed that the other cabeceras municipales also have over half of their population born in other localities although the proportion 19 is not as high as that for Cali. Thus, the population in all Valle is very mobile. 16Group I: 5 munici ios with more than 65,000 inhabitants Group II; 14 munici ios with more than 20,000 and less than 65,000 inhabitants, group III: 19 munici ios with more than 10,000 and less than 20,000 inhabitants. e munici ios of San Pedro, Ulloa and Vijes with less than 10,000 Inhabitants were excluded. 171b1d.. pp. 12-13. 181bid., p. 14. 19Ibid., p. 6. y. 1-4 'l ‘- 26 The different kinds of migratory currents among non- native born population of Cali are shown in Table 2.6. Migrants to the rural area of the municipio of Cali are few: 1.7 percent come from urban areas and 0.8 percent from other rural areas, while migrants to urban Cali are 97.5 percent of the total. Urban-urban migrants have the highest pro- portion among all migrants, 77.6 percent and rural-urban migrants are 19.9 percent. The proportion of women is higher in both urban-urban and rural-urban migrants. The annual increase of the migrant population of working 20 This age arriving in Cali was estimated at about 8 percent. rate is more than three times the estimated rate for all age groups arriving in Valle which, as was said before, is 2.6 percent. Upon comparing the classification by age between the persons born in Cali and those born elsewhere, Table 2.7 shows that 68.8 percent of the native born are under age 15, while only 17.7 percent of the non-natives belong to this age group. On the contrary, 76.8 percent of the non-native born population of Cali is between 15 and 64, which signifies that a large section of those who arrive in Cali are of working age, while only 30.2 percent of the native Calefios fall into this category. 20Planeacién Municipal de Cali, "Estudio Preliminar de la Poblacién de Cali," (Mimeo) 1960, and Universidad del Valle, Estudios Demograficos de Cali, 1969, pp. 16 and 75. 27 .NooH .«Huo .HH .Ho> .oHHo> Hon oquEMuum on Ho whom :wHomHnom on cause .mHDu .oHHa> Hon umvauo>Ho= "mouoom OHm.mom mam.w¢H nmm.ooH «No.m c~o.q wam.¢ wwc.omm mmq.¢mH mam.HoH Hmuoa ¢n¢.mn noo.Hq new.mm nmm.~ o¢~.H qu.H amm.~m Hom.oq w~m.~m AmooN Housmv moomHm Houuo omo.oa~ oom.omH Ono.MMH nw~.e www.~ mmq.m m<5.mw~ wMH.¢nH HHo.m~H HoooN smounv mmmwmmmmm Hmuoe mHmawa mHmz Hmuoe wHaamm onz Hmuoa «Hmama MHmz Hmuoa , mooMHm Hoouo Anson ompuav mmmmmmmmm N 0.5m meom rlll unomwum camauo we woman .ssaa .xom an “Hue mo oOHumHsaom anon o>Humzlooz ouu mo ouoouHmoM uoomoum can anHuo mo ooaHm .o.N oHan 28 Table 2.7. Distribution of the Population of Cali by Place of Origin, Sex and Age Groups, 1969. Age Groups Place of Origin Born in Cali Born Outside Cali Male Female Total Male IFemale TTotal ----------------- Percent------------------ 0-15 Years Old 35.0 33.8 68.8 8.7 9.0 17.7 15-64 Years Old 14.2 16.0 30.2 33.0 43.8 76.8 65 Years and Over 0.4 0.6 1.0 2.3 3.2 5.5 Source: Universidad del Valle, CUIP, "Estudios de Poblacién," Cali, 1970 (Mimeo). The population data presented show a high native popu- lation growth, considerable migration to the department of Valle, different kinds of migratory currents including a strong current of rural-urban migration, and that within valle, Cali is the destination of the majority of the migrants. The high birth rate combined with accelerated migration, determine the extremely high population growth rate in the department and in Cali. The major part of those who migrate to Valle and to Cali are between 15 and 64 years of age, and more of them are female. (See Table 2.7). The proportion of the nonecon- omically active population is high and rising. Valle is a highly urbanized department (70.4 percent), while the rural population is decreasing in both absolute and relative terms. .-~ I 0‘. a... L. o 'a "a. ‘4 M.‘ ‘- -'F 1“ [ID 5|,- ; A" a a I 29 Production in Valle In the period from 1958 to 1969, the gross product, in 1959 prices, grew in Valle at an annual rate of 5.6 percent, while the economically active population grew only 2.2 percent. This rate is even lower than that of total population growth in Valle, which was 3.5 percent annually. Valle produces a high percentage in terms of aggregate value and in terms of gross production, of both Colombian industrial and agricultural goods. The aggregate value of the manufacturing sector of Valle has been calculated by using the figures given by the DANE, and adjusting them.by coefficients calculated on the basis of information from Sociedades Anonimas , since some intermediate consumption such as telephones, insurance, publicity, etc. were not included in the calculations made by the DANE. The figures for the industrial sector are for the period from 1956 to 1968, since the data for more recent years are not yet available. The aggregate real value of the output of the industrial sector of Valle in this period, according to the aggregate value at current prices adjusted by the price index for wholesalers calculated by the Banco de la Repfiblica, grew at an annual rate of 7.5 percent. The real value of production shows a growth rate per year of 7.3 percent. These figures reflect the accelerated development of the industrial sector, which is the same rate proposed as a target for gross internal product by the Development Plan for the period from 1970 to 1973 to which is the 30 required rate to absorb the new labor force.21 However, unemployment has grown in Valle, as we shall demonstrate in Chapter VI. The fact that the growth rate of aggregate value is higher than that of production, indicates the large volume of production of intermediate goods and capital in the department. The number of workers employed by the manufacturing sector increased at a rate of 3.3 percent a year, which is below the rates for industrial production.22 It was not possible to obtain information on the agri- cultural sector for exactly the same period, since information on prices for agricultural goods from 1956 to 1959 are not available. Therefore, we will consider the period 1960 to 1970. Agricultural production grew during this period at an annual rate of 6.5 percent. The value of industrial production is greater than the value of agricultural produc- tion and grows at a higher annual rate. In Table 2.8 the data on the total value of agricultural and manufacturing production in Valle, in 1958 prices, are shown for the 1960 to 1970 period. 21Ministerio de Agriculture, "E1 sector agropecuario frente a1 problema del desempleo," Bogota, 1971, (Mimeo), p. 1 221m Chapters V and VI we examine the fact that the ‘manufacturing sector employs a low percentage of the labor force in Valle. 31 Table 2.8. Total Value of Manufactured and Agrarian Production of Valle Measured in 1958 Constant Prices for the 1960 to 1970 Period. Year Manufacture Agricultural Production1 Production2 Current Colombian Pesos (in thousands) 1960 2,015,163 461,521 1961 2,229,039 525,407 1962 2,583,170 565,430 1963 2,646,803 492,601 1964 2,797,146 602,695 1965 2,798,130 563,293 1966 3,145,903 607,609 1967 3,115,741 706,440 1968 3,542,136 737,148 1969 3,381,879 838,083 1970 --- 851,777 1 The index price of wholesales calculated by Banco de la Repfiblica was used as deflactor. 2The implicit price index of the Colombian national account was used as deflactor. Source: CVC, "Plan de Desarrollo Agropecuario del Valle del Cauca," Unpublished preliminary draft, 1974, Banco de la Repfiblica, and Planeacién Departamental del Valle del Cauca, Unpublished data. 32 Since the population of the rural sector has decreased at the rate of 0.6 percent a year, and production has increased at an annual rate of 6.5 percent, agricultural production per capita has risen and at a faster rate of industrial production per capita. Undoubtedly, this has been due to the increasing use of better technology and mechaniza- tion. However, production per occupied person is low when compared with the industrial sector. In 1964 the value of manufactured goods was 4.6 times the value of agricultural goods, but occupied only one-third the manpower.23 The manufacturing and agricultural sectors of Valle are a dual economy. Dualism.has been defined as a dispersion of firm and farm sizes, output levels and factor proportions.24 There is a highly mechanized modern sector where production per worker is high, and a traditional sector with low pro- ductivity. A brief description of the characteristics of the four subsectors follow.25 The modern agricultural subsector has developed on Valle's flat land with high fertility. The main crops are 23This comparison is made for the year 1964 due to the availability of the census data. The values of industrial and agricultural production are shown in Table 2.8. The population employed in the industrial sector was 50,045, and in the rural sector it was 147,036. 24R. A. Berry, "Special Problems of Policy Formation in a Dualistic Agriculture: Colombia," Department of Economics, University of Western Ontario, London, Ontario, Canada, December 1972, Mimeo, p. 5. 25The characteristics of modern and traditional agri- culture are described for the case of Valle in CVC, "Plan 33 sugar cane, cotton, soybeans, grain,sorghum, corn and rice. Nontraditional inputs are used like chemicals to control diseases, insects, and weeds. Cultivation is by tractors, with little use of hand cultivation, while harvesting is less mechanized. The scale of operation is large in relation to traditional crap culture. In the traditional agricultural subsectors most of the farms are located in the mountainous areas although there are many minifundia areas in the Valle's flat land (See Appendix B.5 and B.6). The main crops are _2223, beans, plantains and cane for panela. Farms are not specialized on production and several crops are mixed on a given piece of land. Inputs are mainly traditional and cul- tivation by hand is the rule. The education and training level of traditional farmers is low. In the modern manufacturing industry the firms are similar to typical firms in the same industry in more developed countries. Some companies in Valle are subsidiaries of U.S. companies with highly mechanized machinery and are managed by foreigners. The firms in the modern manufacturing de Desarrollo Agropecuario del Valle del Cauca," Cali, unpublished preliminary draft, 1974 and for Colombia in L. Jay Atkinson, A ricultural Productivity in Colombia, Foreign Agricultural Economic Report No. 66. washington: U.S. Government Printing Office, 1970, pp. 85-9. The characteristics of modern and traditional manufacturing industry are analyzed in Richard R. Nelson, T. Paul Schultz and Robert L. Slighton, Structural Change ina_Develgping Economy: Colombia's Problems and Prospects. Princeton: ‘Princeton UniversityPress, 1971, pp. 77-156. 34 industry are "somewhat smaller, with somewhat lower value added per worker, capital per worker, and labor quality, than firms in the same industry in a developed country but they were using roughly the same kind of technology."26 The traditional manufacturing industry is composed of "small craft firms using significantly less modern equipment, employing quite different (and less related to formal educa- tion) labor skills, and creating far lower value added per worker. To a considerable degree these two groups produced different products."27 Finally, it is important to point out that Valle has one of the highest economic growth rates among departments of Colombia, and this fact has created an imbalance development process in Colombia. Income Distribution In a market economy, income is formed in the productive process as conctractual income in the form of salaries, interest, rent, and as residual income in the form of benefits. It is possible to analyze income distributions both according to recipients (personal) and function. Personal distribution data on incomes show the total amount received by the persons involved in a given period of time. Despite the fact that this information is valuable, 26 27 Nelson, Schultz and Slighton, 92. gi£., p. 115. Idem. 35 it does not elucidate the source of the income: personal work, property, entrepreneurial activity. By contrast, functional distribution pinpoints the sources of income. In the following section, we will consider the personal distribution of income in Valle as a whole, in the rural and urban zones of Valle, and in the City of Cali. In 1965, 28 Of 1,284 families were interviewed in the department. these, 864 lived in the urban zone and 420 in rural areas. The income distributions according to these interviews are shown in Table 2.9. It may be observed that as income rises, the percentage of families decreases in both the rural and urban sectors. It may also be observed that the rural sector has a much higher percentage of low incomes. The data gathered from the sample taken by a study undertaken by the CVC in July, 1972, confirm the fact that income is highly concentrated in a few families in the rural area. Table 2.10 presents the data showing the distribution of income comparing the flat valley with the mountainous zone of rural Valle. A higher percentage of the high income group is found among those with property in the flat lands. The information pertaining to the income distribution of residents of Cali comes from.a survey conducted by Planeacién Municipal de Cali in August 1969.29 Table 2.11 28DANE, "Encuesta de la Investigacién Nacional de la Morbilidad, 1965," Bogota, DANE. 29Plan General de Desarrollo - Encuesta Urbana de empleos e ingresos del area urbana de C311. Cali, 1969. 36 Table 2.9. Distribution of Family Income in Valle, 1965. Average Family Income Proportion of Families Living In Urban Rural Total Current Colombian Pesos --------- Percent ------ Less than 3,600 17.5 44.8 27.9 3,601 to 6,000 19.0 29.0 22.8 6,001 to 12,000 24.0 10.7 19.0 12,001 to 30,000 12.1 2.6 8.5 More than 30,000 6.4 1.4 4.5 Undetermined 21.0 11.5 17.3 Source: DANE, "Encuesta de la Investigacion Nacional de Morbilidad, 1965," Bogota, DANE. Table 2.10. Distribution of Family Income in the Rural Sector of Valle, According to Sample, July 1972. Average Family Income Proportion of Families With Respect to the Total Living In Flat Region Mountainous Region Current Colombian Pesos ----------- Percent --------- Less than 10,000 26.3 34.7 10,000 to less than 20,000 30.5 29.7 20,000 to less than 30,000 11.0 13.8 30,000 to less than 50,000 9.3 9.4 50,000 to less than 100,000 9.3 12.8 100,000 and over 13.6 2.2 Source: CVC, "Estudio Socio-Econémico del Sector Rural de Valle," Unpublished preliminary draft, 1973. 37 Table 2.11. Distribution of Family Income in Cali, August 1969. Average Family Income Proportion Current Colombian Pesos Percent Less than 1,000 21.5 1,000 to 1,500 31.0 1,501 to 2,000 20.0 2,001 to 3,000 13.8 3,001 to 6,000 4.1 6,001 to 10,000 5.8 More than 10,000 3.8 Source: Planeacion Municipal de Cali, Plan General de Desarrollo, Encuesta Urbana de Empleos e Ingresosfdel Area Urbana de Cali. Gali, I969. illustrates the unequal distribution of income in the city, where it may be noticed that the concentration of income is higher than in the department as a whole. The figures cited clearly show the uneven distribution of income in Cali, and in the department of Valle, and emphasizes the fact that the situation is most acute in the rural zone. We have presented in this chapter selected information relevant to the thesis's subject with respect to the population and the economic activity of Valle and Cali. We showed the high growth rates of the population living in urban areas of Valle, 6.3 percent, and the capital city, Cali, with a rate of 6.4 percent, but the non-native working age population of 38 30 These high rates are, Cali growing at a 8.0 percent rate. in part, the result of strong rural-urban migration. Rural population decreased with a rate of 0.6 percent. Agricultural and industrial growth of Valle is quite satisfactory, 7.5 percent for industrial production and 6.5 percent for agri- cultural production which equal the rates proposed for the country by the national development plan in 1970. The Valle industry produces large volume of intermediate and capital goods. When the industrial growth rate is compared with its rate of manpower absorption, 3.3 percent, it is clear that production increases due mainly to improvements in productivity rather than through utilization of more manpower. Dualism is the rule in the Valle's agricultural and manufacturing industry. Income is highly concentrated in Cali and especially in the rural areas of Valle where a highly mechanized modern sector exists together with a traditional sector with low productivity. 30This figure was given by Planeacién Municipal as we said before. In Chapter IV we estimatedfan increase of 7.6 percent for males non-natives and 7.2 percent for females non-natives. CHAPTER III DECISION MODEL FOR RURAL-URBAN MIGRATION The tremendous increase in Valle's urban population while its rural population decreased, as it was shown in Chapter II, is largely explained by the phenomenon of rural- urban migration. However, we showed in Chapter II that other migratory currents have also taken place in Valle. Over half of the migrants (54.8 percent) went from urban to urban areas while only 17.5 percent went from rural to urban areas. The relative importance of the kinds of migratory currents were computed on the basis of defining urban areas as the cabeceras municipales and as rural areas the other zones of a municipio. These definitions were approximations to the concepts of urban and rural areas in order to be able to identify migra- tory currents using census data. This definition is not conceptually clear since other factors like economic depen- dency on agriculture and the social organization of the community should be taken into account. However, there are not clear cut criteria to decide when an economy is not depending upon agriculture and the community is socially organized in such a way that is more urban than rural. There are difficulties in classifying some productive activities as ‘belonging to the agricultural sector or to the industrial or 39 ‘4 Va 40 service sectors. Therefore, for practical purposes, an arbitrary number of people in an aggregate is used to classify a group as urban or rural. The author believes the figure 1,500 people used by the Colombian census is low; thus has suggested 2,000 and 10,000.1 If a cabecera municipal that has 10,000 is defined as a rural area, urban-urban migration will be less while rural- urban migration will increase. High potential rural-urban migration exists in the Valle's small and medium size municipios as L6pez Toro pointed out (see Chapter II) and even higher potential rural-urban migration exists in the municipios of other departments of Colombia from where the majority of the migrants to Valle come from (See Table 3.4). There were 129,000 urban-rural migrants (17.5 percent of the total migrants of Valle) involving a high proportion of Valle's natives, between 1951 to 1964. This fact together with (a) the high mobility of the Valle's population, and (b) the higher immigration rates observed to smaller and more rural municipios during the period 1958 to 1963 when compared to the period 1951 to 1958 show the important role of the rural violence of the late 19403 and the first part of the 1950-1959 decade. Rural-urban migration, underestimated by the definition of rural and urban areas, should have been high during the 1WilliamL. Flinn, "Rural to Urban Migration: A Colombian Case." Madison: Land Tenure Center, University of Wisconsin, July 1966, p. 18. 41 period of violence but it was not permanent migration for some migrants as discussed above. Mbre recently, with violence disappearing in the countryside, rural-urban migration con- tinues to play an important part in the geographic distribu- tion of the Valle's population determining high increase in its urban population and decrease in the rural population. There are still potential areas for high migration in Valle and other departments, as discussed above. Thus, it is important to determine the factors for this kind of migration to be able to develop policies aimed to help combat unemploy- ment in the urban areas. Factors Affecting_Rural-Urban IMigration There is an extensive literature discussing factors of rural-urban migration for different localities around the world.2 The authors agree that the predominant cause of 2See for instance William L. Flinn, 92. cit.; Harris and Todaro, and Todaro in different articles ciEEd frequently in this study (see bibliography); Derek Byerlee and Carl R. Eicher, "Rural Employment, Migration and Economic Development: Theoretical Issues and Empirical Evidence from Africa," paper presented at a Conference of the International Economic Association on the Place of Agriculture in the Development of Developing Countries, Bad Godesburg, Germany, 1972; Juan F. Gaviria G., Francisco J. Gomez P., and Hugo L6pez C., Centro de Investigaciones Economicas (CIE), Universidad de Antioquia, Contribucion a1 Estudio del Desempleo en Colombia. Bogota: DANE, 1971; Sunday M. Essang and Adewale F. Mabawonku, "Determinants and Impact of Rural-Urban Migration: A Case Study of Selected Communities in Western Nigeria," East Lansing: African Rural Employment Research Network, Depart- ment of Agricultural Economics, Michigan State University, 1974; Derek Byerlee, "Research on Migration in Africa: Past, Present and Future." East Lansing: African Rural Employment 42 rural-urban migration has been economic, in terms of income or wage differentials in favor of urban places or better economic opportunities as perceived by the migrants, rural poverty, and low agriculture productivity. Besides some apparent noneconomic factors are really economic ones like education since returns to education are generally very low in rural areas compared with urban areas and the distance between the sending and receiving areas can be explained in terms of economic factors such as the additional cost of 3 Other factors frequently acquiring information and of moving. mentioned are tendency to migrate to places where the migrants have relatives, and freedom offered by town where the control of older generation is avoided. For the Colombian case William L. Flinn4 interviewed 196 migrant families to the working barrio E1 Carmen of Bogota. The important factors influencing migration were: (1) monetary reasons (wage differentials) as the most fre- quently mentioned by 42.5 percent, (2) violence in rural areas, 13.2 percent (3) dependents, youngsters who migrate with their families 12.3 percent, (4) "other reasons," 10.4 percent. Research Network, Department of Agricultural Economics, Michigan State University, 1972; Josef Gugler, "The Impact of Labour Migration on Society and Economy in Sub-Saharan Africa: Empirical Findings and Theoretical Considerations," African Social Research, 6, December 1968, pp. 463-86; Nelson, Schultz andSlighton, 9p. cit., pp. 45-76. Albert Berry, Development of the Agricultural Sector in Colombia, forthcoming. 3Derek Byerlee, pp. git., p. 11. 4lbii. pp. 6-14. 43 These factors were followed by better living conditions, (5.7 percent) health (3.8 percent), education (3.8 percent), compulsory military service (3.8 percent). No reason or no response from 5.7 percent. As argued before some of the nonmonetary factors are of economic significance; these include: dependency, better living conditions, health, and education among others. Gaviria, G6mez and L6pez, researchers at the Universidad d3 Antioquia argue that the answers given by migrants are only the result of the predominant political ideology and do not reflect what they believe is the real cause of migration, capitalistic agricultural exploitation which takes the land away from peasants.5 Nelson, Schultz and Slighton developed a model of interregional migration in Colombia. Their basic hypothesis is that "people migrate because they have reason to believe that, by migration, they can improve their conditions and that of their family."6 Factors that affect the decision to migrate from rural to urban areas are: (1) interregional inequalities as measured by wage differentials, (2) distance from the rural locality to the receiving city, (3) population pressure in the rural areas, (4) relative educational oppor- tunities: the level of education of rural people and the educational facilities offered by urban areas, and (5) violence. 51bid., p. 59. 61bid., p. 56. 44 They estimated local rates of migration by a linear function of six independent variables: local wage rates in agriculture, the estimated local rate of population increase in the absence of migration, distance to the nearest large town, the level of political violence, and two measures of schooling: school enrollment rate for children between 5 and 9 years and the school enrollment rate for children between 10 and 14 years. The main findings were (1) if rural wages were doubled (and city wages held constant) migration might be reducedlnrabout 70 percent. No effect of rural wages on male migration rates is evident after age 26; (2) 20 percent reduc- tion in surviving fertility in rural areas is associated with a 40 percent reduction in the outmigration of men and a 50 percent reduction in the outmigration of women; (3) the vari- able distance to city helps to account for differences in udgration rates among the cabeceras but does not appear to affect the propensity of the rural population to migrate; (4) violence in rural areas produce outmigration: one death by violence is associated with a net outmigration of approxi- mately 40 persons for the immediate rural area, and (5) rural school enrollment for the 10-14 age is associated with out- migration but, contrary to what was expected, school enrollment from the 5—9 age group was not associated with outmigrat ion . Albert Berry argues that "in the event that commercial farming continues to gain ground relative to small-scale farming, and neither exports nor other rural sources of 45 incomes for small farmers take the pressure off, then the only way out for the small-scale farmer is migration."7 For the Valle case the two most frequent reasons for ndgrating given by prospective migrants in a survey carried out by CVC in the flat and mountainous zone of Valle were 8 Answers (1) marriage and (2) better economic opportunities. of prospective migrants interviewed to test Todaro's model, for the purpose of this study cited better economic oppor- tunities and higher wages, or monetary factors as the more important factors in migration. Thus, most studies have revealed income differences to be the main factor in explaining rural—urban migration though there are other monetary and nonmonetary factors as well which are of socio-economic consequence. Decision Model for Rural-Urban Migtation with Urban Unemployment An economic model which explains rural-urban migration in terms of income differentials when there are high unemploy- ment rates in urban areas, as is the case in urban Valle as a whole and Cali,9 assuming perfect knowledge on the part of the prospective migrants was prOposed by J. R. Harris and 7Development of the Agricultural Sector in Colombia, Chapter I, fOrEhcoming. 8"Estudio Socio-Economico de la Poblacién Rural del Valle," pp. cit. 9See Chapter VI. 46 Michael Todaro.10 It is a model for well-informed rural laborers and poor owner-operators who plan to go to the cities to look for a job. The model is not intended to cover the case of wealthy migrants bringing capital to the urban areas in search of the best opportunity for investment. The fact is that there are reasons to believe that a high proportion of the Valle migrants are poor uneducated people11 in need of a subsistence job and thus the Harris and Todaro model is relevant. We turn now to a brief description of the Harris and Todaro model. The Harris and Todaro model looks at migration as an investment decision where potential migrants choose that alternative which maximizes the present value of their expected real income streams and since the receiving urban places have unemployment the model takes account of this fact and uses the concept of expected income. Harris and Todaro calculate the salary expected in the urban sector, by multi- plying the real salary by the probability of finding a job within a certain period of time. It is assumed that when there are more prospective workers than jobs, a periodic system of chance selection exists. The probability of finding a job is estimated by the relation between the number of workers employed in the city and the number of persons of 10"Urban Employment in East Africa: An Economic Analysis of Policy Alternatives," East African Economic Review, Vol. 4, No. 2 (1968). PP. 17-36. 11 See Chapters 11 and IV. 47 12 They said that if working age resident in the urban area. the rates of unemployment rise, as is the case of Valle and Cali, the expected wage diminishes, migration to the city falls off until reaching an equilibrium level of unemployment. If the level of unemployment falls then the expected wage in the urban area goes up, and if rural wages and living condi~ tions do not change, new migratory currents begin, and the cycle repeats itself until equilibrium is reached again, and migration ceases once more. The fundamental assumption of the model is that rural- urban migration to cities with unemployment occurs when the expected wage in the urban area is a certain amount greater, or simply more than the rural wage. Harris and Todaro explain the model thus: "The crucial assumption to be made in our model is that rural-urban migration will continue so long as the expected urban real income at the margin exceeds real agricultural product--i.e., rural migignts behave as maximizers of expected utility. In symbols, as long as w: > Wr migration will occur. Migration ceases when W3 - Wr = 0. As Harris and Todaro point out, this is arbitrary, and one may suppose that migration will cease when W3 - Wr = a where a may have any 12An analysis of employment probabilities is found in lfichael P. Todaro, "Model of Labor Migration and Urban Unemployment In Less Developed Countries." 92. cit. 13"Migration, Unemployment and Development: A Two- Sector Analysis," 92. cit., p. 127. 48 given value. The value of a depends on the possible dif- ferences in inconveniences and ways of life in city and country, the expenses involved in moving, or preferences for one place or another. Similar reasoning can explain a large part of urban-urban migration when this involves people moving from small or medium-sized urban areas to the metropolis with a greater degree of economic development. In order to test the Harris and Todaro model in Valle, 500 families from the municipio of Palmira, corregimientos d 14 Rozo, Acequia and Latorre were interviewe This locality was chosen in order to take advantage of the fact that professors from the University of Florida, with financial aid from.the World Bank and collaboration from the CVC, were carrying out research on the benefits of rural electri- fication in the area. The questionnaire is included in Appendix D. The head of the family of each farm was interviewed. This fact might cause some distortion in the information, since the migrating persons themselves were not interviewed. Instead of asking the causes for migration already under- taken, the questions were framed for those still on the farm 14Empirical verification of the Todaro model using regression analysis and rural-urban income differentials as the explanatory variable of rural-urban migration have been difficult due to lack of adequate data and the theore- tical problems in measuring the relevant rural and urban income and comparing the two incomes as explained by Byerlee, pp, git., and Essang and Mabawonku, gp. git. 49 in terms of whether they were thinking of migrating, and if so where to and for what reasons. Finding recent arrivals in Cali was a task impossible to undertake at the moment when this study was prepared. The persons interviewed were asked if they were planning to change their residence; if they had information about the wage levels in Cali; how those levels compared with their present income; what information they had about unemployment and the magnitude of unemployment in Cali; and what reasons they would give for migrating. An analysis of the responses to these interviews allows us to venture some judgements about the model under consideration, which follow. Table 3.1 shows that 54 of the total of 447 persons were thinking of moving to the city, which constitutes 12 percent. Of these 91.5 percent had some idea of the existence of unemployment in cities, especially Cali. This percentage is about the same than those for persons who plan rural-rural migration or who are not planning to migrate. Table 3.1 shows also what the prospective migrants know about the relative magnitude of unemployment in the cities. The majority (73.4 percent) said unemployment in the cities is large or little (see Table 3.1) but 16.7 percent said there is not unemployment in the cities, a statement that this study proves is incorrect. ‘With respect to wages Prospective rural-urban migrants are less informed than on unemployment. Only 31 percent had some idea of the wages 50 3.329 ”Soon: sasHou on». a.“ umuoHooH mums “manna no: 36 as: moomuom wH.HH on om.~o Hmm om.OH on mm. PS) increases with rising unemployment rates, provided urban and rural wages hold constant. Another reason explaining the difference between Pa and P: is the cost of moving from rural to urban areas, which change with age, family status, seniority or lack thereof. The wage differential adjusted by the probability of finding jobs in a given period of time is not the only factor for the decision to immigrate. Other factors discussed above are: (l) greater'educationand.health facilities in the cities, (2) relatives living in urban centers, (3) freedom offered by towns to young people, (4) compulsory military service, and (5) recreation facilities. The model takes account of these "other factors" in a variable, say an, which can take any value different from zero. Nonmonetary factors in favor of rural areas are taken into account by a When r' MVP + ar > Pa + “u urban-rural migration will take place. When MVPx + ar < P: labor will move out of agriculture until MVP +0. =Pe. WhenPe MVPx when rural migrants are unemployed for a considerable period of time.22 Support from relatives can overcome this factor. Also, rural laborers who lose their jobs will go to the cities even if they realize the low probability of finding a job suitable to their lack of skills. The reasons for this is that, on the one hand, they can get some free or very cheap services in the city such as medical care and can get some income as streetvendors, yardmen, washmen, etc., and as beggars or by means of "under- ground" activities while, on the other hand, they are unable to provide even food in the country. The value of at is expected to be positive for people with preferences for rural life and positive values of “u are associated with greater education and health facilities in the cities, relatives living in urban centers,23 freedom 22Seventy-three percent of the unskilled labor find a job in a period less than twelve months. OIT .‘ 22. Cit. , p. 404. 23Not necessarily providing economic support to the migrants. 62 offered by town to young people, and recreation facilities. We already discussed evidence found in Colombia for some of these factors, such as the importance of education facilities estimated by Nelson, Schultz and Slighton. The attraction of the city is a real phenomenon as shown by the laborers of the sugar cane plantations who prefer to live in Cali and Palmira even though they could stay in the plantations for less money in a better house than the ones they rent and.which have some of the conveniences found in the cities like electricity.24 Urban-rural migration was an important migratory current during the period 1951 to 1964, as has been pointed out several times before. This migratory currency is explained by (a) the number of people returning to their farms once the violence was over in their rural areas, (b) construction 25 and (c) the displacement of rural physical infrastructure, of extensive cattle raising from the flat lands by agriculture, to crop mainly sugar cane and cotton. Some owner-operators' assets become fixed assets and some farmers themselves have expected salvage values above their own rural labor income find it better to stay in farming since taking their labor and assets together they are fixed resources. 24Presently the Economic Studies Section of CVC is carrying out a study to determine, among other things, dif- ferences in cost and living conditions of sugar cane workers living in Cali and Palmira. 25Alvaro L6pez Toro, gp. cit., p. 13. 63 Young people and females migrate in greater proportion than old people and males, as was established in Chapter II and is again pointed out in Chapter IV. The main reason for the greater tendency of younger rural people to migrate is 26 Older peasants with they are not yet fixed in farming. large families and with some investment in the rural areas and who have lower expected salvage value outside agriculture are less likely to migrate. Females are common among migrants because they have few opportunities in rural areas where their MVPs are lower than their expected salvage values in spite of the fact that their off-farm wages are also quite low. There are some farmers who overestimated their wages in urban jobs and thus decided to migrate. After living for some time in the cities they realize their mistake but find themselves trapped between lines e and f as urban unskilled labor and, thus, cannot get back to their rural areas. Others who take the apparently right decision at the time of migra- tion but are later replaced by labor-saving capital27 cannot go back to the rural areas since they are fixed in the urban area. This is also why many unemployed native urban unskilled laborers or those with low paid jobs cannot move to other urban or rural places in search for jobs. These people have to stay in urban slums barely surviving and engaging in some cases in illegal activities. 26This is consistent with the findings of Nelson, Schultz and Slighton mentioned above. 27See Chapter V. 64 The risk taking attitude of rural laborers has also to be considered in discussing migration. Many poor farmers stay in the rural areas longer because they have less ability to handle uncertain situations. Some prefer to purchase flexi- bility28 and stay in the rural areas at the cost of under- employment and low wage rates while they obtain enough information to decide whether or not to migrate. On the contrary, risk-taking farmers will migrate to the cities and will buy flexibility for him and his family at the cost of the foregoing income and living expenses while looking for jobs in the city. Errors committed during this process may be impossible to correct later, as explained above.29 The investment/disinvestment asset theory as developed here and the concept of flexibility helps us to understand (1) flow of people to areas with more economic growth, (2) migration under imperfect knowledge, (3) high urban unemploy- ment rates, (4) the high proportion of young people and females among migrants, and (5) people living in both rural or urban areas under very difficult economic conditions without being able to move to other places looking for better opportunities. 28"Flexibility is an allocation of resources Egg_which enables an entrepreneur to reallocate his resources more effec- tively later on the basis of improved information, i.e., so as to raise his expected future income in a noncertain situation" as defined by Glenn L. Johnson, "Allocative Efficiency of Agricultural Prices as Affected by Changes in the General Level of Employment," pp. 213., pp. 43-44. 29We also have to reorganize the effect of probabilities. One can make the "right" decision and be "wrong" because the low probability occurred. CHAPTER IV THE URBAN LABOR FORCE AND SUPPLY As a factor of production, the labor force is heter- ogeneous, since the workers and employees have different physical and intellectual characteristics, different tastes and attitudes, and above all, different levels of schooling and preparation. As a result, there are many different types of labor supplied to the market, as many different types as there are homogeneous groups of workers and employees. Here we shall consider two types of labor, skilled and un- skilled, classified according to their special knowledge, or lack of it. Although aggregation of the labor supply in these two groups overlooks other heterogeneous factors present in the labor force, and we recognize the limitations imposed by this process, there are valuable conclusions to be drawn using this level of analysis which justify its use. The labor supply function is defined as the relation between the minimum wages, and the number of hours per time unit that persons between 15 and 64 years of age are willing to work, full—time students and housewives being excluded. The variable "amount of work" may be measured as the number of hours per day or per week that a given person or group of 65 66 persons is willing to work, or as the number of persons who offer their work daily or weekly in a certain time period. Skilled Labor Supply Skilled labor supply on a short-term basis shows a positive curve in the first section, showing that when salaries are low, the employee is willing to work for longer hours, in order to raise his income. In other words, the effect of income predominates over the effect of the substitution of leisure for labor time. Upon reaching a certain income level, which the employee considers high enough, the effect of leisure over income begins to take effect, and the labor supply curve goes backward.1 This second section implies high income; in Cali and Valle the number of persons with sufficiently high incomes for this 2 However, effect to be noticeable is extremely low. according to observations, some relatively poor people upon reaching a higher income level, but far from.being wealthy or rich, like to take things easy and work equal or less time than before. Also there are people getting only their food to survive, like some fishermen, and do not bother to look for more work and additional income. 1Milton Friedman, Price Theory, Chicago: University of Chicago Press, 1962, p. 203. 2See Chapter II for information on income distribution. 67 The supply of skilled labor reflects the investment made in human capital through education and training. This investment in human capital has two parts: The price of education and the opportunity cost3 given by the income which the student could earn while studying. This opportunity cost grows as the person ages, to a certain limit. If the cost of education is held constant (the average price for all types of training), those with few and only low paid opportunities and hence a low opportunity cost, are the youngest. They would require a relatively low wage, while this income would have to increase in order to attract persons with higher opportunity costs. This means that in the long run, the skilled labor supply has a positive slope curve. In order to have a skilled labor force, it is necessary that the discounted income stream over the lifetime of these workers be at least equal to the discounted income stream over the lifetime of the unskilled worker. This equality and the period of negative income of the skilled workers, determines that the wages of the latter must be higher than the wages of the unskilledworker.4 Consequently, 3This concept is defined as the highest income the person would have received if he had not chosen to study instead of work. 4It is assumed that there are no preferences as to type of work, and differences in risks are ignored. The cost of acquiring skill can be seen as negative income. 68 the greater the difference in wages between these two classes of workers, the greater the number of persons desiring training. Where do the persons who enter the training process come from? There are two sources: (1) Persons under 15 years of age, who are not part of the labor market as hypothesized here, and (2) the unskilled workers. The difference in wages offered to skilled workers and unskilled workers and their employment opportunities are not the only condition necessary for these persons to begin training. This depends on how much of the family income must be spent to pay for the education, and on the distribution of income. The price of education is cheaper for those under 15 and for unskilled workers who would be sacrificing the lowest wages. Youth under 15 from families with average and high incomes are the principal source of skilled labor. The rate at which unskilled workers pass over to the group of skilled workers depends on the difference in wages received by the two groups, the relative levels of employ- ment of the unskilled and skilled labor, the relation between the cost of training and family income, the availability of educational and training programs, and the distribution of income. The difference in wage levels tends to produce a high conversion rate but this tendency does not hold for the lowest income group because they cannot cover the cost even though the cost of training is low due to low opportun- ity costs; therefore, the conversion rate of unskilled labor 69 toward skilled labor is quite low. unemployed skilled labor, those with high school, technical or university education, amounts to one-third of the unemployed, but about 82 percent of the unemployed skilled laborers find a job in a period-~twelve months or 1ess--as compared to 73 percent of the unskilled labor.5 The training of per- sonnel already employed, with the cost absorbed by the company or society, seems to be a solution to the education problem and, indeed, the only solution for the lowest income group. In Valle, given the high cost of training and the concentration of income already described in Chapter II, the growth rate for skilled labor is very low and there is a low conversion rate from the ranks of unskilled labor. For example, the education coefficients, defined as the number enrolled with respect to the corresponding age group, are very low. Even at the primary level, the coefficient is only 0.625 at the secondary level it is 0.136.6 The Servicio Nacional de Aprendizaje, SENA, Valle section, trains employees of affiliated companies, but not at a rate significant enough to reduce notably the supply of unskilled labor and, moreover, part of the training is further specialization of skilled workers. 5Oficina Internacional del Trabajo, gp. cit., pp. 401-4. 6Mendoza and Olarte, "Diagnostico de la Educacién Media. Bases de Politics," Cali: Secretaria Departamental de Educaci6n del Valle, 1972. 70 Unskilled Labor Supply Unskilled labor supply is very different from skilled. The relevant wage for the unskilled laborer is the minimum legal wage or lower.7 The total number of persons between the ages of 15 and 64, able and willing to work for this wage, determine for any given moment one point and, thus, the supply of unskilled labor is not defined except as a vertical line.8 There is little considerations for investment in human capital. There is little relation or choice between more or less salary and more or less leisure. It is mainly an "all or nothing" proposition. If the minimum legal wage remains constant from one year to the next, the increase in the unskilled labor supply in a given place, in graphic terms forming a horizon- tal line running left to right, depends on four factors: (1) the rate of conversion of unskilled labor to skilled, (2) the rate of population growth of the native-born, (3) the participation rate in the labor market of those in the relevant age-group, and (4) migration to the place under consideration. 7The effect of unions are discussed later in this chapter. 8Several authors, for example, W. A. Lewis, Economic Development with Unlimited Supplies of Labor, London: Manchester school, 1954; Amartya,Kumar Sen, La Seleccién de Técnicas. Un aspecto de la Teoria del DesarrOllo Econhmico Planificado, Mexico: Fohdo de Cultura Economics, 1969f speak of'a perfectly elastic labor supply at the level of the minimum legal wage. 71 It should be recalled here that the conversion factor was analyzed with relation to the supply of skilled labor, remembering that increases in this factor will tend to decrease the unskilled labor supply. With respect to growth of the native-born population of Valle, the factors which determine this process are birth, mortality and out-migration rates. The mortality rate is higher in the rural area, and in the lowest income group in the city, and lower in the urban area and especially among the highest income group, due to access to better medical care, and generally better health of these groups. During the period between the census, 1951-1964, the native working age population of Valle increased at the rate of 3.4 percent annually. There is a notable difference between the sexes in regard to the rate of participation in the labor market. A great many women dedicate their time to being housewives. In underdeveloped areas, women do not enjoy the advantages of technological development of labor-saving devices for housework. Factors which influence a greater participation of women in the labor market are the necessity of contri- buting to family income, and the possibility of employing cheap domestic help. In 1964, the percentage of male residents in urban areas of Valle between 15 and 64 who were full-time students, retired, or incapacitated was 13.7 percent which means a rate of participation of 86.3 72 percent. The corresponding rates for women are 67.2 percent for housewives, 7.3 percent for students or disabled, and 25.5 percent for the rest of the females.9 Migration is an extremely important factor in the increase in the labor supply of this section of the country. In Chapter II, we described the type of migrant received by Valle and Cali. The calculation of the growth rate of this working-age group may be made on the basis of the data on the change in the non-native population; this will give us an approximation of the impact of migration on the labor market. In Chapter II, we found that the annual increase rate of increase in the non-native population for all of Valle is 2.6 percent. The rate of growth of the working age non-native population for urban areas is considerably higher, 7.1 percent for Valle and 7.4 percent for Cali. The greater rate for Cali is explained by greater migration from outside Valle to Cali and migra- tion from.ather cabeceras municipales of Valle to Cali. In Chapter III we developed a model to explain why uneducated people of the rural and small towns migrate to the cities and we showed that the main reason for migration is wage rate differences. To further examine the fact that the 9The participation rate of men in both urban and rural areas of 0.863, of women, 0.255. Upon comparison with the urban rates, one may conclude that women in rural areas have a low rate of participation in the labor market, and that men in rural areas participatate more than those in urban areas. 73 great majority of persons who migrate from rural areas and small towns are poor uneducated people who arrive in the cities to swell the ranks of the unskilled workers we examine the characteristics of these people with respect toeducation, age and sex.10 ‘ In the flat land, 115 families consisting of 839 persons with an average of 5.4 persons per family were interviewed. In the mountainous zone, 165 families con- sisting of 948 persons with a family average of 5.7 were surveyed. The families from the flat lands were more likely to migrate. Of a total of 245 migrants, 132, or 53.9 percent, were from the valley floor, and 133 from mountainous areas. However, not all migrated to cities. Thirteen point six percent of the migrants from flat lands went to other rural areas, and 23.9 percent of those from hilly sections changed residence within the rural area. Of the total population in the flat lands, 13.6 percent have migrated to their present urban area in the last ten years. Of these, 41.7 percent are between 20 and 10This information of the characteristics of the migrant population of Valle is from.the preliminary analysis of the questionnaire administered by the Department of Economic Studies of CVC, which is part of a socio-economic study of the rural sector of Valle under preparation. The sample was randomly selected to represent both flat and mountainous rural areas of Valle. The head of each house- hold was interviewed. Questions about personal character- istics and employment were asked for both members in the family living in the farm and those who have migrated in the last ten years. CVC, "Estudio Socio-Econdmico del Sector Rural del Valle," 92. git. 74 30 years of age, as is shown in Table 4.1, and 50.7 percent correspond to the age groups of 15-20, and 30-64. Thus 92.4 percent of the migrants are of working age. The greater part of the 15 to 20 year old migrants are women. The figures are similar for the mountainous zone. Of the 92 percent of working age, 48.7 percent are between 20 and 30 years of age. The reasons for higher rates of migration for rural young and females were analyzed in the previous chapter. (See Table 4.2.) With respect to formal education the great majority of the migrants have received little education. While 71.8 percent have received some primary education, most of these had not gone past the third grade. They are barely literate, according to Tables 4.3 and 4.4. The proportion of migrants with any secondary, technical or university training is very low for migrants from either type of rural zone. Of those from the flat land, 20.5 percent had finished more than primary, 7.6 percent had entered a university though very few finished, and 0.8 percent had completed a technical course, as may be observed in Table 4.3. The statistics on the migrants from the mountainous areas indicate even less education, since only 15 percent had advanced beyond primary level. Four point four percent of the migrants from the valley areas and 7.1 percent of those from.the mountains had received no schooling at all. (See Table 4.4.) 75 =.sHHs> use Hesse possum use ceasesoee-oeeom encasemsse emsssasse es en maeH se.ae>use use s acne some "season o.eos «no e.oeH ea o.oeH mm Hence 0 o o o o o Ho>o pom munch me o.- em e.mH ea e.m~ mH eHe enema ee-om A.He mm m.em am m.ee mm ens enema om-e~ a.e~ em ~.sm em m.nH m use enema e~-mH e.a OH e.m e m.eH e see enema we seen new; unmouom Honaoz unmouom Hooaoz unmouom Monauz Houoa ooaoz cox meoouu ow< coon: ou onumeHz .meea on sees messes eHHe> .mw< one sum up oHHm> Honor mo coca umHm sou mo SOHuoHoeom .H.q mHowH 76 Has ouaaecoomeeuom 333m: :.oHHm> Hop Honor Houoom oH noomHHooe on on man oH ho>uom u>u m Scum some ”mouoom e.oeH nae o.eos mm o.oeH ee Hence 0 o o o o o uo>o pom momma no ~.mH ma e.s n e.eH on see enema ee-em e.me mm e.He Nu o.mm mm eHe enema om-e~ H.om em e.em Hm e.H~ no see enema o~-mH o.m e e.e m a.e e was enema ma seen sees unmouom Hooaoz udoouom Hooaoz unmouom amnaoz Hmuoe ooaoz no! mnoouu ow¢ ou moHuousz .ow< new row zo mHHm> Honor mo oo0N moochuoooz moo mo GOHumHomom .meea on eeeH assess eases sense .~.e «Heme Table 4.3. Levels of Education of People From the Rural Flat Zone of Valle, Migrating to Urban Valle During 1964 to 1973. Education Men Women Total Number Percent Number Percent Number Percent Primary 34 58.6 54 73.0 88 66.7 “High School 15 25.9 12 16.2 27 20.5 College 10.3 4 5.4 10 7.6 Technical 1.7 -- -- 1 .8 Without Studies 3.5 4 5.4 6 4.4 Total 58 100.0 74 100.0 132 100.0 Source: Data from a CVC survey in 1973, to be published in "Estudio Socio-Econémico del Sector Rural del Valle." Table 4.4. Levels of Education of People from the Mountainous Zone of Rural Valle, Migrating to Urban Valle During 1964 to 1973. Education Men WOmen Total Number Percent Number Percent Number Percent Primary 48 80.0 40 75.5 88 77.9 High School 10 16.7 7 13.2 17 15.0 College -- -- -- -- -- -- Technical -- —- -- - -- -- Without Studies 2 3.3 6 11.3 8 7.1 Total 60 100.0 53 100.0 113 100.0 Source: Data from a CVC survey in 1973 to be published in "Estudio Socio-Econémico del Sector Rural de Valle." 78 The data show that there is a high rural-urban migra- tion by young, poorly educated people. One may safely conclude that they may be classified as unskilled workers upon arrival in the city where many join the ranks of the unemployed, or make their living working in the small crafts- men firms as the substantial increase in the number of firms with less than nine workers suggests11 (see Table 5.15). One hypothesis which, unfortunately, is impossible to test with the 1964 census data on migration is that out- migration from rural areas is mainly of poor uneducated people that go to intermediate towns while out-migration from these towns are the most educated and trained people. This phenon- enon would lower the human resource quality of intermediate towns . 1'2 Unions' Effect on the Supply of Labor Under Colombian labor legislation, a firm.must employ a minimum.of 25 laborers to be eligible for formation of a labor union. This rule causes a modification in the supply of unskilled labor to the firm. For the first 24 workers, the relevant salary is the minimum legal wage; however, when 25 workers are reached, a higher salary is established on 11Nelson, Schultz and Slighton argue that manufacturing firms with less than 9 workers are craftmen firms, 92. git., pp. 115-120. 12Hypothesis proposed by William.McGreevey, "Causas de la Migracién Interna en Colombia," Empleo y Desem leo en Colombia. Bogota: Universidad de los Andes, CEDE, 1958, quoted 5y Alvaro L6pez Toro, gp. git., p. 12. F4 79 the basis of a collective-bargaining agreement for a period of about one year. This salary is paid, in general, to both the unionized and nonunionized laborers of a given firm. If W1 is the minimum wage and if we suppose that the wage for unionized labor is the same for all firms (Wt), we have a stepped supply function of labor for a firm: W = W1 when X < 25 (X - number of workers) and W = Wu when X‘: 25. However, not all groups of workers with 25 or more members become unionized. In some cases the entrepreneurs prevent the creation of unions by offering salaries above the minimum wage along with other benefits. In practice this wage tends to be equal to Wu, but entrepreneurs have the advantage of facing fewer labor conflicts and strikes. However, the laborers can decide to unionize at any time. We will see later that conflicts, strikes and the expecta- tion of frequent pressure to raise salaries make labor more expensive non-monetarily as well as monetarily for firms employing over 25 persons. The Labor Force in the Urban Zone of Valle One may quantify the contribution of the migrant pop- ulation to the labor force in the urban area of Valle, discriminating by sex and place of birth, in the following manner, using the census year as the basis: XH = the migrant male population between 15 and 64 born in urban areas of Colombia, and living in the urban zone of Valle in 1964. 80 XML: the migrant female population between 15 and 64 born in urban areas of Colombia and living in the urban zones of Valle in 1964. YH = the male population between 15 and 64 born in rural areas of Colombia and living in urban zones of Valle in 1964. YM = the female population between 15 and 64 born in rural areas of Colombia and living in urban areas of Valle in 1964. r. = the weighted growth rate of each of the four populations. H II the rate of participation of each of the four populations. It follows that the population of male migrants from urban zones now (in the year t) living in urban areas of Valle, will be: XHt = T1(1+r1)t XH; the population of women migrants from urban areas residing (in the year t) in urban areas of Valle will be: xMt = T2(l+r2)t XM' The formulas for the other two types of migrants will be YHt? T3(l+r3)t YH and YMt = T4(1+r4)t Labor composed of migrants in the YM. year t is: Mit = XHt + XMt + YHt + YMt' The demographic information analyzed in Chapter II does not directly give us the value of the variables and rates required to calculate the supply of migrant labor in the census year, but it does allow us to approximate these values, as will be explained below. That part of Valle's population which was born in other departments may be taken as representative of the migrant population. In 1964, 76.4 percent (see Table 2.3 ) was between the ages of 15 and 64. The application of this Percentage to the total non-native population residing in ‘.P.. .- in. 1 m; 4“ Na ‘~ 81 urban centers (see Table 2.5) gives us the following values for 1964: XH = 117,902 XM = 131,923 YH — 32,373 YM = 24,468 The participation rates of males, XH and YH are assumed equal, with a value of 0.863, and the rates of participation of women, KM and YM are assumed equal with a value of 0.255, these parameters having been discussed above. The census of 1951 does not contain information about the number of natives and non-natives by sex, age and residence 13 making it impossible to estimate as does the 1964 census, ri for the four kinds of population defined above. A rudimentary way of estimating the urban growth rate of the working age population by sex will be used; it follows: A) weighted growth rate of non-native males between 15 and 64 years living in the urban zones of Valle. The growth rate of Valle's native population between the ages 15 and 64 was 3.4 percent from 1951 to 1964.14 The male population of Valle between 13Valle is the only department of Colombia which has some detail data on migrating population for the census year, thanks to the Universidad del Valle (See Chapter II of this study). 14Computed from data in Table 2.3. 82 the ages of 15 and 64 residing in the urban zone 15 and the proportion 16 grows at a rate of 5.5 percent of males born outside of Valle is 58.8 percent. This gives us: 5.5 (100) = (58.8) (X) + (41.2) (3.4) which gives us X = 7.0 This estimated growth rate of the working age non-native population is the best approximation of the migrant population, adopted as the growth rate for the population XH and YH' B) Weighted growth rate of non-native females between 15 and 64 years living in the urban zone of Valle. The female population of Valle between the ages of 15 and 64 residing in the urban zone grows 17 and the proportion of 18 at the rate of 5.7 percent females born outside of Valle is 59.9 percent. This gives us: 5.7 (100) = (59.9) (X) + (40.1) (3.4) which gives us X = 7.2 percent. This estimated growth rate of the working age non-native population is adopted as the growth rate for the papulation XM and YM. 15 16 17 18 Computed from data on Table A.1 of Appendix A. Computed from data on Table A.5 and A.1 of Appendix A. Computed from data on Table A.1 of Appendix A. Computed from.data on Table A.6 and A.1 of Appendix A. 83 The high growth rates of native urban popula- tion of working age for males and females show the importance of rural-urban migration while the even higher growth rates of non-native urban population of working age for males and females show the importance of migration from people of other Colombian zones to the urban areas of Valle. Migrant females have higher growth rates than males, as was expected. The population of both men and women born in the urban area of Valle, Z and ZM’ between 15 and 64 years of age, is H estimated by subtracting the migrants already estimated, from the total population of this age group. The results are: ZH = 303,616 - 150,275 = 153,341 2M 8 355,361 - 156,391 = 198,970 Rates of participation are assumed constant at 0.863 for men and 0.255 for women. The growth rate for both native populations is 3.4 percent.19 Consequently, the labor force for the urban zone of Valle in the year t is: M = (XH + XM + YH + YM + ZH + ZM) and the components are estimated as follows: xH = 117,902 (.863)(1 + .070)t 131,923 (.255)(1 + .072)t H 32,373 (.863)(l + .070)t 24,468 (.255)(1 + .072)t r404? 19Computed from data on Table 2.3. 84 zH = 153,341 (.863)(1 + .034)t 2M = 198,970 (.255)(l + .034)t The labor force in urban Valle estimated with the above six equations for the period 1964 to 1980 are presented in Table 4.5. The Labor Force in the City of Cali Following the same methodology used to estimate the labor force in Valle to estimate that of Cali, including both native and migrant population, we may proceed as follows: Xfi = migrant men between 15 and 64 born in urban areas of Colombia resident in Cali in 1964. XM = migrant women between 15 and 64 born in urban areas of Colombia resident in Cali in 1964. =‘mi rant men between 15 and 64 born in rural areas of Co ombia resident in Cali in 1964. rd :1: = migrant women between 15 and 64 born in rural areas of Colombia resident in Cali in 1964. *4 I! native male population of Cali between 15 and 64 in 1964. :i“ if native female population of Cali between 15 and 64 in 1964. r1 = the weighted growth rate of each of the six population groups above. Ti = the rate of participation of each of the six population groups. The contributions of manpower in the year t of any of the six groups is estimated in the formula Mit = Mo x Ti (1+r1)t Seventy six point eight percent of the non-native popu- lation of Cali falls in the age group 15 to 64. When this percentage is applied to the population arriving in Cali (see Table 2.6) we get: £35 .Nm women so ooonoo one: 2m .mm one .ow omen no ooaHmou whoa ZN use mN . w .mw .zx .mx oeo.¢Hw Hwo.~oo mH~.¢mH com.n<¢ nun.on nmo.ow ~¢m.n- mnm.mH wn¢.~m num.~0H owm.oon ommH Hum.om ¢m~.~mm www.maH omo.~m¢ www.mns ann.mw NHm.wHN Non.wH mwo.nn Hm<.mm mmn.omN ohmH mon.¢¢ mmH.mon wo~.q¢H mmeane non.~mn qmonHw num.HHN mHm.oH omo.Nn oqo.mm «om.~oN man oOH.o¢ Hmo.¢qn 0mm.mmH Hue qoq nmn.nmo oem.ws mum. as meson noses sass: .n.c oHoeH 86 XH = 99,541 KM = 118,378 YH = 24,866 YM = 30,977 The difference between the total number of men in the urban zone of Cali20 and the migrants of the same age group gives us the values of ZM and ZH' ZM = 187,767 = 149,375 = 87,381 ZH = 156,428 - 124,407 = 32,021 The values of the Z population of Cali, especially the value of 2H appear very low. The value of ZH of Cali is only about one-fifth of the corresponding value of urban Valle and the value of ZM of Cali is only about one-half of the 2M value of urban Valle. The reason for this big difference is that for the case of urban Valle the native urban population migrating between Valle's municipios are classified in the 2H and ZM population while, for the case of Cali, natives of Valle migrating to Cali are classified in the corresponding X and Y populations. The difference between the Z population of urban Valle and Cali indicates the importance of migration within Valle during the period 1951 to 1964. The rate of participation in the labor market (men and women) is the same as that for all of Valle (.863 and .225, respectively). 20See Table A.5 of Appendix A. 87 The growth rate r1, given the lack of historical data permitting precise calculation, is estimated in the same manner as was done for Valle. In Cali migration is a more important factor than the native birth rate, in the growth rate, precisely the inverse of the factors for the Valle as a whole. Migration produces an annual growth of 8 percent in the non-native population of Cali.21 A) Weighted growth rate of the non-native males between 15 and 64 years living in Cali. The male population of Cali between the ages of 15 and 64 grows at a rate of 6.3 percent22 and the proportion of males born outside Cali is 70.1 percent:23 This gives us: 6.3 (100) - 70.1 (x) + 29.1 (3.4) which gives us X = 7.6. This estimated weighted growth rate of the working age non-native population is the best approximation of the migrant population adopted as the growth rate for the population XH and YH' B) Weighted growth rate of the non-native females 21Data on non-native and native population for Cali in 1951 was not available. The growth rate of 8 percent is used in Planeacién Municipal de Cali "Estudio Preliminar de la Poblacion de Cali” (Mimeo) and Universidad de Valle, CUIP, Estudios Demogréficos de Cali, 1969, pp. 16 and 75. 22 23 24 Computed from data on Table A.5 in Appendix A. Computed from Table 2.5 and Table A.5 in Appendix A. Computed from data on Table A.5 in Appendix A. 88 between 15 and 64 years living in Cali. The female population of Cali between the 24 and ages of 15 and 64 grows at a rate of 6.7 the proportion of females born outside Cali is 82.8 This gives us: 6.7 (100) = 82.8 (X) + 21.2 (3.4) which gives us X = 7.2. This estimated growth rate of the working age non-native population is adopted for the growth rate for the population XM and YM. Consequently, the labor force in the urban zone of Cali in the year t is given for: xH a 99,541 (.863)(1 + .076)t 118,378 (.255)(l + 0.72)t Y = 24,866 (.863)(1 + .076)t M 30,997 (.255)(1 + .072)t 32,021 (.863)(1 + .034)t zM = 87,381 (.255)(1 + .034)t i‘ .4 ll er The labor force in Cali estimated with the above six equations for the period 1964 to 1980 are presented in Table 4.6. The high growth rate of the working age urban.popula- tion of Valle, and especially Cali reflect the high migration rates to these areas of Colombia. It also reflects the low out-migration rates as well as high birth and low mortality rates observed during the period 1951 to 1964. It is 24Computed from data on Table A.5 in Appendix A. 89 .Nm owed co ouchmu mama Sm .mm one . s.w owed co noonou ouo3 ZN .mN .2» . w .zx .mx cmH.¢HN www.mmm nmo.m0H Hm¢.omN Non.m¢m ¢¢0.mm NwH.~¢ N¢0.¢N om~.mo hHw.Hm nmm.mn~ ommH ono.omH ooo.-m www.ma ~H¢.-N ono.~Hm mmn.om onc.m¢ n~¢.- own.eo omo.mm ¢m~.nmu oan NHm.noH Nmm.HHm oH¢.om Nwm.nH~ mom.mn¢ www.mm OMH.¢¢ Hum.o~ mmm.mm mom.mn mqm.mmu wan H¢©.~¢H mmm.Hom o¢N.mm mm¢.wo~ «mm.mqq mH¢.¢m anc.mq mHn.mH NHo.nn Hmn.¢n 8N0.NNN nan mmo.m~H HHm.Hm~ ¢~H.om «mo.Ho~ aor.o~¢ Hm~.mm mn~.H< mom.wH «mo.Hn www.mo mmm.oo~ oan nqo.NHH oH~.NwN mom.nm moo.mmH mo~.¢mm NHH.Nm me.mm Nmm.oH «mo.wo onw.¢o www.mmH nnoH ¢w¢.o¢ omm.~n~ qu.¢m mmm.mmH o~¢.mom mNH.Hm noo.mm qu.mH qu.¢¢ oom.oo «on.QNH ean ~o~.~m Noa.mem won.Hm «om.~wH ¢-.oem moH.om omm.nm nnn.eH mm¢.H¢ om¢.om ~wo.ooH man om~.mo www.mmm oww.wn oom.o~H Nom.¢~m mHH.mN wOH.om mmN.MH www.mm o¢o.~m Hmm.¢mH Nan m¢¢.~m wmm.o¢~ NmN.en oom.o~H Hmm.¢om me.w~ Hmm.¢m mmm.NH cnm.mm OHH.a¢ m¢¢.n¢H Han Noo.oq ao~.wm~ mwm.mm owm.eoH qu.nm~ ~m~.- mun.mm mm¢.HH mom.mm HHw.mc “Hm.mmH oan nmw.om me.om~ 5mm.Hn Hom.mmH mmn.~o~ omm.o~ ~oo.~n omH.HH Hn¢.om mm~.~¢ Hom.mNH ooeH mem.n~ mmm.m- HHo.ao «Hm.¢nH ¢n~.Hn~ o~¢.m~ www.Hm wme.oH mo~.w~ som.am meH.nHH wooH ohm.oH www.mHN ~¢N.oo oq~.m¢H www.mmm mmo.¢~ onm.om nmm.m mmn.o~ an.nm oHo.noH womH mom.~H omo.mo~ nqm.¢o mmm.¢¢H mq¢.H- mmw.m~ nqm.mm mwo.m mom.¢m mmo.¢m wm<.mm oomH omo.m NHo.No~ m~¢.No mmm.mmH mom.~o~ oqo.m~ cnn.m~ mn¢.w omo.m~ mmm.~m Nm¢.No mooH o mom.mmH Num.oo emm.qu mom.mmH www.mm «mo.- som.u one.HN omH.om ¢0¢.mm comH He - we. we an me an. nu mm a» me xx :x ooooHomMHn Hmuoh\‘ Hooch coHuowsz uoonuHS oouom HoooH owns: GOHuonHz nuHS oouom uOAMH soon: now» .ome cu «omH uoHuom oou wcHHon cOHuousz uoooqu one nqu HHno cH oouom HOQMH amok: .o.¢ oHomH 90 unfortunate that geometirc growth rates have to be computed from long intercensus periods of time and that recorded data differ from census to census making it impossible to calculate growth rates for subpopulations and age groups. However, the above equations are useful to assess unemploy- ment rates in urban Valle and Cali. The demographic analysis and the decision model on migration presented in the previous two chapters allow us to make the following points with respect to future migration: 1. Potential areas for urban-urban and rural-urban outmigration are decreasing, although there are a good number of municipios in both Valle and other departments with large growing populations. 2. The growth of modern agriculture with lower labor input and the pressure of the modern sector on the small farm.owners tend to drive people out of agriculture. 3. The greater educational, health and amusement facilities in the large cities tend to attract people from other areas. 4. High unemployment rates in the large urban center 25 tend to diminish or abolish the gap between urban of Valle and rural earnings and, thus, to slow down or eliminate its effect on rural out-migration. 25Urban unemployment in Valle and Cali are between 17 and 20 percent, see Chapter VII. 91 Some of the above factors tend to maintain or accelerate outmigration while others tend to slow down migration. We do not have enough information to be able to establish possible changes in the equation parameters from 1964 to date of the equations for estimating the labor force in urban Valle and Cali. However, we will use the above equations to estimate the labor force in urban Valle (See Table 4.4) and in Cali (see Table 4.5) for the period 1964 to 1980. We want to make it quite clear that the above equations of population growth using weighted growth rates are more useful for short- run projections. For instance, the native populations of Valle and Cali, given a constant migration rate, will tend to increase when descendents of non-native population are born there, changing the weights used and the the composition of the estimated population. Using the above equations we will project the popula- tion of urban Valle and Cali assuming no migration. We will also examine the differences with respect to the above estimates when there is migration in order to isolate the effect of migration on the urban labor force.26 The procedure used to estimate the labor force of Valle and Cali without migration is: (a) aggregate the work age 26For example, the four points made above with respect to future migration act in such a way that migratory currents do not change. In Chapter VI we examine the reliability of the equations with migration rates equal to those in the past and with no migration. 92 male populations (XH, YH and Zn) of Valle and Cali for 1964 in one group, PH’ and the working age female population (XM, YM and ZM) of Valle and Cali for 1964 in another group, P M’ and (b) apply the native Valle's population of working age growth rate from 1951 to 1964 which is 3.4 percent. Thus, the equations are: A) For Valle PH 303,616 (.863)(1 + .034)t M 355,361 (.255)(1 + .034)t B) For Cali P PH 156,428 (.863)(1 + .034)t PM = 236,756 (.255)(1 + .034)t The labor force PH and PM estimated for 1964 to 1980 and the difference with previous estimates for labor force with migration are presented in Table 4.5 for Valle and Table 4.6 for Cali. The estimated labor force under two alternative hypotheses ((a) equal migratory currents observed in the past will hold until 1980, and (b) no migration will take place in the future) will be analyzed with respect to the effect on levels of employment and unemployment (Chapter VII) and the possible need and design of employment policies (Chapter VIII). In conclusion, high urban unskilled labor growth rates were observed during the period 1951 to 1964. Factors causing these high growth rates were (1) low conversion rates from 93 unskilled to skilled labor and (2) high birth and migration rates. In Chapter VI we will examine possible changes in the migration rates from 1965 up to the present. The urban unskilled labor force tends to concentrate in Cali. Participation rates in the labor market are high for men of working age (15 to 64 years) and low for women of working age. CHAPTER V URBAN LABOR ABSORPTION A great number of people are settled in the urban areas of Valle, especially in the capital city, Cali. Average expected urban earnings above the average rural earnings and the structural changes occurring in the rural area of Valle where a growing modern sector cannot absorb all rural laborers are causing small farmers to migrate out of agriculture. The relative low earnings of Valle's neighboring departments and the attraction of the cities are all factors determining migration to urban Valle, as dis- cussed in the previous two chapters. The purpose of this chapter is to examine the absorptive capacity of Valle's manufacturing industry for unskilled labor. In Valle the urban labor employment in the manufac- turing industry is very low with respect to the other sectors and has fluctuated between 12.5 and 14.5 percent Of the total urban employment with no tendencies toward changes during -the period 1960 to 1969, as shown in Table 5.1.1 The annual 1DANE defines a manufacturing firm with less than 5 laborers as craftsmen. Nelson, Shultz and Slighton classified firms with less than 9 workers as craftsmen. If this defin- ition is adopted the proportion of the urban force employed in the manufacturing industry will drop about two percentage points. 94 95 .wooH>uom Horuo one uooaouo>ow .oOHuooHooaaoo .QOHuouuonooouu nonoHooH .NeeHIemeH .mz mo ecuuom soon: onu oH unoa%0Hasm unsound: Hmuoa .H.n oHan 96 increase in industrial employment has fluctuated from 200 to 4,000 persons, disregarding the year 1967 when there was a decrease.2 The average annual growth of employment in the Valle's manufacturing sector was 3.3 percent. In Table 5.2 the Valle's industry is subdivided into three groups: a) the industrial branches producing consump- tion goods like food, beverages, clothing, tobacco, wooden furniture; b) industrial branches producing intermediate goods, like wood, paper, textiles, leather, rubber products, chemicals, non-metallic minerals, and basic metals; and c) industrial branches producing durable and capital goods like metallic products, non-electric machinery, electric machinery, and transportation materials. Most manufacturing laborers are employed in the consumption goods subsector, followed by the intermediate goods subsector. The annual growth of employment was 2.3 percent in the consumption goods branches, 4.0 percent in the intermediate goods branches, and 8.8 percent in the durable and capital goods branches. Foodstuffs has the most employees among all Valle's manufacturing branches absorbing 22 percent of all industrial workers. Chemical products follow with 11 percent, metallic products with 8 percent, textiles with 7 percent, clothing 2The figures on the annual increases in urban employ- ment vary between 13,000 and 17,000 persons with a slight upward trend between 1960 and 1969. (See Table 5.1.) 97 one mHH one MHH .moz ooHummooumm on moHoomooz wooHuoHom .mzm mo moooovomooo o mH H.m oHooH one oHoeu mHnu consume ooHoon Iooo oOHuom sou oH mucouomeo 0:8 .wuoo mo xomH on one monoomuo HHo mooHooH woo moon H can OMH.H ow omm.m mHm.om Noo.wH momH HmH I man I mom.HI qw~.m wwm.aH me.NH momH new eHH MNH mm¢.m own.mH m-.mH oomH Now «mm qu 05H.m www.mH omo.mH momH HmN on mm wom.m omn.wH nom.wH oomH com on man NHm.m moo.mH www.mH momH NNN.H mmH.H «no NHm.w NmH.wH sow.NH NomH mwu mmq.H emu mam.“ mom.oH mmH.mH HemH Noe mmo oHH.H mom.o won.mH omm.oH oooH mom Ham mum.H eoH.o mnm.¢H omw.mH mmmH mHm. now. own I me.m mmn.¢H me. mo mooou moHooooum HouHmuw one .oumHooauoucH .GOHueaomooo sou oH uooBMOHeam unsound: Houoa .N.m oHomH 98 with 7 percent, paper and derivative products 6 percent. In Table 5.3, the annual rates of growth of industrial employ- ment are shown. Industries with the highest growth rates of employment are the industries producing intermediate, durable and capital producing goods. The relative importance of the different industrial centers of the department of Valle in relation to utilization of manpower is presented in Table 5.4. It may be noted that industry located in Cali represents about 60 percent of industrial employment in Valle as a whole. The cities of Cali, Palmira and Yumbo may be considered as one industrial complex. Together, they absorb nearly 83 percent of all industrial employment in Valle. The great majority of these workers live in Cali. Thus, the Valle's manufacturing sector, largely concen- trated in the cities of Cali and the neighboring cities of Palmira and Yumbo, shows very little absorption of new laborers during the period 1956-1969, below the growth rate of the urban labor force. The industrial branches producing durable and capital goods are absorbing labor at a faster rate than industrial branches producing intermediate and consumption goods. We turn now to the objective of this chapter, an examin- ation of the reasons for the low absorption labor rate of Valle's industry. First of all we present some theoretical ideas. 99 Table 5.3. Employment Growth Rates in the Industrial Sector of Valle, 1956 to 1968. Industrial Sector Annual Rates of Growth Basic metals 14.40 Metal products 10.70 Non-Electric machinery 10.30 Electric machinery 10.10 Diverse industries 7.70 Woods 7.60 Paper 6.40 Printing and publishing 6.30 Chemistry 6.00 Non-metallic minerals 4.40 Transportation materials 4.20 Food 4.00 Rubber products 1.65 Leather, leather products except shoes 1.18 Clothing 0.90 Textiles 0.86 Beverages 0.25 WOoden furniture 0.18 Tobacco -3.50 Source: DANE, Anuario General de Eatadistica Bogota, DANE, 1958 to 1972. 100 Table 5.4. Percentage of Total Employment Absorbed by the Principal Industrial Centers of Valle, 1965 to 1969. Cities Years 1965 1966 1967 1968 1969 ------------------- Percent--------------- Cali 61.2 60.2 58.0 59.3 61.7 Yumbo 12.3 14.0 15.0 14.4 15.0 Palmira 8.6 8.5 9.2 9.0 8.5 Buga 6.4 4.3 4.3 3.7 3.7 Tulua 1.3 1.3 1.3 1.0 1.7 Buenaventura 0.7 0.5 0.6 0.7 0.9 Cartago 0.7 0.4 0.7 0.7 0.3 Source: CVC, "La Industria fabril del Valle del Cauca," Unpublished preliminary draft, 1974. Demand for Labor As mentioned previously, labor as a factor of produc- tion is very heterogeneous and there are as many different demands for labor as there are kinds of labor used in the productive process of a community. Though this chapter treats laborers as a homogeneous group, considerable atten- tion is paid to the unskilled laborer. In markets for factors of production, those making the demands are the producers; in the case of labor, those who produce offer labor services are consumers as well as suppliers. This is an important difference from the markets for goods where consumers demand and producers supply. Traditional economic theory assumes that all producers try to maximize earnings (not excluding the non-monetary) by 101 using their productive resources as well as possible, subject to the technical restrictions given by the production function. If the firm acts in a perfectly competitive market of products and factors, prices of both productive factors and products are fixed parameters to the firm and maximiza- tion determines the optimal amount of the productive factors to use. The production function shows the physical relation between the final product and the factors of production: mathematically, Y = f(X), where Y is the final product and X the vector of the factors of production. The equation of the firmis earnings is: G = Py - f(X) - Px - X, where G = earnings, Py is the price of the final product, and Px is the vector of the costs of the factors of production. Maximizing this function with respect to X, we obtain a. .1 = ° o = Py ax Px’ that is Py MPx Px’ in other words, the marginal value product of the factor X (VMPx) equals its price. This expression indicates the maximum price the producers will pay for a given amount of the factor X. The amount of the factor X a firm.wants to buy is a function of the factor price and is the amount which equates Px with Py - MPx. The demand function for a given factor, say labor, when all other factors are fixed is X = g(VMPx). This function is negative sloping, in general. The marginal value product curve has negative slope in the rational stage of production, 102 so that the more manpower used, the less productive the last man hired. This explains why unlimited quantities of labor or of any other factor, are not used in the production process. If the firms have a monopoly in the market of their products, they pay the factors according to their marginal revenue curve (MR) also a function of the price of the product and as a result the demand function for labor is given by X = C(MR . MPx) or X = C(MRP) where MRP is the marginal revenue product of the factor X which is equal to dp dp P - MPx + ail (Y - MPx) where HEX is the change in P Y y resulting from using more of X to produce dy more of Y. Py and Y are measured after dx more of X has been used and hence 3% = MPx has to be subtracted from Y. To derive the demand curve for labOr we assumed price of labor fixed, independent of the number of workers a firm hires. What will be the effect on the demand curve for labor if workers can unionize when there are a minimum number of workers in the firm, as the Colombian labor legislation has established? We will expect cost of labor to increase. Owners of firms when faced with unions consider the cost of a) conflicts, b) strikes, c) assuring job security to the union members, and d) frequent pressure for wage raises. Unskilled labor has, then, two prices or 'wages; one for unionized laborers and another lower rate for non-unionized laborers. An entrepreneur has to face these two prices for labor. Which of these prices he will have to 103 pay depends on the number of workers he hires. The real price of unionized labor is the nominal salary the worker will be asked for which will be above the minimum legal wage in the great majority of the cases, a difference increased even more when the unions press for extra-legal payments for vacations, Christmas and mid-year payments, etc. not to mention other real costs like the abhorence of unionization and organized union pressure implying processes of negotiation ‘with high possibilities of open conflicts and strikes. Besides the unions offer job security to their members. This means that the entrepreneur will be forced to maintain a labor force some of whom, at times, will be paid above their marginal value products. The situation an owner is facing in the labor market is illustrated in Figure 2 where Wfi is the minimum legal wage and Wfi is the cost of labor which includes all the items we have just discussed. If the demand for labor X = f(MVP) shifts up or down due to changes in the marginal productivity of labor or changes in the price of the product but does not go above point A or below point B there will not be changes in the number of workers desired. This analysis brings about an important modification in the demand curve for labor derived at the beginning of this chapter. That modification will be discussed in the following section. 104 Cost of Labor Wu \\\\\\\: \\\\\MVP Wm B Number of workers Figure 2. The prices of labor for unionized and non-unionized laborers and the marginal value product curve of labor. Modifications in the Demand for Labor The industrial demand curve for labor defined as X,= f(MVP) is affected by a) other variable factors of production and labor-saving capital-oriented sub-production functions, and b) labor price differentials, of the kind discussed above. In what follows we will first discuss the effects of capital as a variable factor of production, the sub-production functions oriented toward capital which substitutes for unskilled labor and, second, labor price differentials. The demand curve for labor of a given firm X - f(MVP), 'was derived from the production function Y = f(X/K,Z,. . .) where all factors but labor were held constant. 'We now treat capital (K) as a variable factor to determine the effect of changes in capital on the demand for labor. We must consider the relationship between labor and capital. 105 Two production factors are defined3 as substitutes when the relations: AMPXIIAX2 < 0 and AMPxZIAX1 < 0, that is, if the use of one of these factors is increased (+AX2), the marginal product of the other factor will decrease (-AMPX1). Two factors are complementary when these same relations are positive, that is, if the use of one factor is decreased (-AX2) the marginal product of the other factor is decreased (-AMPx1). If more capital is available, in the form of machinery and equipment which substitute for labor, the marginal productivity of labor will decline and the entre- preneur will use less of it at any given level of output. 0n the contrary, if more capital goods are available, and these are complementary with labor, the latter will have greater marginal productivity and the entrepreneur will demand more labor. Thus, when a country is creating and/or importing labor substituting capital goods the marginal productivity of labor decreases from MVPI to MVPZ in Figure 3-I.4 If the original price of labor was P1 the firm is in equilibrium using the quantity 0A. When the price falls to P2 the firm equates P2 and MVP2 to arrive at the amount OB of labor. The curve dd, joining equilibrium points describes 3We recognize the definitions of substitutes and com- plements inputs apply to Stage 11, see Irving Morrissett, 'Some Recent Uses of Elasticities of Substitution-—A Survey," Econometrics, Vol. 21, No. l, (1953). 4We assume that capital is substitute of labor for all quantities of labor used, i.e., the shift in the MVP curves is parallel. 106 the demand for labor when the capital substitute for labor is also variable. We see that this demand for labor is more inelastic and at the price P2 less labor will be used, OB instead of 0C. When the factors are complementary the MVP of labor increases when capital is increased. The resulting demand for labor is more elastic and at the price P2 more labor will be used, as it is shown in Figure 3-II. Px 1. Substitute Case PK 11. Complement Case (1 (:1 P1 pz --2_.... MVPl I ‘ 2 l. . ”I “8* 1 - l d 0 A B c x 0 A c B Figure 3. Demand for labor when capital is variable. We now consider two kinds of capital, labor-saving <=3‘E’1tal (Ks)' neutral-traditional capital (1%) and two kinds 0f labor, skilled labor (x8) and unskilled labor (xu). In that follows we analyze the demand of labor of four hypo- 1:h-etzi.cal groups of firms with different production conditions. Group I: this group is composed of very small firms mulch use unskilled labor and neutral-traditional capital. The owners' economic resources are limited and thus an insignificant increase in size of these firms is expected. The firms utilize few workers and, therefore, do not con- t ell“Plate the possibility of unions in the firms even if they 107 can increase production. Their demand for unskilled labor is of the type of demand drawn in Figure 3-1 or 3-II. The aggregate demand of labor of these firms is low and they employ a limited number of unskilled laborers. Group 11: medium size firms belong to this group. These firms use unskilled labor and neutral-traditional capital. The owners do not know of the existence of labor- saving capital and/or do not have the money to buy it. These firms have at least two expansion paths, one for non-unionized labor and another for unionized labor for a given price of capital (see Figure 4). If the firm wants to produce Y4 11:11 ts it will need to increase the number of workers above the minimum number, N, necessary to form a union with wages equal to the minimum legal wage and so, he will go from P0 int A to A' , but the laborers can form a union which likely will ask for higher salaries and increase the cost of labor in other ways. To increase production using the same amount of labor at A and being in equilibrium will require the price of capital to fall but this is unlikely to happen at a time when more capital is being demanded. So » a point like B is not feasible. Point C is not relevant at this time since there are no reasons to expect wages to go up as less than the number of laborers to form a union will be employed. Also the price of capital is not expected to decrease when more of it is demanded. Thus, the maximum v0 1% of production that will be considered by entrepreneurs W ho do not want to deal with large numbers of workers is Y3. 108 If a producer can and want to expand production, he will hire a number of laborers above N to be able to produce, His relevant expansion path is bb. say Y6. medium size firms owned and managed by Group III: producers who know and are able to buy modern equipment We assume that these firms initially belong to this group. 'produce with unskilled laborers and neutral-traditional capital but are accumulating resources at their present .leevel of production, and are interested in acquiring labor— The change in technology depends on: a) s aving capital. circa expectations about the market for their products, b) “tries acquisition price of the labor-saving capital and the .saajlsvage value of the neutral-traditional capital, and c) the expected changes in the cost of labor as a result of using leessas unskilled labor and hiring skilled labor. N4 11 0 Expansion paths of firms Figure 4. for non-unionized and unionized labor. 109 Consider the situation of a firm described in Figure 5. Initially the firm operates with the production function Y = f(Xu,Kn|Z,. . .,XS = 0, KS = 0),5 as described by iso- quants numbered 1 in Figure 5, and the firm is producing Yo units at point A. If the firm wants to produce Y1 units it will have to move to B but more than N workers are needed and the firm will be confronted with expectations of rising wages and other non-monetary costs associated with unioniza- tion. If these cost expectations are greater than the net cost of labor-saving capital the firm will acquire labor- saving capital and shifts to the production function Y = f(Xs, Xu’ KSIZ,. . .,Kn = 0) will produce Y1 units of C where less than 25 laborers are hired. Average salaries for workers are higher than before because skilled workers have been hired. Also some unskilled laborers will see their MVP increase6 and demand higher salaries while other unskilled workers will see their MVPs decrease and lose their job. Finally, production and consequently the use of labor and installed capacity of the firm will be limited if the high profit point is at C where the number of workers is below N. _ 5 Xu’ XS, Kn and KS as defined on page 106. 6The labor saving capital is complementary for the first workers generally skilled to operate the capital, 1- e- , the MVP of labor curves before and after labor-saving is used cross each other. xu’xs N+ 0 Figure 5. Change in technology and expansion paths for a firm. Group IV: big firms organized as stock companies, some of them with foreign capital, form this group. In the majority of the cases the companies were initiated with a large amount of machinery and their operation requires a large number of workers well above the number necessary for unionization. These companies demand substantial numbers of skilled and unskilled laborers; however, the skilled laborers are complementary with labor-saving capital, and together, the two replace unskilled labor in these firms. We will show that production conditions of the four groups of firms as developed above explain factor proportion variations among the Valle's manufacturing industry. But first, we will examine the relation between the relative Prices of labor and capital7 and the capital-labor use ratio. ; 7The price of capital can be defined according to its acquisition price or the salvage value or the shadow price (If\rl?s through the capital good life-time). The price of cal-">1 tal is the discounting rate equalling, say, the acquisition 111 After that we shall examine (l) the effect of a change in the composition of the industrial demand on manufacturing employment, (2) labor employment by size of firms in the Valle's manufacturing industry, (3) use of installed capacity, and (4) growth of the manufacturing sector. Elasticities of Substitution A price relation which favors labor over capital will determine higher rates of employment only if there exists the technical possibility of substituting labor for capital. This substitution possibility is measured by the elasticity of substitution S, which is defined as the percentage change in the ratio of the two factors with respect to the percent- age change of the marginal rate of substitution (MRS) of the factors when output is constant.8 The elasticity of substitution is symmetric, that is the elasticity of sub- stitution of labor for capital is equal to the elasticity of substitution of capital for labor.9 price plus repairs with the present value of its product share throughout the capital good lifetime. The commercial interest rate is generally used as the price of capital, Which is the opportunity cost of the funds invested in the capital good. 8The marginal rate of substitution between two factors 1&3 defined as the relation between their marginal products MszlMle. This is the slope of an isoquant and shows how, 353 one resource is reduced in the process of being substituted by another, the quantities of the latter must be increased constantly in order to maintain the volume of production constant. mthematicaly the elasticity of substitution is expressed as follows: 112 Elasticities of substitution for 20 industrial branches of Colombia were computed by Planeacidn Nacional10 and by 11 Gaviria, Gomez and Lopez using different production relations. Planeacidn Nacional estimated the relationship between the logarithm of output per worker (dependent variable) and the logarithm of real wage as Arrow, Chenery, Minhas and as} s a [a] .1; d oy/axz. = S = “Y x1 LWJ d aylaxz] X1 oyfoxz LGYIO'X]. Y; @735; Since the partial derivative of a function of production with respect to a factor is the marginal product of the factor (MPx) we have: ' \ El MPxZ x2 Mle MPxZ g; Mle x2 L F 9The elasticity ofrsubstitution can also be written as X X d log —l 2 dlogaX—L [2 rx SYImnetry is proven by noting that d log 2—21-1= - d ugly-12?] and d {3.1.}: - [(1le logldxzj d logla-x—l—J 10 Departamento Nacional de Planeacion, "Breve Esquema Sobre el Problema del Desempleo en Colombia," UPC-002, junio 30. 1970. 1192. cit., pp. 140-152. 113 12 Solow did. The data used were time series data (1956 to 1968) published by DANE. High elasticities of substitution were found for all 20 industrial branches except tobacco. Gaviria, Gomez and L6pez used the same data but followed 13 Dhrymes who does not assume constant returns to scale nor prefectly competitive markets obtained very low elasticities of substitution.14 The contradictory results of these two estimates would be explained in what follows. Arrow, et al., wanted to relate productivity (value added per worker to differences in the capital-labor ratio, but capital stock data were scarce. Given a general pro- 15 the assumption that the price of labor duction function, is equal to its MVP, and that there is an increasing rela- tionship between the wage rates and the capital-labor ratios,16 the Arrow, et al., model relates productivity to wage rates.17 Nelson, Schultz and Slighton showed that if there are other 12"Capital—Labor Substitution and Economic Efficiency," Egview of Economics and Statistics, Vol. 43, No. 3, (1961). 13Phoebus J. Dhrymes, "Some Extensions and Tests for the CES Class of Production Functions," Review of Economics and Statistics, Vol. 47, No. 4, (1965). 14They estimated the function log L = a + 8 log (W/P) '* c log Y where L = labor, W/P = real wage, and Y = output. 15y - f(K,L) and Y/L = f(K/L,l). 16m? = g, g' > 0 17Y/L = f [gCK/L)]. then Y/L = f [g-1(W)] or Y/L = h(W). 114 factors that vary across countries18 which affect value added per worker and MVPs of labor, the direct relationship between value added per worker and wages used by Arrow, et al., does 19 This not change a good fit for the equation is obtained. is a case of specification errors in a model as relevant explanatory variables are omitted. If one variable is omitted and is positively correlated with the included variable the elasticity of substitution will be biased?0 Positive correlation with wage rates (included variable) is likely to be the case if more skilled labor and other changes in the production process to achieve better efficiency21 when time series data are used. Thus, we will expect the elasticities of substitution estimated by Planeacion Nacional to be biased upward. Gaviria, Gomez and Lopez estimated the equation log (X) = a + b log (W/P) + log Y.22 They appear to have a poor fit for the equations with no statistically significant (different from zero at 10 percent level) elasticities of substitution in 11 equations. All elasticities of substitution 18Arrow, 35 al., wanted to show that variations in the capital labor ratio across countries explained productivity differences among them. 1993. 515., pp. 98-99. 20See Jan Kmenta, Elements of Econometrics, New York: ITIGB Macmillan Co., 1971, pp. 391-5. 21 22X is number of laborers, W/P real wage, and Y total ‘ZEIJL‘Je added. We will examine this further in the following pages. 115 are substantially lower than the ones estimated by Planeacion Nacional. They probably havemulticollinearity among indepen- dent variables. With a high degree of multicollinearity, estimates of the regression coefficients are highly imprecise.23 Thus, there are serious statistical problems in the elasticities of substitution estimated by Planeacién Nacional and Gaviria, Gomez and Lopez of the Universidad de Antioquia. There are other problems in estimating elasticities of substitution for industrial branches. There are different kinds of labor like unskilled and skilled laborers; there are several types of capita labor-saving, labor-intensive and 24 machinery and equipment are near perfect comple- neutral; ments for some workers and near perfect substitutes for others; shifts on production functions occur as a result of changes in technology; production increases or decreases for different firms, and there are differences in managerial abilities. When time series data are used additional diffi- culties in measuring the degree of substitution between factors of production are: output is not constant and changes in technology, kinds of capital, and labor training take place 23Kmenta, 92. cit., pp. 387-9. 24These differences in the kinds of capital as well as technology are described by the four hypothetical groups 0f firms discussed above. Furthermore, dualism is character- ized by technological and kinds of inputs differences. 116 over time. Wages bring market conditions into the estimates. This is a real limitation since we know there are definite imperfections in the labor market of Colombia. In spite of all these difficulties we will compute elasticities of substitution for Valle's l7 industrial branches (see Table 5.5 for a list) using a different equa- tion. Since we shall use time series data we will make the efficiency paramter‘Y a function of time, like Y = ext, and since 7 affects in equal form labor and capita the parameter A reflects neutral technology. The distribution parameter 9 reflects non-neutral technology since it affects capital and labor in a different manner. When y is a function of time the equation to be estimated is: Log (Y/XZ) = A + S Log 25 Also, we will estimate the same equations (W/P) + B t + U. estimated by Planeacién Nacional, which is the same as above excludingthe time variable t (that is, the efficiency para- 'meter does not change over time) to be able to compare the results of the two sets of equations and study the statistical problems discussed above. The data is for Valle during the 25When y = ext, we have W/P - (l - 6)(eAt)-p(Y/X )p+1, ‘wage equal to its MVP, using the CES production functi . Thgn, Log (W/P) = - Log (1 - 6) - plt + (p + 1) Log (Y/Xz), an - 1 1- A Log (Y/Xz) - - 3:1 Log (1 - 6) + 511'- Log (W/P) + BET t and therefore, =-1 - =_1_ gal. 9A 3:1 Log (1 6), S p+1 and B 0+1 . The equation estimated by Planeacién Nacional was Log (Y/XZ) = A 7 S Log W P . 117 Table 5.5. Regression Coefficients of the CES Production Functions When the Efficiency Parameters Change Over Time1 for Valle's Industrial Branches. Industrial 2 2 Correlation Branches A S B R F D.W Coefficient3 frood 3.744 .119 .031 .80 20.00 1.48 .976 (3.937) (.312) (.030) Beverages 8.956 -.285 .072 .79 19.10 2.22 .958 (4.750) (.432) (.030) 1P<5bacco 6.650 -.047 .0150 .17 1.05 1.34 .650 (2.602) (.248) (.0124) fireaxtiles - .074 .416 -.015 .20 1.21 2.46 .845 (3.221) (.269) (.011) (Zilothing 6.819 -.229 .012 .46 4.21 1.31 .338 (1.375) (.118) (.004) Vitaod 1.267 .295 -.028 .07 ”37 2.11 .947 (3.458) (.358) (.033) VJ<>oden Furniture -2.858 .658 .053 .75 14.92 1.28 -.575 (1.820) (.180) (.010) ' I?£1per 5.066 -.006 .076 .83 24.00 1.68 .981 (4.065) (.358) (.056) I.¢2ather -2.929 .698 .008 .73 13.39 1.66 -.985 (6.187) (.699) (.071) Flaabber -5.121 .841 .011 .88 38.02 1.97 .982 ‘ (7.073) (.603) (.047) (Etiemicals -4.l74 .804 -.039 .84 26.30 .99 .986 (4.257) (.354) (.036) E'crn-metallic 6.643 -.183 .020 .03 .14 1.78 .977 Minerals (9 . 433) (. 845) (. 062) liansic metals 2.945 .159 .057 .59 7.11 1.75 .410 ( .493) (.064) (.037) Metallic products 4.824 -.030 .052 .91 48.09 2.45 .953 (1.268) (.115) (.016) Non-electric 1.554 .270 .024 .55 6.06 .93 .954 chinery (4. 213) (.457) (.053) Electric Machinery 8.438 -.351 .116 .91 51.43 2.12 .963 (1.805) (.175) (.025) Transportation .391 .327 .029 . 83 23. 76 2. 25 .424 terials (1. 142) (.102) (.007) Iitxmbers in parenthesis are the standard errors of coefficients. 1. 1-<>g (Y/XZ) - A + S Log (W/P) + BT + y; Y - value added, X2 - number of workers,1UP - ‘feaal wages, T - time variable, 1956 - 1, 1957 - 2,. . .,1968 - 13. :1 lDumbin-Watson statistics, dL - 0.82 and du - 1.75, 1 percent. COrrelation coefficient between the independent variable, W/P and t. 118 period 1956 to 1968. Value added and wages used in the equations are in 1958 prices.26 The results are presented in Table 5.5 and 5.6. When the efficiency parameter (y) is a function of time we noted high correlation coefficients between the two inde- pendent variables (see Table 5.5) with the exception of three industrial branches, with correlation coefficients below 0.5. With greater efficiency from more trained laborers, among other considerations, wages are expected to increase. When there are high degrees of multicollinearity, the regression coefficients are imprecise, as noted above for the case of Gaviria, Cdmez and Lopez estimates. The ifidustrial branches for metallic products and transportation materials, which have low correlation coefficients between the two independent variables, have relatively low elastici- t115—es of substitution. The elasticity of substitution for Clothing,27 which have the lowest correlation coefficient, have a negative sign, contrary to what was expected. Negative elasticity of substitution implies that the substitution Parameter (p) of the CES production function is less than -1 and the isoquants will have the wrong curvature. \ 26The index price used is the implicit index price in the national social accounts. This index as well as the Innnber of laborers employed in each industrial branch, Value added and wages in current Colombian pesos are given 11 Appendix B.4. t 27The elasticities of substitution of metallic products, dransportation materials and clothing are statistically ferent from zero at the 5 percent level. 119 Table 5.6. Regression Coefficients of the CES Production Function When the Efficiency Parameter is Constant1 for Valle's Industrial Branches. Industrial 2 Branches A S R F D.W Food - .0960 .425 .80 38.99 1.41 (.903) (.068) Beverages -1.867 .703 . .67 22.78 1.96 (1.689) (.147) Tobacco 4.660 .149 .05 .60 1.16 (2.062) (.192) Textiles 3.723 .096 .04 .41 2.35 (1.832) (.150) Clothing 5.642 -.122 .06 .76 0.93 (1.632) (.139) Wood 4.057 .004 .00 .00 1.65 (1.159) (.113) Wooden 3.061 .097' .01 .12 0.50 Furniture (2.757) (.279) Paper - .293 .481 .80 42.98 1.74 ( .888) (.073) Leather -2.239 .620 .73 29.41 1.64 ( .934) (.114) Rubber -6.790 .984 .88 83.10 2.08 (1.323) (.108) Chemicals .387 .423 .82 50.56 1.03 ( .758) (.060) Non-metallic 3.739 .078 .02 .20 1.83 Minerals (2.010) (.172) Basic Metals 3.010 .200 .49 10.42 1.46 ( .522) (.062) Metallic .972 .322 .81 46.39 1.50 Products ( .567) (.047) Non-electric - .290 .472 .54 12.83 1.00 Machinery (1.316) (.132) Electric .471 .429 .72 28.58 .67 Machinery ( . 905) (. 080) Transportation -1.45 .56 .53 12.24 1.44 Materials (1.655) (.145) Numbers in parentheses are the standard errors of coefficients. 1Log (Y/X ) - A + S Log (W/P) + U; Y - value added, X - number of workers, W/P a real wages, T = time variable, 1956 - 1, 1957 - 2,. . .,1968 - l3. 2Durbin-Watson Statistics. d1 - 0.82 and du = 1.75, 1 percent. 120 Autocorrelated disturbances were not found for any industrial branches when the Durbin-Watson test was applied. However, for some industrial branches this test is inconclusive . 28 A negative correlation coefficient between real wages and time was obServed for wooden furniture and leather. This means that real wages have decreased in these two industrial branches. Probably small manufacturing and handcraft firms are numerous in these industries. In the simple regression model, where real wages is the independent variable the elasticities of substitution differed widely (Table 5.6) from the ones estimated in the Previous case. Elasticities of substitution for tobacco, textiles, clothing, wood, wooden furniture and non-metallic Ini‘lrxerals are not statistically different from zero at the 5 percent levels and poor fits were obtained, with R23 less than .07. The estimates for elasticities of substitution of all other industrial branches except leather and chemicals are greater than the corresponding estimated elasticities of substitution obtained in the previous case. However, we calf‘rnot conclude that the elasticities of substitution are greater since with specification errors in the model and wi 1211 positive correlation between the included and the this sing variables, the estimates are biased upward, as \ e 28The Durbin-Watson test is inconclusive when the Stimated Durbin-Watson statistics falls between d.L and du' 121 explained before. Autocorrelated disturbances were found for electric machinery and wooden furniture. The Durbin- Watson test did not give conclusive results for some industrial branches. Our estimates using the first model (the efficiency parameter a function of time) are still similar to the ones obtained by Gaviria, Gomez and L6pez and our estimates with the second model are more similar, 29 but lower, than those obtained by Planeacion Nacional. In conclusion, we cannot agree with Gaviria, deez and L6pez when they claim that, contrary to Planeacién Nacional, there are low elasticities of substitution in Colombia's industry and, thus, policies oriented to modify relative 30 Prices are not effective. Important statistical problems Wetc—e found in the models used which do not allow us to k1'l<>w the shape of isoquants. Research on industrial pro- chula-lion functions should be carried out in the future. Labor Input-Output Analysis We use input-output analysis to determine possible 6Effects of composition changes in the aggregate demand for \ 29Lower estimates for Valle than for Colombia could mean more fixed proportion in Valle's industry. However, 1?th results are not totally comparable since Planeacién tacional used price indices of the corresponding manufac- turing Branch to compute real wages while we used implicit price indices of the national social accomts to obtain real wages. 30 We will discuss price policies in Chapter VII. 122 industrial products over the level of employment. Input- output tables are used to study interrelations among sectors of a given economy. If the aggregate demand increases or if its composition is changed, we can trace by means of input-output analysis changes in inputs used and possible bottlenecks in the production process. An input-output table can be transformed to show labor input only and allow 11s; to determine the effect of increases or composition changes in the aggregate demand over the level of employment. Input- output analysis assumes fixed proportions in the production 31 process and for this reason is used for short term analysis. In what follows, we analyze the effect on employment of a change in the composition of the aggregate demand when more c=<>nsumption goods and less durable intermediate and capital goods are wanted. Populatims with low income levels need final products produced with high proportions of labor and domestically produced materials the production of which eIIlploys much manpower, while the high income populations demand goods which contain little labor and a high proportion of imported materials. To examine the effect of a greater demand for consump- tiOn goods and less intermediate and capital goods, an input-output table of the Colombian manufacturing sector Will be utilized transformed to show the labor input of \ 31The factors are perfect complements, i.e. , zero Substitution. 123 each of the industrial sectors. The coefficients of the principal diagonal of the matrix shows direct employment that which an industrial sector generates through the utilization of its own products. The sum of the employment indicated by one line of the matrix gives the employment in the respective industry generated by the industrial sector; and thus subtracting the element of the diagonal from this line yields indirect employment. In order to estimate the composition of the modified demand for final products we need to know the income elasticity for groups or levels of income in each of the economic sectors. Unfortunately there is no available information which permits the estimation of income elasti- cities by industrial sectors. For this reason, the impact on employment of alternative structures of consumption constructed by increasing final consumption goods and decreasing consumption of durable, intermediate and capital goods is examined. The national planning department constructed an input- output matrix in 1956. It was not used due to the untrust- worthiness of its coefficients and the lack of qualified technical personnel. Ten years later, 1966, the department constructed a new table taking advantage of the efforts made by DANE to construct a scheme of industrial relations between the 20 manufacturing branches, to which 11 more branches were later added, covering the national economy: 124 5 branches in the primary sector, 21 in the secondary sector, and 5 in the tertiary sector. The figures were corrected by the accounting of inputs and by the cost of transportation and marketing margins which were included in the prices of the products, since it was not always possible to obtain producers prices, thus underestimating the coefficients which reflect the requirements of inputs in production and underestimating the services of transportation and commerce. The percentages the value of these services represent in the gross produc- tion of each branch of the economy were calculated to make this correction. The Colombian input-output matrix corresponding to the manfuacturing sector was used except for petroleum derivatives. The part of the matrix used is considered applicable to Valle given Valle's high participation in the country's industrial production and its similarity with the industries of Antioquia and Bogota, which together contri- bute nearly 90 percent of the national industrial production. As this is not true for the primary and tertiary sectors, they were excluded. The matrix l-A where A is the matrix of input-output is presented in Table 5.7. Each column of this matrix shows the elements necessary to produce one unit from the respective economic branch for final demand. The inverse of the matrix l-A is called "Leontief's matrix." Since the purpose is to analyze the level of employment, a 125 one. . Hmo. - qu.m n mom. u qnw. . «oo. . Hmo. u msa. n NRA. u mm usowuouuz coauuuuoomauua was. - omH. . Hoo.H u moo.a u Nos. . nNH. u n nmo. u «so. u on unusuaoua ofluuomam oum.a u omo. . o~o.¢~ n «oa.~ u NN~.H . moo. . - -uu nao. u mm knocwnoua owuuooaoucoz omw.m n moH.H . Hom.H¢ u Nan.HH- oo~.n . m¢n.~ - wma. . wo¢.mm . nuw.n u an muuaooum aqua: ooo.n 1 Hon. . m¢a.n u oa¢.m u mno.H . mmH.H . I «No. u ooH. : mm masons oanum -o. . moo. n ¢H¢.m : mmn.a . oHN. . HRH. - u ~m~.~H u «Ho.~ u Nm maunosfia owaaouoansoz ~m¢.¢n . nun.om n m~o.nc n wNH.omu ouo.¢ - com.oc . mmn.~a u omw.oa u coo.NHu Hm aaoofiaonu -H.~ u now. a Non.ma u Nun. u nww.o . Nmm.m - u ooo. u moo. n on ouoaoouo Honoam moH.~ u Hon. . Nm¢.¢ . one. u moo.~¢ . Nam. u s nu- «ma. - on Ammonm unooxov Hoauooq ema.wom omm.¢u . Non.¢ . Nnm.N u NNH.N . ooH.H : nna.nH u mo¢.< u “Ho.m n ma wawusaum oo~.HNHu Hon.nwu ono.a u «no. . «no.5 u Noa.m . o~¢.¢a n moo.~ . oam.nau mm woman own. u moH.¢ u mw¢.mmm oom.H . as». . MHH.H u mmo. . oqa. u owo.a u om unauwousm woos nun. n mmm. . Nom.am - men.¢mo «no. . man. u nmo. u q~a.ma ., moH. . mu poo: mm“. : oHN. . Hmm.ma u moo. u ooo.ooo.a m~H.H . - an: moo. n om wcfinuoao moa. u mo¢.~ . moo.¢m u «an. n on¢.oo¢ . mwo.Hmn Hoo. . omo. u m¢~.ma- mu aoHHuon .2 .3. 3- --- -2. --- 35.03 --- --- mm 8038. moo. u was. . mma.a n mow. . wmo. . omo. u Hoo. u mm~.o¢o mom. u HN mowouo>om moo. . son.m~ - ooN. . «Hm.H - nNo. . Nnm. u on. u oon.mm - mom.~am oN pooh mm mm om mm om mm «N am oN opou _ oauz nosoum wauumsvaH gunman HuwuuonunH .mwoaoaou mo Aom wom.H u oma. u mag. I «No. - wno. : Haa. a ooH.HH u moo. u mno.m u oN pooh km on mm on mm NM an on ow ovoo «Biz noaeum HoauummwcH soauum HowuuoavnH .vossHuaoo .~.n manna 127 factor of transformation (B) must be determined which con- verts Matrix A into coefficients of labor utilization. The coefficients of the labor utilization were calcu- lated multiplying each of the elements of the direct and indirect inputs by the fraction of labor used in the year 1966 over the value of the respective production measured in 1966 prices. These fractions for each industrial branch were examined and it was found that they had remained quite stable during the 1956 to 1969 period. These coefficients are presented in Table B.7 of Appendix B. The result of multiplying the input-output matrix A of the year 1966 by the coefficients of the use of labor (the transformation factor B) is matrix C which is the natrix of coefficients of labor utilization in the different industrial branches. This matrix is presented in Table 5.8. If matrix (l-A) is mutliplied by the value of the production of the different industrial branches, the result is the total demand of other sectors for industrial goods (l-A) X = d. If income is redistributed, the composition of "d" will vary and we will have, for example, do a (l-A)'1Xo and X0 may be obtained by (l-A) do' Utilizing the transformation factor B we can resolve the above relations in terms of employment. Total indus- trial employment with the present aggregate demand will be: CX + Bd and industrial employment with the modified aggregate 1128 ooo. Hoo. omo. Hoo. omo. --- moo. moo. ooo. mm omomuoooa comuouuooocous omo. moo. omo. mmo. ooo. moo. --- --- moo. om muocmaooa omuuoomm moo. ooo. omm.o mmm. omo. --- --- --- moo. mm muoamoooa omuuoomouaoz omm. omo. omm.m omm. omo. mmo. moo. mmo.m omm. om ouoaoouo moon: mom. Hmo. omm. mom. mmo. mmo. --- moo. moo. mm omouoa omoom omo. mmo. Hoo. omo. moo. moo. --- mom. omm. mm omouoama omHHouoanaoz omm. ooo. omo. omo. ooo. omm. oom. mmm. mmm. mm omoomaonu omo. ooo. mmm. moo. omo. omo. --- --- moo. om oposoouo poooam moo. mmo. mmm. mmo. omm.m mmo. --- --- moo. om flooooo uoooxoo Honuooq mom. mmo. mmm. mmo. omo. mmo. mom. omm. omo. mm wcmuaauo moo.m omo.m ooo. moo. moo. omo. omm. mmo. oom. mm nooom mmo. omm. mmm. mom. moo. ooo. moo. mmo. ooo. om ousomouao coo: mmo. moo. omm.o mom.m omo. omo. moo. mom. omo. mm ooo: omo. mmo. mom. mmo. --- omo. --- --- --- om womnoomo moo. moo. moo. mmo. omo.m oom.o --- moo. mmm. mm oomaouom nu: uuu -I- nun nu- In: moo. nun nun NN oooonoa moo. moo. omo. moo. --- --- --- moo. ooo. mm oomouo>om --- mom. moo. mmo. --- moo. --- moo. mmo. om ooom mm mm om mm om mm mm mm om oooo oaoz mooou nocoum HomuuoaocH accoum Homuuunvau oHHo> mo Houoom HomuuosocH onu cm ucoahoaqam .muoxuoB mo oooomoommooo moomonoom mo xmuooz .o.m omoom 2129 omm.~ on“. oom. oom. ooo. omo. omm. ooo. omo. mm mmomuouuz somuouuomusoua ooo. mom. «mm. oom. «mm. woo. moo. Noo. mmo. on huoamnqma omuuoomu omo. mmm. oom.m omm.m mmo. Noo. omo. woo. woo. on humomnoua omuooomousoz mmo. oo~.m mom. mom.~ mmo. mom. omo. Now. mow. on ouoooouo moon: oom. aom. mmo.~ omo.m moo.o oom. mmo. omo. moo. Mm mmuuoa omaom mom. «on. mmm. mmm. mom. oom.~ moo. omo. Nam.m ~m umouonma ommmouoauaoz ooN. omm. mom. mom. moo. mom. «no. now. one. mm umoomaonu mmm. ooo. mmo. moo. woo. woo. mmo. one. omo. on muoaoouo nonnam omo. mmo. ooo. mmo. Noo. ooo. mmo. won. mmo.m mm mucosa uoooxov nonuoom moo. moo. woo. mom. uuu mmo. ooN. ooo. ooo. mm womusmum moo. omo. moo. omo. moo. omN. omm. moo. omo. mm uooum mmm. mum.m omo. own. moo. omm. oon. omo. omo. om oudumouam coo: mmoo omm. non. mom. Noo. omo. ooo. moo. on.m nu woos mmm. omo. wow. omo. nun moo. moo. moo. mom. om wcmnuomo omo. mmo. mmo. mmm. nu- moo. mmm. mom. moo. mm mommuxoa nu- nu- ..u --u uuu --- nu- .. n.: «N oooonoa --- --- moo. mmo. --- --- --- moo. moo. mm oomouo>om mmo. woo. moo. nu- moo. moo. mum. moo. mmo.. o~ coon mm on on on mm mm mm on ma ovoo oaoz mooou Susana momuumaocm noooum momuuuooam .oosomuoou .m.m «Home 130 demand will be: c(l-A) ‘1d0 + Bdo' The difference (CXo + Bdo) - (CX - Bd) will show the change in the level of industrial employment. It was assumed four different changes in the total aggregate demand. The first one consisted of a 20 percent increase in the demand for food and beverages and a 10 percent increase in the demand for textiles and clothes. In order to hold total effective demand constant, all the other industrial branches were reduced in proportion to their participation in the demand. (See do in Table 5.9.) The other three aggregate demands, d1, d2, and d3, resulted from a 10, 20 and 30 percent increase in the demand for food and beverages, respectively, a 5, 10, 15 percent increase in the demand for textiles and clothing, respectively, and no change in the demand for tobacco, wood and wooden furniture. All other industrial branches were reduced in proportion to their participation in the aggregate demand. The present (d) and modified demands (do, d1, d2, d3) for industrial products are presented in Table 5.9. The industrial employment level with the present aggregate demand is given in Table 5.10. Employment within the industrial sector was determined by the interindustrial relations differentiating direct and indirect employment and industrial employment caused by the demand for indus- trial products external to the industrial sector. The first type of employment, that caused by the demand for industrial products from the sector itself, was calculated by the 131 .uouomno mmnu cm oocmmmmxo mm nonosm ocu an ooummnommo mp3 muosooua momuumdoam How ocmaoo oommmooz .mzom omm.mmm mom.mmm mmo.mmm mom.mmm omm.omo oooo mo mo mo oo o mocmBoD oommmooz ocmaoa ucomoume socmum momuumsocm oommmooz vow odomoum .momm .ommo> .mpamaoo oumwouww< “muosooum mmmuumnocm mow muouoom nonuo Scum panama mocmm .o.m oman Table 5.10. 132 Aggregate Demand, 1969. Industrial Employment in Valle with Present Industrial Industrial Branch Employment In Industrial Sector (ABX) Other Sectors Direct Indirect Total (BD) Total Food 964.4 1147.6 2112.0 9778.2 11890.2 Beverages 65.7 6.0 71.7 1148.1 1219.8 Tobacco 2.0 0.0 2.0 598.8 600.7 Textiles 746.9 1834.2 2581.2 1211.6 3792.8 Clothing 0.0 138.7 138.7 4097.4 4236.1 Wood 93.7 202.4 296.1 991.5 1287.5 WOoden furniture 1.8 58.0 59.8 582.3 642.1 Paper 800.4 650.9 1451.3 1829.6 3281.0 Printing 112.7 264.0 376.7 2977.6 3354.2 Leather (except shoes) 39.7 58.3 97.9 511.3 609.2 Rubber Products 96.6 184.1 280.6 2667.4 2948.1 Chemicals 326.4 1741.5 2067.9 3370.7 5438.6 Non-metallic minerals 103.2 176.4 279.6 787.3 1066.9 Basic metals 560.8 362.2 923.0 902.9 1825.9 Metallic products 321.1 1170.2 1491.3 3549.2 5040.4 an—electric machinery 43.6 100.0 143.5 992.7 1136.3 Electric machinery 100.8 135.8 236.6 1785.4 2022.1 Transportation Materials 105.6 99.6 205.2 1942.3 2147.5 Totals 4485.2 8329.9 12815.1 39724.2' 52539.2 133 expression ABX, and the second type is given by BD, as explained above.32 Tables 5.11, 5.12, 5.13 and 5.14 present the data on employment with the modified aggregate demands discriminated in the same form as Table 5.10. A comparison of Tables 5.10 to 5.14 shows that employ- ment diminishes slightly as a consequence of the change in the structure of the demand to reflect higher demand for consumption goods. This slight reduction is due to the demand for industrial products from the industrial sector itself. The demand of the other sectors for industrial pro- ducts caused no variation whatever in employment. Similar results were observed when the demands for food, beverages, textiles and clothes were increased by higher rates. The conclusion is that higher demand for manufacturing consumption goods and an equal decrease in the demand for intermediate, durable and capital goods has no major impact on industrial employment. We conclude that the Valle's manufacturing firms are using labor-saving capital extensively in all manufacturing branches and thus they produce large proportion of goods.33 32The total industrial employment with the present aggregate industrial demand in 1969 given for the labor input-output table shows a decrease in the number of laborers of 3,827 (See Tables 5.1 and 5.10). Changes in the industrial employment by branches can be seen by comparing Table 5.10 and Appendix B, Table B.2. 33If small firms with other kinds of capital were pro- ducing substantial amounts of manufacturing consumption goods, of production more accessible to small firms, labor employ- uent should be increased when there is a large demand for consumption goods. Table 5.11. 134- Aggregate Demand do, 1969. Industrial Employment in Valle with Mbdified Industrial Industrial Branch Employment In Industrial Sector (ABX) Other Sectors Direct Indirect Total (BD) Total Food 1147.8 1365.8 2513.6 11704.0 14217.6 Beverages 78.7 7.2 85.9 1377.7 1463.6 Tobacco 1.7 0.0 1.7 529.0 530.7 Textiles 808.8 1986.3 2795.1 1332.8 4127.9 Clothing 0.0 152.2 152.2 4507.1 4659.2 Wood 84.9 183.5 268.5 872.3 1140.8 Wooden furniture 1.6 52.3 53.9 513.0 566.9 Paper 718.0 584.0 1302.0 1612.3 2914.3 Printing 100.1 234.7 334.8 2622.7 2957.5 Leather except shoes 36.3 53.3 89.5 450.4 539.9 Rubber Products 85.3 162.5 247.8 2351.0 2598.8 Chemicals 292.4 1559.9 1852.3 2970.4 4822.7 an-metallic minerals 93.9 160.5 254.4 692.0 946.3 Basic metals 516.5 333.7 850.2 836.0 1686.2 Metallic products 290.9 1060.1 1351.0 3128.9 4479.9 an-electric machinery 38.6 88.4 127.0 873.9 1000.9 Electric machinery 88.9 119.8 208.7 1573.1 1781.9 Transportation Materials 93.3 88.0 181.3 1712.3 1893.5 Totals 4477.8 8192.1 12670.0 39658.8‘ 52328.8 Table 5.12. 135 Aggregate Demand d1, 1969. Industrial Employment in Valle with Modified Industrial Industrial Branch Employment In Industrial Sector (ABX) Other Sector Direct Indirect Total (BD) Total Food 1057.4 1258.3 2315.7 10756.0 13071.7 Beverages 72.2 6.6 78.8 1262.9 1341.7 Tobacco 2.0 0.0 2.0 598.8 600.7 Textiles 777.8 1910.1 2687.9 1272.3 3960.2 Clothing 0.0 145.4 145.4 4302.2 4447.7 WOod 93.6 202.3 295.9 991.5 1287.4 WOOden furniture 1.8 56.8 58.6 582.3 640.9 Paper 755.6 614.5 1370.2 1711.0 3081.2 Printing 106.5 249.4 355.9 2801.8 3157.7 Leather (except shoes) 38.2 56.1 94.2 484.8 579.0 Rubber products 90.5 172.4 262.9 2496.2 2759.1 Chemicals 308.0 1643.5 1951.5 3153.7 5105.2 NOn-metallic minerals 98.7 168.6 267.3 742.8 1010.0 Basic metals 524.2 338.6 862.7 840.0 1702.7 Metallic products 306.3 1116.3 1422.5 3343.2 4765.8 Non-electric machinery 41.1 94.2 135.3 932.7 1067.9 Electric machinery 94.9 127.9 222.9 1681.6 1904.5 Transportation materials 98.6 93.1 191.7 1811.4 2003.1 Total 4467.3 8254.1 12721.4 39765.2 52486.6 Table 5.13. 136 Aggregate Demand, d2, 1969. Industrial Employment in valle with Modified Industrial Industrial Branch Employment In Industrial Sector (ABX) Other Sectors Direct Indirect Total (BD) Total Food 1150.5 1369.0 2519.4 11733.8 14253.3 Beverages 78.7 7.0 85.9 1377.7 1463.6 Tobacco 2.0 0.0 2.0 598.8 600.7 Textiles 808.6 1985.8 2794.4 1332.8 4127.2 Clothing 0.0 152.2 152.2 4507.1 4659.2 Wbod 93.6 202.2 295.8 991.5 1287.3 wooden furniture 1.7 55.6 57.4 582.3 639.7 Paper 710.9 578.1 1289.0 1592.4 2881.4 Printing 100.2 234.9 335.1 2626.0 2961.2 Leather (except shoes) 36.7 53.9 90.5 458.4 548.9 Rubber products 84.4 160.8 245.2 2325.1 2570.3 Chemicals. 289.7 1545.5 1835.2 2936.7 4771.8 Non-metallic minerals 94.1 160.8 255.0 698.2 953.2 Basic metals 487.5 314.9 802.5 777.1 1579.6 Metallic products 291.5 1062.3 1353.8 3137.4 4491.2 an-electric machinery 38.6 88.4 127.0 872.5 999.5 Electric machinery 89.1 120.1 209.2 1577.8 1786.9 Transportation materials 91.7 86.5 178.1 1680.5 1858.7 Total 4449.4 8178.2 12627.6 39806.1 52433.7 Table 5.14. 137 39 1969. Industrial Employment in Valle with Mbdified Industrial Aggregate Demand, d Industrial Branch Employment In Industrial Sector (ABX) Other Sectors Direct Indirect Total (BD) Total Food 1243.5 1479.7 2723.2 12711.6 15434.8 Beverages 85.3 7.8 93.0 1492.5 1585.6 Tobacco 2.0 0.0 2.0 598.8 600.7 Textiles ‘ 839.5 2061.6 2901.1. 1393.4 4294.4 Clothing 0.0 158.9 158.9 4711.9 4870.8 WOod 93.5 202.1 295.7 991.5 1287.1 Wooden furniture 1.7 54.5 56.2 582.3 638.5 Paper 666.1 541.8 1207.9 1473.8 2681.7 Printing 94.0 220.3 314.4 2450.3 2764.6 Leather (except shoes) 35.2 51.6 86.8 431.9 518.7 Rubber products 78.3 149.2 227.4 2153.9 2381.4 Chemicals 271.3 1447.5 1718.8 2719.7 4438.5 Non-metallic minerals 89.6 153.1 242.6 653.7 896.3 Basic metals 450.9 291.3 742.2 714.3 1456.5 Metallic products 276.7 1008.4 1285.1 2931.5 4216.6 Non-electric machinery 36.0 82.7 118.7 812.4 931.2 Electric machinery 83.2 112.2 195.4 1473.9 1669.4 Transportation materials 84.7 79.9 164.6 1549.7 1714.3 Total 4431.5 8102.4 12533.9 39847.0 52381.0 138 Now we study the labor employment by size of firms in the Valle's manufacturing sector. Labor Employment and Size of Firms 34 Our hypothetical four groups of firms stressed: (1) limited monetary funds for smaller firms, (2) higher labor costs as firm size increases, and (3) different kinds of tech- nology between large and small firms. Many large firms using modern technology are stock companies, some of them with foreign capital. They have access to credit at reasonable terms since they are able to fulfill bank requirements. As noted by Richard Cooper35 large firms "tend to get strong effective protection for their products, hence charge very high prices, and their high value added per worker largely reflects monopoly rents." Real net rates of return on capital of 30 to 40 percent or more36 provide these firms with substantial amounts of funds to reinvest. In Table 5.15 we see that large firms of Valle, those with 50 or more workers, have expanded from 1963 to 1966 and from 1966 to 196937 at a rate of 5 percent and 1963 34See pages 105-110. 35See Nelson, Schultz and Slighton, QB. cit., p. 122. 36 37Data relating number of firms to size were only available for 1962, 1966 and 1969. Ibid., p. 123. 139 to 1969 at a rate of 10 percent while medium size firms have increased in number at lower rates or tend to decrease. Data on wage rates by firm size were not available for Valle. Nelson, Schultz, and Slighton published this kind of data for industrial branches of Colombia for the year 1958.38 The data show a consistent tendency to increase as the firm size increases for all branches. Unfortunately the firm size groups are less disaggregated than the size groups we were able to use in Table 5.15. Furthermore, changes in wages only take into account monetary wages not including the non-monetary costs of unionized labor discussed above. In Colombia the minimum number of workers to form a union is 25. If the non-monetary costs of unionized labor is substantial and this combines with the impossibility of firms of about this size (15 to 30 workers) to lower other costs of production (i.e., using labor-saving capital), firms will try to keep the number of workers below 25 or go out of business. Data relating numbers of firms to size on Table 5.15 show that a decreasing number of firms only occurred for firms with 15 to 19 and 19 to 24 laborers between 1963 and 1966. In the other firm sizes, larger numbers of firms with l to 4, 5 to 9, and 10 to 14 were found and, as we already explained, the increase in the craftsmen and small firms are associated with migration to the urban areas; largest firms (more than 50 workers) have 3892. cit., Table 23, p. 119. 140 Table 5.15. Size of Valle's Craftsmen and Manufacturing Enterprises According to the Number of Employees, 1963, 1966 and 1969. Number of Number of Enterprises in Years WOrkers * ' 1963 1966 1969 Increase Increase 1963-1966 1966-1969 ----- Number------- -------Percent-------- 1 to 4 469 493 NA1 5.1 --- 5 to 9 534 594 NA 1.1 --- 10 to 14 221 296 NA 3.4 --- 15 to 19 73 58 NA -2.1 --- 20 to 24 73 45 NA -3.8 --- 25 to 49 132 142 NA 7.6 --- 50 to 74 56 49 58 1.3 18.4 75 to 99 26 39 40 50.0 2.6 100 to 199 55 55 49 --- —12.2 200 and over 48 51 57 6.3 11.8 1Not available. Source: Computed from unpublished data of DANE. increased in about 10 percent from.l963 to 1969. During the period 1966 to 1969 a decrease of six firms with 100 to 199 employees while the firms with 200 or more employees increased in exactly the same number. . Furthermore, we note there is a decay curve relating number of firms to size and that there are fewer firms as number of employees increases. However, for firms with 15 to 24 workers the curve levels off for firms with 15 to 24 workers, as the 1963 curve shows. A sharp decrease in the number of firms with just over 25 workers was expected. 141 Unfortunately information was not available for the group of companies with 25 to 29 workers. Laursen and Taylor using a model for dual technology39 (firms with more than 100 employees using modern technology and firms with less than nine laborers using traditional technology) showed that: (a) in the modern sector a 3.5 percent growth in labor employment requires equilibrium growth of 7.8 percent for product and capital, and a 5 percent growth in labor employment requires equilibrium growth of 10 percent for product and capital; (b) in the traditional sector a 3.5 percent growth in labor employment requires equilibrium growth of 1.9 percent for product and capital, and a one percent growth in labor employment requires equilibrium growth of 0.5 percent for product and capital.40 Thus the marginal labor productivity and marginal capital-labor ratio are 0.5 for the traditional sector and 2 for the modern sector. With respect to equilibrium rates of change on wages in the two sub-sectors Laursen and Taylor estimated that (a) 39Karsten Laursen and Lester Taylor, "Desempleo, Productividad y Crecimiento en Colombia, " Revista* de Planeacidg y Desarrollo, Vol. 1, No. 2. Bogota, Departamento Nacional de Planeacion, Julio 1969. Laursen and Taylor used Cobb- -Douglas 1production functions with technological change, Y a ae°t(L“K1'a) The values of the parameters are: (a) for the modern sector, labor share a = 0.27, economies of scale A = 1.10 and technological changes 0 = 0.005, and (b) for the traditional sector, a = 0. 50, A = 0.75 ando . 4092. cit., pp. 76-77. 142 in the modern sector a 3.5 percent growth in labor employ- ment requires an increase on wages of 4.6 percent, and a 5 percent growth in labor employment requires an increase on wages of 5 percent; (b) in the traditional sector a 3.5 percent growth in labor employment requires a decrease on wages of 1.6 percent, and a one percent growth rate of labor employment requires a decrease of 0.5 percent in the traditional sector.41 PIMUR“2 found that small and medium size firms of Valle producing consumption goods (clothes, shoes, domestic goods) have serious problems with the cost of unionized labor. The large firms advantage situation, the large number of craftsmen and the decreasing number of medium size firms have established a high degree dualism in Valle's industry, a phenomenon consistent with the uneven income distribu- tion in the urban areas including Cali, discussed in Chapter II. Small and medium sized firms frequently find labor cost relatively expensive, that is, the ratio between MVP and MFC of labor appears to be less than the ratio between MVP and MFC of capital. 41 42Proyecto de Mercadeo Integrado Urbano-Rural del Valle, PIMUR, Technical Report No. 4, The Production and Distribution of Selected Consumer Goods in the Cali Area, 1969, pp. 42'and 98 of Spanish version. Idem. 143 Use of Industrial Installed Capacity The National Planning Department estimated the rate of utilization for the whole country through a sample of industries.43 The Planning Office found that except for some lines of textile production, there was a low utiliza- tion of installed capacity. In this study, the rate of utilization of equipment was calculated on the basis of the maximum number of shifts worked in each enterprise. These rates are presented in Table 5.16. A low rate of utiliza- tion of equipment is the rule for all the industrial sectors listed, especially in the industrial branches of foods, and capital producing goods: metal products, non-electric machinery and transportation material. The rates of utilization of equipment are even lower if the maximum.number of work shifts is considered. Little excess capacity was found in the construction sector by Planeacion Nacional and the Instituto de Crédito 44 Territorial as reported by Urrutia. Full use capacity was found in cement and glass industry. 43Departamento Nacional de Planeacion, DNP, "Comentar- ios Preliminares de Los Resultados de la Encuesta, Sobre La Utilizacion de la Capacidad Instalada, en la Industria Manufacturera Fabril de Colombia," Bogota, DNP, UE14-009-DI,I August 10, 1971. - 44Miguel Urrutia Montoya, "Problemas de Oferta de Factores de Produccion en la Nueva Estrategia de Desarrollo in CORP, Controversia sobre el Plan de Desarrollo. Bogota: Editorial La Oveja Negra, 1972, pp. 102-106. 144 Table 5.16. Rates of Utilization of Colombian Industrial Equipment Based on the Maximum Number of Shifts Worked, by Industry, 1971. FOODSTUFFS Percentage Preparation and conservation of meats and fish 44 Milk products 60 Bottling and conservation of fruits and vegetables 59 Production and refining of sugar 67 Vegetable and animal oils and fats 67 Bread, cookies 79 Sweets, with and without chocolate 74 TEXTILES Cotton threads 100 Cotton weaves and manufactures 100 Pure wool cloth and articles 100 Cloth and articles of artificial and other fibers 88 Handwoven cloth, manufacture of pants and nylon stockings 75 Manufacture of underwear 70 Cloth and manufactures of silk 81 SHOES CLOTHING AND OTHER TEXTILE PRODUCTS Shoes except rubber shoes 63 Mens wool suits 100 Linen, silk, cotton suits 55 Manufacture of clothes, bluejeans, slacks, pijamas, housecoats, work clothes, men's children's clothes 61 Manufacture of shirts 79 Women's clothing 62 WOOD AND LUMBER INDUSTRY EXCEPT FURNITURE Sawmill, wood polishing, fabrication of strips and white molding 74 Fabrication of wooden furniture and accessories 88 FABRICATION OF PAPER AND PAPER PRODUCTS Wood pulp. paper and cardboard 62 Paper and cardboard articles 70 PRINTING, PUBLISHINQLgAND RELATED INDUSTRIES Typography, lithograph's, publishing and related fields 69 LEATHER INDUSTRY AND LEATHER PRODUCTS EXCEPT SHOES AND CLOTHING Tanneries and Accounting 67 Fabrication of substances and chemical products 67 FABRICATION OF NON-METALLIC MINERAL PRODUCTS Clay products for construction 62 Glass and glass products 83 Cement 9O Cement and asbestos artifacts 71 FABRICATION OF METAL PRODUCTS EXCEPT MACHINERY AND EQUIPMENT FOR TRANSPORTATION Tinplate and other metal articles 38 Handmachinery and accessories for agricultural use 67 Razor blades and domestic knives 74 Fabrication of gas and petroleum stoves 70 Aluminum articles for domestic use 55 Clamps, wire hangers for clothes and hair, zippers, clips, screws, locks, etc. 52 Metallic furniture and equipment for office, home, schools 57 145 Table 5.16. Continued. NON-ELECTRIC MACHINERY Percentage Agricultural machinery 49 Machinery and pumps, wood working machines, sewing machines, spare parts and replacements for industrial machinery 61 Machinery for the transportation industry 77 ELECTRIC MACHINERY Transformers, rheostats, current chargers, radioreceptive apparatus, electric cables 80 Electric articles for the home, including refrigerators 30 Transportation material 56 Source: Departamento Nacional de Planeacion, "Comentarios Preliminares de los Resultados de la Encuesta sobre Utilizacién de la Capacidad instalada en la Industria manufacturers fabril de Colombia,” Bogota: DNR, Documento UEIA-OO9—DI, August 10, 1971. 146 There are two studies of the utilization of installed 45 carried out in 1969 capacity in Valle. The first study examined firms in the following six industries: (1) textiles, (2) ready-made clothes, (3) electrical appliances, (4) shoes, (5) agricultural hand tools, and (6) domestic utensils. Substantial excess capacity was found in all industries with the exception of the textile industry. The textile industry was being operated at 85 to 100 percent of three- shift capacity, the ready-made clothing industry at about 40 percent of one-shift capacity, the electrical appliance industry at only 25 percent of one-shift capacity, the agri- cultural hand tools industry at 40 percent of one-shift capacity, and domestic utensils industry at 40 percent to 50 percent of one-shift capacity. Data on plant capacity for the shoes industry were not possible to secure. The 46 was carried out in 1973. The utilization second study of the installed capacity of the metal machinery industry was examined. It found that the lathes were utilized only 50 percent of the possible time. It was calculated that 76 percent of the enterprises only worked one shift, and that 80 percent of these worked 8 hours a day. 45Harold Riley, et al., Market Coordination_in the Development of the Cauca ValleyfRegion-Colombia. Research Report—No. 5, Latin American Studies Center, Michigan State University, pp. 295-297. See also PIMUR, Technical Report No. 4, 92. gig. 46Yesid Isaza, FEDEMETAL, seccional del Valle, "Estudio de la Capacidad Instalada de la Industria Metalmecénica en el Valle del Cauca," FEDEMETAL, abril 1973. 147 The reasons for underutilization of industrial equip- ment in all industrial branches but textiles indicated most frequently by the owners of these enterprises, in the study carried out by the National Planning Office, are the following: scarcity and irregular flux of primary materials, scarcity of bottles and packages, lack of capital, delays in the approving of licenses for importation of supplies, labor legislation which makes more than one shift too expensive, scarcity of techniques and qualified personnel, bottlenecks with some final products necessary for the construction of others, errors in the planning of the size of the plant, lack of internal demand, and difficulties for export. Unutilized capital capacity in industrial installa- tions is profitable in our Group III and Group IV firms. When the entrepreneurs have expectations of future increases in demand, the cost of installed capacity decreases through time. Under uncertain conditions of demand, the entrepreneur 47 at the opportunity cost of may decide to buy flexibility unutilized capacity. Economies of scale tend to favor unutilized capacity when there are expectations of increas- ing demand for the product. When the cost of labor increases *with firm size, as shown above, and when the cost of labor is more expensive for night work the owner is led to install labor-saving equipment even though he does not plan to use it to full capacity since the fixed cost of equipment is 47This concept was defined in Chapter III. 148 less than for unionized labor. It is well known that most labor-saving equipment is produced in developed countries where the wage rates are high and labor relatively scarce. The lack of an internal market is in part then, a conse- quence of importing labor-saving equipment with a great productive capacity, and the real cause of its low utiliza- tion is the relative high cost of labor (monetary and non-monetary) which led some owner to buy such equipment. When more than one work shift is used, the monetary and non-monetary cost of labor is increased since the law requires extra payment for night work and daytime overtime and triple pay for Sundays and other holidays. With more than one shift some firms will need more than 24 workers and will have to deal with unionized labor and, hence, higher wages. The marginal value product of additional manpower is high since new laborers utilize idle capacity, but may be lower than the monetary and nonmonetary cost of unionized labor. The increased cost of labor for additional shifts does imply the employment of fewer workers than that which would be the case if the salaries and wages were maintained at the same level as those of the first shift. Thus, the decision is not whether or not to use additional shifts, but rather how many workers to employ on those shifts. Why not contract additional shifts of fewer workers? Again, the answer is the high cost of labor when the workers can form.unions. It is also possible that new shifts require 149 skilled labor which is not available48 and/or the increase in production causes a decrease in the price of the product and with an inelastic demand for the product it does not pay to contract additional shifts. Other Factors Affecting the Manufacturing Industry Growth Rates Factors determining the rate of growth of manufac- turing industry, especially the modern sector, have been 49 We have seen analyzed extensively in different studies. how the craftsmen industry increased substantially from 1963 to 1966, the slow growth rate of the small and medium size firms and the relatively high growth rate of the large modern sectors from.l963 to 1969. Other factors for industrial growth not considered above are summarized in what follows. Low growth rates of the manufacturing industry during some periods of time have been associated with scarce foreign exchange earnings and unfavorable terms of trade for the country which seriously limited the purchase of capital equipment and intermediate goods.50 This factor, the small 48This is not likely to be the case as explained in the following section. 49Nelson, Schultz, and Slighton, 9%. £15., Oficina Internacional del Trabajo, 92. cit., Gav ria, Gomez and L6pez, 22. £15., and CORP, 92. gig. 50ILO, estimated the following equation log IM. = 3.51 + 1.036 log Zt + 1.508 log At, R2 = 0.8 where IMt = invest- ment in machinery in year t, Zt = stock of foreign exchange earnings in year t, A = terms of trade in year t. (See Appendix A of this study.) 150 internal markets and the high cost of some supplier require- ments appear as the three most important factors determining the growth rate of modern industry. In conclusion, low labor absorption on the part of the Valle's industry results from labor-saving capital being used in the majority of the industrial branches, especially intermediate, durable and capital manufactured goods as well as some consumption goods. The number of craftsmen industry has increased as a result of the high number of migrants trying to make a living. Small and medium size manufacturing industries occupy a very low proportion of the labor force. Some of these industries with limited access to credit, cannot survive the pressure of higher wages, especially when labor is unionized, or hold the number of workers below 25 to avoid higher cost of labor or use of labor-saving technology. Large firms operate with a higher capital-labor ratio than small firms and pay higher wages. Industrial dualism is the rule in Valle with its associated uneven income distribution. Part of the industrial capital remains underutilized because of labor-saving capital installed has a capacity in excess of what the markets can absorb. Consumption goods and the construction sector have low excess capacity. MOst of the unskilled urban labor force is employed in the service sector or is unemployed as shown in the following chapter. CHAPTER VI URBAN UNEMPLOYMENT RATES Employment levels and equilibrium wage rates are determined by the intersection of the supply and demand functions for labor. In the case of unskilled labor,1 the supply is determined by the minimum real wage and the number of persons from 15 to 64 willing to work. The demand for unskilled labor is low in the industrial sector and most of the urban labor is absorbed in the service sector. In Figure 6 the market for unskilled workers is illustred. P xu S D A \E S3 B D 0 5X u Figure 6. Function of supply and demand in the unskilled labor market. 1The union's effects are ignored here. 151 152 Segment AE is employed workers and segment EB unemployed workers, at the minimum wage level. The high rate of unemploy- ment in Valle and Cali is explained in the first place by the abundant supply of labor; this abundance is caused by the high natural population growth rates of Cali and Valle, the low rates of transfer from the unskilled to the skilled category, and migration. Second, there is a low demand for industrial workers, as discussed in Chapter V. Other economic sectors of the urban areas of Valle, especially the service sector absorb substantial amounts of laborers (see Table 5.1) but not enough to absorb all those willing to work in the urban areas. The sectors employing the greatest number of persons in Cali are manufacturers, commerce and services. In 1969, 71.5 percent of all persons employed were working in these three sectors. In the same year, the construction industry employed only 6.1 percent.2 Unemployment Rates, 1963-1971 By open unemployment we understand the excess of the supply of labor at the prevailing wage in the labor market. Thus unemployment of unskilled labor is segment EB in Figure 6. Open unemployment is usually divided into structural and Keynesian unemployment. The second type is that which is 2Planeacion Municipal de Cali, Plan General de Desarrollo. Encuesta Urbana de Empleos e Ingresos del Area Ufbana as Cali. Ca1i,71969. 153 originated by an insufficient aggregate demand and may be corrected by the proper use of fiscal and monetary policy. Structural unemployment is that which is created by the incapacity of the economic system to absorb the excess of labor given the characteristics of the labor market. The great mass of unemployment in Cali and Valle is unskilled labor, and the present productive structure is unable to utilize it. Thus unemployment in Cali and Valle is struc- tural unemployment caused by factors discussed in the previous chapters. MOreover, many persons considered "employed" are working as street vendors, shoeshine boys, occasional watchmen of cars parked in public streets, etc. This is partial employment, and less than the legal minimum wages are earned and thus, these persons should really be included in the group of unemployed at the legal wage which is greater than the equilibrium wage rate. Thus the figures on unemployment are underestimated. ILO3 estimated occupied persons but with incomes beIOW'ZOO pg§g§_monthly (one dollar is equal to 27 pgggg) at 12 percent of the active urban labor force. unemployment has also been subdivided in open unemploy- nent and disguised unemployment. Open unemployment applies to unemployed openly seeking for jobs and disguised unemploy- ment to those persons without work and who would probably 3 Oficina Internacional del Tarabajo, OIT, op, cit., p. 18. 154 4 ILO estimated seek it if unemployment were much lower. open unemployment at 14 percent of the Colombian active urban labor force and disguised unemployment at 7 percent 5 Thus, of the Colombian active urban labor force in 1967. total unemployment, that is open plus disguised unemployment, is 21 percent of the Colombian active urban labor force. Riley, et al., found that in Cali 20 percent of the active labor force was unemployed in 1969.6 Unemployment can also be subdivided in frictional and nonfriCtional. Frictional unemployment describes unemployed persons in the process of changing jobs. Seasonal jobs exist in both urban and rural areas. In the city of Cali there are seasonal jobs such as those connectedwith Christmas and the Fair in the month of December. Those who are employed for these jobs often suffer open unemployment most of the year. Unemployment rates for Cali for different months of a year have not been measured. For Bogota surveys on unemployment have been carried out every three months for several years. The data showed lower levels of unemployment in December.7 4Idem. 5 6Riley, et. al., 2%. 315., p. 16. See also, Proyecto de Mercadeo Integrado Ur ano-Rural del Valle, PIMUR, Tech- nical Report No. 16, An Economic Analysis of Residential Construction in Cali, 1969. Idem. 7Universidad de Los Andes, CEDE, Encuesta Urbana de Empleo y Desempleo. Bogota: Universidad de Los Andes, 1969. 155 Another phenomenon faced by the labor force is under- employment, that is, those laborers working less than a given number of hours a week or less than a given number of months per year. ILO estimated open underemployment, persons working less than 32 hours per week and seeking to work longer, at 2 percent of the Colombian active urban labor force; and disguised underemployment, persons working less than 32 hours per week, who would probably seek longer hours if the opportunity were available, at 3 percent of the Colombian active urban labor force.8 In Table 6.1 we estimated the weighted average of months worked in Valle during the year 1964 by the econo- mically active population based on data reported in the 1964 census. This average fluctuates between 8 and 8.5 months depending on the type of work, and the average is higher for men than for women. Unemployment rates, the ratio between those unemployed openly seeking jobs and the urban labor force, have been estimated for twelve major cities of Colombia during the period 1963 to 1969.9 These unemployment rates have 8Oficina Internacional del Trabajo, OIT, 92, cit., p. 18. 9Ibid., Table 5, pp. 398-9. The unemployment rates were estimated with surveys carried out by different govern- ment and university personnel, like Centro de Estudios de Desarrollo Econdmico, CEDE, Empleo_y Desempleo, Bogota: Universidad de los Andes, 1968; Centro He Investigaciones de Desrrollo Economico, CIDE, Encuesta de Em leos e Ingresos, Cali: Universidad deI VaIIe, 1965; PIaneacién 156 .meaa .mzaa ”mnemom =.eo=mo Hop mHHm> we soaamom .eoma .smwowaoom ow Hmsowomz omaoo HHHN: .mzs omucwfioz omo.ooa mam.mm~ mma.mmm owm.e moa.mea mmn.mas oea.eoa sma.-e eem.m~m sauce em~.He Ham.mea sem.e- awo.e mum.moa mae.aoa mam.me amm.we~ ~HN.emm masses messes eHe.~ ewa.a ooe.e Hem Hma.aa ~ea.~a New.~ mmm.ws Nea.H~ nausea seesaw mm~.m eHe.a e~w.oa Has moe.m owe.o~ see.m m-.aa sem.o~ masses use ewe.a mam.m soo.m em ewe.~ oaa.~ ~am.~ Nes.e eaa.a messes meez mam.~ mme.a mmm.oa ewN -m.m HH~.e ewa.m cae.ma ema.ea ensues eases Hms.m ooe.m Hmo.~H com sam.s s~o.m Hue.m sas.ma mao.aa messes em>mm mo~.m omo.e~ mm~.mm ode mee.ea Hm~.as HNm.a mae.oe eam.on masses sew m~m.~ emm.e mms.a sea aoo.m mmm.~ mae.~ mmm.a oao.oa ensues seem aao.m oun.e ome.m HHN «Ho.m ~o~.m mw~.n emm.HH -e.ea ensues seem eo~.m mee.m Nam.m Hes eNN.~ omm.~ mem.m amm.a ~e~.HH messes amuse mHa.~ aHm.m mme.e ems New.” H~a.a aam.~ mae.m emm.m masses 0:9 oaw.e nma.a moo.ea mNH H-.~ eem.~ mmm.e eHe.HH «mm.ea nudes use doaoz :0: Homes auaoz aux Hmuoa doaos so: Hmuoa muouoom oaaoaoom mouuumsvaH mumaaum coaumasaom o>fiuu< nonuo ca moomoaaam aw noomoamam mHHmofiaosoou Houoa voxuos mafia .voxuos QEHE was .xom .mua>«uo< mo wonky ou mawvuooo< oHHo> mo soaumasmom o>wuo< hHHooflaocoom .ecmH .H.o manna 157 fluctuated from 6.7 percent (Bogota, March, 1964) up to 18.4 percent (Barranquilla, October 1967). The unemploy- ment rates of Cali, Medellin and Barranquilla show a tendency to increase while the unemployment rates of Bogota increased from March 1963 to March 1968 and decreased thereafter. The unemployment rates for Cali are given in Table 6.2.10 In 1965, Cali's unemployment rate was 13.2 percent, in 1968 it was 14.9 percent, in 1969 it was 18.3 percent, and went down to 12.4 percent in 1971. The rates calculated for the city of Cali reflect an increase in unemployment for the period from 1965 to 1969, and a sharp drop in 1971. This drop is explained by the fact that Cali was the host city for the Pan American Games in 1971 and therefore does not imply a change in the tendency toward increasing 11 We unemployment which is clear in the previous years. also observe that disguised unemployment, that is, the proportion of people unemployed and not looking for jobs, Municipal de Cali; Plan General de Desarrollo. Encuesta Urbana de Empleos e Ingresos del Area Urbana de a , Cali, 1969. 10Includes an estimation of the unemployment rate of 1971, which was not available to ILO. DANE, Encuesta de Hogares in Cali. Bogota; DANE, 1973. 11The tendency toward an increase in the rates of unemployment found in Cali result the same as those rates for Colombia, according to the Oficina Internacional del Trabajo, OIT, 22. cit. ,‘whose methodology is reporduced in AppendixB , and according to Departamento Nacional de Planeacidn. "El empleo en Colombia: Diagndstico y Recomendaciones de Politica," Revista de Planeaciéngy Desarrollo, Bogota, June, 1970. 158 Table 6.2. Active and Non-Active, Employed and Unemployed Population of Cali in Different Years. Items Years 1965 1968 1969 1971 Percent Economically Active Population 34.4 32.5 32.2 47.7 Economically Non-Active Population 65.6 67.5 67.8 52.3 100.0 100.0 100.0 100.0 Employed 86.8 85.1 81.71 87.6 Unemployed 13.2 14.9 18.3 12.4 100.0 100.0 100.0 100.0 Open unemployed 72.0 62.4 58.7 68.3 Disguised Unemployment 28.0 37.6 41.3 31.7 100.0 100.0 100.0 100.0 1Riley, et al., estimated an unemployment rate of 20 percent in 1969, ‘22. cit., p. 16. Source: Universidad del Valle, CIDE, Encuesta de Empleo e Ingresos. Cali, Universidad del Valle, 1965. Universidad de Los Andes, CEDE, Encuesta de Empleo e Ingresos. Bogota, Universidad de Los Andes, 1968. Planeacion Municipal de Cali, Plan General de Desarrollo, Encuesta Urbana de leos e Ingresos del Area Urbana de Cali. Cali, 1969. DANE, Encuesta de Hogares en Cali,fiBogota, DANE, 1971. 159 increases when the unemployment rates increase. The inter- pretation of this direct relationship is that people get discouraged in searching for work and lose interest in entering the labor force but are expected to do so when unemployment rates decrease. Economic dependency increases in such a situation. Comparing the unemployment situation in Cali with other cities in Colombia, we observe that Cali has the third highest unemployment rate, after Bogota and Barranquilla. (See Table 6.3.) The unemployed, classified by regional origin and sub-classes of unemployed (unemployed who were looking for work) are presented in Table 6.4. The highest rates of unemployment among Valle's non-native population were suffered by persons from Cauca, Tolima, and Nariho. A large proportion of the unemployed are natives of Cali, 49.4 percent. This phenomenon can be explained by (1) family support to the unemployed members,12 (2) greater 13 job selectivity, and (3) perhaps outmigration among Cali's 12Higher unemployment rates are found among young people. They are more likely to receive family support. 13Greater job selectivity is important in domestic service. Census data have shown over 70 percent of the domestic servants are born outside the city of Cali. The author interviewed 32 families during the 1973 population census and found all domestic servants were born in rural valle or rural and small towns of other departments, mainly from.NariHo, Cauca, Huila and Bogota. 160 Table 6.3. Percentage Classification of Population Employed and unemployed in the Major Cities of Colombia, 1964. City Employed Total Unemployed Economically Open Disguised gzn;:::::: employment Unemployment p Percent Barranquilla 23.6 3.4 1.9 71.1 Bogotél 28. 7 3. 3 2. 1 65. 8 Cali: 27.7 3.0 1.8 67.5 Ibagfie3 27.2 4.2 1.0 67.6 Manizales 26.1 4.1 1.4 68.4 Medellin 27. o 3. 1 1. 5 68. 4 Popayan 31.2 2.1 1.7 65.0 1 April 1967. 2May 1968. 3May 1967. Source: Planeacién Municipal de Cali, Plan General de Desarrollo. Encuesta Urbana de Empleos e Ingresos del Area Urbana de Cali. Cali, 1969. 161 Table 6.4. Distribution of the Unemployed of Cali According to Origins and Characteristics, 1969. Department Open Disguised Total of Origin Unemployment Unemployment Unemployed ---------------- Percent------------------ Cali (city) 26.2 23.3 49.4 Valle 8.1 9.3 17.4 Antioquia 2.3 1.2 3.5 Atléntico 0.6 -- 0.6 Caldas 1.2 1.2 2.3 Cauca 5.8 1.2 7.0 C6rdoba 0.6 -- 0.6 Cundinamarca 1.7 -- 1.7 Choc6 0.6 -- 0.6 Huila 0.6 -- 0.6 Nariflo 2.3 1.7 4.1 Quindio 2.3 1.2 3.5 Risaralda 2.3 1.2 3.5 Santander del Sur 0.6 -- 0.6 Tolima 3.5 1.2 4.6 Total 58.7 41.5 100.0 Source: Planeacion Municipal de Cali. Plan General de Dgsgrrollo. Encuesta Urbana de_Empleos efilngresos del Area Urbana de Cali. Cali, 1969. 162 non-native persons who cannot find a job in relative long periods of time. We examine now the reliability of the equations for estimating the labor force in Cali. One study of Cali's employment situation in 1969 found an unemployment rate of 18.3 percent and 47,192 unemployed.14 These figures imply a total labor force of 257,874. The estimated labor force for Cali in 1969, as computed in Chapter IV, is 267,775, which is about 10,000 more than Planeacion Municipal figures.15 Cali's active labor force using the figures estimated by Planeacion Municipal is 210,682 and the proportion of unemployed with respect to active labor force is 22.4 percent. Cali's active labor force using the total labor force estimated in Chapter IV and the total unemployed figures given by Planeacion Municipal is 220,583 and the proportion of unemployed with respect to active labor force is 21.3 percent. This preportion is two percentage points above the estimated ILO's proportion for Colombian urban areas and one percentage point above the estimated 16 PIMUR's proportion for Cali. If after the Pan American 142192§§21§Q,M§p1§ip§1 de Cali. "Plan General de Desarrollo del Area Metropolitana de Cali," Cali, 1974. 15Planeacién Departamental estimated an unemployment rate of 14.9 percent and 185,176 employed this latter figure is too low according to the data by Planeacién Municipal and our estimate of the labor force. 16The ILO and PIMUR estimates were given on page 153. 163 games the unemployment rate increased to its previous level, about 20 percent, and if the labor force was 346,224 in 1973, as estimated in Chapter IV, the total unemployed in Cali is 69,245.17 Unemployment rates have not been estimated for the total area of any department of Colombia. We can compute for Valle the ratio between total urban force (estimated in Chapter IV) minus the labor force employed in urban Valle and total urban labor force. The ratio is 3.5 percent for the census year of 1964 (see Table 6.5). This ratio tends to increase from 1964 to 1967 if the urban labor force of Valle is estimated with the equations with migra- tion rates equal to those observed fOr the period 1951 to 1964. If the ratios are computed with the estimated labor force assuming no migration, the ratio shows a tendency to decrease, inconsistent with the upward trend in the Cali's unemployment rates estimated for the same period. To estimate the unemployed in urban Valle in 1973 we assumed: (a) migration rates from 1964 to 1973 were equal to rates observed from 1951 to 1964, (b) the native popula- tion growth rate continues to be 3.4 percent, (c) the employed labor force will continue to grow at the same rate observed from 1960 to 1967, that is 3.3 percent for the 17This figure is 22,000 above the total unemployed found by Planeacion Municipal in 1969. 164 can»: Hmuou was sowumadmom o>Huow haamofiaoaooo anode oouom Honda cons: smmsumn ofiumm .oouom Honda a .mouom Momma pomoaeao paw mouom Momma amps: Hmuou newsman monouommwnm .mH.n wands Scum ma cowma «cog umooxo .H.m magma aoum mounwwmm .eo ammo Homao so ma ammo How:50% mum one mcomuom oohoamaog mo coauuoaoum one How unsound ou unmouom 1» .3 pompouoaw m3 0.3mm. Scum mounwwha nun nun w.m qmw.om nqo.om¢ mu¢.mmq Hum.mn¢ moma In: we m.o mmo.om mna.maq HNN.mH~ Ho~.m¢¢ moma o.o om¢.~ o.m omm.mm omm.mo¢ om¢.mo¢ oom.o~¢ mama N.H Hmn.¢ m.¢ www.ma mmm.mwm soa.mmm mmo.mo¢ coma a.a mmo.¢ ~.N mmm.oa oHH.mnm Ham.mnm aeo.mmm moma m.m www.ma m.m wNN.NH mn¢.mnm H¢N.oom H¢~.oom omma spasm umcasz «comm noneaz soHumusz coaumumwz usoanB .muH3 aowumuwwz.usonufiz. cowumuwwz nufi3 mouom momma mmooconommwo mvohoamam woos: HmuOH snow ooauom on» madman oHHm> mo mou< cons: osu CH pohoamaosb mo puma was Hmnadz .moma ou «oma 6.0 0.3.3.. 165 Manufacturing sector and 4.1 percent for other sectors, and (d) people under 15 and above 64 will continue to be 4 per- cent of the labor force. Thus, the total labor force is 582,741 (see Table44,5), the employed is 510,067 and unemployed 72,764. This figure is approximately 3,500 above the estimated unemployed figure for Cali.. So, it appears that the large majority of the unemployed are located in Cali. Additional Jobs Needed in Cali To estimate the number of jobs that need to be created to prevent the unemployment rate from.increasing during the period 1974 to 1980 in Cali, we establish three alternative assumptions: (1) migration rates from.l964 to 1980 equal to rates observed from 1951 to 1964, (2) migration rates from 1964 to 1973 were equal to rates observed from.l951 to 1964 and from 1974 to 1980 migration rates equal one-half the rates from 1951 to 1964, and (3) migration rates from 1964 to 1973 equal to rates observed from 1951 to 1964 and from 1974 to 1980 migration rates equal one-fourth the rates from 1951 to 1964. The second and third hypothesis imply that high unemployment rates have lowered the expected urban earnings and fewer people will migrate to urban areas and/or proper policies on rural employment and development will be taken. Also, we assume (a) natural population growth rate will be 3.4 percent as in the past, and (b) people under 15 and 166 above 64 will continue to be 4 percent of the labor force. The unemployed in 1980 will be under (a) assumption one, 547,702 (.20) = 109,540; (b) assumption two, 440,635 (.20) = 88,127; and (c) assumption three, 387,102 (.20) = 77,420. Thus, given that 69,245 were the unemployed in 1973, the additional number of jobs needed in Cali to prevent unemploy- ment from increasing will be 40,295 under assumption one, 18,882 under assumption two, and 8,175 under assumption three. To estimate the number of underemployed and persons occupied in low-paid jobs we assume the proportion of people working in jobs with incomes below 200 pgggg per month estimated by ILO for urban Colombia can be applied to Cali. Thus, at the end of 1973 the number of under- employed and persons with low-paid jobs in Cali was (346,224 - 69,245) (.12) = 33,237. Similarly, the low-paid jobs in 1980 will be 52,579 under assumption one, 42,301 under assumption two, and 37,162 under assumption three. To estimate additional number of jobs needed in Cali to give everyone of the projected labor force the oppor- tunity to find a job from 1974 to 1980 are summarized in Table 6.6. The creation of such a number of jobs does not mean that unemployment and underemployment will disappear since there always exist frictional unemployment and over- commitment of people in both rural and urban areas. 167 .qoma ou anma scum «mums onu sunfloMuoao assoo moumu soauonwaa_omma ou smma Bonn was soma ou amma Bonn om>uomno momma ou asamo who? mama on «oma scum momma soaumuwflZm .soma ou amma scum momma msu mamnaoao assoo momma coaumuwaa omma ou qnma scan has «oma ou amma Scum oo>uomno woman on asses mums mnma ou soma.aoum momma soaumuwazm . .eema as amma scum oo>uomno nouns ou amamo on aaa? one who? owma ou soma Bonn mouwu soaumuwaza ooa.¢aa oom.mma ooo.aoa amuoa oom.m ooo.m oom.ma omwouoaa ou anon mama 36a was usoahoaeamuooss uao>oum ou mQOn amaoauaoo< oom.m oom.ma oom.o¢ 36am on unmahoamaons uso>oum ou ween assoauaoo< ooo.mm ooo.mm ooo.mm Amomom oom swam mmoav mama ca mean pawn 30a mo Hogans ooumaoasood ooo.mo ooo.mo ooo.mo mmma ca pomoamaoss pomaswmap was ammo mo Hogans voumasasoo< msoaumabmm< Ncoaumanmm< aaoaumasmm< maoua .owma ou dnma Bosh aaoo ca vouooz anon mo Honeuz commaaumm .o.o manna 168 We will analyze in the following chapter policies to create urban and rural jobs and the effect of lowering urban unemployment rates on rural-urban migration. CHAPTER VII ALTERNATIVE POLICIES TO REDUCE RURAL-URBAN MIGRATION AND INCREASE PRODUCTIVE EMPLOYMENT Rising urban unemployment rates, especially in Cali, and considerable rural-urban migration (from relatively small towns to cities) have characterized the Valle's economy during the 19603. The growth of mechanized agricul- ture caused small farmers and rural laborers to migrate to the cities. Dualistic systems in both urban and rural areas are more acute in Colombia than in other coutries and within Colombia appears more striking in Valle than in other depart- ments.1 With policies oriented toward production growth the dualistic systems produce (1) high unemployment and under- employment rates, and (2) very uneven income distribution. It has been noted that the growing rates of unemployment in Valle and Cali occur simultaneously with an economic growth rate considered by many to be satisfactory and, indeed, high enough to meet the goals of many development plans. Conse- quently, the pursuit and achievement of high growth rates of the gross product is not sufficient to assure the 1See Chapter II and V. 169 170 absorption of labor, the most abundant production resource in the region. What is required is a policy to increase the demand for labor relative to other inputs. Jobs must be created faster than the rates of increase in the labor supply in order to decrease the present level of unemployment. National policies and development plans now reflect an increasing commitment to the idea that the process of modernization should be carefully directed to create more jobs and more even distribution of income. It is our view that a more even distribution of income is important along with more production. Unfortunately, the two are not necessarily complements; in fact, the meaning of aggregate product itself is a function of the weights used to aggregate individual products into a total, those weights being functions of the distribution of income. This makes policy decisions and evaluation difficult. We must understand that policies that benefit the advanced sectors may hurt the traditional sectors. With dualistic production conditions, policies should take into account the characteristics of the modern and the traditional subsectors. The Valle's policy makers seem to have recognized this with the creation of institu- tions specialized in assisting both the modern and the traditional sectors. Examples are CVC's extension services for medium and large farmers while ICA gives extension to small farmers. In manufacturing, SENA gives technical assistance to small enterprises (a relatively new program) 171 and Banco Popular and Corporacibn Financiera Popular (new credit agency) provide credit to small manufacturers. The problem seems to be that governmental officials do not recognize the great importance of helping small enterprises and hence do not provide enough resources and assistance to small enterprises through the institutions which exist. Helping small and medium-sized rural and urban enterprises could lead to more employment and less uneven distribution of income. In Chapter IV we explained that, give a minimum wage, the increase of unskilled labor in a given place depends on the factors which determine, (a) the conversion rate from unskilled to skilled labor, (b) the growth rate of the native population between 15 and 64 years of age, (c) the rate of participation of those from.lS to 64 in the labor market, and (d) migration to and from the given place. The causes affecting each of these four factors discussed in Chapters II to IV, as well as the determination of the minimum wage, suggest policies which might be followed to combat unemployment. These policies are: (1) revision of labor code especially as it affects unskilled laborers and small and medium-sized enterprises, (2) reduction of growth of total labor force, (3) changing the composition of labor force and upgrading of skills, (4) slowing down the rate of migration to the large urban centers such as Cali with incentives to stay in farming and rural population centers, 172 (5) improving the information system with respect to job opportunities and wages (the existence of both urban unemploy- ment and rural underemployment desite high migration rates to the urban areas of Valle suggest the need for improving the performance of the labor market of Valle with a better information system), and finally, (6) the creation of additional jobs in urban areas for large pools of unskilled laborers is very urgent in Valle. The industrial sector of Valle and of Cali absorbs a low proportion of labor relative to the service sector although an increase of 14,146 workers in the industrial sector was observed during the years 1960 to 1969.2 The service sector is the principal source of employment in the urban areas of Valle but a high proportion of labor it absorbs is employed under conditions of uncertain employment, underemployment, and low productivity. The big industries absorb the majority of the skilled industrial manpower but new opportunities for employment offered by them each year are low compared with the increase in the unskilled urban labor force since these industries use labor-saving capital, as shown in Chapter V. Revision of the Labor Code Especially as it Affects Unskilled Laborers and Small and MediumrSized Enterprises Raising the minimum legal urban wage has different effects on unskilled and skilled laborers and on unskilled 2See Chapter V. 173 workers employed by small, medium, and large enterprises. Skilled laborers have wages above the minimum wage because their MVPs are high. This kind of labor is complementary with capital which saves unskilled labor. Thus, minimum wages do not affect the level of employment of skilled laborers. High minimum wages can and do make unskilled laborers more expensive than labor-saving capital, that is the ratio between MVP and MFC of labor may be less than the correspondent ratio to labor-saving capital; therefore, raising minimum wages can diminish the employment of unskilled laborers while skilled laborers do not loose their jobs and a few unskilled laborers get higher wages. In large enterprises, the effect of raising the real minimum wage is nil since these firms are paying wages for unskilled laborers above the legal minimum, as was shown in Chapter V. In small and medium enterprises it is more likely that increases in minimum.wages will establish minimum wages above the equilibrium.wage for unskilled laborers and thus the utilization of unskilled manpower will be less. However, labor 1egislation.makes it diffi- cult to discharge workers and fewer laborers are dismissed than otherwise would be. The firms then have to decrease other costs of production, obtain higher prices for their products or go out of business. Urban wages for large and small enterprises are regulated by the same legislation. Large enterprises can pay much higher wages because they have higher productivity 174 than small firms, as discussed in Chapter V. Union actions to raise wages and other benefits in small and medium-sized firms with relatively low productivity increase the cost of production. If this action combines with low price for the products of these firms, limited credit and a relatively high cost of capital, the small and mediumrsized firms find growth impossible. Under these conditions, lowering the cost of labor for small and medium.firms would increase employment. Two possibilities of lowering cost of labor for Valle's small and medium firms are (l) revision of the law requiring extra payment for night work and daytime overtime, and triple pay for Sundays and holidays, and (2) labor legislation with respect to the minimum number of workers required to form a union. Extra payment for night work, daytime overtime, and triple pay for Sundays and holidays were some of the factors causing low use of installed capacity,3 reported in Chapter V. Firms with about 25 workers, the minimum legal number of workers to form a union, are confronted with expectations of rising 3Rile et a1. , suggested the elimination or modifi- cation of the requirement for 35 percent extra payment for *work.at night as well as other regulations of the labor code. They argue "And the code, while seemingly helpful tx: those already employed may be a significant barrier to employment for those entering the labor force," 0 . cit. I’- 323. Similar suggestions were made by ILO. O EIEo suggested revision of regulations protecting against laborers dismissal. Oficina Internacional del Trabajo, OIT, . cit., pp. 218-224. On the contrary, Gaviria, Gomez and ‘Egpez—Tomd that some entrepreneurs believed that such revision of dismissals will open the opportunity to substitute labor-saving capital for unskilled labor, pp. 23;. , p. 160. 175 salaries and other non-monetary cost associated.with union- ization. Thus, the union actions seeking higher wages, perhaps wage levels comparable to those paid by large enterprises where the employed unskilled labor have higher ‘MVPs, caused low growth for some small and medium firms, or some firms shifting toward labor-saving capital or firms going out of business. Increasing the minimum.number of laborers to form unions will increase employment in some firms. Those firms whose demand for labor intercepts the stepped supply of labor in the vertical segment, at A in Figure 7, when the minimum.number of workers to form.unions is 25 will inter- cept the supply at B if the minimum number of workers legally required to form unions is increased from 25 to X', or will intercept the supply at C if the minimum number of workers legally required to form unions is increased from 25 to X". 'W \ 25 X' X" X Figure 7. Labor employment when the minimum number of workers legally required to form.unions is increased. 176 Additionally, creation of special labor courts and legislation to deal with wages and labor conflicts in small and medium.firms will lower the monetary and non-monetary cost of labor. Raising wages according to productivity in these firms under labor court supervision is an alternative. Increases in urban wages and employment in urban areas tend to attract labor from the rural areas as was argued in Chapter 111. Thus rural minimum wages have to be increased at least at the same rate that increases in urban wages as well as improvement in living conditions and job opportun- ities. When discussing rural-urban migration policy we will refer to the necessity of improving conditions for small and medium-sized farmers. If these farmers have better economic situations they will be able to pay higher wages to their rural laborers as well as receive better returns for their own labor. Reduction in Growth of Total Labor Force The growth rate of the native population between 15 and 64 years of age is already determined for the next 15 or 16 years, due to high birth rates in the recent past. Currently, 44.4 percent of the total population of Valle auui 41.8 percent in Cali are under 15 years of age. Mortality rates will continue to decline. Birth control, the promotion of which is very slow in a population with low educational levels, is a long run program which can do little to significantly affect population growth in the near future . 177 Nelson, Schultz and Slighton argued in favor of population growth control4 although they recognize rapid increase in the labor force for the next decade or two. They said that rapid population growth (1) retards the growth of physical capital, (2) depresses average private rates of savings, (3) decreases the public sector's oppor- tunities to save and invest its resources in tangible pro- ductive assets, and (4) causes a "smaller fraction of the population" to reach "the progressively higher rungs on the educational ladder, and improvements in health proceed less rapidly." We will refer to the potential effects of these factors on the demand for labor in the following section. Population control programs must be implemented in both rural and urban areas. High rural birth rates will create population pressure and rural—urban migration will take place, as urban unemployment rates diminish due to decreases in urban birth rates. Qhanging_Composition of Labor Force and Upgrading Skills The conversion rate of unskilled labor is low due to the lack of educational and training programs and the uneven 4gp. 9_i_t_., pp. 263-275. They claimed that effective control of population growth can be obtained by distribution of birth control information and devices, and influencing family size goals. Family size goals, that is, the number of children families want are affected by factors like (1) avail- ability of primary schooling for children and (2) obstacles to child labor and employment opportunities. 178 distribution of income. Estimates have recently been made of how much the Colombian government would have to spend in order to fulfill the constitutional mandate to offer free primary education to all Colombian children. To matriculate all the children who had no chance of entering primary schools in 1972 would have taken 36 percent of the national budget,5 excluding the investment in physical plants, equipment, etc. This percentage would increase each succeeding year. Twenty percent of the children now matriculated in primary schools are studying in private schools.6 The government seems unlikely to provide instruction for all Colombian children. Income redistribution could permit more families to take on the burden of part of the expenses demanded by their childrens' education. The government also could collect additional taxes in order to broaden public education. Educational reforms which provide more opportunities for the school age population would also diminish their partici- 7 pation in the labor market. Training and the unskilled 5Centro de Recursos para la Ensehanza, CREE, "Estra- tegias curriculares combinadas para un sistema educativo no formal." Cali, Universidad del Valle, 1972, p. 7. 6Mendoza y Olarte, pp. £15., p. 35. 7In Chapter VI we saw that minors (under 15) and older people (over 64) constitute 4 percent of the labor force. The Universidad del Valle and some national organizations are studying ossible reor anizations and new systems such as ‘nonrforma education w ich is defined as "a systematic educational structure, non-scholastic and consisting of training activities of a relatively short duration in which the responsible agencies search for concrete behavior changes in a great variety of groups." Centro de Recursos para La Ensei’ianza, CREE, 9p. 9_i_t_. 179 laborers will change them and increase their productivity. As a different kind of labor, they will experience a new and higher demand for their labor. A national institution which carries out an important task in the improvement of the labor force is the Servicio Nacional de Aprendizaje, SENA, financed by private industries which give a percentage of their budget to maintain the SENA. SENA encourages employers to give workers the time required for training, and some remuneration while they are studying. Although the number of students attending courses in the SENA is growing, they still constitute a small pro- portion of the total labor force. The incentives for unskilled labor to undertake training include in higher wages and lower rates of unemployment. Training the unskilled laborers will change them and increase the productivity of their labor while decreasing the supply of unskilled labor. Furthermore, more education and training for the people helps to achieve a more even income distribution. Slow Down the Rate of Migration to the Large Urban Centers Such as Cali A policy of rural-urban migration control is necessary when the urban areas find it impossible to absorb all the available labor and, hence, urban unemployment and low paid jobs tend to increase, as is the case of urban Valle, especially in Cali. High migration from rural small towns to cities-where unemployment has been high clearly indicates very low earnings in rural areas compared with those in urban 180 areas even when urban earnings have been adjusted by the probability of finding jobs in a given period of time. Reduction of migration from the rural areas to the cities is very difficult given the excess of manpower in the country— side which, as mentioned before, is produced by high birth rates and, in part, by the concentration of rural property ownership and income in a few families. Under these condi- tions, the decision of those without property to move from the rural area to the city is rational despite the high rate of unemployment in the cities. ‘Consequently, a reduction of rural-urban migration will only be achieved by improving the incomes of rural workers and small farmers, perhaps through policies of land redistribution, subsidized credit, training, infrastructure, and favorable prices to help small farmers more adequately exploit his property and raise his income. Price protection for agricultural goods generally produced by smaller farmers might receive more attention. In Valle, large farmers produce the export crops other than coffee. They benefit from.higher prices in the international market (sugar, cotton, soy beans) and receive in addition a 15 percent export incentive pay— ment from.the government. Production from small producers goes through the local markets where lower prices are paid. Public credit could be provided to small farmers. Caja Agraria which was created with the objective of giving credit to these farmers has many difficult requirements and 181 8 many of its loans go to large farmers while other agricul- tural public institutions loan many of their funds to large enterprises.9 Providing funds, tools and technical advice to small farmers allows them to create different kinds of capital in rural places like fences, land improvement, livestock, etc., working individually or in organized groups at Juntas de Accién Comunal. Such self-generated capital raises productivity, helps the development of relatively backward areas and, hence reduces migration. With respect to the redistribution of rural property, the Colombian Institute of Agrarian Reform, INCORA, has accomplished very little. INCORA estimated that there were 190,000 landless families, and 745,000 families living at bare subsistence levels in Colombia. A study made by COFISEL10 for the Corporacién Financiera del Valle, where the INCORA data is cited, urges the stimulation of the "national purpose to carry out an agrarian reform which stresses prominently the distribution of land to poor farmers.‘ However the immense difficulties faced by such a program are pointed out. The text reads: 8Albert Berry, "Special Problems of Policy Formation in a Dualistic Agriculture: Colombia," pp. cit., Table 3A. 9Fondo Financiero Agrario awarded 90 percent of its funds to moaern relatively large farmers in 1969. Gavira, Gomez and Lopez, op. cit., p. 75. 10Paper for the Vallecaucano meeting. Cali, unpublished Preliminary Draft, 1973. 182 "In the first place, much of the land included in the largest properties is not apt for inten— sive agricultural use, which is required to give employment and income in small units. This is demonstrated by the experience of INCORA, which has anulled the land titles of more than 3,000,000 hectares and tried to give them to the small farmers from 1962 to 1971, with little effect on employment and the creation of small productive units. The second problem caused by the democratic means of distribution of land, is the cost of acquiring the land, and the difficulties encountered in the process of making the new owners at least as efficient as the owners of well-exploited large farms. On the one hand then, financial resources so large that they are outside the national possibilities must be obtained, or else legal norms must be adopted which permit the expropriation of the needed amount of land with very small payment to the previous owners, which easily causes panic among the investors and short-term lowering of production. All this, added to the enormous resources needed to assure foundation of capital, techniques, investments required by the existing exports, forms a true laberinth." The concentration of land ownership in Valle observed during the period 1959 to 1970 has caused an increase in the concentration of rural income despite the fact that INCORA has Several programs in Valle. A study on the effects of the agrarian reform in Valle shows greater concentration of rural property, and the construction of Lorenz curves for the years 1957 and 1970 show increased concentration 11 of income in 1970. More inputs and land resources to small farmers would produce a more even distribution of rural 11CVC, "Reforma Agraria en el Valle," pp. cit. See also Appendix B.5 and B.6. 183 income. In Valle there is a limited amount of flat fertile land that is inadequately cultivated and which does not require expensive investments in reclaiming it. Therefore, unemployment can not be significantly affected if land redistribution affects only land presently adequately cultivated. There are possibilities of reclaiming and improving about 120,000 hectares of Valle's flat land presently subject to flooding by the Cauca River, or affected for high levels of underground water or poor drainage. The author estimated that reclaimation of this land in a CVC project called Proyecto de Regulacidn del Rio Cauca would generate about 30,000 additional jobs in the area once under agricultural exploitation.12 The construction of the project will start at the beginning of 1975 and will take about ten years to complete. The importance of this project for labor absorption can more fully be understood if we recall that the urban-rural migration which occurred in 1951-64 was, in part, associated.with the replacement of extensive cattle raising in the flat lands with sugar cane and cotton (see Chapter III). Replacing extensive cattle raising in land now affected by flooding will be replaced by 12CVC, Pro ecto de Regulacién del Rio Cauca. Cali: CVC informe 72-15, 1972. The number of new jobs was cal- culated on the basis of projection of land use of the area sun! the labor inputs of the different craps without changing land ownership. 184 cropping13 will permit some urban-rural migration to take place again and/or a decrease in the rural-urban migration in Valle. The creation of 30,000 jobs by land reclaimation and the other agricultural measures suggested above could prevent or further diminish the decline in the rural force of Valle in the next ten years. The ILO's plan to lower unemployment rates of Colombia to 5 percent by 1985 needs a 1.8 percent increase in agricultural labor force,14 a goal not applicable to the Valle's case, but important for other departments, those from which most of the Valle's non-native population come. The policy adopted by the government, of promoting the development of the so-called intermediate cities by channeling resources and credit through favorable discount rates is a positive measure to control migration to Cali. IMiddle-sized cities serving rural communities can employ people in agricultural processing wholesale and retailing, transportation, traditional manufacturing and repair services, among others. The development of middle-sized cities and 13The expected crops for the area are soybeans, rice, sorghum, corn and sugar cane. The sugar industry of Valle vflhich combines sugar cane crapping and sugar manufacturing in the rural areas generates a considerable number of jobs auui illustrates the employment potential of this type of enterprise in rural areas. 14 p. 418. Oficina Internacional del Trabajo, OIT, pp. cit., 185 their neighboring departments should be stressed to further discourage migration by stimulating and facilitating the opening of new lines of production. Also, such policy would help attain a more balanced income distribution among the different regions of Colombia. In conclusion, controlling growth of the unskilled labor force in the urban areas implies the creation of numerous education and training opportunities and of various employment possibilities in the rural area. These might be achieved by (1) changing the distribution of rural property in favor of the poor farmers, (2) assisting them with its exploitation and (3) the reclaimation of land. The lowering of the urban unemployment rate may also be accomplished by programs for rural development to reduce migratory flows from the rural to the urban areas. In the long run, birth control will decrease the growth of the labor force, but increased female participation in the urban labor market will increase the labor force over the next several years. The ILO's recommended policies gave importance to the development of rural areas and focused on small farmers and labor intensive agricultural porducts as a means to increase rural employment as we suggest in this study. While the National Economic Development Plan concentrates on the creation of urban jobs, the proposed ILO's plan to lower unemployment rates of Colombia to 5 percent by 1985 contemplates a 1.8 percent increase in the agricultural 186 labor force, a goal not applicable to the Valle's case but important for other departments which are the source of most of the Valle's non-native population. Creation of Jobs in Urban Areas for Large Pool of Unskilled Workers Factors affecting the number of jobs in urban areas of Valle for unskilled laborers discussed in this study are: (1) relative price of labor and capital, (2) factor pro- portions used by size of enterprises, (3) the fuller use of installed capacity, and (4) amounts and kinds of labor absorbed by different products produced for internal and external markets. In what follows we analyze policies aimed at affecting these factors to increase the rate of trowth of unskilled jobs in urban Valle. When discussing labor and price policy we must dis- tinguish different kinds of labor (skilled and unskilled) and different kinds of capital( labor-neutral, labor-saving and labor-intensive). With high unemployment rates among unskilled laborers the use of labor-intensive capital and technological processes using factor proportions in favor of labor are called for. Thus, we have to refer to two aspects: (1) the kinds of capital that are and will be used in the country, and (2) given the kinds of capital being used the factor proportions or technological process that can be used. In relation to the kinds of capital that are used in the country we showed in Chapter V that labor-saving capital 187 is used in the majority of the industrial branches of Valle, especially in the production of intermediate, durable and capital goods. Labor-intensive capital is more common in the consumption goods (food, beverages, clothing) where more labor is employed and less excess capacity is observed but labor-saving capital is also used in the consumer goods. Low cost for labor-intensive and labor-neutral capital could increase unskilled labor employment. If labor- intensive and labor-neutral capital and unskilled labor are complements, a decrease in the price of these kinds of capital holding wages constant will be associated with an 15 and in the demand for unskilled labor expansion of scale and relatively less expansion in the demand for skilled labor. Low cost of labor-intensive capital will also be associated with an expansion of scale but while the demand for skilled laborers will increase (labor-intensive capital and skilled labor are complements) many unskilled laborers will be displaced. This is of importance in choosing technology for future enterprises and reorganization of small and medium firms. Lowering the cost of labor-intensive capital relative to the cost of labor-neutral and labor- saving capital will increase absorption of unskilled laborers. 15A reduction of the price of capital shifts the marginal cost curve downward for all scale levels. 188 The structure of financial markets is important to consider, especially as it affects small producers. It is true that the rates of interest of the banks in Colombia have been maintained low artificially, and that imports of equipment have been subsidized, as the national government 16 The money market has controlled exchange in this market. at banks is characterized by an excess of demand at existing interest rates. The marginal investor must resort to the nonbanking market where, of course, the interest rates are much higher. Some of the available money in the nonbanking market is supplied by bank borrowers, thus provoking a redistribution of income in their favor, since they borrow from the banks at a low interest rate and lend in the nonbanking market at a higher rate. Subsidizing the import of capital goods causes similar effects if they are accompanied by rationing of foreign exchange. Those with access to import quotas are able to sell equipment, once imported, at much higher prices.17 However, import subsidies can be used so that the technology imported requires an important amount of labor and removing 6In some years there has been a fixed rate of exchange and in other years the government intervened in the market buying and selling foreign exchange to avoid price rises. 17An analysis of this situation may be found in a ‘mimeo by Buillermo Calvo, "The Rate of Labor Absorption and the Cost of Capital," Bogota, Universidad de los Andes, 1971. 189 import subsidies would encourage use of more labor-intensive technologies. Research on labor intensive technology could be promoted in the country and in joint programs with other less developed countries. In relation to factor proportion future changes in the relative prices of labor and capital can be directed to change prices in favor of labor. Since the MVPs of unskilled labor is expected to be lower than the MVPs of labor-neutral and labor-saving capital the relative price of these kinds of capital and unskilled labor should be directed in favor of unskilled labor. Furthermore, the prices of the three kinds of capital (labor-intensive, labor-neutral, and labor-saving capital) should be directed in such a way that labor-intensive and labor-neutral capital becomes less expensive than the cost of labor-saving capital, that is, the ratio between the MVP and MFG of labor-intensive capital that actually is less than the correspondent ratios for labor-neutral and labor-saving capital should be reversed. These features will induce firm owners to change their technological processes or maintain labor-intensive ones and to use more unskilled laborers. With high elasticities of substitution between any two factors small changes in relative prices in favor of any one of the factors will determine a large substitution of the relatively expensive 18 factor for the relative cheap one. Thus, when high 18Even with high elasticities of substitution some entrepreneurs may believe the production functions they 190 elasticities of substitution exist small changes in relative prices in favor of labor-intensive capital and unskilled laborers will increase unskilled employment levels. Unfor- tunately, elasticities of substitution computed for Colombian industries and the elasticities computed in this study for Valle did not distinguish between different kinds of labor and capital due to lack of data, and besides, have very serious statistical problems. When hourly wages increases with firm size some owners are led to install labor-saving capital even though he does not plan to use it to full capacity since the fixed cost of equipment is less than the cost of unionized labor. Thus, changes in the relative prices of labor and kinds of capital, as discussed above, will tend toward more intense use of the installed capacity. Also some revision can be made in wages paid for night work, Sundays and holidays, but this proposal would have to be studied to see if this would have an impor- tant impact on use of installed capacity and employment levels. Educational and trained policy, discussed above, are using are characterized by fixed preportions. R. S. Eckaus has treated this aspect in his article "The Factor Preportion Problem in Underdeveloped Areas," American Economic Review, Vol. VI, September 1955. The entrepreneur may think the isoquants are right angles when there are substitution possibilities. They do not change the combina- tion of factors when prices change. In a survey made in Colombia by Gaviria, GOmez and L6pez it was found that the majority of owners see very little flexibility in factor proportions. 9p, pip., anexo. 191 should be complemented by measures aimed at increasing the urban demand for skilled laborers and highly trained people. One way to obtain this is by technical assistance, reorgani- zation and growth of small and medium firms producing for the internal and external markets, as it is discussed in what follows. Helping small and medium-sized firms industry is needed to stop the decreasing number of these size manu- facturing firms which employ higher proportions of labor— intensive capital and unskilled labor and to promote the development of those which stay in business but cannot grow and, hence to increase the number of urban jobs, especially for unskilled laborers. Wages for large and small enterprises are regulated by the same legislation. Large enterprises can be paid :much higher wages for unskilled labor because they have higher productivity than small firms, as shown in Chapter V. union actions to raise wages for unskilled labor and other benefits in relatively low productive industries increase the cost of production. If this action combines ‘with.low prices for the products of these firms, limited credit and a relatively high cost of capital, the small firms find growth impossible and fewer jobs are available for unskilled laborers. Riley, et. al., noted that in Valle "the most unsatisfied credit needs exist among the more labor-intensive small and.mediumrsized consumer goods producers (e.g., 192 clothes, kitchen utensils, leather shoes, hand tools, and kerosene stoves) and among food processors of the same size. The lack of working capital is a frequent complaint."19 Credit, technical and financial assistance for such labor-intensive small and medium-sized industries could help combat high unemployment rates and could create jobs for the increasing urban unskilled labor force. ILO strongly urges the more rapid expansion of labor-intensive industries,20 which as we showed in Chapter V are, mainly, the small and medium-sized firms. Riley, et al., noted that some small and medium-sized firms are poor credit risks and, thus, need supervised credit and technical assistance.21 Craftsmen and small manufacturing firms contributed a great deal to the Japanese development process. Urrutia and v111a1ba22 have pointed out that the importance of traditional sectors in Japan was due to effective relations between the powerful financing institutions, trade agencies and small producers. The financing institutions organized trade agencies to assist producers, to buy parts and inter- mediate goods from small producers and sell them to other 199p. 933., p. 342. 20Oficina Internacional del Trabajo, OIT, pp, cit., p. 123. 219p. £35., p. 343. 22Miguel Urrutia Montoya and Clara Elsa Villalba, “E1 Sector Artesanal en el Desarrollo Econ6mico," Revista de Planeacion y Desarrollo, Vol. 1, No. 3, Bogota, 193 small producers who will make a final product, to control quality, and to sell the final product to both the internal and the international market. Exports of goods produced by craftsmen and small manufacturing firms with traditional technology, low capital-labor ratio and no economies of scale were important in Japan's foreign trade. Thus, these Japanese experiences should be explored to see it's feasi- bility for Colombia, especially in relation to trade among Latin American market agreements: ALALC and Zona Andina. Construction sector growth policies have received prime importance in recent Colombian Economic Development Plans. the central idea was the massive creation of con- struction jobs which will provide new incomes to laborers and, therefore, stimulate industrial growth and growth of other sectors as the internal market is broadened. The plan proposes a massive investment in the construction industry in those cities, capable of absorbing a large 23 proportion of unskilled labor. In order to channel savings toward the construction sector, units of constant 23In a study on the construction sector of Cali in 1967, SENA found 80 percent of the sector labor force are immigrants to Valle and 61 percent of them came from other departments. SENA also found that in Medellin, one of the largest cities of Colombia, 61 percent of the labor in the construction sector had up to three years of elementary school and illiterates were 8 percent. See Eduardo Sarmiento Palacio, Departamento Nacional de Planeacién, DNP, Aspectos Cuantitativos del Plan de Desarrollo, Bogota, Imprenta Canal Ramirex-Antares, 1974 and PIMUR, Technical Report No. 16,'pp.'p$p. 194 buying power (Unidades de Poder Adquisitivo Constante, UPAC) which pay a guaranteed interest rate plus the rise in the cost of living were issued, in 1973. The UPAC has been successful in transferring large amounts of money to the construction sector. The ILO's plan contemplates an increase in the con- struction sector growth rate of labor employment from 8.1 to 8.9 percent while the manufacturing sector growth rate of labor employment increases from 2.4 to 5.0 percent.24 The increased growth rate of labor employment in the construc- tion sector was apparently judged by Planeacién Nacional too low to achieve desired employment goals and decided, contrary to ILO's proposal to give the construction sector the leading role among all economic sectors of Colombia in absorbing unskilled urban labor force. The reasons to choose the construction sector as the leading sector for absorbing labor were (1) the underutilized production potential in the unemployed, underemployed and 25 (2) and (3) the construc- poorly employed unskilled urban and rural labor, enormous potential demand for housing,26 tion industry needs relatively low capital per worker because 24Oficina Internacional del Trabajo, OIT, pp, cit., pp. 415-7. 25Departamento Nacional de Planeacion, Guidelines for a New Strategy. Bogota: Italgraf, 1972, p. 13. 26 Ibid., p. 47. 195 labor-intensive capital has traditionally been used and little imported machinery or raw material is needed. With the growth in the construction sector it is expected that employment will grow at a faster rate in other sectors of the economy since there will be a larger internal market as a result of the construction sector expansion.27 In a recent study28 made by Planeacion Nacional, where detailed quantitative analysis of the plan is made, it was estimated that the reduction of unemployment from ten to five percent in Colombia in ten years would require the creation of 70,000 new jobs annually in the construction sector.29 They pointed out that the number of jobs the construction sector can create depends on the supply of inputs and the effective demand for houses. The national planning office has identified four critical inputs. They are cement, brick, glass and wood. The effective demand for houses depends on the funds available to buy the product. There is a lack of clarity in the plan about the source of savings30 to sustain the demand for houses needed to accomplish 27The proposed growth of the construction sector will produce both a net increase in the aggregate demand and a change in its composition. 28Eduardo Sarmiento P., DNP, Aspecto Cuantitativos del Desarrollo, pp. cit., pp. 16-18. 291bid., pp. 66-67. 30Idem. The Corporaciones de Ahorro Y Vivienda, new credit agencies for the construction sector have had excess funds in 1973 and 1974. Governmental officials argue that this is a temporary phenomenon caused by lack of experience (If the new credit agencies in handling loans. 196 the goal of lowering unemployment to 5 percent in ten years but the national planning office is confident the construc- tion sector can create 50,000 additional jobs annually to prevent increasing unemployment rates.31 It is very possible that the resources provided to the construction sector in the last two years have created many additional jobs in Valle but adequate data are not available to confirm this. In 1969 it was estimated that a total of 356,000 new houses were needed in Cali in order to eliminate slums and congestion and to provide for popu- lation growth during the period 1969 to 1989, that is an average of 17,000 new units be built each year.32 The direct employment effect was estimated to be 17,500 man-years annually in active construction and an additional 17,500 in production materials.33 Barriers to expanded housing construction listed by Riley, et al.,34 are (1) lack of effective demand, (2) lack of risk capital due to low interest rates on second mortages, (3) high cost of land and urbanization and (4) low labor productivity and in6fficiency in construction tasks.35 31Idem. 32Harold Riley, et al., pp. cit., p. 306. See also, PIMUR, Technical Report No. 16, pp. cit. ”pp. _c_:_i_§., p. 307. 34Construction activities and employment declines in ‘Valle during the period 1960 to 1980. Labor absorption decrease for 1964 to 1968. In 1964 there were 12,088 employees; 111,170 in 1965; 11,147 in 1966; 9,997 in 1967; and 8,132 in 1968. PIMUR, Technical Report No. 16, pp. _c_i__t_., p. 86. 197 The need for housing which, as we said before, was one of the reasons to choose the construction sector as the leading sector for absorbing labor is difficult to transform into effective demand due to (1) shortage of savings which need to increase 4 percent in order to decrease unemployment from 10 to 5 percent in Colombia during the first six years 37 ~of the Development Plan,36 (2) poverty and uneven distri- bution of income which makes down payments for houses impossible for the majority of the population and (3) low investment returns which imply houses are seen as consumption 38 goods rather than financial assets. Even if these factors affecting the effective demand for houses can be overcome by means of new savings,39 foreign loans, lower down payments, and more even income distribution, there are still the important questions of inputs bottlenecks and productivity. The actual expansion of the construction sector has raised the price of many inputs, especially iron, steel and cement.40 35 36Eduardo Sarmiento P., DNP, Aspectos Cuantitativos del Plan de Desarrollo, pp. cit., p. 71. To prevent unemployment to increase savings must increase by two percentage points. 37 38Miguel Urrutia Montoya, "Problemas de Oferta de Factores de Produccion en la Nueva Estrategia de Desarrollo," in CORP, pp, cit., pp. 114-115. 39We recalled that Nelson, Schultz and Slighton pointed out that birth control results in higher private rates of saving. 40 Harold Riley, et. al., pp, cit., p. 3112314. Harold Riley, et. al., pp. cit., pp. 311-312. Before the actual Development Plan the construction 198 Credit funds are being given to these industries to expand existing facilities or establish new ones.41 The number of skilled labor and the costs of training programs have been estimated by the national planning office42 and suggestions have been made for specific training programs.43 NOw, if we assume barriers for expanding the construc- tion sectors are eliminated and that according to Riley, et al., estimate that 17,000 new units will be built each year, 35,000 new jobs will be created. This figure is above one- third of the accumulated number of unemployed and low paid jobs estimated for Cali in 1973. The new construction jobs and the multiplier effects could result in full employment before 1980. Full employment would raise expected urban earnings and influence the level of migration to Cali. Thus, housing needs and employment opportunities will be affected44 but it appears that the growth of the construction sector has limitations arising from the lack of effective demand sector of Bogota grew at a fast rate during the period 1966 to 1970 but did not grow during the period 1962 to 1965. However, the price index for construction sector rose at a faster rate than the general price index for consumer goods in Bogota during the period 1950 to 1970. See Miguel Urrutia M., "Problemas de Oferta de Factores de Produccion en la Nueva Estrategia de Desarrollo," pp, pip., pp. 105-8. 41In Chapter V we pointed out that the construction sector does not have excess capacity. 42Eduardo Sarmiento P., DNP, Aspectos Cuantitativos del Plan de Desarrollo, pp. cit., pp. 88-120fi 43Harold Riley, et al., pp. pip., p. 313. 44Similar conclusion was reached by Riley, et al., pp. cit., p. 311. 199 for houses and the rising cost of production which, in turn, decreases houses purchased. Promotion of exports is another development strategy 45 Nontraditional exports, that is contemplated in the plan. exports other than coffee, bananas and oil,46 have increased at rates superior to those planned. The availability of foreign exchange is an important factor in economic growth plans and for employment, due to the elimination of bottle- necks in the productive sector. It is clear that the most abundant factor of production possessed by Colombia is manpower, thus labor-intensive exportable lines of production 47 Recently a short study was published48 should be shought. which analyzes the existing possibilities not only for minor exports, but also for labor intensive exports. They show, by the presentation of four new manufacturing firms established by the private sector, the possibilities for the generation of foreign exchange, absorption of labor and private profits 45The actual Development Plan contemplates four strate- gies. We already mentioned the construction sector and exports. The other two are agricultural productivity and distribution of income. See: Departamento Nacional de Planeacidn, Guidelines for a New Strategy, pp. pip. 46Agricultural exports from Valle other than coffee are: cotton, sugar cane, soybeans and beans. 47Gabriel Turbay M., Fedesarrollo, op. cit. In this essay the feasibility and difficulties encountered in devel- oping this type of industry are amply analyzed. 48Eduardo Wiesner y Daniel Vargas, Fundacién para la Educacion Superior y el Desarrollo. Las exportaciones y el ‘Empleo una Perspectiva para Colombia, Bogota,'Fedésarrollo, 71972? 200 of the entrepreneurs. The industries are: flowers grown on the central plain of Bogota, three-dimensional books made in Popayén, women's coats sewn in the zona franca of Barranquilla, and a factory producing toys for dogs. The analysis of each case shows the rapidity with which they developed and the economic convenience both for the country and for those involved in this activity. Incentives to establish this kind of activity have been (1) higher effective exchange rates,49 (2) promotion and information 50 of external market opportunities, and (3) creation of zonas francas where few restrictions on imports and exports exist. Nelson, Schultz and Slighton argue that capital-labor ratio is positively related to the ratio between the real ‘wages and the rate of exchange and negatively related to the 51 If the exchange rate increases with real interest rate. 'wages and interest rate constant, the capital-labor ratio will fall and more labor will be employed to produce the same 49The exchange rate has risen annually 8.85 percent since 1967 when the CAT was created and the consumer price index in Cali has risen 10.6 percent annually. CAT stimu- lates exports as shown by John Sheahan and Sara Clark, "The Response of Colombian Exports to Variations in Effective Exchange Rates," (Mimeo). Colombian exports respond favorably to rises in the rate of effective exchange. 50The government created an office to promote exports PROEXPO, and the private sector has several organizations like EXPASA. 5192. pi_t., p. 186. 201 amount of output. The same result will be observed if the interest rate increases. If these relations are valid we see that policies oriented toward increasing the price of imported goods and capital, lead to production methods which favor labor-intensive and domestic capital. Improve the Performance of the Labor Market The lack of information among prospective rural-urban migrants on urban wages and opportunities for employment (see Chapter III) suggest the need for creating information services for rural areas. The simultaneous existence of unemployment, underemployment and low paid jobs in both urban and rural areas indicates the need for measures aimed at eliminating or decreasing the difference between acquisi- tion price and salvage value.52 These measures should be taken at the proper time, that is before factors like age and seniority that tend to increase the gap make it impossible to reduce the difference between acquisition price and salvage value. One possible measure could be to decrease transportation costs and the cost involved in living while unemployed, this latter cost being directly related to the time involved in finding a job and the creation of employment and settling in public agencies. Policies aimed at affecting urban employment which, if successful, will modify the expected salvage value of rural 52These terms were defined in Chapter III. 202 laborers, that is the probability of finding an urban job in a given period of time multiplied by the salvage value of rural laborers will diminish the difference between the acquisition price and expected salvage value. In conclusion the lack of effective demand for the output of the construction sector is a serious barrier to the expansion of this sector to achieve lower unemployment while absorbing the increasing unskilled urban labor force of Valle. Though the construction sector has important possibilities of absorbing unskilled labor force, it appears necessary to promote labor-intensive small and medium-sized firms mainly producing consumer goods for both the internal and external markets. The growth of these kinds of indus- tries could be helped through the proper revision of wage regulations, credit and other marketing facilities, technical assistance, and relative prices in favor of labor-intensive capital and unskilled labor. A more balanced development between the rural and urban areas should receive priority. This can be accomplished by providing funds and technical assistance to the traditional small farmers rural sub-sector and a relatively higher level of resources to the rural sector as a whole in order to maintain rural wages, productivity and levels of living relative to the urban sector. In this way it is expected that rural-urban migration would not accelerate when the urban unemployment rates decrease as a result of the new jobs 203 created by the construction and small and medium-sized manufacturing firms. Actions to help create employment and the development of rural and urban small and medium-sized enterprises will result in a more even income distribution. Training the labor force and the control of population growth will also result in a more even distribution of income and less unemployment and underemployment in both rural and urban areas. Information services on urban wages and employment opportunities in the rural areas as well as assistance for moving from rual areas for those who should move (P: + an > Pa + ar’ see Chapter III), like lower transportation cost, shortening the period of the time of arrival to the first payday, and other services from employment and settling in public agencies are called for. The same applies to those who should move from the cities to other areas of the country. CHAPTER VIII SUMMARY AND CONCLUSIONS The general objective of this study was to analyze rural-urban migration and its relation to unemployment in the urban areas of Valle, especially of Cali. The specific objectives were: (1) to formulate a decision model on rural-urban migration under conditions of urban unemployment and imperfect knowledge on the part of the prospective migrants, (2) to formulate a model for estimating the labor force of urban Valle and particularly of Cali, (3) to develop a theoretical framework that explains the limited labor absorption of Valle's manufacturing sector for unskilled labor, (4) to estimate the number of jobs needed (a) to prevent a further rise in unemployment or (b) to lower unemployment rates in urban Valle during the period 1974 to 1980, and (S) to identify and examine policy alternatives ‘with.respect to labor employment. Analysis of the characteristics of the population (Chapter II) of Valle and Cali shows the high growth rate (If the population, 3.5 percent per year in the department, and 6.4 percent in Cali. This enormous annual increase of the population of Cali has been due in part to the strong migratory current from both urban and rural areas. 204 205 Valle has a high percentage of its population concen- trated in urban centers (cabeceras municipales), 70.4 percent; the urban population had a high growth rate of 6.3 percent, while the rural population decreased at an annual rate of 0.6 percent between the years 1951 and 1964. Estimates for 1973 indicate that nearly half the population of the depart- ment lives in Cali. The population of Valle and Cali is very young, with a high proportion of persons under age 15. Valle receives a large number of persons born in other departments of Colombia. In 1964, 37.5 percent of the resident population was born in other departments. The growth rate of the non-native population between the census years was 2.6 percent annually. The migrant population coming from municipal cabeceras constituted 69.3 percent of the total migrant population and 77 percent of these migrants settled in urban zones or cabeceras municipales. The pro- portion of Cali's population born elsewhere is 57.3 percent. Urban-urban and rural-urban migration to Cali is distributed as follows: 77.6 percent for the former, and 19.9 percent for the latter. Nearly 77 percent of the population which migrates to Cali is of working age (15 to 64 years old). Agricultural and industrial growth of Valle are 6.5 and 7.5 percent, respectively, equalling rates proposed for the country by some national development plans. When these growth rates are compared with the rates of absorption of manpower, 3.3 percent growth for the manufacturing labor 206 force while agricultural labor force declines, it is clear that production increases were due to utilization of capital and improved technology rather than the utilization of more manpower. The tremendous increase in Valle's urban pOpulation is largely explained by immigration of people from other depart- ments of Colombia and by rural-urban migration within Valle. The decrease in Valle's rural population is due to migration to the urban areas. The importance of rural-urban migration is easily overlooked because the definition of urban areas includes many rural places as urban. Factors determining urban-urban (from small towns to large cities) and rural-urban migration are: (l) wages or MVPs below their expected net earnings after migration. Wages and MVPs are very low in Colombian small towns and rural areas due to underemployment and the small holdings that the majority of the farmers have. If the urban unemploy- ment rate decreases, the expected salvage value of rural people will increase and more rural-urban migration will take place. There were high migration rates in the receiving areas like Cali; (2) differences in living conditions between the rural areas and small towns with respect to the more developed areas of Valle; and (3) high population growth rates in rural areas. Prospective rural-urban migrants of Valle are poorly informed with respect to wages and employment levels in 207 urban areas as shown by the survey conducted in the municipio of Palmira, corregimientos Rozo, Acequia and Latorre. Young people and females migrate more than older people and males in Valle as shown in Chapter II. Young people migrate more because they are not yet fixed in farming like older people. Females migrate more than males because women have fewer opportunities in rural areas where their MVPs are less than their expected salvage value in spite of the fact that their off farm wages are also quite low (Chapter III). The unskilled labor force in Valle and Cali in a given time is principally determined by (a) the high growth rate 3.4 percent, of the native p0pulation; (b) the rate of par- ticipation in the labor market which is 86.3 percent for men of working age, and 25.5 percent for women of working age; (c) migration to the urban area of Valle especially to Cali; and (d) the low conversion rate from unskilled to skilled laborers. This rate has been low due to few available training institutions and high cost of programs in education and training, and the uneven distribution of income. The labor force in the year t was estimated by sex and place of origin through equations of the form Xt = T(l + r)th, where X0 is the initial population, T the rate of participa- tion in the labor market, and r the population's growth rate (Chapter IV). Labor absorption in the urban sector of Valle and Cali is analyzed in Chapter V, where sources of employment 208 are divided into industrial and other sectors. Very limited amounts of labor are used by the industrial sector which accounts for less than 15 percent of the labor force in the urban sector. The low absorption rate for unskilled labor on the part of the industrial sector of Valle is explained by the growth of a modern industrial subsector, composed by large enterprises, which use labor-saving capital, extensively, and, therefore, have a high capital- labor ratio. The growing importance of large enterprises producing intermediate, durable, and capital goods uses labor-saving capital will further lower the absorption of unskilled labor in the modern industrial subsector. 0n the other hand, small and medium size enterprises find it difficult to expand due to (a) limited funds, (b) relative higher costs of labor with respect to its productivity than those for large firms, that is the ratio between MVP of labor and its marginal factor cost (MFC) is less for small than for large enterprises, and (c) lower profits due to higher cost of production and lower prices for its products. It was observed a 10 percent increase in the number of large firms, those hiring more than 50 laborers, during the period 1963 to 1969 while the number of firms with 15 to 24 laborers decreased from 1963 to 1966.1 Handcraft (firms with less than 10 workers) and very small 1No data were available for other years. 209 manufacturing firms (10 to 14 workers) increased in number reflecting the pressure of people looking for something to do in order to make their living. These growth differences among industries of different sizes show a clear tendency toward greater dualism, that is a few large and numerous very small firms and a few medium size firms. Under these circumstances, there is a group of skilled and unskilled laborers working in the large enterprises with relative high salaries while a great number of people working in small firms or in the service sector with low salaries or unemployed. An uneven income distribution is associated with such a situation. With respect to the cost of unionized labor we stated that once workers are unionized the expected monetary cost of labor increases substantially. Some medium-sized firm owners faced with 25 workers or more, and hence, unionization, consider the cost of dealing and managing a large number of workers, the time consumption and the burden of conflicts, losses caused by eventual strikes, and the cost of assuring job stability to union members. Thus, the low cost of capital which substitutes for unskilled labor relative to the total (monetary plus non-monetary) cost of unskilled labor increase the use of labor-saving capital and reduces use of unskilled laborers. The data showed, as we noted above, that there is a drop in the number of firms classified according to the number of workers hired once the limits of III, \III] ill-l! All" 210 15 to 19 and 20 to 24 workers are reached and, probably, a lower number of firms with 25 to 29 workers. There is a ‘decay curve relating number of firms to size in a given year. However, for firms with 15 to 24 workers the curve levels off as the 1963 curve showed. A sharp decrease in the number of firms with just over 25 workers was expected. Unfortunately, information was not available for the group of companies with 25 to 29 workers. It is frequently argued that industrial branches pro- ducing consumption goods for the domestic market use higher proportion of labor along with domestically produced materials the production of which employs much manpower. We examined the effect of a greater demand for consumption goods and less demand for durable, intermediate and capital goods holding constant the total aggregate demand for industrial products. Such a change in the composition of the aggregate demand would not increase employment. The consumption goods industrial branches seem to be dominated by large firms using high capital-labor ratio as well. There is an excess of installed capacity in Valle's industry. Low utilization of installed capacity is profit- able for Valle's relatively large firms. When entrepreneurs ‘have expectations of future increase in demand the cost of installed.capacity decreases through time. .Also, under uncertain conditions of demand the entrepreneur may decide tn) buy flexibility. Economies of scale tend to favor 211 unutilized capacity when there are expectations of increasing demand for the product. When the cost of labor increases with firm size and when labor is more expensive for night work owners install labor-saving equipment even though it is not used to full capacity as the cost of equipment is less than for unionized labor. Low excess capacity was found in the construction sector and in some industries producing consumer goods. Two studies on Colombian unemployment obtained con- tradictory estimates for elasticities of substitution between labor and capital in 20 Colombian manufacturing branches. We showed that grave statistical problems are affecting both sets of estimates, namely specification errors and multicollinearity which lead to upward biased and imprecise estimates of the elasticities. Further research on the nature of industrial production functions should be carried out to be able to evaluate input price policies (Chapter V). Unemployment rates were seen to rise over time in Cali as shown in Chapter VI. In the year 1969, Cali had an unemployment rate of 18.3 percent. In 1971 this figure was lowered to 12.4 percent due to the exceptional amount of work generated by the Pan American games. Open under- employment and disguised unemployment are frequently found in Valle and Cali. High migration rates to urban Valle, especially to Cali, were responsible for the growing unemployment rates found up to 1969. It appears that the 212 high rates of unemployment have slowed down migration to urban Valle, including Cali. The unemployed population in urban Valle and Cali were estimated at 72,800 and 69,000, respectively, at the end of 1973. Laborers with low paid jobs in Cali were estimated at 33,000 at the end of the same year. If the natural population growth rate continues to be 3.4 percent as in the past, and migration rates from 1964 to 1973 are equal to the observed rates from 1951 to 1963, the additional total number of jobs for Cali during the period 1974 to 1980 depends on the rates of migration in the same period. If the migration rates from 1974 to 1980 will be equal to rates observed from 1951 to 1964 (Assumption 1) the additional total number of jobs for Cali to prevent increases in unemployment are estimated at 40,300. If the migration rates from 1974 to 1980 will be one-half the rates observed from 1951 to 1964 (Assumption 2) the additional total number of jobs for Cali to prevent unemployment increase are estimated at 18,900 for the period 1974 to 1980. If the migration rates from 1974 to 1980 are one- quarter the rates observed from 1951 to 1964 (Assumption 3) the additional total number of jobs for Cali to prevent unemployment increase are estimated at 3,900. The number of jobs needed to prevent underemployment and low paid jobs from increasing are 19,300 under Assumption 1, 9,000 under Assumption 2, and 3,900 under Assumption 3. Thus, the 213 additional number of jobs needed in Cali to give everyone of the projected labor force the opportunity to find a job from 1974 to 1980 are 161,600 under Assumption 1, 129,900 under Assumption 2, and 114,100 under Assumption 3 (Chapter VI). Policies oriented toward production growth and pro- ductivity have produced (1) high unemployment and under- employment rates, and (2) very uneven income distribution. Agricultural and manufacturing sectors of Colombia are characterized by a dualistic system, more acute in Colombia than in other countries and within Colombia more striking in Valle than in other Colombian departments. Under dualism policies oriented to increase agricultural and manufacturing production in the modern sectors can hurt the traditional sectors. For instance, a given measure can be very successful in increasing production and raising productivity in the modern sectors while small and 'medium size firms in the traditional sector are driven out of business and laborers out of employment. Growing rates of unemployment in Valle and Cali occurred simultaneously ‘with agricultural and industrial production growth rates high enough to meet the goals of many development plans. We discussed the following policies aimed at affecting the amounts and kinds of labor at the large urban centers of Valle: (l) revision of labor code especially as it affects skilled and unskilled laborers and enterprises of 214 different sizes, (2) reduction in growth of total labor force, (3) changing composition of labor force and upgrading of skills, and (4) slow down the rate of migration to the large urban centers such as Cali with incentives to stay in farming and rural population centers and (5) an improved information system with respect to job opportunities and wages. Also, the existence of both urban unemployment and rural underemployment despite high migration rates to the urban areas of Valle suggest the need for improving the performance of the labor market with a better information system. A sixth policy, the creation of additional jobs in urban areas for large pools of unskilled laborers is very urgent in Valle. The control of the number of unskilled laborers in the urban zones of Valle, especially in Cali, will be obtained through the creation of numerous educational and training opportunities and of various employment possibilities in the rural area which might be achieved by changing the distribution of rural property in favor of the poor farmers, assisting them.with their exploitation, and the reclaiming of land. The lowering of the urban unemploy- ment rate may also be accomplished by programs for rural development to reduce migratory flows from the rural to the urban areas. In the long run birth control will decrease the growth of the labor force but increased female participa- tion in the urban labor market will increase the labor force. 215 The ILO's policies gave importance to the development of rural areas and focused on small farmers and labor intensive agricultural products as a means for increasing the rural labor force as we suggest in this study. While the Economic Development Plan concentrates in the creation of urban jobs, the proposed ILO's plan to lower unemployment rates of Colombia to 5 percent by 1985 contemplates a 1.8 percent increase in agricultural labor force a goal not applicable to the Valle's case but important for departments from which most of the Valle's non-native population comes. The creation of additional jobs in urban areas for these large pools of unskilled laborers is very urgent in Valle. Factors affecting the number of jobs for unskilled laborers discussed in this study and which are subject to public policy are: (1) relative price of different kinds of labor and capital, (2) factor proportions used by size of enterprises, (3) use of installed capacity, and (4) amounts and kinds of labor absorbed by different products producing for both the internal and external markets. Although the current national policy is to stimulate urban construction, the lack of effective demand for houses appears to be a barrier to the expansion of the construction sector as a means to lower unskilled unemployment rates. The growth of labor-intensive small and medium-sized manufacturing industries producing mainly consumer goods for both the internal and external markets seems a necessary 216 complement to the construction sector stimulation. Small and medium-sized enterprises could be helped through the proper revision of wage regulations, credit and other market- ing facilities, relative prices in favor of labor-intensive capital, and technical assistance. Finally, the existence of urban unemployment and rural underemployment despite high migrations rates to the urban areas of Valle suggested the need for improving the per- formance of the labor market of Valle. To imporve such performance it was recommended the creation of public agencies to provide information services on urban wages and employment opportunities, to assist prospective migrants in finding jobs and settling in urban areas (Chapter VII). Future Research The study of migratory currents was made possible for Valle due to the efforts of the Universidad del Valle to process the 1964 population census data in more detail for the non-native population. In 1973 a new population census was carried out. The data processing system used for the census of 1964, and this recent information will allow us to determine more recent trends in migration to and within Valle. Furthermore, the 1974 population census will give us important demographic and economic data valuable to study employment characteristics. Previous estimates of elasticities of substitution made by Planeacion Nacional and Gaviria,G6mez and L6pez 217 did not distinguish between different kinds of labor and different kinds of capital and, besides, have serious statistical problems and the estimates are of little value, if any. Investigations are needed to improve estimates of these elasticities for use in making policy decisions on input prices. Furthermore, we have to study the capital market to see how it is affecting the price of different kinds of capital paid by enterprises of different sizes. Possibilities of substituting imported inputs and new materials for domestically produced ones should be studied as well as the role that the prices of foreign exchange, wages and internal interest rates play in the above substitutability process. We need to know the kind of goods being produced by small farmers and small manufacturing firms, the level and fluctuation of prices of these products and how they affect the incomes of these firms. Also, it appears that economies of scale are not commonplace in agriculture2 but we do not know what is the case for manufacturing industry. It will be important to determine the effect of the present National Development Plan on urban employment since 1972, as soon as data becomes available, since the plan gives the construction sector the leading role to create new jobs for the unemployed. 2Albert Berry, Development of the Agricultural Sector in Colombia, forthcoming. 218 Finally, performance evaluation of public institutions created to help small farmers and small manufacturing firms should be carried out in the future. The problem seems to be that governmental officials do not recognize the great importance of helping small enterprises and hence do not provide enough resources and assistance to small enterprises through the institutions which exist. APPENDICES APPENDIX A SOME CHARACTERISTICS OF THE POPULATION OF VALLE AND CALI 219 .qoaa was Hmma .maaaoaoo mu amauwanom as omaoo .mzo was me m.Hu o.m~ nmo.om~ mmm.~HH AHH.msH ~.ms Hwo.mom New.w~H sHN.omH souma m.o m.Hm mmm.se~ smq.mHH ano.o~a o.mm moo.sm~ Nam.~aa omn.a~a «Hams ma amsu mmoq o.ou o.m~ oH~.mHm osa.om~ o-.o- m.on mom.omm mac.~s~ e-.aom Hanna m.n o.ma nma.mm nHH.o~ Nso.mfl m.nm owm.kfl oaa.oa cad.“ uo>o was me o.m o.- Nam.mmo Hom.nmm oao.mom m.Hn msm.m~m ago.mnfl ama.~ma «sums s.a m.ms mmo.m~m amm.qo~ ow~.ao~ o.~q s~s.ha~ s-.soa owa.moa ammo» ma non» mama m.o s.o~ amm.a-.a mow.mmo sha.m~m h.ms wmm.omm amo.mm~ aflm.~o~ can“: m.m o.ooa mam.ss m~o.m~ ~aa.- o.ooa coa.am NwN.ms cno.sa uo>o van no m.~ o.ooa ~mo.mam ¢m~.mos mm~.osq o.ooa smm.smo oaa.aom mas.~mm sauna s.s o.ooa oo~.oan me.~mm mum.hmm o.ooa Nms.ass om~.oa~ ooN.s~N mumus ma noon mama m.m o.ooa mmo.mm~.a mow.m~m om~.omm o.ooH -m.eOH.H ~ma.mmm mmh.aam Hmuoa Illuflmuuwmll llucwouomi II I HUEZIIIIIIIIII lluamuuwml II MOQQZIII'IIIII sowuuoaoum kuoa uoao3 no: coauuoaoum Hmuoa coaos # so: nusouu mo ooaowfimom was mum“ Hmsaa< soma Hmsa masons mw< .qoma was Han .xom van mw< .modovqmmm ou waavuooo< suaouu we wumm cam oHHm> mo manomm mo umnaaz .H.< manna .«oma pom Hmma .mwoaoaoo mu coaomaoom ow oncoo .mzauo< ounaocoomH 0.0m qqm.mmm mmo.mmm.a 0.0m ou~.wmm nmm.ooa.a Hmuoe «.ma ooa.ooa mom.onm o.HH mom.ao NmH.nmm c6803 ¢.m¢ «mm.mmq omm.mmm a.mm Hom.omm mmn.amm so: ucmuuom, uuuuuuuuu Hmoadz ........ unmouom uuuuuuuu Hoomoz uuuuuuuu cowumaoaom aowumaamom coauuoooum H.m.<.m Hmuoa aowuuomoum H.m.<.m Hmuoh «oma Hmma BouH .83 was. $3 .33, mo 83338 $33. owaocoom .N.< 3an Table A.3. 221 Relation Between Migrant Population and Total Economic Active Population of Valle del Cauca, by Age and Economic Sectors, 1964. Economic Sectors Economic Active Total Economic Proportion Migrant Population Active Population - Number- -Percent-- Primary_Sector1 81,728 178,606 45.8 Less than 15 years 1,899 8,774 21.5 15 - 64 75,836 163,040 46.5 65 and over 4,003 6,792 58.9 Secondary Sector2 67,133 130,804 51.3 Less than 15 years 539 1,458 37.0 15 - 64 65,531 127,217 51.5 65 and over 1,063 2,129 49.9 Terciary Sector 119,671 220,134 54.4 Less than 15 years 2,653 6,321 42.0 15 - 64 114,449 209,359 54.7 65 and over 2,569 4,454 57.7 12531 268,532 529,544 50.7 Less than 15 years 5,081 16,553 30.7 15 - 64 225,816 499,616 51.2 65 and over 7,635 13,375 57.1 1 . Agriculture and extractive industries. 2Transformed goods and construction. Commerce, transportation, communication, public services, etc. DANE. Cauca. Source: XIII Censo Nacional de Poblacion Resumen del Valle del Bogota: DANE, 1967. 222 .ssas was Hmofl .msnaosoo ms assumaaom as omaou .mzsa “muuaom m.s- s.m smm osN smm m.~s mas sss son uo>o saw ms m.au s.~ oso.oa asm.s mus.n ~.ss oso.m~ smm.os shs.ss so . ms w.su s.m wms.a Hmm.s amh.s N.ss ssm.ss os~.m ssn.m spam» ms cuss smog ~.s- H.m ss~.ms mss.s an.os H.ms mNm.~s o-.mfl smn.m~ Hanan N.n s.ss -m.ss NNm.m ooa.s n.5m sms.s sms.s oom.~ um>o saw as s.s ~.nm maH.ssm ssh.~ms mas.sms m.mm mso.HmH m~s.om smn.o~ so . ms o.a s.sm mas.~m~ mss.s~s mmo.m~s m.mw mss.mm Hmm.ss Asa.ss mumms mH amfiu mmmA m.~ m.sm ma~.mas ~mo.-m mms.ss~ m.sm hmm.ss~ som.s~s Hmm.ssg away: s.s o.oos wmo.~s ~so.oa ssa.s o.ooH ass.“ oss.s soo.m um>o can ms m.n o.OOH mm~.smm smm.~ss Hoa.sss o.oos mso.sas mso.sm mmc.nw ss - ms m.~ o.oos sms.ss~ s~m.mms Nam.Nms o.ooH sss.oos HNH.om msm.om mums» . ms sass mama s.s o.oos s~s.ams o-.smm mo~.Hom o.oos sms.sm~ sua.mss oss.wmj Hosea uflmuuwm udwoumm llllllldllHflQE—aZIIIIIIIII udfloumm ,Iluwngzwllllllll aoauuomoum H.309 _ son—o3 _ no: aowuuomoum Hooch _ con—03 _ so: .3395 no moaowfimom was «sum saunas ssss Haas masons was .ssma sum snas .xmm sum was .muamssmom an smsmsmmsau “Ham mo coaumasaom .s.< magma .«oma was Hmma .wHoEoHou mu cowomanom mo omaoo .mz «0 Ego—.3 3 mo g gawk, 2nd 0.33. 225 .mzom Nam.HH mao.HH mom.HH «NN.HH mow.o~ mm¢.oa unm.m cm~.m n¢m.m nmw.w OOH.» anm.~ Hum.n noon mama nmma coma mama «oma moma Noma Homa coma mmmH wnma mama onma mummy Seaman acupunsvcu .moma o» ommfl .moauu Hon oaag> no munoamum HmfluumswaH NH a“ uoaoflaam muuxuoz mo nunauz .~.m magma 226 .mz4a mp vouomaaoo muwv wonmaanancoc was mNa .maa .qoa .mm .moz coauamvuumm «a umamamcmz mocauwaom .mzom mmm.nm~ om¢.m¢u mmo.wa~ amn.nna mmo.~ma mum.maa mmn.om m~5.oo am~.ao Noo.w¢ mmw.~m oqm.n~ mma.o~ voom mama noma coma mema «oma moma Nwma aoma ooma mmma wmma nnma omma mung» noawum amauuwuvca .Amonmm vGMmDOSHV woma cu omma .mosmu amv maam> mo mmsuamum amauumstH ma ca vwmoamam muouonwa mo uwaz awaan< .m.m maQMH 227 ‘Table B.4. ‘Manufacturing Sector Price Index and Implicit GNP Price Index, 1956 to 1968. Year Manufacturing Implicit GNP Price Index Price Index 1956 76.5 75.3 1957 90.0 88.2 1958 100.0 100.0 1959 106.8 106.6 1960 119.6 115.7 1961 129.2 125.2 1962 143.4 133.5 1963 184.7 164.5 1964 198.9 191.4 1965 220.3 208.9 1966 245.4 240.1 1967 261.0 260.0 1968 275.5 284.4 Source: Computed from data in Revista de1 Banco del 1a Repfib1ica, 1958-1970 and national social accounts. 228 Table B.5. Number and Area of Farms Classified by Size, 1959. Farm Size (Hectares Number of Farms Area (Hectares) Number Percent Number Percent Less than 1 10,692 21.0 5,091 0.4 l to less than 2 6,649 13.1 10,415 0.9 2 to less than 3 3,467 6.8 8,762 0.8 3 to less than 4 3,973 7.8 13,927 1.2 4 to less than 5 1,728 3.4 7,724 0.7 5 to less than 10 8,991 17.7 63,929 5.5 10 to less than 20 6,480 12.8 93,527 8.0 20 to less than 40 3,843 7.6 110,759 9.5 40 to less than 50 755 1.5 33,761 2.9 50 to less than 100 2,069 4.1 142,257 12.2 100 to less than 200 1,166 2.3 170,120 14.6 200 to less than 500 717 1.4 220,045 18.8 500 to less than 1000 215 0.4 143,672 12.3 1000 to less than 2500 67 0.1 92,089 7.8 2500 and over 11 --- 51,869 4.4 Source: Universidad del Valle: Censo Agropecuario del Valle del Cauca, 1959. Cali, 1963. 229 Table B.6. Number and Area of Farms Classified by Size, 1970. Farm Size (Hectares) Number of Farms Area (Hectares) Number Percent Number Percent Less than 1 12,550 28.5 5,173.2 0.5 l to less than 2 5,263 12.0 7,564.0 0.7 2 to less than 3 2,709 6.2 6,613.2 0.6 3 to less than 4 2,374 5.4 8,143.9 0.7 4 to less than 5 1,370 3.1 5,981.7 0.5 5 to less than 10 5,976 13.6 42,256.l 3.7 10 to less than 20 5,060 11.5 72,255.5 6.4 20 to less than 30 2,101 4.8 50,990.l 4.5 30 to less than 40 1,470 3.3 50,302.6 4.4 40 to less than 50 799 1.8 35,205.8 3.1 50 to less than 100 2,080 4.7 l41,l33.6 12.5 100 to less than 200 1,134 2.6 159,060.4 14.1 200 to less than 500 783 1.8 235,995.8 20.9 500 to less than 1000 205 0.5 129,918.6 11.5 1000 to less than 2500 83 0.2 122,698.0 10.8 2500 and over 13 --— 58,292.6 5.1 Source: DANE, Censo Argopecuario 1970—1971. APPENDIX C ILO'S MODEL FOR PROJECTION OF EMPLOYMENT IN COLOMBIA 230 386 886 86 886 886 .886 886 886 886 «386 886 ~m86 886 886 36.333 §u3u8§ , 3.86 386 886 886 886 886 -86 886 386 886 886 886 886 886 8558 3588 886 86 886 886 886 .886 86 886 ««86 .386 .886 88.6 886 386 83.5.3 0383015 886 -86 386 886 886 886 886 «866 886 86 886 886 386 886 3960.5 3888.3 «~86 m~86 886 «866 «886 886 886 886 886 .3. 886 886 886 886 338 38 886 886 8.86 886 886 886 8.86 886 386 386 886 886 «866 886 8.855 63358.82 886 .386 386 386 386 886 «36.6 886 8.86 886 886 6366 886 886 88 8.86 886 886 886 «366 886 8.86 886 886 886 «866 886 886 886 3960.8 88 88.~ 886 886 88.6 0.866 .886 886 «~86 386 886 386 886 386 886 woos» ”.898 .8583 .886 886 886 «686 886 886 886 386 886 886 6866 886 886 886 8388 886 886 886 886 886 8.86 886 886 886 8.86 «366 886 886 8.86 8&8 886 886 886 886 ««86 886 886 886 886 886 886 886 8.36 886 «H58 5683 8.86 886 886 886 886 R86 886 886 886 .886 386 886 «886 886 303 886 886 886 386 886 -86 886 886 886 886 ««86 «~86 886 886 8338 88 6 8.8 6 n~8 6 88 6 38 6 836 6 88 6 H86 6 8.8 6 88 6 88 . o «68 6 88 6 ~86 .6 83860.3. 88 6 3.8 6 «n8 6 88 6 88 6 88 6 88 6 88 6 m«8 6 8.8 6 88 6 88 6 8.8 6 38 6 8030.3. m86 «686 m~86 8.86 886 886 886 «~86 m~86 886 886 886 886 8.86 888 88 6 88 6 886 38 6 88 6 88 6 88 6 m8 .6 «H8 6 8.8 6 .886 38 6 38 6 88 6 68m 83 83 83 83 83 «o3 83 83 83 83 83 83 83 83 was.» 885 1.5865 .88 83 6588 5 838on mo 83> «>3qu «5 8m 33, mo .8985 6mm .3 3.58: gm: mo ~55. st 8253 8338 Sn 3an APPENDIX C ILO'S MODEL FOR PROJECTION OF EMPLOYMENT IN COLOMBIA1 1. Here the model utilized for the projection of growth of employment on the basis of certain hypotheses is described. We adopted the following classification of employment, in four sectors: Agricultural Sector (agriculture, cattle raising, fishing, forestry). Other Sectors: Sector alfa, with great density of capital and highly skilled labor (mining, modern industry, public services, transportation). Sector beta, with low density of capital, few skilled workers (construction, handcraft, commerce, restau- rants, personal services). Sector gamma, with density of professional specializa- tion (banking, insurance, public administration, other services). 1Translated to the English language from Oficina Internacional del Trabajo, QE- cit., pp. 411-3. 231 232 2. The diverse characteristics of the nonagrarian sectors may be summarized approximately as follows: Characteristics .Alfa Beta (knnma Pnflkesflmxfl.quakUficatnxw Gtaflxm' Leastban Gmamer thmiuean ‘manl flmrlmami (hmiufl. lfigh M9fi£mtx> low low Champszkxldxnrpnxhctbdty Impnxent Sanxe: auuce Champszhipmodxnfion Impmnant Smmxe 'Mafinm Prdbability of entrance of rural migrant in the sector Low High Low The projections are extrapolated from the tendencies. This is carried out through the use of a simple econometric model, whose basic idea is that the income of foreign capital, that is, exports and imports of capital, have in great part determined the evolution of the Colombian economy between 1950 and 19661: 1. The income produced by these two factors has determined the level of capital formation for transportation equipment and machinery by means of these imports. Thus this type of invest-investments (IMt) is a function of: a) Monetary resources in foreign exchange (Zt), that is exports (Xt 1Banco de la Repfiblica: Bogota, 1969. exports of coffee + exports of Cuentas Nacionales 1950-1967. 233 unrefined oil + "minor exports") plus (At) net income of foreign capital. b) Exchange relations (that is, fit = export prices divided by import prices): Log IMt = -3.51 + 1.036 Log Zt + 1.508 Log At (.09) (.02) Toward Full Employment 2. The formation of capital in construction form (IC) is almost independent of foreign exchange resources and may be estimated as follows: Log 10 = 0.280 + 0.921 Log ICt-l (0.121) R2 = 0.892 The total formation of fixed capital (I) is equal to IM + 10. Changing some of the estimates of the taking off of the formation of capital in the four sectors,1 and leaving a margin for the incremental capital/pro- duct relation (CPR)2 it is possible to obtain the 0 Gross Internal Product (GIP) for sectors at the cost of the factors. It is possible to estimate the employment level (number of man-years) for each sector with a basis on the changes of labor productivity by sectors (that is, GIP at the cost of the factors for man-year) 1Estimate made for 1966 in collaboration with the DANE. 2 ICOR. 234 calculated for the 1951-1964 period. 5. Different hypotheses have been established about the projected values of the variables and exogenous parameter: a) Exports and Net Income of Foreign Capital Rates of Annual Growth 1970-1985: Minimum Medium Maximum Hyp. Hyp. Hyp. Coffee exports 1.5 2.0 3.0 Unrefined oil exports 1.5 2.0 2.5 Minor exports 6.0 8.0 9.0 Capital income 3.5 4.5 5.5 b) Terms of trade (1958 = 100). Due to the increase in the price of coffee in the last few years, it is probable that the terms of trade index will equal 100 in 1970. Consequently two hypotheses have been established: 1970 1975 1985 Minimum Hypothesis 100 95 85 Maximum Hypothesis 100 100 100 c) Capital/product relation, CPR. Between 1950 and 1966 the general measure of the CPR has been 5, but since the data is not sufficient, we have based our estimate of the CP sectorial on 235 international experience, except for the indus- trial sector, where Colombian figures have been used. Two bases for calculation were used, depending on whether they gave high or low pro- jections with respect to income and employment. The corresponding projections are given below. Sectors 1970-1975 1975-1985 Minimum Maximum Minimum Maximum Hypothesis Hypothesis Hypothesis Hypothesis Primary 6.0 5.5 5.5 5.0 Alfal 5.0 4.5 _ 4.5 4.0 Beta 2.5 2.0 2.3 1.8 Gamma 1.5 1.0 1.4 0.9 1 services. Account taken of the high CPR in transportation and public For example, in 1949-1959 the figures relative to Greece are 6.2 and 10, respectively. d) The productivity of labor has been projected on the hypothesis that the tendencies will be maintained during the period, except in the case of agriculture, where we have hypothesized a greater increase than in the past; thus the following percentages of growth in labor pro- ductivity between 1970 and 1985 have been cal- culated: agricultural sector, 2.0 alfa, 3.5; beta, 1.4 and Gamma, 1.8. Among the different results of the projections for 1970-1985, we will only present two as 236 possible, both based on the maximum hypothesis relative to the CPR (which gives the highest GIP level, and therefore, highest employment level). With respect to the hypothesis on exports, capital income and exchange relations, the first projection is based on the minimum hypothesis. Employment Projections (In Thousands of Man-Years). Sectors 1970 Maximum Minimum Hypothesis Hypothesis 1975 1985 1975 1985 Agricultural 2,550 2,720 3,290 2,680 3,100 Alfa 590 640 810 620 730 Beta 1,350 1,790 2,460 1,610 2,220 Gamma 560 690 940 620 800 Total 5,050 5,840 7,500 5,530 6,850 Active Work Force1 6,480 7,740 10,790 7,740 10,790 unemployment Rate 21.0 24.5 30.5 28.5 36.5 1According to the Centro de Estudios sobre el Desarrollo Econ6mico de la Universidad de los Andes. 2Subemployment included (converted into equivalence in unemployment. These projections are made on the basis of the assump- tion that there is an unlimited supply of labor in Colombia; that there is no limiting factor as far as the labor situation is concerned, whether in terms of quantity or in terms of 237 quality. It must be born in mind that this means that the professional qualifications necessary for reaching the corres- ponding production level will be found in Colombia or will be imported. However, the two possibilities show very high levels of open unemployment toward 1985 and probably toward 1975. APPENDIX D QUESTIONNAIRE ON MIGRATION FOR RURAL INHABITANTS, MUNICIPIO PALMIRA, CORREGIMIENTOS OF ROZO, ACEQUIA, AND LA TORRE APPENDIX D QUESTIONNAIRE ON MIGRATION FOR RURAL INHABITANTS, MUNICIPIO PALMIRA, CORREGIMIENTOS OF ROZO, ACEQUIA, AND LA TORRE Tendency Toward Migration 162. Do you intend to spend many more years in the country? No Yes 164 163. Do you intend to remain in this same place? No Yes 166 164. Where would you like to go? Please give me the name of the place. DEpartment Municipio Vereda 165. What do you intend to do in that place? 166. Do you prefer to abandon the country or stay in the country? Stay Leave Why 167. Do you know what salaries are being paid in Cali or in another city? No 170 Yes 168. If you go to Cali or another city, how much do you think you could earn per week? 3 238 239 169. Would that salary be less than, equal to, more than, or much more than you earn now? Less Equal More Much more 170. Do you think that in Cali or other cities that there is no unemployment, little unemployment, or much unemployment? None Little A lot 171. Have you ever visited Cali? NO 173 Yes 172. How long ago did you visit Cali? 173. If you were to go and live in Cali, who would help you to find a job? (note whether interviewee gives name of neighbor, member of family, employer, etc. 174. Do you think you could find a job quickly in Cali or another city? No Yes 176 175. Why are you going to Cali or another city even when you know that finding a job will take some time? 176. How long can you live in Cali or in another city without working? Months BIBLIOGRAPHY BIBLIOGRAPHY Arrow, Chenery, Minhas and Solow. "Capital-Labor Substitu- tion and Economic Efficiency," Review of Economics and Statistics, Vol. 43, No. 2, I961} Atkinson, L. Jay. Agricultural Productivity in Colombia. Foreign Agricultural Economic Report No. 66. Washing- ton, D.C.: U.S. Government Printing Office, 1970. Berry, R. Albert. Development of the Agricultural Sector in Colombia. Forthcoming. . "Special Problems of Policy Formation in a Dualistic Agriculture: Colombia." Department of Economics, University of Western Ontario, Canada. December 1972. (Mimeo). Byerlee, Derek. "Research on Migration in Africa: Past, Present and Future. East Lansing: African Rural Development Research Network. Department of Agricul- tural Economics, Michigan State University. 1972. and Eicher, Carl K. "Rural Employment, Migration and Economic Development: Theoretical Issues and Empirical Evidence From.Africa." Paper presented at a Conference of the International Economic Association on the Place of Agriculture in the Development of Developing Countries, Bad Godesburg, Germany, 1972. Calvo, Guillermo. "The Rate of Labor Absorption on the Cost of Capital: A Note.‘ Bogota, 1971. (Mimeo). CVC. "Plan de Desarrollo Agropecuario del Valle del Cauca. " Cali: unpublished preliminary draft. 1974. . Proyecto de Regulacion del Rio Cauca. Cali: CVC, informe 72-10. 1972. "Reforma Agraria en el Valle. Cali: unpub- Ii shed preliminary draft. 1973. . "Estudio Socio-Economico del Sector Rural del Valle." Cali: unpublished preliminary draft. 1973. 240 241 . "La industria Fabril del Valle del Cauca.” Cali: unpublished preliminary draft. 1974. COFISEL. Paper for the Vallecaucano Meeting. Cali: unpublished preliminary draft. 1973. CORP. Controversia sobre el Plan de Desarrollo. Bogota: CORP, 1972. DANE. Analisis del Desempleo en Colombia. Bogota: DANE, 1970. . "Censo de Poblacién de Colombia, 1951." Bogota: DANE, 1954. . "Censo de Poblacién de Colombia, 1964." Cali: Universidad del Valle, 1968. . "XIII Censo Nacional de Poblacidn Resumen del Valle del Cauca." Bogota: DANE, 1967. . "Encuesta de la Investigacién Nacional de la Morbilidad, 1965." Bogota: DANE. . "Encuesta de Hogares en Colombia." Bogota: DANE, 1973. . "Encuesta de Hogares en Cali." Bogota: DANE, 1973. Anuario General de Estadistica. Bogota: DANE, __—1'9'59-I972. . Boletines Mensuales de Estadistica. Ns. 113 and 1182 Bogota: DANE. Departamento Nacional de Planeacién, DNP. "Comentarios Preliminares de los resultados de la encuesta, sobre utilizacién de la capacidad instalada, en la industria manufacturera fabril Colombiana." Bogota: DNP, Documento UELA-OOO-DI, Agosto 10 de 971. . "Estimativos Preliminares del coeficiente de 1ntensidad del Capital." Bogota: DNP. Documento UDRU-l77-URH. Marzo 17 de 1970. . "Analisis Estructural de la Economia." Bogota: DNP. Documento 644-UPG-UCP. Octubre 2 de 1970. Las Cuatro Estrategias. Bogota: DNP, 1973. 242 "El Empleo en Colombia: Diagndstico y Recomendaciones de Polit1ca," Revista de Planeacidn y_Desarrollo. Bogota: Junio HE 1973. . Guidelines for a New Strategy. Bogota: Italgraf, 72. . "Breve Esquema sobre el Desempleo en Colombia." UPC-002, Junio 30 de 1970. Dhrymes, Phoebus J. "Some Extensions and Tests for the CES Class of Production Functions," Review of Economics and Statistics. Vol. 47, No. 4, 1965. Eckaus, R. S. "The Factor Proportions Problems in Under- developed Areas," American Economic Review. Vol. 45, No. 3, 1955. Essang, Sunday M. and Mabawonku, Adewale F. "Determinants and Impact of Rural-Urban Migration: A Case Study of Selected Communities in Western Nigeria." East Lansing: African Rural Development Research Network Department of Agricultural Economics, Michigan State University. 1974. Flinn, William L. "Rural to Urban Migration: A Colombian Case." Madison: Land Tenure Center, University of Wisconsin, July 1966. Friedman, John and Sullivan, Flora. "The Absorption of Labor in the Urban Economy: The Case of Developing Countries," Economic Development and Cultural Change. Vol. 22, No.PI. 1974} Friedman, Milton. Price Theory. Chicago: University of Chicago Press. 1962. Gaviria G., Juan F., G6mez P., Francisco J., L6pez C., Hugo, Centro de Investigaciones Ecndmicas, Universi- dad de Antioquia. Contribucidn al Estudio del Desempleo en Colombia. Bogota: DANE, I971. Gugler, Josef. "The Impact of Labour Migration on Society and Economy in Sub-Saharan Africa: Empirical Findings and Theoretical Considerations," African Social Research. December 6, 1968. Harris, John R. and Todaro, Michael P. "Migration, unemployment and Development: A Two-Sector Analysis," American Economic Review. Vol. 60, No. l, 1970. . "Urban Unemployment in East Africa: An Economic Analysis of Policy Alternatives," East African Economic Review. Vol. 4, No. 2, 1968. 243 Hathaway, Dale E. Government and Agriculture: Public Policy in a Democratic Society. INEw York: Macmillan Co., 1963: Isaza, Yesid. Fedemetal, Seccional Valle. ”Estudio de la capacidad instalada de la industria metalmeéanica en el Valle del Cauca." Cali: Fedemetal. Abril 12 de 1973. Johnson, Glenn L. "Allocative Efficiency of Agricultural Prices as Affected by Changes in the General Level of Employment." Unpublished Ph.D. dissertation, University of Chicago. 1949. and Quance, Leroy (Editors). The Overproduction Trap in U.S. Agriculture. Baltimore: Johnson Hopkins University Press, 1972. Jurgen, Jaksch Hans. "Modelos Macroeconémicos para Proyecciones a Corto Plazo." Bogota: September 1971. (Mimeo). Kmenta, Jan. Elements of Econometrics. New York: The Macmillan Co., 1971. Laursen, Karsten and Taylor, Lester. "Desempleo, Produc- tividad y Creciminento en Colombia," Revista de Planeacién Desarrollo. Vol. L, No.42, Bogota: Departamento Nacional de Planeacién, Julio 1969. Lewis, W. A. Economic Development with Unlimited Supplies of Labour. Londani—The Manchester Schbol. 1954. L6pez Toro, Alvaro. "Nota Sobre los Fen6menos Migratorios del Valle del Cauca." Bogota. (Mimeo). Mendoza and Olarte; "Diagnéstico de la Educaci6n Media; Bases de Politica." Cali: Secretaria de Educacion Departamental del Valle, 1972. Ministerio de Agricultura. "E1 Sector Agropecuario frente a1 Problema del Desempleo." Bogota, 1971 (Mimeo). Morrissett, Irving. "Some Recent Uses of Elasticities of Substitution: A Survey," Econometrica. Vol. 21, No. 1, 1953. Nelson, Richard R., Schultz, T. Paul, and Slighton, Robert L. Structural Change in a Developing_Economy; Colombia's Problems and Prospects. Princeton: Princeton University Press. 1971. 244 Oficina Internacional del Trabajo, OIT. Hacia e1 Pleno Empleo. Bogota: Imprenta del Banco Popular,l970. Planeacién Departamental del Valle. "Area Metropolitana de Cali, Aspectos Economicos." Cali, 1969. Planeaci6n Municipal de Cali. Plan General de Desarrollo. Encuesta Urbana de Empleos e Ingresos del Area Urbana as Cali} Cali: 1969. . "Estudio Preliminar de la Poblacién de Cali." Cali. 1960 (Mimeo). . "Plan General de Desarrollo del Area Metro- politana de Cali." Cali, 1971. Proyecto de Mercadeo Integrado Urbano-Rural del Valle, PIMUR, Technical Report No. 4, The Production and Distribution of Selected Consumer Goods in Ehe Cali Area. 1969. . Technical Report, No. 16. An Economic Analysis of Residential Construction in Cali. 71969. Riley, Harold, et al., Market Coordination in the_Develgp- ment of the Cauca Valley Region-Colombia. Research Report No. 5, Latin American Studies Center, Michigan State University. Sarmiento Palacio, Eduardo, Departamento Nacional de Planeacién. Aspectos Cuantitativos del Plan de Desarrollo. Bogota: Imprenta Canal Ramirez- Antarex, 1974. Sen, Amartya Kumar. La Selecci6n de Técnicas. Un Aspecto de la Teoria del Desarrollo Econémico Planificado. Mexico: Fondo de Cultura Econ5mica. 1969. Sheahan, John and Clark, Sara. "The Response of Colombian Exports to Variations in Effective Exchange Rates." (Mimeo). Todaro, Michael P. "A Model of Labor Migration and Urban Unemployment in Less Developed Countries,” American Economic Review, Vol. 59, No. 1, 1969. Turbay, M. Gabriel, Fundacién para la Educacion Superior y el Desarrollo. Una Politica Industrial ara Estimular las Exportacionesyy Fomentar eliEmfiIeo. Bogota: Fedesarrollo, Mayo 1972. 245 Universidad de los Andes, CEDE, Empleoyy Desempleo en Colombia. Bogota: Universidad de los Andes, 1958, 1968 and 1969. Universidad del Valle, CUIP, "Estudios de Poblacién," Cali, 1970 (Mimeo). . Estudios Demograficos de Cali. Cali: Univer- sidad déIPVaIIe, 1969. . Censo de Poblaci6n pgra e1 Departamento del Valle. 0511, 1968. Uhiversidad del Valle, CIDE, Encuesta de Empleos e Ingresos. Cali: Universidad del Valle, 1965. Universidad del Valle, CREE, "Estrategias Curriculares Combinadas para un Sistema Educativo no Formal." Cali: Universidad del Valle, 1972. Urrutia Montoya, Miguel and Villalba, Clara Elsa. "E1 Sector Artesanal en el Desarrollo Econémico" Revista de Planeacién y Desarrollo, Vol. 1, No. 1, Bogota: Departamento Nacional dePPlaneacién, Julio 1969. Wiesner, Eduardo and Vargas, Daniel. Fundacién para la Educacién Superior y el Desarrollo, Las Exportaciones y el Empleo una Perspectiva para Colombia. Bogota: Fédesarrollo, 1972. ”'Tflfl‘llfllljjflhj}MI®EDiJflfiJMfMES