URBANIZATION AND INDUSTRIAL DEVELOPMENT Stylianos Athanassiou IN LATIN AMERICAN COUNTRIES Serie A, N° 125 Julio de 1974 Centro Latinoamericano de Demografia | "•»Ktlf»-™'» J, Santiago de Chile ERRATA Page 44 44. 46 47 48 48 48 56 56 59 69 91 c i It Reierence & reads 2483.1 In the l i n e a r equation 4 ^ In the logarithmic e q u a t i o n 1 the standard error of estimate should appear 56 000 line 6 Line 19? in the f o r m u l a of income elasticity 10.65 Line 14 10.65 L i n e 15 10.31 L i n e 18 Colchagua L i n e 11 3981 Line 21 . In the linear equation 3 10.23 Line 3 ' L i n e 29» add G u a t e m a l a a f t e r Costa Rica •i-y-i Uni 2 du. • It should read -2483.1 (0.1810) 12 000 • 10 16.5 13.1 Aconcagua 4891 + 12.3 Distribución Interna Stylianos Athanassiou $ Fri, I v - ^ w E S S 'WTIhGC \ A \O/ x&-, '-hi/-s ..ossv Serie A, N°125 Julio, 150 1974 URBñNIZATION AND INDUSTRIAL DEVELOPMENT IN LATIN AMERICAN COUNTRIES The views arid opinions expressed therein arc those of th-j author and do not necessarily r e f l e c t those of ih; L a t i n American Demographic Centre (CQ.ADE). Acknowledgment I would like to express r y warm thanks to the Director n of the Centre, Miss Carmen A. Mirò and the Deputy Director of the Centre, Professor Juan Carlos Elizaga, both for their idea of the study of the problem in question as also for their contribution in the completion and presentation of the work. I particularly express my thanks to Professor Juan Carlos Elizaga for his guidance and his outlining of the general purposes of the study and his constructive remarks during the procedure of the work. Furthermore, I wish to thank to the colleagues Messrs, R. ifellon, J. Arévalo, J. Alberts, A. Fucaraccio, 17. Pecht who participated in the conference of 31st October 1973, conducted by Professor J. Elizaga for their remarks, Which were taken into consideration during the investigation of the problem under study. I also feel obligation to my assistant, Mr. Guillermo Mundt, who carried out the heavy task of collection and elaboration of the statistical data, and express my thanks to him. The calculations were carried out at the electronic computer of the Centre and, on this occasion, I express my thanks to the manager of the section, Mr. Santiago Vasquez and Mr. Juan José Calderón, who showed the greatest willingness to serve me. I extend my thanks to Mr. Carmelo Soria, Chief, Publication Department of the*-Centre, for his assistance and cooperation with the author, throughout the various stages of preparing the first draft for publication. Finally, many thanks are due to Mrs. Erica VJulff for typing up the first draft and to Miss Anita Esquivel for typing up the final text of the present work. ABSTRACT The present work is a contribution of the author to the investigation of the problem of urbanization and industrial development for the Latin American countries. As we mention in the determination of the problem under study, urbanization, which is a world phenomenon, and the industrialization of the country, are problems for each country to solve, targets of the economic plans and state policy. Furthermore, the development of these two magnitudes is an index of socio-economic growth of the country. VJhat is the role of industrial development which, in its broader sense, means economic development, in urbanization? VJhat are the economic-danographic factors which influence urbanization? Does the development of these two dimensions occur concurrently and to what degree? What are the consequences in the case of negative answers to the previous question? Are population redistributions expected between the Latin American countries? And what extension and what countries? Such questions and other similar questions arise during the analysis of the aforementioned problem. The need to investigate this problem, as regards the influence of the industrial development on urbanization and the existing relationship between them, from the point of view of their development, became evident for the Latin American countries, where urbanization experienced a large extension in the last decade 1960-70, while these countries were in the stage of economic growth. Similar works have been undertaken for the European countries. Ctost correctly, therefore, did the Centre manifest the desire to realize such a study. Further to the answers which will be given to the above questions and from which those components will be able to exercise the policy indicated on urbanization and industrialization, the conclusions drawn can be probably used by other colleagues of this Centre who study subjects similar to the above, and also by the author during the study of the problem of economic growth by constructing an economicdemographic model. Finally, the comparative data regarding the degree of development of urbanization-industrialization among the Latin American countries will be useful for international comparisons. I N D E X Page I. T H E PROBLEM 1. In general 2. Specification of the problem. Its purposes 3. Usefulness of the study. Its structure II. THEORETICAL CONSIDERATIONS. HYPOTHESES. METHODOLOGY. MODELS 1. General aspects for a theoretical consideration of the problem 2. Basic hypotheses.. Methodology 3. Models used III. THE DEVELOPMENT OF URBANIZATION OF THE CHILEAN POPULATION DURING THE SAMPLE PERIOD 1952-70 1. Introduction 2. Urbanization 3. Degree of urbanization 4. The urban areas IV. URBANIZATION AND INDUSTRIAL DEVELOPMENT IN THE CASE OF CHILE 1. The influence of industrial development and urbanization 2. Regional economic growth and urbanization V. RELATIONSHIP OF URBANIZATION AND INDUSTRIALIZATION IN LATIN AMERICAN COUNTRIES. APPLICATION OF THE SYMMETRY MODEL 1. Testing of urbanization of the Chilean population and industrialization at a National and Regional level ... 2. The development of urbanization and industrialization from the symmetrical point of view in the remaining Latin American countries at the national level. Their classification based on the model applied 3. Reliability of the model 4. International comparisons based on the symmetry model. Conclusions drawn 5. Population movement anticipated in Latin American countries and probable redistribution of their labour force VI. FINDINGS AND CONCLUSIONS 1. A review of findings and conclusions 2. Epilogue APPENDIX A BIBLIOGRAPHY 1 1 2 4 6 6 8 10 15 15 15 25 28 40 40 54 74 74 79 84 89 93 95 95 97 99 .... 107 ) ii ( Index of Tables Chapter III 2.2.1 Development of urbanization in Chile during the sample period, 1052-1970 16 2.3.1 Distribution of the provinces of the country based on the proportion of their urban population during the sample period, 1952-1970 20 Distribution of provinces based on the percentage of degrees of their rural population during the period 1952-1970 23 The annual rate of growth of the urban population of the country during the sanple period 1952-70 26 3.2.2 The degree of urbanization of the Chilean population by region, during the sanple period 1952-70 .......... 27 4.1.1 Distribution of urban and rural areas by provinces, during the sample period 1952-1970 30 4.2.1 Distribution of urban areas by population size-classes: 1952, 1960 and 1970 33 4.3.1 The population development by size of city in the period 1952-70 ..*.. 36 4.3.2 The demography development of the main cities of the country during the sanple period 1952-70 39 2.3.2 3.2.1 Chapter IV 1.4.2.1 Elasticities with respect to the explanatory variables of the linear equations tried 49 1.4.2.2 Elasticities with respect to the explanatory variables of the logarithmic equations tried .. .. . 49 1.5.1.1 Development of determinative factors of urbanization in 1975 51 1.5.2.1 The development of the urbanization of the country in the year 1975 " 52 2.2.1 The per capita income by province during the year 1960 and 1970, at constant 1965 prices 55 2.2.2 The index of industrialization by province during the census year 1952, 1960 and 1970 61 2.2.3 The population outflows towards the larger urban centres in the year 1970 63 2.3.1 The population outflows of the province towards the metropolitan area of Santiago during the period 1966-1970 and the per capita income by region in 1970 65 2.4.1 2.4.2 The best fitting regressions Elasticities with respect to the explanatory variables of the equations tried 68 68 ) ill ( Page 2.4.3 Assumptions and the development of the explanatory variables in the year 1975 69 2.4.4 The development of the population of the metropolitan area of Santiago in the year 1975 70 Chapter V 1.1 2.1 The symmetry coefficient of urbanization of the population of Chile and industrialization at a regional level, during the census year of the sample period 1952-70 75 Indices of urbanization and industrialization, as also the ratio, in various Latin American countries during the census year of the period 1960-1970 81 3.1 The number of cities by population size and their population for certain countries of Latin America 88 4.1 Indices of urbanization and Industrialization, as well as the symmetry model for certain countries, during the period 1960-70 90 The symmetry coefficient and the per capita income of certain countries during the period 1960-1970 92 4.2 Index of Charts and Diagrams Charts 2.2.1 The development of the urban and rural population during the sample period 1952-1970 17 2.3.1 Regional distribution (percentage) of urban and rural population in 1970 21 2 The development of labour force in Chile by branch of economy in the sample period 1952-1970 77 2.1 The degree of industrialization and urbanization of the Latin American countries in the year 1970 82 The symmetry coefficient of urbanization and industrial development in Latin American countries in the decade 1960-1970 85 2.2 Diagrams 1-4 Diagrams of the linear equations tested; observed and estimated values of the urbanization of the population in Chile, their deviations in percentage and the influence of the explanatory variables of each equation 57 5-6 Diagrams of the linear equations tested s observed and estimated values of urbanization of the population of the metropolitan area of Santiago, their deviations in percentage and the influence of the explanatory variables of each equation 71 I. 'THE PROBLEM j 1. '' 'l i • * In general : • » • ' Since the end of; World War II, and especially during the past two decades (1950-70) , economic growth hàs beccane the nain problem for solution of economic plans and state policy of all countries,, regardless of their economic progress and théir socio-economic standard of living.. Economic growth, in its broader sense, means the maximization of socio-economic prosperity. As it is known, the main "dimension" of the degras of economic growth is the size of per capita income.—^ Furthermore, it is known that the rise in the level of economic development is mainly achieved by the development of the industrial sector of the economy, which has greater possibilities. It has been proved that there is a high degree of correlation between the per capita incòtte and income from the industrial branch of the economy. Therefore, the increase of industrial production, both in the economically developed countries and in the less-cfe- veloped ones, is a principal aim of the responsible state aoències. More 2/ specifically, in the less-developed countries,— the wholè problem of industrial development is of greater importance, and consists in the objective aim of all the efforts made by the state. This is obvious, because in these countries, the per capita income is at a low level, arid" industry is in the first stages of development. These countries desire the most rapidly possible 1/ This index, of course, must be supplfartented with other economic magnitudès, such as the ccrrpositionof consumption, income distribution, employment, productivity and education. This index of economic growth' presents many difficulties as regards the statistical estimation of elements, referring both to thè national income as. also1 tò the population, but has international acceptance. 2/ As a rule, less-developed countries have a per capita income of less than 25 per cent of the corresponding per capita income of the U.S.A. These countries, mainly Africa, Asia and Latin America, contain twothirds of idle vrorld!s population. • * ) 2( increase in per capita income, which is achieved, as aforementioned, mainly by the increase of industrial production and entry into the area of the economically developed countries. Industrial development, as we shall see below, created the primary incentive for urbanization of the population. As it is known, urbanization of the population means, mainly, a movement of the agricultural labour force into the industrial sector of the economy. To this movement of the population, there also contributed, of course, the development of the tertiary sector of the 3/ economy, which is also the result of industrial development.— Consequently, the urbanization of the population in each country, apart from certain exceptions, had an extent proportional to the degree of industrial development. On the basis of the above, we can say that the urbanization of the population 4/ is a function of industrialization.— Finally, if the movement of the agricultural labour force towards the industrial branch of the economy, is made to cover the employment opportuni- ties created in the secondary production, then we can say that the urbaniza— tion of the population of the country takes place symmetrically with industrial development. This can be said in a general point of view, that the proportion of the urban population, particularly that employed in the secondary sector of the economy, consists of an indication of the economic growth of the country being effected. 2w ' Specification of the problem. Its purposes In the previous section we said that, econcmic growth, speaking broadly, means industrial development which creates employment opportunities in accordance with its rate of growth. . Furthermore, we accepted that industrialization of the country greatly influences the urbanization of the population 37 There is a positive correlation of these two branches of the economy. 4/ Of course, urbanization is also attributable, apart from the industrial development realized, to other reasons. These reasons are social, cultural, etc. In addition to the above causes, the mechanization of the agricultural economy and the increase of agricultural productivity,; contributed to the movenent of the farmers towards the urban centres of the country, in search of employment, on account of the increase in unemployment in that sector of the economy. ) 3 ( of the country. In the general description of the problem under study, we also said that if urbanization covers the needs created in labour force in the secondary branch of the economy, then it can be said that the development of urbanization is symmetrical with industrial development. This can be said in general as followss if the movement of the agricultural population towards the urban centres (urbanization) is proportional to the extent of the secondary production, then urbanization can be considered as an ideal kind of domestic mobility of the population, because it facilitates the industrialization of the country. Finally, we said that the symmetrical development of urbanization and industrialization can be considered as an index of economic growth of the country, and that the %er capita income is used as a measure of the degree of socio-economic prosperity of the country, disregarding other characteristics, like social conditions, cultural patterns and so forth. Eased on these conclusions, we can determine the purposes of this work which, in general, are the followings a) Hie analysis of the existing functional relationship between urbanization and industrialization. In other words, the estimation of the influence of the industrial development on the increase of the urban population of the country (or a geographical region). b) The testing of the development of urbanization and industrialization from a symmetrical point of view. More specifically: The analysis of the functional relationship between urbanization and industrialization will refer to the case of Chile at the national level. Furthermore, we will investigate the aforementioned relationship from the point of view of the regional problem of the country. The testing of the development of urbanization and industrialization will be realized for all the Latin American countries, at a national level and, in the case of Chile, this investigation will expand to the regional level. ) 4( 3. Usefulness of the study. Its structure The usefulness of the work, on the basis of the conclusions derived from the analysis of the whole problem of urbanization and industrialization/ will consist of the following s a) In the determination of the main economic factors which contribute to' the population movements of the rural areas towards the urban centres and which factors, in the broader sense consist of an index of the industrial development of the country. Furthermore, in the exercise of the indicated policy on the basis of the degree of the influence of these factors on the urbanization of the country. b) In the determination of the main factors (demographic and economic) prevailing in the regional problem of the country and the investiga- tion of the correlation between these factors. c) In the making of forecasts as regards the development of the urban population of the country, based on economic factors. The points of (a), (b) and (c) will refer to the case of Chile, as mentioned in the previous section. d) In the determination of the countries of Latin America, where the development of urbanization is parallel to industrial development. Finally, e) In the investigation of the probable case of the anticipated population movements in the coning years in Latin American countries, based on their rate of growth and especially on the basis , of the development of urbanization and industrialization. The above will provide a general view of the problem of urbanization and industrial development in Latin American countries and the possibility of outlining the indicated policy, on this problem, at a national level by the responsible state agencies. Regarding the specific case of Chile, because of analysis of the profclon at a regional level also, this policy can be escpanded by regie»!. This policy will mainly consist of the advancement of industrial development in those countries where urbanization preceded the country's industrialization and the acceleration of urbanization in those instances in which urbanization does not develop symmetrically with industrial development, with ) 5( the aim of achieving, in both the above mentioned cases, an equilibrium between urbanization and industrialization. In other words, the social and eco- nomic measures to be adopted by the state, will refer to the retardation or acceleration of the urbanization of the population of each latin American country, according to the development already realized or anticipated in the secondary branch of the economy of each country, as also regional development in the case of Chile. The work consists of six chapters. In the first chapter, we describe the problem to be investigated in its general and specific form, as well as the purposes and the usefulness of the study. In the second chapter, the hypotheses are given, the methodology to be applied for testing the hypotheses, as also the general economic aspects of the problan. Furthermore, we describe, on the one hand, the econometric model which will be applied for testing the relationship between industrial development and urbanization? on the other hand, the symmetry model fron the point of view of its use and its broad interpretation. Hie analysis of the development of the urban population in Chile at a national and regional level is realized in the third chapter. This analysis will facilitate us in the investigation of the functional relationship between urbanization and industrialization. In the fourth part, the statistical results obtained fron the investigation of the relationship between urbanization and industrial development, in the case of Chile, are presented. Furthermore, in this chapter we examine the regional problem of the country, by de~ t. termining the main, factors which prevail in it, and by investigating the existing relationship between regional economic development and urbanization. In the fifth chapter, we apply the symmetry model for latin American countries. Moreover, the results obtained from the application of the aforementioned model are included in this chapter. Finally, in the last Chapter (sixth chapter) a review of findings is made, and a summary of general conclusions is presented. ) 6( II. TiiBORETICAL OHlSIDERATia^S. HYPOffiESES^-lEDSODOLOGY. 1CDELS 1. General aspects for a theoretical consideration of the problem v Urbanization and the increase of the earth's population are the two most important trends which are ascertained in the Demographic Yearbook of the United Nations in the last decades. The urban population of most countries, economically developed and loss-developed, was doubled during the period 1940-70. Ihis very rapid growth of urbanization of the population on an international scale is the most revolutionary phenomenon of modern times. People nowadays prefer to live in the cities and especially in the metropolitan regions and 5/ the capitals of countries, where the largest population movements are realized.— i The migrants of urban areas come mainly from Mechanization of > the agricultural regions. agricultural work and the increase of productivity resulted in a large part of the agricultural labour force being underarrployed. The »suitsof this underemployment was the small increase in the farmer's income but also of the total national income. Furthermore, it is known that the output of the work of residents in the agricultural regions is not proportional to the effort made. It is estimated that returns fran work in agricultural regions is less than half the returns from work in the industrial centres. A consequence of the employment conditions, the return for work and the income of the agricultural areas, is the movanent of the population of these areas to/ards urban areas. Of course, employment and income can be considered the most important factors which drive the population from these areas, but not the only factors. Apart frcm the aforementioned main factors, other factors J/ The population of Buenos Aires amounts to 8.3 million inhabitants, and represents 35.9 per cent of the total population of the country. Santiago de Chile has 2.7 million inhabitants (30.1 per cent of the total population of the country), while the population of iiontevideo consists of 43.3 per cent of the total papulation of the country (1.2 million inhabitants), during the census year 1970. ) 7 ( have also contributed to the mobility of the agricultural peculation towards the industrial areas. These factors are social, political, cultural and demographic, g / the analysis of which is not included in the purposes of this study.— . On the other hand, the secondary branch of the econcny shows a rapid growth in ail the economies. Its rate of growth follows an upwards trend. A result of the development of this branch is an increase in job opportunities, which means an increase of. demand for labour force in secondary production as stated above, by the supply of agricultural labour force, which moves towards the industrial sector of the economy. Furthermore, industrial development causes à rapid increase of the per capita income of this sector of the econoiry. In fact, in the industrial centres, the per capita income is at a higher level than that of the agricultural sector, and its rate of growth is larger than the corresponding rate of growth of the agricultural income. Furthermore, as it is known, the high income results in formulation of high rates of wages. • In general, it cari 1 * be said that the size cf the labour market in the urban centres; the high level of wages and the differentiation of social and cultural conditions between the ruial and urban areas, are the basic causès attracting labour force from the rural àreas to the lirban areas. Finally, it must not be overlooked that the supply of labour hands to the secondary branch of the econoiry has increased not only industrial productivity, but also the national income of the country. 6/ Of course, the urbanization of the population of the country creates many problems, social, traffic, housing, atmosphere polution, etc. In spite of all of this, we aid as much as possible the entry of farmers to the industrial centres. This is because the economic growth of the country depends mainly on industrial development. Furthermore, it is known that the rate of growth of a country is determined; apart from other factors, as for instance per capita income which is considered the main dimension of econcmic development of a country, by the proportion of the rural population to the total population. In other words, the hicjher this proportion- is the more economically underdevelopment is the country. Of the European countries the proportion of rural population to the total population has as follows s Austria (50 per cent), France (30 per cent), Greece (43 per cent), Netherlands (20 per cent), Sweden (22,2 per cent), Turkey (70 per cent), United Kingdom (22 per cent). In certain Latin American countries is as follows: Argentina (26,3 per cent), Chile (31,8 per cent), Venezuela (32,6 per cent), in the year 1960. ) S ( In conclusion, we can say, following what has been outlined, above, that the urbanization of the population, which consists of a movenent mainly of.. economically active population from rural regions to the : industrial centres -which are the urban centres of a country-. was caused by industrial development .and the application of purely economic laws, as regards the return for work andèmployment conditions » Of course, the size of urbanization differs from o i , country to the other, and in many cases, is not proportional to rs industrial development, insofar, as has been stressed, the concentration of population in the metropolitan areas is also attributable to other reasons, not econcmic ones. 2. Basic hypotheses. Methodology It vas previously mentioned that the size of the labour, market, particularly as regards labour demand in thé secondary sector of the econany, continuously increases, bacause of "the development of this sector. Furthermore, employment conditions in-the industrial centres are better, in comparison with the rural regions, and nowadays are improving. Finally, the per capita income of the industrial sector is quite high, and its rate of growth is greater than the rate of growth of the agricultural income. t7e have conclusively accepted that the above situations prèvailing in the secondary branch of the economy,, as also the further anticipated improvements in these situations, are the main causes of attraction of the agricultural population to industrial production. On the basis of the above, we can fannulate the following hypothesis as regards urbanization and, industrial development; that, industrial development (cause) influences the evolution of urbanization (effect). In other words, the increase of urbanization of the population of a country is the result of the industrialization of the country. Furthermore, given this relationship, the question arises, what is the degree of influence of industrial development upon population movement towards the secdndary sector of the economy. To test this hypothesis, repression techniques will be used and data /time series (upon the observations of the years 1960-70)„ Kote; As a dependent variable, The terms "economically active population" and "labour force1' are usei with the same meaning in the present work. ) 9 ( Internal migration towards the urban centres will be used, and as independent variables, certain factors which are considered as corresponding to the reasons of urbanization and refer as indices of the secondary sector of the economy, by means of which its development is determined. The factors are, i.e., the per capita income, the ratio of industrial wages to agricultural wages, the labour demand of the secondary branch of the economy, unemployment in the agricultural sector, etc. As regards urbanization, we have accepted that, apart from the aforementioned reasons, it is attributable to other reasons also; social, cultural, political, etc. Consequently, it is natural that the development of the urbanization of each country does not proportionally follow industrial development. In many countries, urbanization precedes the industrialization of the country, and for this reason we have an enlargement of the tertiary sartor of the economy of these countries. However, it is indispensable to know the situation already cre^hed, from this point of view of movement of the agri- cultural population towards the urban centres of the country, in relation to industrial development realize!, that is, in which cases urbanization proceeds symmetrically with industrial development and to what degree is the development of these two magnitudes normal, with the aim of implementing the policy indicated. On the basis of these thoughts, we can formulate the following hypothesis as regards the normal evolution of urbanization and industrial developments that the development of urbanization is symmetrical to industrialization, if the movement of the agricultural population to the urban centres (urbanization) covers the needs of the secondary production in labour force. If there is no coverage of the supply of labour, force (urbanization) by the industrial sector, then the development of urbanization in this instance is calisi asymmetrical. In other words, this hypothesis can be stated as follows ; if the supply of labour force is in balance with the demand for labour force in the second branch of thè economy, then the urbanization can be considered to have developed symmetrically. In order to test the above hypothesis, the "symmetry model" will be applied, which refers to the indices of urbanization and industrialization and, more specifically, the model in question refers to the relationship of the ratio of these indices. The symmetry model was determined cai an ) 10 ( international level and, consequently, the conclusions therefrom are generally accepted. Further details, as regards the manner of its construction, its broader importance from an economic and demographic point of view, as also its use, are provided in the relevant section same. Finally, the regional problem which appears in each country, is related to the problem under study* Therefore, in the case of Chile, we will try, a) to determine the main factors prevailing in it, as well as the existing correlation between themj b) to estimate the level of economic development of each geographical area based on per capita income, aid tl_e index of regional inequalities for the whole of the country. 3* Models used 3.1. Econometric model 3.1.1. In general Models of different kinds are frequently used in economic analysis and as a tool in solving quantitative problems. Models may be described verbally, graphically and by means of mathematical functions and simple equations or simultaneous equations. Generally, equations describe the economic relationship which exists between economic variables. The exact formula of an equation-model corresponds to the original data and the development of the variable under study. Therefore, the mathematical form of a relationship should be based on economic and statistical criteria. By means of the models we calculate the statistical estimate of the parameters, and, consequently, the interpretation of the determinants, with respect to their influence on the variable under study. From this point of view, the accuracy of a model -equation- is of great importance. The economic and statistical criteria will verify if the adopted model would be the appropriate one. On many occasions the impression is created that the unsatisfactory results are due to the fact that all the determinants have not been taken into consideration in a model. Of course, this claim may also be true, but this is mainly due to the fact that the mathematical form of the relationship is not the suitable one. Such a model is not indicated for expressing forecasts. ) 11 ( As it is known, the consideration of all factors influencing the variable under study, is not possible. This is either due to the technical difficulties, for instance, the nature of factors -qualitative factors- statistical material, measurements, etc. or due to ignorance of the factors. Therefore, these factors which are omitted in the relationships, are involved in the so-called unexplained part of the equation. From this point of view, the equations are called stochastic models, i.e. the functional relationships which have an explained part. This is the meaning of the residual term frcm an economic point of view. The stochastic models include some assumptions about the probable distribution of the disturbance term, e^. They are the following, the mean to be zero, E (e.) = 0 and the variance to be fixed, V (efc) » s g . And, it is also assumed that the random variable, e^, is independently distributed with respect to explanatory variables as well as with respect to itself lagged. Moreover, explanatory variables are assumed to be distributed independently of each other. Finally, the facility with which the constants and coefficients of an equation may be calculated, the simplicity of the model and its a priori validity, the hemoscedasticity of residuals, i.e. the same variance of the terms, consist of the points, frcm an economic point of view, i^iich will contribute to the determination of the appropriate model. 3.1.2. Equations indicated After the formulation of the functional relationships of urbanization and industrial development, which is realized in the next Chapter, an attempt must be made to define the equations to be tried. The defining of the appropriate equations in each case, should be based on the economic theory, the statistical criteria mentioned in the previous section and the empirical data. Following the way of graphic presentation of the empirical data and the statistical criteria, we are led to the following two forms. The linear and the curvelinear -exponential- equations. These regression equations -modelsexplain the increase of urbanization during the1 sampling period 1960-70, by defining its changes due to the influence of the determinative factors taken. Furthermore, they provide a simplicity in their calculations and an easy interpretation of the results obtain©?, fron an economic point of view. In addition, the application of such models for purposes of prediction, in ) 12 ( short-time» is indicated, since the parameters of these equations have been estimated. Finally, the traditional method of the least squares is easy to be applied for the estimation of the parameters of the above equations. 3.2. Symmetry model 3.2.1. Description and use of the model In section 2, it was said that the symmetry model refers to the relationship of the indices of "urbanization" and "industrialization". Therefore, in order to formulate this model, the calculation of the two aforementioned indices is necessary. Thè urbanization index refers to the proportion of the population which by definition, is characterized as "urban population", to the total population of the country. That is Z..=ÌL-100 11 <1> p - whéres • • s Urban population of a country1 or of a region P a i3 • s Total population of a country or of a region ? P • Index of urbanization of a country or of a region ij i ; Country j . s (i » 1, 2, 3 ... n) Region of a country (j * 1; 2, 3 ... k) • The industrialization index is the ratio of that part of the labour force which is employed in the secondary production, to thei total ecóncmicaliy active population of the country. This index can be written as follows , s I . , = i f i i - in ' P . -, " . . . ... e 100 (2) i > where? . P® s Economically active, population employed in the industrial sector J of a country or of a region ) 13 ( Pg ; Total econatdcally active population of a country or of a ^ region s ij Index of industrialisation of a country or of a region ; As in the relationship (1). The ratio of these two indices, multiplied by 100, consists of the sym~ metry model. In other words, the symmetrical coefficient, as the symmetry model is called, can be presented as followss (3) where; Z .. s as in the relationships (2) and (3) s symmetry model. Thus, insofar as the relationship of the ratio of the two indices of industrialization and urbanization -symmetry model- approaches the limit of 100, this means that the urbanization of the population was absorbed by the secondary branch of the economy, that is, in other words, the urbanization was a consequence of the employment opportunities created in the secondary production. On the contrary, if the level of this index is low, then it can be said that the development of urbanization was asymmetrical, that is, it was not the result of needs for labour force creates in the capitalistic sector of the economy. Finally, the limit of 100, of the symmetry model, is the point of equilibrium, in the sense that the possibilities for still further supply of labour have been exhausted. In general, we can say that the level of the symmetrical coeficient is a clear indication of the symmetry of the urbanization realized in relation to the industrial development of a country. 3.2.2. The broader importance of the model It was previously mentioned that the point of equilibrium of the indexmodel is the limit of 100. In the economically developed countries this index is at the point of equilibrium, which means that the supply of labour force (urbanization) mainly from primary production,, has been ftally covered by the ) 14 ( secondary sector of the economy and, consequently, there are no possibilities for further increase in the supply of labour force (exhaustion of resources of labour force, such as the agricultural population). In certain developed countries, this index has exceeded the limit of 100, a fact which can be interpreted to mean that the employment opportunities, which were created in the capitalistic sector of the economy, exceeded the limits of the potentiality of the domestic labour force to be employed.; Hie problem is confronted by external migration from other countries, where there is a surplus of supply of labour force. This ascertainment is of great importance, particularly for developing countries which at the same time have an emigration problem, because they may confront scarcity of labour force in the coming years. Furthermore, this will reflect upon the rate of growth of the population of the "migratory" countries. In certain less-developed countries, the numerical limits of the symmetrical coefficient are less than 25 per cent. This means a delay in the development of secondary production in accordance with the development of urbanization. Generally, the. deviation of the numerical value of the symmetry model between developed and less-developed countries is reflected in two situations; a) in tiie developed countries, that urbanization was caused by the continuously increase of employment opportunities in the industrial sector and, b) in the less-developed countries, that the inflow of internal migrants in the urban areas of the country preceded the creation of needs for labour force in the industrial branch of the economy, resulting in an enlargement of the service sectors. In this last case, we have a form of unemployment in a large part of urban population which is called "invisible" unemployment. All the above situations in the developing countries create many difficulties in the rapid economic growth of the country. Finally, we must not forget that the movement of labour force from the tertiary sector to the secondary secitar of the economy, for covering the demand for labour force, is more difficult than the movement of agricultural population to the industrial areas for the same purpose. ) 15 ( III. THE DEVELOPMEOT OF UFI^LJIZATION OF THE CÎJILEAIJ POPULATION DURH-iG THE SAMPLE PERIOD 1952-70 1. Introduction In the first chapter, during the determination of the problem under study, we said that internal population movenents, which take place in each country, are a very recent phenomenon, for investigation by the researchers of social sciences. In this chapter, we will analyse the development of the urban popul- ation of Chile, during the two last decades 1950-70. This analysis has two main objectives. The first is to establish, through census data, the volume and direction of urbanization of the population of the country. This will show us the role of urbanization in the population increase of the already existing urban centres of the country, the creation of new such centres and the redistribution of urban and rural population by region. Hie second abjective is to assess the degree of urbanization at a regional level. This m i l permit us to see the expected trends cf urbanization in the coming years. Furthermore, to study the development of the main city of the country, for investigation of geographical poles of concentration of population of the country. The urbanization of the population is closely related to industrial development, both these factors consist of the problem under study. Therefore, the analysis of urbanization vdLll facilitate us in the investigation of this problem. Finally, urbanization will possibly be a factor of the regional problem of the country and, consequently, the data as regards its development will be useful in tliis effort. For the analysis of the development of urbanization of the population in the past years, vie will use descriptive statistics, that is, by means of the indicated statiscical tables we will try to achieve the objectives of the present chapter. 2. Urbanization 2.1. Definitions. Characteristics Éefore we mention the urban population of the country and its development up to the present, it is consider«! advisable to give the statistical concept of urban population. The meaning and statistical vise of the concept of urban ) if: i population varies m d e l y between countries, in t*JO main aspects. The first aspect can be related to the administrative organization of the country, public services, industrial areas, etc., an! the second aspect clearly has as criterion 7/ the variation on the r.dnimum population.— In the case of Chile, as regards the cletenriination of the urban population, 8/ it is rather the first aspect which is adopted.—' 2.2. The urban population On the basis of the above adopted statistical aspect, the urban population in Giile, during the 1970 census, amounted to 6 660 thousand inhabitants and was 75.1 per cent of the total population, with a total of 517 ur'~an centres. The development of the urban population of the country from 1952 up to, the year of the most recent census (1970), is shown in the foHowing Table 2.2.1. and in Giart 2.2.1. Table 2.2.1. DEVEDOE'Effi OF UREANIZATIOM £•! Cffi: DURE 13 HIE S K S K J PERIOD 1252-1970 Population (in thousands) Year 1952 5 028 1970 6 660 Source; 77 y Percentage Total 60.2 3 573 1960 Rural 2 360 39.8 5 933 60. 2 2 346 31.3 7 374 75.1 2 209 24.9 . 369 8 Percentage . Urban 5 Statistical Yearbooks, 1952, 1960 and 1970. Institute, Santiago, Chile. National Statistical The internationally acceptad limit for the characterisation of a population as urban, is 2 000 persons and over. This limit varies fren country to country, according to the socio-economic development of the country, arid the reason for this, is that in the determination of part of the population of a country as urban, there also enter racio-economic factors. The determination of tie population centres, as urban, is aade upon the , basis of deirographic and administrative elements, as also thé esdstence in same of public services. Furthermore, certain industrial centres are characterized as urban regions (Servicio Nacional de Estadística y Censos, XII Censo General de Población, 1952, Santiago de Chile and III Censo de Población, 1960, Santiago de Chile). ) 17 ( Chart 2.2.1 THE DEVELOPMENT OF THE URBAN MO RURAL POPULATION DURING THE SAMPLE PERIOO 1952 - 1970 (Percentage) Chile population x 1(T & 9 000 SNA 1 8 000 7 000 N J Urban population ' I R U r a i population _ 6 000 > 75.10 % 5 000 4 000 3 000 2 000 - 1 000 -O Years Sources; Dirección de Estadística y Censos, Censo General de Población (para 1940, 1952-1960), Dirección de Estadística y Censos, Santiago de Chile (1943, 1956, 1963), Dirección de Estadística y Censos, Entidades de Población (para 1970) (de Tarapacá a Magallanes),Santiago de Chile 1972 ) 19 ( From the above table, it can be seen that the urbanisation of the population, during the aforementioned period, experienced a considerable increase and formed, in 1970, three-quarters of the Chilean people, •tore specifically, the urban population, in the last census (1970), was increase:! by 86.4 per cent and 32.5 percent, in comparison with the years 1952 and I960 respectively. A further increase is anticipated in the urban population in tlie caring years. 9/ On the contrary, the population of the rural centres-' sliowed a considerable decrease during the surveyed period 1952-70. Thus, on the basis of the 3ata of the 1970 census, the rural population of the country amounted to 2 209 thousand persons (24.9 per cent of the total population) compared with 2 360 thousand inhabitants (39.8 per cent) in the year 1952, sharing a drop of 6 per cent approximately, ^s a consequence of the continuing increase of the urban population of the country, the development of the rural population will folia-/ a dc* .Trwards trend. Frcra the above, we can be led to the general conclusion that the population development in Chile is characterised by a considerable enlargement or urban population, probably attributable to tlie socio-economic grath of the country, in the sense of its transition from an agricultural to an industrial country. 2.3. Regional distribution of urban and rural population The distribution of the urban and rural population of tlie country by province, during the last census (1970) is sham in Chart 2.3.1. This percentage distribution varies in each province.—^ There are provinces whose near total (over 90 per cent) population is urban, as for example in tlie provinces of Tarapac£, lintofagasta, Valparaiso and Santiago. In the provinces of Atacama, 9/ Rural centres are characterized as those population centres which do not have public services and whose population mainly depends upon primary production (see sources of footnote 8). 10/ In Table 2 of Appendix I, we give tlie proportion of the urban and rural population of each province in tlie census years 1952 and 1960. Furthermore, in order to facilitate the reader of the present work, one of tlie letters of the Latin alphabet was given, to each province. (See table 2 of Appendix A). ) 19 ( Concepción and Magallanes, over 80 per cent of the population resides in urban areas, while in 16 of the 25 provinces of the country, the urban population is more than one half of the total population of the provinces. In Table 2.3.1., we classify the provinces on the basis of the proportion of urban population to the total population of the country. From this table, it appears that the number of provinces whose urban population exceeds 90 per cent, was doubled during the period 1960-1970, and the -urban population of these four provinces was 61.5 per cent of the total urban population of the country in the year 1970. Furthermore, the provinces whose urban population ranged from 70 to 90 per cent during the same as above census year, cover 11.3 per cent of the urban population of the country and the number of the provinces in question was reduced to tliree provinces in 1970, while the proportion of the urban population of the five provinces which are included in this size™ C1Q_S during the years of the previous censuses, 1952 and 1960, was 69.1 per cent and 24.7 per cent respectively. Fran this point of view we observe a drop of the proportion of the urban population, in this size-cj.ass (70-90 per cent), during the period 1952-60 eventhough the number of provinces was not changed. In the medium size class (50 per cent to 70 per cent) we have a considerable increase in the number of provinces (fron two provinces to nine provinces), during the period under examination, 1952-70, and the proportion of the urban population of these nine provinces was 15.8 per cent in the year 1970. The papulation of the provinces whose proportion of the urban population is more than 50 per cent is 7 184 thousand inhabitants, (81 per cent of the total population of the country). The same can he said from the point of view of the existence of a considerable number of provinces in the 40 per cent to 50 par cent size class in the year 1970, eventhough the urban population of these seven provinces covers only 9.9 per cent of the total urban population of the country. In this size class we have a duplication, both of the number of prov- inces as also percentage of the urban population in the semi-period 1952-60. Finally, the provinces whose urban population is less than 40 per cent of the total population, showed a decrease, both from the point of view of number of provinces as also frcm the point of view of proportion of urban population. The number of provinces decreased from fifteen provinces to two provinces, with Table 2.3.1 DISTRIBUTION OP THE PBDVECES OF THE CCUt-TRY EASED ON HIE PROPORTION OF TtiEIR URBAN POPULATION DURHIG THE SAMPLE PERIOD, 1952-1970 Number of the c. Wumber of provinces e class 1952 1960 Peculation Percentage of urban population £ 1970 1952 1960 1970 2 397 963 4 094 780 1952 1960 1970 « 47.69 61.48 2 4 3 2 469 853 1 242 790 754 143 69.12 24.72 11.32 2 .5 A 102 824 396 430 1 049 187 2.88 7.88 15.75 40-49.9 3 6 9 7 152 277 462 441 663 403 4.26 9.20 9.96 0-39.9 15 8 2 848 168 528 436 99 016 23.74 10.51 1.49 Total 25 25 25 3 573 122 5 028 060 6 660 529 100.00 100.00 100.00 7 374 115 0 869 166 60.22 68.19 75.09 1 90 + - 2 70-89.9 5 3 50-69.9 4 5 • ~ Population of 5 932 995 the country Sourcess Dirección de Estadística y Censos, Dirección de Estadística y CensosDirección de Estadística y Censos, Tarapacá y Magallanes).. Dirección 1972. Censos de Población, Tomo I para 1952 y 1960. Santiago de Chile, 1959, 1969. Entidades de Población para 1970 (de de Estadística y Censos, Santiago de Chile, ) 21 ( Chart 2.3.1. REGIONAL DISTRIBUTION (PERCENTAGE) OF URBAN AND RURAL POPULATION IN 1970 Ff * U i i ^ L / 7 ßtßx E ^ f ^ P V i 85.2/ p V " ^ m . "A W / ~s> Rur*l population Urban population vi 11 ii'i --V i •Y ^ v I > . CHILE: -3 . Source: Dirección de Estadística y Censos, ( I n s t i t u t o Nacional de Estadística) Entidades de Población 1970 } Tarapac5 a Magallanes. Dirección de Estadística y Censos Santiago de Chile, 1972. ) 19 ( the result that the urban population decreased by 03„3 per cent. This is of great importance, from the point of view of the internal population movements in Chile during the period under study (1952-70). The increase of the urban population of the country in general, during the period 1952-70, the fact of the increase of the population of those provinces that can be considered as being pure urban a r e a s , — i n conjunction with decrease of the number of provinces whose urban population is less than 40 per cent, leads us to the conclusion that, in addition to the movement of rural population to the large urban centres, we also have a movement of urban population of the regions with a small proportion of urban population, towards the large urban areas. This movement of the urban population is called urban-urban migration and it is of greater importance than the rural-urban migration. Furthermore the considerable increase (386.4 per cent) of the urban population, dutxng the period 1952-70, in the provinces whose urban population is more or less by 10 per cent than half of their population, as also the increase of the number of these provinces, is a favourable point, from a demographic and economic viewpoint. This is because the population increase of provinces i^hich cannot be considered as large urban areas of the country and whose urban population is half of the total population, presupposes economic and social development of the regions in question. Finally, from attached Table 2, in Appendix I, we see that provinces showed an increase of their urban population, as also an increase in the proportion of urban population to rural population, during the period under examination 1952-70. In conclusion, we can say that the continuing trend of urbanization of the Chilean population, is followed by all the country's provinces. As regards the development of the rural population, by province, it can be said that in certain provinces it followed an upwards trend, while in others the rural population decreased, during the period 1952-70. Thus, in sixteen provinces the rural population, probably on account of its movonent towards the urban areas, showed a drop, whereas in the remaining nine provinces, it 11/ Pure urban areas are considered those areas whose urban population exceeds 90 per cent. ) 19 ( increased at a slow rate. Of course, the decrease of rural population in these sixteen provinces varies in each province, and does not exceed 12.5 per cent of their rural population.—^ In the following Table 2.3.2. we give the distribution of the provinces, on the basis of the percentage of decrease of the rural population, as also the provinces that showed an increase of their rural population in the period under study (1952-70). Table 2.3.2 DISTRIBUTION OF PROVINCES EASED 021 TH3 PERCEOTAGE OF DECREASE OF THEIR RURAL POPULATION, DURKG THE PERIOD 1952-1970 Uumber of the clasp Size of class Population Percentage of rural population Number—/ of provinces 1952 1970 1952 1370 1 60 + 1 41 406 13 769 1.75 0.62 2 50-59.9 1 19 319 0 457 0.84 0.38 3 20-49.9 1 38 672 25 502 1.64 1.15 4 15-19.9 1 77 346 63 731 3.28 2 a 39 5 10-14.9 5 720 805 632 254 30.54 28.63 6 5- 9.9 2 315 714 291 842 13.38 13.21 7 0- 4.9 5 453 101 447 C03 19.20 20.28 693 010 725 279 29.37 32.84 Increase of the rural population Total 25 2 359 873 2 208 637 100.00 100.00 Sourcess Dirección de Estadística y Censos, XII Censo General de Población y I de Vivienda, Tomo I, Dirección de Estadística y Censos, Santiago de Chile, 1959. Dirección de Estadística y Censos, Entidades de Población (de Tarapacá a Magallanes), Dirección de Estadística y Censos, Santiago de Chile, 1972. a/ The number of provinces did not change as regards the classification of classes, during the census years 1952 and 1970. 127 The rural population of the sixteen provinces amounted to 1 483 thousand persons in 1970. ) 19 ( Thus, fran the above table, it appears that in one province the decrease of the rural population was considerably important, insofar, as it exceeded 60 per cent. This province was Tarapacá, and its rural population consists of 0.62 per cent of the total rural population of the country. The sarce can be said for the province of Antofagasta, where its rural population decreased by 55 per cent approximately . On the contrary, in the provinces of Ilaule, fíuble, Concepción, Arauco and Fio-5io, the drop in the rural population ivas negligible, less than 5 per cent, while the proportion of rural population of these provinces to total rural population of the country, remained virtually unchanged (20 per cent approximately) in the census years 1952 and 1970. Sim- ilarly unaltered remained the proportion of the rural population to total rural population, (13 per cent approximately), in the two provinces (Santiago and Llanquihu*s), belonging to the 5-10 per cent size-class, during the period 1952-1970. The provinces whose rural population was reduced frcm 10 per cent to 15 per cent are five (Coquimbo, Valparaíso, '^alleco, Cautín and Valdivia) with a total rural population of 632 thousand persons, which was 28.6 per cent of the total rural population during the year 1970. Finally, there are two provinces, Aconcagua and Atacama, which showed a drop in rural population by 17.5 pea: cent and 34.1 per cent respectively, during the period between the two censuses in the years 1952 and 1970, and cover 4 per cent of the total rural population of the country. As regards the provinces which showed an increase of the rural population, we have to mention the following; of the nine provinces, the provinces of Aysén and Magallanes experienced an increase of their rural population by 1C.5 per cent, on an average, during the sample period 1952-70 while this increase (of rural population) in the provinces of Linares and Curicó, was more than 10 per cent. In the remaining five provinces, the increase of their rural population was less than 5 per cent and, consequently, can be considerea insignificant. As a conclusion, we can say that during the period under examination (195270) a redistribution took place of the urban and rural population in the country' s provinces. This redistribution varies in each province. Generally, in all provinces, we had an increase of the urban population and in certain provinces this increase of urbanization was considerable. As a consequence of this fact, the decrease of the rural population of 16 provinces took place. Beyond ) 19 ( the decrease of rural population in these provinces, we had an increase of the rural population, if we compare the rural population of the census year 1952 to the remaining nine provinces (4.65 per cent). Probably, all these population changes are the consequences of the population movements and, specifically, the redistribution of the labour force of the country and of the areas of the country, from the point of view of rural and urban areas, due to the industrialization and, in general, the socio-economic growth of the country. 3. Degree of urbanization 3.1. In general The great increase of the urban population of the country, its redistribution by region, during the sample period 1952-70, the increase anticipated in the coming years and the role of urbanization in the economic growth of the 13/ country, makes necessary the need to assess the degree of urbanization.— In the present work, by the degree of urbanization of the Chilean population, we will be facilitated in estimating the symmetry model at a national and regional level which, as we said previously, refers to the development of the urban population of the country and its industrialization. on the basis of the assessed degree of urbanization, Furthermore, in conjunction with the anticipated economic development of the country, we shall be able to indicate certain points regarding the policy on urbanization. Finally, the degree of urbanization and the rate of growth of the urban population of the country in a period of time, are basic elements of the investigation of regional economic problems of the country. 3.2. Evaluation of degree of urbanization In Table 3.2.2, vie refer to the already realized degree of urbanization and the rate of growth of urban population by region, during the period 1952-70. 13/ From this table, the increase in the degree of urbanization in all the The degree of urbanization is the proportion (percentage) of the urban population to the total population of the country, ©lis relationship, refers to a specific point of time, and consequently it dees not give the degree of development of the urban population of the country during a period of time. For this reason, the rate of growth is evaluated. ) 19 ( provinces of the country, during the three census years 1952, 1960 and 1970, is ascertainedo high.—^ In certain provinces, the degree of urbanization is extremely The degree of urbanization is between 17.97 per cent and 89.28 per cent in the year 1952, and between 31.58 per cent and 96.64 per cent in the last census (1970). From this point of view we observe an increase of both the lower and higher limits of urbanization. Finally, the degree of urbanization of the Chilean population, at a national level, followed an upwards trend during the aforementioned period (1952-70). Regarding the rate of growth of the urban population of the country, by province, we can say in the following Table 3.2.1. Table 3.2.1 . THE ANTCUAL RATE OF GSOWTK OF TKE URBAN POPULATION OF THE COUNTRY DURHiG THE SAtSPLE PERIOD 1952-70 Urban population 1970 Percentage on the urban population Annual rate of growth 1952-70 Hunter of provinces 1-2 1 37 015 0.56 7 1 559 339 23.41 3-4 11 4 496 688 67.51 4-5 3 247 753 3.72 5+ 3 319 729 4.80 25 6 660 529 100.00 2-3 Total Sources? : Dirección General de Estadística y Censos, Entidades de Población (de Tarapacá a Magallanes) para 1970. Dirección de Estadística y Censos, Santiago de Chile, 1972. There are eleven provinces whose rate of growth of urban population of the country is between 3 and 4 per cent, and their population consists of 67.5 per cent of the total urban population of the country. In the provinces of Tarapacá, Atacama and Aysen, the rate of growth of the urban population exceeds 14/ For further details, as far as the number of provinces which have a great degree of urbanization in the aforementioned census years, see Table 3.2.2. ) 19 ( Table 3.2.2 THE DEGREE OF URBANIZATION OF THE CHILEAN POPULATION BY REGION, DURING THE SAMPLE PERIOD 1952-70 / Province 1952 1960 Tarapacá Antofagasta Atacama Coquimbo Aconcagua Valparaíso Santiago O'Higgins Colchagua Curicó Talca Maule Linares fiuble Concepción Arauco EicHBio Malleco Cautín Valdivia Osorno Llanquihue Chiloé Aysén Magallanes 59.72 89.28 51.73 39.38 39.75 85.26 86.72 40.44 26.16 37.27 39.45 36.69 30.27 34.99 76.12 28.63 31.92 38.67 33.12 37.03 40.45 32.28 17.97 44.46 81.37 87.11 94.79 73.52 51.83 55.50 88.82 90.01 53.31 32.69 40.93 43.57 39.84 36.33 39.62 81.70 35.71 37.12 44.95 38.74 43.94 46.12 41.96 22 013 52.86 02.93 92.14 96.64 83.30 59.53 60.53 91.44 93.33 55.31 37.91 45.71 52.53 44.76 40.24 48.31 85.18 49.61 49.11 51.88 43.93 52.53 53.68 52.19 31.53 64.36 86.30 60.22 68.19 75.10 o f tiae 1970 Chile 1. 2. 3. 4. 5. 6. 7. 8. 0 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22 0 23. 24. 25. of Degree of urbanization Sourcess 52-60 . ^ (Pontage) ? ^ ^ ^ a n population 60-70 52-70 6.97 2.65 9.05 5.49 5.30 3.19 4.57 5.26 4.38 3.28 3.39 2.27 4.26 3.15 4.27 5.42 4.37 2.99 2.93 3.51 3.61 5.53 2 0 42 6.70 3.79 4.09 1.75 3.98 2.32 2.25 2.03 3.19 2.03 2.10 1.90 3.00 1.53 2.01 2.68 2.21 4.27 4.17 1.59 2.98 2.46 2.59 3.89 4.69 4.45 2.37 5.37 2.15 6.23 3.73 3.61 2.55 3.80 3.47 3.11 2.52 3.17 1.86 3.01 2.89 3.12 4.78 4.26 2.21 2.96 2.93 3.04 4.62 3.68 5.45 3.01 4.27 2.81 3.46 Dirección General de Estadística y Cansos, Censos de poblacioñ (1952-1960). Dirección de Estadística y Censos, Santiago de Chile, 1959', 1969. Dirección General de Estadística y Censos, Entidades de Población (de Tarapacá. a ffogallanes) 1970. Dirección de Estadística y Censos, Santiago de Chile, 1972. ) 19 ( 5 per cent. Similarly, the rate of growth (4-5 per cent) can be considered high in three provinces, those of Arauco, Bio-Bio. and Llanquihue. The urban population of these six provinces consists of 4.30 per cent of the urban population of the country, on the average. Finally, in the size-class 2-3 per cent, there are "seven provinces,"and their population amounted to 1 560 thousand inhabitants in the year 1970. In general, we can say that great deviations by province are ascertained around the annuail rate of growth of the urban popula. tion at the national level. The rate of growth of urban population of the country showed a drop in nearly all the provinces during the semi-period 1960-70, in certain of which this downwards trend was considerable as, for example, in the provinces of Atacama, Aconcagua, etc. 4. 4.1. Urban and rural areas. The urban areas Their regional distribution In the second section, referring to the urban population, we ascertained that it was increased by 86.4 per cent during the past twenty years (1952-70). Furthermore, we were led to the conclusion that the increase of urban population came, mainly, from the movement of rural population to the urban areas and the movement of urban population of small urban centres to the metropolitan centres of the country -urban-turban migration. In addition to the aforemen- tioned reasons, as far as the urbanization of Chilean population is concerned, the inclusion of rural regions in the urban areas caused a further increase of 15/ urban population of the country, as also the increase of urban areas.— The increase of urban areas, as was expected, caused their redistribution among the provinces of the country. Therefore, it is considered advisable to give a picture of the development of the number of urban and rural regions, as well as the redistribution of these regions effected in each province during the sample period 1952-70.. Table 4.1.1. shows the number of urban and rural areas, and their distribution by province, during the census years 1952, 1960 and 1970. 15/ Fran the table, it appears that the number of urban centres in the The characterization of certain rural regions as urban centres was imposed by reasons of i socio-economic and administrative nature. ) 19 ( year 1960, amounted to 655, showing an Increase by 83.5 per cent, compared with the 357 urban centres in 1952. In the last census (1970), the urban areas were 517, showing a decrease by 21.1 per cent, in comparison with the previous census year (I960). This reduction of the urban centres, in the last decade 1960-70, is mainly attributed to the unification of certain urban areas. Similar to the development of the number of urban centres, was the development in the number of rural regions. Thus, during the first decade (1952-60), the number of rural regions was increase:! by 14.4 per cent while during the semi-period 1960-70, it showed a decrease, in comparison with previous sani-period (1952-60), by 8.9 per cent. The decrease of the rural regions during the period 1960-70 is attributed, on the one hand to the inclusion of certain rural regions in the urban regions and, on the other hand, to the unification of certain rural regions. Finally, Table 4.1.1. shows the provinces where the number of urban and rural areas experienced an increase and a decrease during the census years under examination. Thus, of the 25 provinces of the country, 13 showed a decrease of urban areas between the years 1960-70 and, as a matter of fact, the reduction of urban areas in certain provinces was exceedingly important, such as for example, in the provinces of Santiago, Coquiirfoo, etc. Similarly, during the same period, nearly all the provinces showed a reduction of the number of rural regions, with the exception of 6 provinces which showed an increase of rural areas. These provinces are, Santiago, O'Eiggins, Colchagua, Talca, AysSn and Magallanes. In general, we can say that the increase and/or decrease of urban and rural areas varies in each province, during the two semi-periods 1952-60 and 1960-70. Furthermore, the decrease of the rural areas, for reasons which have been explained previously, can be considered an encouraging element, from the point of view of economic growth of the country and regional economic development. Finally, the decrease of the number of urban centres in certain provinces and in the country as a total, in the semi-period 1960-70, is not in contrast to the result obtained in Section 2, as far as the development of urban population in the same period in concerned, because this reduction of the urban centres came by the unification of certain small urban areas, a fact which can be also considered a favourable point in the whole effort of economic development of the country. ) 19 ( Table 4.1.1 DISTRIBUTION OF URDAN AI© RURAL AREAS BY PROVINCES, DURIKG THE u> ) 19 ( economic development in the future and, consequently, these areas should come in for particular attention, because the demographic development of an area is one of the fundamental prerequisites of its economic growth» Finally, worthy of particular attention, as regards the population development of the above size-classes, is the intermediate size class (5 000 to 10 000 inhabitants) where there is an increase both of the number of the urban centres of this group, as also of the population, in both the periods under examination, 1952-60 and 1960-70. The population of this size class increased by 33 o 6 per cent, during the period 1960-70, while the increase of the population was 2.1 per cent during the previous decade (1952-60). The opposite can be said from the point of view of the two semi-periods (1952-60 and 1960-70), regarding the size-class 20 000 to 50 000 inhabitants, vAiere the increase of the population amounted to 29.1 per cent during the first semi-period 1952-60, whereas during the period 1360-70 the population increase of this class was realized at very slow rate (3 per cent). In conclusion, we can say, as regards the demographic development of the intermediate size-classes, that it can be considered satisfactory, and this is an encouraging point from the aspect of the economic growth of the various regions of the country. And this is because the population increase of urban centres, of a relatively small population size, mainly presupposes industrial development of the region. Furthermore, the economic development of such regions, as above, with manifest signs, not only of retaining their population, but also of increasing it, by the movement towards them or rural and even urban population, considerably limit astifilia, that is the movement of rural and urban population towards the metropolitan areas, such as -in this case- the area of Santiago. Regarding these urban areas, their population development in conjunction with the: urbanization of the population of the country, we shall deal with them in detail in the next section. 4*3. The population development of the cities The population growth and the rapid rate of urbanization in Chile, increased the population of the cities which, as is lenawn, are considered as being nearly pure urban centres. This population increase, a result of the ) 19 ( movement of population from rural areas and urban centres with a small population, is greater in those cities which can be characterized as metropolitan areas, such as -for example- the country's capital, Santiago, or the largest port in the country, Valparaiso. The cities are administrative centres of certain regions, because the main administrative services are concentrated in them. Furthermore, the cities are centres of economic, educational (Universities, Professional Schools, etc.), social and diplomatic services. religious administration. Finally, many of the cities are centres of It is, therefore, natural, that the cities form pop- ulation poles and their demographic development, greater in comparison to the other centres, urban or rural, presents particular interest, in the analysis of the problem of urbanization and industrial development. On the other hand, this population concentration is an index of the urbanization of the country's populación being realizedp and the industrial development of these areas, in the sense of demand for labour force. 18/ According to the census criteria feee Table 4.3.1), the cities— with "urban characteristics" were 92, with a population of 3 003 thousand inhabitants, which was 50.7 per cent of the total population of the country in the year 1952. During the period 1952-70, and at the end of this period, the number of cities increased to 98 and their population showed an increase by 93.5 per cent and consists of 65.6 per cent of the total population of the country. This con- centration of the population of the country in the cities, as also their demographic development, in the period voider examination 1952-1970, is given in the table 4.3.1. Furthermore, the cities are also classified according to popula- tion size. Frcm this table, it can be seen that of the 98 cities, a population of more than one million inhabitants. Santiago has The speed of Santiago's growth, during the period 1952-70, can be considered high, although during the second semi-period 1960-70 the rate of growth experienced a slight drop. 187 In the "Ciudad; Gran conglomerado continuo de viviendas, que posee todas las características urbanas en muy alto grado. Tienen este título las cabeceras de departamento y los centros urbanos que por disposición de la ley reciben tal denominación". ) 19 ( Table 4.3.1 THE POPULATION DEVELOPMENT BY SIZE OF CITY. IN THE PERIOD 1952r70 c. Number of cities e I -f of the the city class (in thousands) 1952 ^"¡¡LSÍ6 of growth, _ (percentage Population (in M thousands) • v v ^ 1960 1970 1952 1960 1970 52-60 60-70 1 More than 1 000 1 1 1 213 1 907 2 729 5.77 3.65 2 150-500 1 3 219 253 611 1.82 9.22 3 100-150 2 3 132 264 384 9.04 3.83 4 50-100 7 9 12 451 591 . 876 3.43 4.01 5 30-50 7 11 11 277 430 458 5.65 0.63 6 20- 30 8 11 11 201 280 277 4.25 -0.11 i t 15- 20 3 9 3 143 154 129 0.93 -1.74 8 10- 15 10 13 14 121 156 167 3.23 0.69 49 32 35 246 222 190 -1.27 -1.54 92 89 98 3 00Ö 4 257 5 821 4.28 3.18 Total Population 5 933 7 598 8 869 Percentage of cities' population to total population 50.70 56.03 65.63 9 Less than 10 Totals Sources Dirección de Estadística y Censos, Entidades de Población (de , Tarapacá a Magallanes) para 1952, 1960 y 1970. Dirección de Estadística y Censos. Santiago de Chile, 1955 - 1963 - 1972. size-class of 150-500 thousand inhabitants, the cities with larger population, of course after the capital of the country, are included. cities was tripled in the period under study (1952-70). The number of these The fact of the. in- crease of the number of cities and the high rate of growth of their population (9.22 per cent in the decade 1960-70) confirms the existing trend for concentration of the population in the larger cities of the country. be said for the next two population size-classes. The same can The number of cities in the fifth and sixth size-class was unchanged in the census years I960 and 1970. We have a considerable decrease of the number of cities in the last size-class (less than 10 000 inhabitants) and, consequently, & decrease in their population. ) 19 ( In contrast to the decreasing trend of the aforementioned population group, in the 10-15 thousand inhabitant size-class, we have an increase in the number of cities. It is possible that this fact may reflect census reclassifications or, more probably, the inclusion of certain cities of less than 10 thousand inhabitants, on account of an increase of the population of the cities of this group. This, of course, does not reduce the importance of the fact that cities with a small population can develop demographically, which means that the internal migrants can move themselves to medium population size-class urban centres, instead of going towards large cities. As it is known, this demographic devel- opment of medium size-class cities, presupposes economic and social growth. Fran the preceding analysis of the data of the table, we are led to the following conclusionss a) An increase in the number of cities with a large population, with a corresponding decrease of the cities with a small population, ©lis can be interpreted as a movement of the population of the cities with a snail size population towards the large urban centres, verifying yet once more the trend far concentration of the population in the metropolitan areas of the country. b) Cities with a small population can develop, as was the case in the size-class 10 to 15 thousand inhabitants. c) The rate of growth of the capital of the country showed a drop during the second decade of the period 1952-70. d) An increase in the total number of cities by 6 cities, while their population was increased by 93.5 per cent during the period under examination (1952-70). We said previously that cities with a large population, (over 50 thousand inhabitants), show a demographic development. This event is made more specific 19/ by Table 4.3.2, which includes the main cities— of the country with the 19/ As the population of the cities in Table 4.3.2, the total population of the area of the city was taken, without separation of the urban centres which consist of the city, from the rural centres of the area. This was done, because the percentage of urban centres of these areas is very small and, consequently, the cities can be considered pure urban centres. ) 19 ( characteristic of an upwards demographic evolution* Based on the data of Table 4.3.2, we see that the population of the cities of Santiago, Vina dial Mar and Antofagasta was doubled during i±e periodl952-70, and the annual r a t e o f growth of these three cities, although different in each decade, wag 3.91,per cent, 4.21 per cent and 3.88 per cent respectively. The samecan be said for the city, of Temuco, where the annual rate of growth wast 4.23 per cent. The largest population increase occurred in the city of Arica, where the population increased -by 363.16 per cent and the rate of growth was 8.51 per cent during the latest eighteen years (1952-70). The population increase of this city was perpendicular, if it be taken into account that the annuâl r'àte of grcwtft was 0.5 per pent during the period 1930-52. .Jfe also have a considerable increase of population in the city of Talcahuano, where the popula--. tion increased by 169.10 per cent. In this case, the. annual rate of growth followed an upwards trend between the decades of the period 1952-70,.—^ On the contrary, the increase of the population of the city of Valparaiso can be considered small, within the period 1952-70, (14,61 per cent) and, as a matter of fact, during the decade 1960-70 we had a slight decrease of its population. Finally, from the aforementioned table it appears that the population of the 10 cities of the table which are surveyed, amounts to 3 993 thousand inhabitants, consists of 45.02 per cent of the total population.of the country and experienced an increase by 92.90 per cent during the period under study 1952-70. As a conclusion, we can say that the demographic development of these cities, being different between each city, followed a différait trend in two decades of the period examined, 1952-70 with an annual rate of growth of 3¡.65 per cent. 20/ In the cities of Temuco, Talca and C h i l l & i . a l s o have an upward trend of their rate of growth during both decades, 1952-60 and 1960-70. ) 19 ( Table 4.3.2 THE DEMOGRAPHIC DEVELOPMENT OF THE MAIN CITIES OF THE COUNTRY DURING THE SAMPLE PERIOD 1952-70 n / /n —.. <~xtx Population (in thousands) Annual rate of growth (percentage) 1952 1960 1970 52-60 1 350 1 907 2 729 4.32 219 253 251 1.81 85 115 182 1 Santiago 2 Valparaíso 3 Viña del Mar 4 Concepción 120 148 5 Talcahuano 55 6 Antofagasta 7 60-70 52-70 Increase Absolute Relative (in thou- (Percenssnds) tage) 52-70 52-70 3.91 1 379 102.15 -0.10 0.75 31 14.61 3.79 4.55 4.21 97 114.12 178 2.62 1.85 2.19 58 48.33 84 148 5.29 5.73 5.53 93 169.09 62 88 125 4.30 3.53 3.83 63 101.61 Temuco 52 72 110 4.21 4.25 4.23 59 111.54. 8 Talca 55 68 94 2.67 3.25 2.99 39 70.91 9 Arica 19 43 88 10.34 7.05 8.51 69 363.16 Chillán 53 65 88 2.67 2.96 2.83 35 66.04 Total 2 070 2 844 3 993 3.97 3.39 3.65 1 923 92.90 Total population of Chile 5 933 7 374 8 869 2.72 1.85 2.23 2 936 49.49 10 Sources: 3.58. Dirección de Estadística y censos. Censos de Población . _ _ Dirección de Estadística y Censos, Santiago de Chile, 1959-1969. Dirección General de Estadística y Censos, Entidades de Población (de Tarapacá. a Magallanes) para 1970. Dirección áe Estadística y Censos, Santiago de Chile, 1972. • ) 40 ( IV. URBANIZATION A N D INDUSTRIAL DEVELOPMENT IN TEE CASE O F CHILE 1. The Influence of industrial development and urbanization 1.1. The functional relationships of urbanization As a basic hypothesis, as regards the urbanization of the population of the country and industrial development, we accepted that the latter -industrial development- influences the development of urbanization. In other words the increase of urbanization is a function of the industrial development of the country. This can be written, in a mathematical form, as follows: NJ: = f (It) (1) where: N£ S urban population of a country, at time, t. It : industrial development of a country, at time, t. f : Symbol of function, t : time. The functional relationship (1) expresses the influence of industrial development on urbanization. It was said, however, in the determination of the problem under study and its theoretical consideration that, apart from the aforementioned factor of industrialization, other factors also contribute to the increase of internal migration towards the urban areas and, particularly, to the movement of the agricultural labour force to the industrial centres of the country which are characterized as urban centres. These factors, on the basis of what we mentioned in the brief description of an econometric model and its statistical criteria in Chapter II, Section 3.1, are included in the unexplained part of the equation. As a consequence, relationship (1) can be written in the following form: N£ = f (It) + e t where: N*. I, f, t : e : as in relationship (1) unexplained part of the relationship. (2) ) 19 ( Furthermore, in the relevant chapter concerning hypotheses and methodology, we said that as independent variables in the functional relationship of urbanization, we will use the per capita income of the urban population, the demand for labour in the secondary branch of the economy, the wage of industry and 21/ unemployment in agricultural p r o d u c t i o n . — T h e reason for this is that the above factors are economic and demographic factors, whose investigation is included in the aims of the present work, and they correspond more with the causes of internal migration towards the industrial centres and can be considered, in a broad sense, as indices of the industrial development of the areas. Based on the above, relationship (2). can be transformed and takes the following form, on the basis of the aforementioned determinative factors -causes- of the increase of urbanization of the population of the country % = f (YJ/^) + e t (3) wheres % urban population of a country, at time, t. Y^/t-r^ s per capita urban income, at time, t. e 21/ i residual of the relationship. Of course, apart from the aforementioned determinative factors, there are other factors. These factors (e.g. communication - information, advertising over TV and radio, etc.) on account of their qualitative character, cannot be introduced as explanatory variables in the functional relationships of the urbanization of the population under study. Further to this, the influence of these qualitative factors on urbanization, is included in certain variables of a quantitative character, which are introduced in the equations to be tried. For instance, in the case of communication, advertising, etc., the factors cause the increase of demand for labour force in the secondary production, and, consequently, the influence of the demand for labour force. Finally, the existing difference between the per capita urban and agricultural income was taken as an explanatory variable cf the development of the urban population, but the results obtained were almost similar to the equation (1) in the Section 1.2 at the present Chapter. ) 19 ( The functional relationship (3) expresses the influence of the changes of the urban per capita income on the size of the urban population of the country. Moreover, this relationship, by the elimination of the neutral part both of national income and total population of the country -agricultural income and rural population- corresponds more satisfactory to the hypothesis adopted regarding the relationship between urbanization and the per capita income of the urban areas. Another important factor which influences urbanization and can be introduced as an explanatory variable in the functional relationship between the variable under study and economic and demographic factors, is the demand for labour force, in the secondary branch of the economy. Thus, in this case we will haves = f (D™) + e t (4) wherej D^ % demand for labour force in the secondary production, at time t. : as in the relationship (1). This relationship can be supplemented by the introduction, as an independent variable, of unemployment in the agricultural sector of the economy, which can also be considered a determinative factor of the variable under study. Thus, we will haves Nj: - f (d£, uj) + e t (5) where? uj ° unemployment in the primary production of the economy at time, t. w ' as in the relationship (4). Finally, the worker's wage in industry was said to be considered a factor for the movement of the agro-labour force to the industrial sector of the economy. Therefore, we can formulate the following functional relationship with the explanatory variable of the worker's w a g e . — ^ 22/ Apart from the labour wage in industry, w W , as an explanatory variable in the above functional relationship, the existing difference between the industry wage and the agricultural wage was introduced- (Remark of Prof. J. Elizaga during the conference on 31st October 1973), however, the statistical results were not satisfactory. Similarly, in this same functional relationship, the ratio of the industrial wage and the agricultural wage was tested as an independent variable, but in this instance also we did not have better luck, as regards the statistical results obtained. ) 19 ( = f fwf) + e t (6) where: labour wage in industry, at time, t. After the formulation of the above six functional relationships of the urbanization of the population of the country, they are tested in indicated forms of equations, that is, linear and curvilinear form. The following tables include the models, i.e. they incorporate their mathematical forms, the parameter estimates with their standard errors in parenthesis, the determination 2 coefficients (R ) and the von Newman criterion. The comparison of the results to be obtained in these ¡models, will allow us the selection of those models for the expression of forecasts. Furthermore, the elasticities, as far as the explanatory variables are concerned, are included in Section 1.4 of the present chapter. Finally, the sampling period (1961-70), from the point of view of the number of observations, can be considered satisfactory, because the Changes of the exogenous variables during this period will be involved in the results of the statistical estimates. 1.3. The analysis of the statistical results obtained In the Section 1.2, the best fitting regressions are given in linear and logarithmic form. These regression equations explain the change which the variable under study undergoes through the influence of the determinants taken during the sample period, 1960-71. number. These equations are numbered by a nominal The comparison of the results obtained, will allow the selection of those equations, for the expression of forecasts. A detailed analysis of the results obtained by the regression equations, on the basis of statistical criteria, is attempted hereunder. The following are observed as statistical criteria of the equations tried: a) The sign of the regression coefficients should correspond to the economic theory, on the basis of which the model was formulated. The sign of the the per capita income is positive in the equations, as was expected. Similarly, the sign of the demand for labour force, unemployment in the agricultural branch of the economy and the wages in secondary production is positive based on the ) 44 ( 1 . 2 . Jhs best f i t t i n g roqrassions 1.2.1. The l i n e a r form Percentage of f variance explained by regression 0) (2) (3) N r - - 3 8 5 6 , 6 + 3.509 YT/HJ (0.417) 1 1 94.2 a 887 0.1820 N r - -1952.2 4 12.12 0? (1.53) 1 96.2 a 925 0.2307 ii r - a + b ti" Nr - 88.7 0.786 0.8607 f Nr - 2483.1 • 7.48 oj + 35a 54 U? 99.9 0.999 0.2307 94.3 0.890 0.1820 n/n 1* Estimate of parameters Kode] r ;.,P Nj - a + b Y j / N j - a + b Dj - a + b of + Ua t c 549.2 + 56.83 WH (17.12) (1.11) 1 & Von Newman K2 (18.48) 1.2.2. The logarithmic form 1* Log Nj - Log a+b Log yJ/nJ Log 9175+1.6513 Log Y j / N j 2* Log Nj - Log a+b Log o j Log f.- a0675+1.3175 Log d" t 1 (a 1741) 96.1 0.924 O 2307 L 3* Log f { - Log a+b Log WH Log Nj- 1.9897+0.9024 Log n j 88,9 0.790 a 8607 99. 9 0.999 0.2307 x Log f , » Log a+b Log 0%cLog U® t t t Log t (0.0110) 0.8469+0.8285 Log Dj + t (QJ212) 0.5922 Log U t ( a 0315) ) 19 ( général hypothesis. In other words, the positive sign of the regression coef- ficients mëans that an increase of the determinative factor creates an increase of urbanization of the population of the country and, especially, an increase of the labour force in the secondary production, by moving the agricultural latour force towards the industrial centres of the country. 2 b) The coefficient of determination, R . This coefficient determines the percentage of the variance of the dependent variable explained by the independent variables. Fran column (1) of the econometric model tables, it is seen that the fitting regressions have explained a very high percentage of the total variance in the dependent variable. Of the 8 regression equations, 6 have ex- plained more than 90 per cent of the variance of the regression, and the fit of the corresponding lines to the data observed must be considered very good. average correlation coefficient is found to be about 0.90. The In two cases -(2) and v4) equations in both forms- the correlation coefficient is greater than 0.95. The criterion "F of Snedecor" is taken into account on time series, for 2 testing the statistical significance of the coefficient of determination, R , and it is investigated at a level of 5 per cent or less. From such a point of view, the regressions are considered statistically significant. In conclusion, we can say that the coefficient of determination, as a measure of the good fit of empirical data, shows that the aforementioned regressions can be considered the best fitting ones. Moreover, we verify the prob- able existing correlation between the variable under study and independent variable of the functional relationships. d) The standard error of estimate. The reliability of the statistical parameter estimates of an equation is judged by the standard error. In other • words, the standard error is the measurement of the fit of the data of a regression equation formulated. The criterion of t-student is taken into consid- eration for testing of the statistical significance of the parameters estimates, at a level of 5 par cent or less. On the basis of this criterion, the regression coefficients of the equations tried are statistically significant. e) Regression coefficients. test fitting regressions. They were included in the tables of the Erch regression coefficient measures the change of the esqalained variable -urbanization- caused by the unique change of the ) 19 ( independent variable. Thus, the coefficient 3.509 on the per capita urban income in equation (1) of the linear form, means that if the per capita income rose by 100 escudos, the urban population would rise by 350 thousand persons approximately. In the linear equation (2), its coefficient shows that if the demand for labour force rises by 1 000 persons, the urbanization of the population would increase by 56 000 inhabitants approximately, and so on. f) Autocorrelation s As regards the investigation of autocorrelation between the successive observations of the sane time-series of the independent 23/ variables, von Newman's criterion was applied.—• was made at the five per cent level. Ihe test, of significance, The calculated value, Kg, is less than the corresponding value of the table (K^) and consequently, it is concluded that a positive autocorrelation exists. Having analysed the results obtained, we come to the following conclusionss i) Hie regression equations tested can be considered as those best fitting to the empirical data regarding the explanatory variables, as 2 can be verified by the coefficient of determination, R . ii) The forms of the equations can also be considered the appropriate ones, in accordance with the standard error of estimate, iii) The signs of the regression coefficients were what was indicated by the economic theory. Therefore, all the equations can be vised for forecasts of the development of urbanizations in the coming years, and for this reason are marked by*. 23/ v , 2 z - -As- s where: d2 The average of the sum of the squares of the differences between the successive values of the variable, (X t + 1 " Xt)2' * 1,2 s 2 , The variance of the sample. n ) 19 ( 1.4. Elasticities , 1.4.1. Hie concept of elasticity In a quantitative economic analysis, the existing relationship between two economic magnitudes is measured by the elasticity and, consequently, its investigation is deemed necessary in the present work. Before proceeding with this, we represent its mathematical concept. A mathematical function has the property of elasticity. Thus, given a function f(x, y, z) ,its elasticity with respect to the variable, x, is defined as plus or minus its logarithmic derivative, with respect to x. We denote the elasticity by e f according to the sign, x d lQ g f = + illw e f . + x ~ d logTx ~ f dx The elasticity with respect to income, demand etc,, is briefly called income elasticity, demand elasticity, etc. Besides the sign, the elasticity consists of its amount. Finally, the elasticity is independent of measuring units of the variables. 1.4.2. Average and point elasticities According to formula (1) of the preceding section, the income elasticity is defined as follows: e„r *Y V dl/ . d(YVMr) ^ where; i /tr • income elasticity Y^fiF, N37 s as in the list of variables i.e. the income elasticity is equal to the product of the derivative of urbanization, with respect to the per capita urban income at a point of the curve, by the ratio /Nr where; Y , „ and vF are the per capita urban income and urbanization at that point. 'W ) 19 ( This elasticity is called point elasticity. The average elasticity is derived by taking the average values of the variables, and N* in the formula álf ^ " d( Yr / w r) ' ^ and it refers to a period of time. The elasticities, with respect to the explanatory variable of the logarithmic equations, are included in the following Table 1.4.2.2., while the elasticities of linear form estimated are included in Table 1.4.2.1. The analysis of the elasticity is based on the sign and its amount. The sign shows the direction of the change of the variable under study -urbanizationcaused by each explanatory variable, while the amount shows how much. thi 0 From point of view of sign, the elasticities coefficients drawn have the sign indicated by economic theory. Regarding the amount, the average income elas- ticity is 1.65 in the linear equation (1). That is, if the average income increases by 10.65 per cent the urbanization of the population will increase by 10o65 per cent. In the linear equations (2) and (4), the average demand for labour force elasticity is 1.3 and 0.81 respectively. This means that if the labour force of the secondary production increases by 10 per cent, the urban population will rise by 10.31 per cent and 8.10 per cent respectively. In the same manner, we can explain the meaning of the other explanatory variables. Finally, the elasticities of the linear equations are not constant at different points of the curve. These changes of elasticities during the sample period 1960-70 mean that the influence of each explanatory variable and urbanization was different at the various points -years- of this period of time. In conclusion, we can say that inccme and the demand for labour force in the secondary branch of the economy (first and second equations) can be considered the most important determinative factor of the urbanization of the Chilean population. The wages for labour in secondary production and unemployment in agricultural production also influence the increase of the population of the urban centres, but at a lower level than the previous two main factors. ) 19 ( Table 1 . 4 . 2 , 1 ELASTICITIES WITH RESPECT TO THE EXPLANATORY VARIABLES OF THE LINEAR EQUATIONS TRIED Number of Variable equation lx Per capita Income, v r . /„ r /N Averaige elasticity 1.65 2s Demand f o r l a bour force, Dw 1.31 3* Wages for workers, ^ a 81 Unemployment, U 1960 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1.74 1.74 (189 Demand for l a bour force, Dw Point e l a s t i c i t i e s - a 92 - - 1.29 - 1.28 - . a80 0,79 - 0,79 - 0.60 a 51 - 0.53 - 1.70 1.36 - - 0.93 - 0.88 0.84 - - Q.59 - 1.59 1.30 - 0156 - 1.52 1.69 - - - - Q.S3 Table 1 A 2 . 2 ELASTICITIES WITH RESPECT TO THE EXPLANATORY VARIABLES OF THE LOGARITHMIC EQUATIONS TRIED Number of equation Average elasticity Variable Per capita income, 2* Demand for labour force, D 3s Wage for workers, Ww 4* w Demand for labour force, D a 83 4* Unemployment, l) 0.59 a V r N 1,65 1* W' 1.32 ' a 90 ) 19 ( Similar remarks can also be expressed for the logarithmic form of the equations, as far as the elasticities are concerned. 1.5. Forecasts 1.5.1. Techniques. Period of forecasting. Assumptions Forecasting is a more or less successful prediction of the future, on the basis of the observed regularities in the past, i.e., the inference from the past in the future. In Chapter III, we analysed the development of the urban» ization of the Chilean population in the past, while in Section 1 at the present chapter, we determined the main economic factors which influence the increase of the urban population of the country and we selected the most appropriate regression equations for forecasts marked by In this section we will make forecasts as far as the development of urbanization in the projected period of time 1971-75. In other words, the dependent variable -urbanization- will be predicted on the basis of selected regression equations and certain assumptions as far as the determinative factors are concerned. This technique of fore- casting, which is called forecasting by functional extrapolation, will be used in the present case. Furthermore, the projected period of time is five years approximately (1971-1975) and, consequently, we will make long-run forecasting. As regards the assumptions, they refer to the factors which influence urbanization. More specifically, these assumptions are the following! a) The development of the explanatory variables, as this is presented in the respective tables, will indeed take place and their properties will remain unchanged. b) The determinative factors defined during the sample period 1960-71 will continue exercising the same influence on the urbanization of the pop- ulation of the country in the forecasting period. c) New systematic factors affecting the variable under study will not appear during the projection period 1971-75. It is known that the validity of the assumptions mainly depends on the "elongation" of the period of forecasts, while the accuracy of a forecast depends on the realisation of the assunptions adopted. In general, we can say that it is not easy to forecast accurately, because there are many difficulties, ) 19 ( however, the deviations between what is realized and what is evaluated by forecasting will facilitate the improvement of the models, by introducing new determinative factors, type of equation, period of forecasting, etc. Finally, a comparison of the results obtained by this technique of forecasting and those drawn by CELADE will be realized hereunder. Table 1.5.1-1 DEVELOPMENT OF DETEPMn^TIVE FACTORS OF URBANIZATION IN 1975 Variable n/n Assumptions Rate of growth 1971-75 Estimations 1975 794.7 2.5 810.4 (C) 3.0 826.4 Pessimistic (A) 2.0 103 7—/ Conservative (B) 4.0 118.8 Optimistic (C) 6.0 135.8 Pessimistic (A) 1.0 10.8 Conservative(B) 1.4 11.0 Optimistic a/ 2.0 Optimistic Note: 1.6 2.0 Conservative(B) if Pessimistic (A) Conservative (B) Optimistic (C) Pessimistic (A) Y W 1*8 11.2 (C) 2.5 3 108 3 157 3 220 in escudos in escudos in escudos in thousand persons in thousand persons in thousand persons t ^l0usand persons in thousand persons in thousand persons in thousand persons in thousand persons in thousand persons pess:umstlc' assumption of the aforementioned variables corresponds to their rate of growth during the past period 1960-71. Base year 1962»100 and period of time 1968-75 for the rate of growth. 1.5.2. Evaluation of forecasts comparisons In Table 1.5,2.1., we present the arithmetical results which refer to the increase of the urban population of the country in the year 1975. Further- more, this table includes the number and the type of the equations tested, as ) 19 ( well as the three assurtptions as regards the development of the determinative factors of the variable under study -urbanization- in the same year (1975). Table 1.5.2.1 THE DEVELOPMENT OF THE URBANIZATION OF THE POPULATE® OF THE COUNTRY IN THE YEAR 1975 Number of equation Type of equation Assumptions Urban population (in thousands) 1975 1 Linear A B C 7 049 7 221 7 442 2 Linear A B C 7 630 7 370 8 064 3 Linear B C 7 301 8 267 4 Linear A B C 7 247 7 435 7 625 1 Logarithmic A B C 7 074 7 258 7 500 2 logarithmic A B C 7 736 7 938 8 145 3 Logarithmic B C 7 278 8 212 4 logarithmic A B C 7 273 7 473 7 676 Finally, we compare the results obtained, regarding the development of the urban population of the country in 1975, with the urban population forecast in the same year (1975), made by the Latin American Demographic Centre (CEIADE) 24/ La proyección media para 1975, de que se dispone después del Censo de 1970, es para la población total (CEIADE, Boletín Demográfico N°ll, Santiago de Chile, enero 1973). Para obtener la proyección de la población urbana para 1975, se aplicó el porcentaje de la población urbana, estimado por la Corporación de Fomento de la Producción (Perspectivas de Crecimiento da la Población Chilena 1970-1965, Corporación de Fomento de la Producción, publicación N°10-A 370. Santiago de Chile, 1970). ) 19 ( CELADE and equations applied Assumption Urban population (in thousands) 1975 Conservative OB) Optimistic (C) 8 140 7 442 698 8.6 Conservative (B) Optimistic (O 8 140 8 064 76 0.9 Conservative (B) Optimistic (O 8 140 . 8 267 -127 -1.6 CELADE Conservative (B) Optimistic (C) S 140 7 625 515 6.3 4* equation linear Conservative (B) 8 140 640 7.9 CELADE 1* equation logarithmic CELADE 2* equation logarithmic CELADE 3* equation logarithmic CELADE 4* equation logarithmic Optimistic (C) 7 500 Conservative (B) 8 140 5 -0.1 Optimistic 8 145 - 72 -0.9 464 5.7 rFTADK 1* equation linear CELADE 2* equation linear CELADE 3* equation linear (O Conservative (B) 8 140 Optimistic (C) 8 140 Optimistic - 8 212 Conservative (B) Deviations Absolute Per cent 7 676 (O From the above data it appears that the use of the variable of demand for labour force in the secondary branch of the economy, as a deterrcdnative factor of the increase of the urban population of the country, has been the roost successful. The same can be said for the other determinative factors. The rela- tively higher deviation between the forecasts of urban population made by CELADE and the models appears in the first equation which incorporates as an explanatory variable, the per capita income. This can be attributable to the non-continuous upwards trend of the national, income in the period under study 1960-70. This fact does not reduce the meaning of the per capita income as regards its influence on the population movements of the rural areas towards 25/ the urban centres.—' ¿¡5/ fhe factor, per- capita income in the case of increase of urban population of the metropolitan area of Santiago, gave very satisfactory results, verifying in this manner what we previously stated. ) 19 ( Consequently, on the; basis of the above comparative results obtained by the Latin American Demographic Centre (CELADE) and the models of the present work, regarding the development of the urban population of the country in the year 1975, we can say that their deviations can be considered negligible. Therefore, the application of such models, from the point of view of making urban population forecasts, is indicated. In the diagrams 1-4, we give the observed and estimated values of the urbanization of the Chilean population, their deviations in percentage and the influence of the explanatory variables on the variable under study of linear equations which, with their logarithmic ones, are used for forecasts. 2. Regional economic growth and urbanization 2.1. In general In the first chapter, we said that the problem of economic growth appears more intense in less-developed countries. In these countries, the difference in income levels by geographical regions are, at tiroes, great. As a rule, the more less-developed the economy of a country is, the greater is the inequality of development among the various regions of the country. The regional problem, as the inequality of income distribution among the different regions of a country is more broadly called, is of different intensity and form in each country, and is characterized by that factor which appears more intensively. These fac- tors, apart from income, are employment, urbanization, industrialization, external emigration, etc. These factors are closely related between themselves, so that the appearance of one of them presupposes the existence of the others. The per capita income of the inhabitants of a region, is the measure for the estimation of the degree of economic growth of each geographical region. Fur" thermore, for the total of regions, as a measure of the existing differences, the index of regional inequalities, introduced by Williamson, can be applied. This index is a reliable criterion for the measurement of total differences during a period of time. Apart from the above comparative usefulness, this index can be utilised to comparisons between countries as regards existing inequalities among the regions of each country, but this vise of the index is limited. ) 19 ( 2.2. The regional problem in Chile The regional problem in the case of Chile, appears in the following form; Large income differences exist between the provinces. development differs considerably in each province. in internal population movements. "regional differences". following The degree of industrial There are also differences Hereunder, we realize an analysis of these Thu3, as regards the income and from the data of the Table 2.2.1., we can say that, in the provinces of O'Higgins and Table 2.2.1 THE PER CAPITA INCOME BY PROVINCE DURING THE YEABS 1960 AND 1970, AT CONSTANT 1965 PRICES "/ Region n Per capita income (in Escudos) 1960 1970 Tarapacá 2 192 3 064 2 p Antofagasta 4 130 5 983 1. 3. Atacama and Coquimbo 1 732 2 413 4. Aconcagua and Valparaíso 2 214 2 391 5. Santiago 2 312 2 984 6. O'Higgins and Colcliagua 1 903 2 247 7. Curioó, Talca, Maule and Linares 1 265 1 560 3. Ñuble, Concepción, Arauco, Bío-Bío and Malleco 1 418 1 697 9. Cautín 961 1 092 1.382 .1'243 1 614 10. Valdivia and O s o m o JLXo Llanquihue, Chiloe and Aysén 12. Magallanes 3 832 4. 758 Chile 1 912 2 437 Sourcess i) ii) 1 532 Oficina de Planificación Nacional (ODEPLAN). Unidad de Estadísticas Básicas e Indicadores Regionales. Producto Geográfico Bruto 1960-1970. (Cifras Provisorias). (Mimao), Santiago de Chile, 1973. Instituto Nacional de Estadística, -Censo de Población 1960, TOKIO A , Santiago ds Chile, 1969. -Censo de Población 1970. Muestra de Adelanto de Cifras Censales, Santiago de Chile, 1972. ) 19 ( Colchagua, the per capita income was nearly the sane as the per capita income of the country (1 900 Escudos) in the year I960, while the per capita income of the provinces of Antofagasta and Magallanes was double. Of the remaining 21 prov- inces, 4 provinces (Tarapaca, Aconcagua, Valparaiso and Santiago) had a per capita income larger than E°1 900, while the per capita income of the remaining 17 provinces was equal to E°1 300 on the average during the same year (1960). The per capita income of the country, in the year 1970, reached E°2 450, thus showing an increase by 27.5 per cent compared with the per capita income for the year 1960. The position of the provinces, in their distribution according to the per capita income, changed in the year 1970. Thus, in four provinces (Atacama, Coquimbo, Colchagua and Valparaiso) the per capita income did not differ from the per capita income at the national level (E°2 400). The per capita income cf the provinces of Antofagasta and Magalianes was E°5 980 and E°4 758 respectively. In other words, the per capita income of these provinces continues being double the average income of the country, as in the year 1960. To these provinces, with a per capita income larger thin E°2 450, can also be added the provinces of Tarapaca and Santiago. The smallest per capita income of the country is that of the province of Cautln (E°l 100), and the difference of the per capita income of this province from that of the per capita income of the country and the province of Antofagasta, which has the largest per capita • income, amounts to E°1 343 and E°3 891 respectively. This ascertainment, as far as the differences of income levels concerned, is verified by the index of regional inequalities. — ^ The value of this index was calculated to be 38.0 in the year 1970 and it is characteristic of the differences existing between the provinces, as regards income, in the case of Chile. Of course, this index showed a slight improvement in comparison with 1960, which means that a reduction occurred in the income differences of geographical regions of the country. 26/ Index of regional inequalities where; f.. s population of the geographic region,.! n : total population of the country Y i ! per capita income of the region, i Y % per capita income of the country ) 19 ( Diagram 1 : 1 s t . LINEAR EQUATION TESTED: OBSERVED AND ESTIMATED VALUES OF THE URBANIZATION OF THE POPULATION IN ' CHILE, THEIR DEVIATIONS IN PERCENTAGE AMD THE INFLUENCE OF THE EXPLANATORY VARIABLES OF THE EQUATION Per capita urban iricone ( i n oscudos) 3 OOO A 2 800 X Y V 2 600 2 400 I 1960 ! i i ' i. _j 65 < 3 j > 70 Years Urbanization * 10 7 000 4L t - -3856.2 • 3.509 Y^ I ^ 6 500 6 000 h 5 500 5 000 J 1960 (Percentage) 100 • I I !_ -I 65 1 1 L —1 o 70 Years ) 58 ( Diagram 6: 2nd. LINEAR EQUATION TESTED: OBSERVED AND ESTIMATED VALUES .OF URBANIZATION OF THE POPULATION OF THEMETROPOLITANAREAOFSANTIAGO,THEIR DEVIATIONS IN PERCENTAGE AND THE INFLUENCE OF TIE EXPLANATORY VARIABLES OF THE EQUATION Demand f o r 3 labour force x 10 7404 700 650 1965 66 67 68 69 70 71 70 71 Urbanization x 10" 7 000£_ ï j - -1952.15 • 12.12 D j 6 500 6 000 5 500 5 000 -J 1965 (Percentage) I 66 67 J 68 L 69 Years ) 59 ( Diagram 6: 2nd. LINEAR EQUATION TESTED: OBSERVED AND ESTIMATED VALUES .OF URBANIZATION OF THE POPULATION OF THEMETROPOLITANAREAOFSANTIAGO,THEIR DEVIATIONS IN PERCENTAGE AND THE INFLUENCE OF TIE EXPLANATORY VARIABLES OF THE EQUATION Wage f o r workers loot 90 V J 1963 ! _ 64 65 J 66 L 67 "Years1"' Urbanization x 10" 7 oodt. J J - 549.24 - 56.83 HJ 6 500! 6 OOOi 5 500 5 000 J 1963 (Percentage) 10$- L 64 65 J 66 L 67 Years ) 60 ( Diagram 6: 2nd. LINEAR EQUATION TESTED: OBSERVED AND ESTIMATED VALUES .OF URBANIZATION OF THE POPULATION OF THEMETROPOLITANAREAOFSANTIAGO,THEIR DEVIATIONS IN PERCENTAGE AND THE INFLUENCE OF TIE EXPLANATORY VARIABLES OF THE EQUATION Unemployment x 10 3 11.1 10J 10.3 9.9 9.5! ,i 1965 I L 66 67 68 Years Urbanization x 10v 7 000 i j - -2483.07 + 7.48 D j + 350.54 6 500 6 000 5 500 5 000 J 1965 t ! _ 56 67 68 Years (Percentage) -o Years ) 19 ( As regards industrial development by region, we can also say that industrialization differs in each province. In certain provinces, the industrial- ization index can be considered as being at a satisfactory level, while in other provinces this index is at a low level. In greater detail, the indus- trialization index, on the basis of the data of Table 2.2.2., has as followss Table 2.2.2 THE INDEX OF INDUSTRIALI ZATIQM BY PROVINCE DURING THE CENSUS YEARS 1952, 1960 AND 1970 Index of Industrialization-7' Provinces n 1. Tarapacá 1952 1960 1970 39.1 33.1 26.3 2. Antofagasta 59.5 50.6 37.5 3. Atacama and Coquimbo 35.5 34.8 32.8 4. Aconcagua and Valparaíso 29.2 26.8 24.5 5. Santiago 37.3 34.7 30.3 6. O'Higgins and Colchagua 19.4 18.9 21.9 7. Curicó, Talca, Maule and Linares 16.2 14.8 13.9 8. füíuble, Concepción, Arauco, Bío-Bío and Malleco 25.8 26.9 24.2 9. Cautín 14.8 13.8 12.5 10. Valdivia and Osorno 21.9 21.7 16.4 11. Llanquihue, Chiloé and Aysén 14.4 17.0 12.9 12. Magallanes 31.8 29.9 24.2 Total 29.4 28.2 25.3 Souroess a/ i) Dirección de Estadística y Censos, XII Censo de Población y Vivienda, 1952. Santiago de Chile, 1959. ii) Dirección áe Estadística y Censos, XIII Censo de Población, 1960. Santiago de Chile, 1969. iii) instituto Nacional de Estadística, XIV Censo de Población y Vivienda, 1970. Muestra de Adelanto de Cifras Censales, Santiago de Chile, 1972. Incluye Población Económicamente Activa ens Industria Manufacturera, Minas y Canteras, Construcción, Luz-gas-agua y servicios sanitarios. Mas adelante, en el Capítulo V, Tabla 1.1, el índice de industrialización incluye solamente Industrias Manufactureras. ) 19 ( In the provinces of Antofagasta and Cautín, the industrialization index was 50.6 per cent and 13.8 per cent respectively, during the year 1960. We observe, therefore, that a considerable difference exists, amounting to 36.8 per cent, between the two industrialization indices of these two provinces which, as can be seen in the table, hold the extreme but opposite, positions, from the industrial development view. There are only six provinces (Tarapacá, Antofagasta, Atacama, Coquimbo, Santiago and Magallanes) whose index of industrialization is more than the index at the national level (28.2 per cent). Finally, in 10 provinces (O'Higgins, Colchagua, Curicóf Talca, Maule, Linares, Cautín, Llanquihue, Chiloé and Aysén) the index of industrialization is 16.1 per cent on the average and, consequently, it can be considered as being at a low level. The index of industrialization has not changed considerably in the years 1952 and 1960, while it showed a drop in the census year 1970, (25.3 per cent). in this year (1970), all the provinces (with the exception of the provinces of O'Higgins and Colchagua) experienced a decrease in the index of industrialization and in certain provinces, as for instance in the province of Antofagasta, this decrease of the index was considerable. As regards the exist- ing differentiation of the number of provinces, from the point of view of industrialization, the situation, has not changed considerably since 1960. Thus there are also five provinces whose index of industrialization is more than 25.3 per cent, while in 10 provinces, mentioned above, this index was 15.3 per cent on the average. •. . inally, as regards population movements and Darticularly the movements f of the population to the urban centres, we can repeat in a sunroary the conclusions obtained during the analysis of the problem in the previous Chapter III, Section 3-4. These results are the following s a) The degree of urbanization of the population of the country can be considered high, and it differs in each province. The same can be said for the rate of growth of urban population by province; b) A redistribution took place of the urban and rural population in the country's provinces; c) The trend of concentration of population in the metropolitan areas. In the following table, we give a general picture of the population outflows to the larger urban centres, on the basis of the 1970 census. ) 63 ( Table 2.2.3 . THE POPULATION OUTFLOWS TOSfcRDS THE LARGER URBAN CENTRES Hi THE YEAR 1370 Internal migration n ; / Urban Centres Absolute number (in thousands) Percentage on the population of the province 228 3.3 1 Santiago 2 Concepción 37 7.0 3 Valparaiso 52 8.1 4 Antofagasta 28 13.0 5 Tarapacá 24 16.2 Total 369 8.6 In conclusion, we can say that in the case of Chile there is a regional problem. The main factors prevailing in this problem are the unequal distri- bution of income, the low level of industrial development of the country and 27/ the differentiation of industrialisation by, region.— Furthermore, the concen- tration of the population in the larger urban centres and in general the great rate of urbanization composes the regional problem of the country. Finally, the results obtained as regards the industrial development by region are in agreement with the conclusions drawn, as regards regional economic growth of the country on the basis of the per capita income. The reasons of these eco- nomic differences of the provinces of the country are the natural resources, such as i.e. in Antofagasta the mines, in Magallanes the oil, the concentration of industries in Santiago, the climate, historical reasons etc., whose investigation on the one hand is frequenly difficult and it needs much time; on the other hand, they are not included in the purposes of this study. 27/ To these factors there can be added the factors of wages in the secondary branch of the economy, unemployment in agricultural production,, etc., the investigation of which was realize! in the analysis of influence of industrial development and urbanization, (Section 1 of the present chapter). ) 19 ( 2.3. The relationship of per capita income by region and the urbanization of metropolitan areas. The case of Santiago. In the previous Chapter III, it was said that there is a concentration of the population in the larger urban centres and particularly in the metropolitan area of Santiago. Furthermore, it was ascertained in the analysis of devel- opment of urbanization in Chile, that the population outflows of the provinces mainly towards the capital of the country, Santiago, differ in each province. Finally, we were led to the conclusion that the differentiation of income by region is the main factor of such population movements. In other words, this last ascertainment indicates that there is a correlation between urbanization and the degree of regional economic development. Hie degree of the economic growth of each province can be determined by the per capita income of the inhabitants of the region, whilst as an index of the urbanization of an urban centre of country, we will take the proportion of the population outflows of all provinces towards a certain urban area, in relation to their population. 98/ we will examine the metropolitan area of Santiago.—' Here, In the following Table 2.3.1, all the data regarding the investigation of the relationship between the per capita income by region and the urbanization of the metropolitan area of Santiago are included. The statistical data, as regards the internal migrants from each province to Santiago, were procured on the basis of the sample of the census in 1970. The correlation coefficient of the two variables, per capita income by region and population outflows of the provinces towards the area of Santiago, 29/ on the basis of the above data, was calculated to be equal to minus 0.48.— The size of the coefficient indicates that there is a correlation between the aforementioned variables. 28/ 29/ This correlation has the meaning of the influencing A similar ¿analysis, as regards the population movements from various geographical areas towards the larger urban centres of the country, can be made for the areas of Antofagasta, Valparaiso, etc., and it is believed that the same conclusions can be obtained as in the case of Santiago. The same analysis was made in Sweden. The correlation coefficient was equal to minus 0.67. This means that the main reason which caused the population outflows of the regions towards the capital of the country, Stockholm, was the differentiation of income between the capital and the rest of the geographical areas of the country. ) 19 ( Table 2.3.1 j* - * TEE POPULATION OUTFLOWS OF THE PROVINCES TOWARDS TEE METROPOLITAN A!SA OF SANTIAGO DURING THE PERIOD 1966-1970 AITO THE PER CAPTTA TNCQfH RV REGION Usi 1970 Population in 1970 (More than 5 years) Population movements towards Santiago (More than 5 years) Proportion of population movements to the total population of region Per capita incane n/n Region 1. Tarapaca 148 600 5 940 4.0 3 064 2 o Antofagasta 212 600 6 540 3.1 5 983 3. Atacama and Coquimbo 408 900 14 120 3.5 2 413 4. Aconcagua and Valparaiso 786 080 30 060 3.8 2 391 5«. O'Higgins and Colchagua 402 100 27 780 6.9 2 247 510 520 30 480 6.0 1.560 1 180 020 48 980 4.2 1 697 6. Caricòe Talea, Maule and Linares 7. Suble, Concepcicn, Arauco, Bio-Bio and Malleco 8. Cautin 350 020 21 740 6.2 1 092 9. Valdivia and Osorno 356 900 17 840 5.1 1 614 10. Llanquihue, Chiloe and Aysen 294 500 7 100 2.4 1 532 11. Magallanes 78 520 2 560 3.3 4 758 4.8 2 437 Abroad Total Sources; 15 260 4 728 760 228 400 Centra Latinoamericano de Demografia (CEIADE). Banco de Datos. Muestra del 5 por ciento. Censo de Chile, 1970. of one variable by the other. Furthermore, the correlation of the two variables is negative, in the sense that the influence of the per capita income on the outflow of emigrants, from each geographical region, to Santiago, is inverse, that is, in other words, the smaller per capita income of the provinces causes an increase of the population movements towards the larger urban areas, as for . ) 19 ( instance in the case of Santiago. This ascertainment is in agreement with the fundamental hypothesis of the influence of the level of per capita income on urbanization. In fact, as we observe from the data of Table 2.3.1., the largest outflow of internal migrants is realized in those geographical regions which have the smaller per capita income, and vice versa. Thus, the province of Cautín, with the smallest per capita income (1 092 Escudos) indicates a proportion of emigrants' outflow in relation to their population, amounting to 6.2 per cent, which can be considered high. Similarly, the provinces of Corico, Talca, Maule and Linares, considered as being one geographical region, have a high degree of population outflow (6.0 per cent) towards the metropolitan area of Santiago, while the per capita income of these four provinces was 1 560 Escudos on the average, in the year 1970. Hie same can be said for the two groups of provinces, 0'Biggins and Colchagua, as also Valdivia and Osorno, whose percentage of internal migrants was 6.9 per cent and 5.1 per cent respectively, while their per capita income is at a low level. On the contrary, the provinces of Antofagasta and Magallanes, which have the highest per capita income among the geographical areas of the country (E°6 000 and E°4 800 respectively), experienced a small population movement towards the capital of the country, which was 3.1 per cent and 3.3 per cent respectively in the sane census year (1970), Following the above, we are led to the conclusion that among the main reasons which cause the movement of the population from various geographical areas -provinces- to the metropolitan area of Santiago is the income of each province. In other words, the low income of the residents of certain regions of the country is the reason for migration of part of the population of these regions to the capital of Santiago. 2.4. Factors influencing the population movements towards the metropolitan area of Santiago. Their statistical estimates During the analysis of the development of urbanization in Chile, Chapter III, we ascertained that the greater part of the internal population movement goes to the metropolitan area of Santiago. In the previous section, we were led to the conclusion that there exists a correlation between the per capita income by region and the increase of the urban population of this area. Fur- thermore, we saw that the index of urbanization of the area of Santiago is at high level and the demand for labour force continuously increases. Thus, we ) 19 ( see that among the factors which caused the population concentration in the capital area of Santiago are also the economic factors and, consequently, the estimation of their influence on the urbanization of the population of the area is of great interest to the present work. Based on the general hypothesis, as regards the factors influencing the urbanization of the population and what we said previously, we will try to estimate here the influence of the two factors, the per capita income of the region of Santiago, and the demand for labour force in the secondary branch of the economy, D , of this area, on the increase of the urban population of the metropolitan area of Santiago during the last decade 1960-70. The best fitting regressions which are included in table 2.4.1, refer to the aforementioned determinative factors. Moreover, in this table, the criteria ot the statistical reliability of the equations, i.e., the coef- 2 fici~nt of determination, R', standard error, etc., are included. Etcm the analysis of the obtained statistical results of the above equations we see that; i) Hie sign of the regression coefficients corresponds to the economic 2 theory, and the coefficient of determination, R„ is statistically significant in all equation, ii) iii) The standard error of estimate is statistically significant at a level of 5 per cent or less. The results as regards the criterion "F of Snedecor" for the 2 testing of the coefficient, R and tie von Nesanan's criterion fcr the existing of autocorrelation can be considered satisfactory. Therefore, the aforementioned regressions can be considered the appropriate ones for forecasting, on the basis cf the said criteria, and they are marked by *. Furthermore, regarding the average and point elasticities, with respect to the explanatory variables of the aforementioned equations, which are included in Table 2.4.2., we see that; the sign of the elasticities of the per capita income and of the demand for labour force is positive, as it is indicated by the economic theory, on the basis of which the hypothesis was formulated. Regard- ing the amount of the elasticities, the elasticity of the per capita income is ) 68 ( Table 2.4.1 THE BEST FITTING REGRESSIONS Percentage Model n/n of N r t 3 variance explained by regression Estimate of paramoters . — x 2 X b Nrt+s-a N ™ . - 5 9 1 . 7 2 • 1.117 Y^/H™ L N; + s - b DWt+s + 1 b. a17g ( R Von Newman K2 ) L 219.67 • 9.921 0 W+S (0.882) i2)0.810 96.6 Yr*$/CS (1) 90.0 - • The l i n e a r form 1 tm a 933 0.1799 90.0 0.811 0.1495 96.0 0.926 CU799 (3) a 1495 The logarithmic form 1 * Log H ™ - Log a+b Log t 2 * Log < + s t « Log a*b Log D! 4 S t 1 t Log N^ +s - -4791016 • 1.21681 Log y T + S / C + s i t (CL17844) * Log N^ +s - 1.25653 • 0*9069 Log D? +s X * (0.0852) Table 2 A 2 ELASTICITIES WITH RESPECT TO THE EXPLANATORY VARIABLESiOF THE EQUATIONS TRIED " o r of equation ^ 2 , kl Variable Type of _ . . equation Average -elasticities. Per capita income of the area of Santiago Li near Logarithmic Demand for labour force of the area of Santiago Li near Logarithmic u 1960 Point e l a s t i c i t i e s 1 962 1964 1966 1968 1970 1.23 1.22 1,25 1,28 1,12 0.92 0.91 0,96 1,25 1.30 1,12 * 0,89 0.89 0.94 0.92 0.90 ) 19 ( more than the one, a fact that means that, if the income rises by 10 per cent, the urbanization of the population of the area of Santiago will increase by 10.23 per cent. This elasticity is not constant in the whole period under study (1960-70). The higher value is in the year 1966 (1.30), and after this period of time it follows a downwards trend. As regards the elasticity of demand for labour force, it is 0.92 and its value changes during the period 1960-70. From the point of view of the sign and the amount of the elasticities, the factors, per capita income and demand for labour force, can be considered the most important economic factors influencing the urbanization of the population of tfo capital of Santiago, during the period 1960-70. Finally for the development of the urbanization of the population of the metropolitan area of Santiago in the projected period, 1971-75, it is necessary to adopts i) Hie same basic assumptions as those mentioned in this chapter, Section 1.5.1., as regards the determinative factors of the urbanisation of population, the appearance new ones, etc. ii) Certain assumptions, as regards the expected development of two main economic factors of the increase of the population of the metropolitan area of Santiago, in the year 1975. These assumptions are included in the following table. Table 2.4.3 ASSUMPTIONS AND THE DEVELOPMENT OF THE EXPLANATORY VARIABLES IN THE YEAR 1975 n/n Variable 1 Yr+S/Nr+S Rate of growth 1970-75 Assumptions Es timations 1975 3 455 in Escudos 3.5 3 627 in Escudos Optimistic (C) 4.5 3 806 in Escudos Pessimistic (A) 3.0 320.9 in thousand persons Conservative (B) 4.0 336.8 in thousand persons Optimistic Note; 2.5 Conservative (B) 2 (A) 4.5 345.0 in thousand persons Pessimistic (C) The pessimistic assumptions of the aforementioned variables correspond approximately to their rate of growth during the past period, 1960-70. ) 19 ( In Table 2.4.4., we give the development of the urban population of the metropolitan area of Santiago based on the equations tried, as well as the population projections of this area made by the Latin American Demographic Centre (CELflDE), by commenting on the results obtained. Thus, the deviations as regards the forecasting of the development of the urban population of the area of Santiago in the year 1975, made by the models and by projections of CELADE, are not significant and, consequently, the results obtained by application of economic models for the estimation of the increase of the urban population of this area, can be considered satisfactory. This conclusion is in agreement with the general conclusion of Chapter IV, Section 1.5.2., as regards the possibility of consideration, apart from the demographic factors, of economic factors also for forecasts of the urban population of a country, or an area of the country Table 2.4.4 THE DEVELOPMENT OP TEE POPULATION OP THE METROPOLITAN AREA OF SANTIAGO B1 THE YEAR 1975 Number and type of equations Population of the area of Santiago Assumption By equations h Y a/ CELADE - Deviations Absolute Per cent 1st. linear equation C 3 660 3 831 171 4.5 1st.logarithmic equation C 3 679 3 831 152 4.0 2nd.linear equation C 3 642 3 831 189 4.9 a/ urban population projections by CELADE are based on the manual "Perspectivas de Crecimiento de la Población Chilena 1970-1985". Publicación de la Corporación de Fomento de la Producción 'íí°10 - A/70. Santiago de Chile, 1970. In the diagrams 5 - 6 , we give the graphical presentation of the linear equations tried, from the point of view of theoretical and observed values of the variable under study, their deviations and the influence of the determinative factors of the increase of urban population of the metropolitan area of Santiago. ) 71 ( iDiagrara 5: 1 s t ; L I HEAR EQUATION. TESTED: "OBSERVED Af.'D ESTIMATED VALUES OF URBANIZATION OF THE POPULATION OF THE METROPOLITAN AREA OF SANTIAGO, THEIR DEVIATIONS IN PERCENTAGE AND THE INFLUENCE OF THE EXPLANATORY VARIABLES OF THE EQUATION Par capita urban fncocie (Percentage) ) 72 ( Diagram 6: 2nd. LINEAR EQUATION TESTED: OBSERVED AND ESTIMATED VALUES .OF URBANIZATION OF THE POPULATION OF THE METROPOLITAN AREA OF SANTIAGO, THEIR DEVIATIONS IN PERCENTAGE AND THE INFLUENCE OF TIE EXPLANATORY VARIABLES OF THE EQUATION Demand f o r „ labour force x 10 280t 1960 Urbanization x 10 (Percentage) 65 3 70 Years ) 19 ( 2.5» Regional differences and the theoretical consideration of the problem In the theoretical consideration of the problem of urbanization and industrial development, Chapter II, Section 1, we said that the movement of the inhabitants of the rural areas to the urban centres was caused by the technological revolution, which occurred in agriculture, the result of which was the unemployment of the farmers, etc. Furthermore, the rapid development of the industrial sector of the economy, the Increase in job opportunities in the urban centres, the higher wages, etc., were the incentives of attraction of the agrolabour force to the urban areas. Of course, to the aforementioned reasons there should be also added, the development of trade, and the cultural, social and political reasons which contributed to the population movements, without of course, overlooking the better living conditions, and the recreation etc., prevailing in the rapidly developing urban centres. Finally, we were led to the conclusion that the urbanization of the population of the country can be attributed to the application of purely economic laws. In the analysis of the regional problem in Chile we saw that the regional differences refer to the per capita income, the differentiation of industrialization, unemployment, etc. Moreover, we were led to the conclusion, during the study of the problem of concentration of the population in the metropolitan area of Santiago, Section 2.3 of the present chapter, that the higher per capita income of the area was the main reason of the population movement towards the aforementioned urban area. Based on the above, we can say that the basic thoughts, as regards the whole problem under study, are in agreement with what we ascertained during the analysis of the regional problem in the case of Chile. All the conclusions as regards the creation of a regional problem of the country, verify the basic points of theoretical consideration of the problem. From this point of view, and beyond the statistical verification of the hypotheses which were adopted on the basis of the theoretical consideration of the problem in the case of Chile, both for the total urban population and the urban population of the capital area of Santiago, we can say that both theory and practice are in parallel in the case of the regional problem of the country. ) 19 ( V. RELATIONSHIP OF URBANIZATION AND INDUSTRIALIZATION IN. LATUM MERICÄN COUNTRIES. APPLICATION OF THE SYIMETH? MODEL 1. Testing of urbanization of the Chilean population and industrialization at a national and regional level In Chapter III» we undertook the analysis of the development of urbanization of the population of the country during the sanpie period 1952-70, analysis extends to the regional level. This The investigation, in this manner, of the mobility of the population of the country towards the urban centres, will facilitate us in the application of the symmetry model by province. Thus, on the basis of the data of the table 3,2.2. of Chapter III, Section 3.2., concerning the degree of urbanization of the population by province, and the data as regards the economically active population and, specifically, the population employed 30/ in the industrial sartor,— we formulate the indices of "urbanization" and "indnstrialization" at certain time-periods and specifically in the 1952, 1960 and 1970 census years. Furthermore, based on these indices we construct the symmetry model which, as we mentioned in the relevant chapter, consists of the ratio of these two indices, expressed in percentage. In the following table, the urbanization and industrialization indices are given at a national and regional level, Z. ., I.., as also the ratio of their relationship, in perij ij ,, , centage, which consists of the syirmetry coefficient—• -model- S... Fran Table 1.1 the following can be. seen. The numerical limits of the symmetry coefficient at the national level are less than half of 100, during the census years of the period 1952-70. This tells us, on the basis of what we have accepted as regards the interpretation of the syrrmetry model in Chapter II, Section 3.2, that the country, from this point of view, i.e. the development of the urbanization of its population aid industrialization, belongs to the less-developed countries. More specifically, the estimation of the symmetry coefficient at the level of 21.2 per cent during the last 30/ 31/ See Table 3 of Appendix I. The grouping of certain provinces was imposed by the fact of the lack of statistical data, as regards the labour force according to the branch of economic activity, for the census year 1970., This, however, does not alter the iiiporfcance of the conclusions drawn, a s regards the application of the symmetry model at a regional level. Table 1.1 THE SYMMETRY COEFFICIENT OF URBANIZATION OF THE POPULATION OF CHILE AND INDUSTRIALIZATION AT A REGIONAL LEVEL, CURING THE CENSUS YEARS OF "HE SAflPLE PERIOD 1952-70 n/n Coun-ry and fcy province Index of urbanization I.. 13 1. Tarapacá 2. Antofagasta 3. Atacama and Coquirabo 4. Aconcagua and Valparaíso 5» Santiago 6, O'Higgins and Colchagua 7. Curicó, Talca, Maule and Li nares 8. Nuble, Concepción, Arauco, Bio-Bio and Malleco % Cautín ia Valdivia and Osorno i i . llanquihue Chiloé and Aysón 12o Magallanes Chile Sourc9s: fr) 1970 1960 1952 Index of Industriali z a n on !.. ij Index of urbani z a t i on 7 index of industrialization i Index of Industrialization Index of urbanization \ Symmetry Coefficient S i 1952. 1960 1970 14.8 ia9 6.7 14.3 22.88.2 17.6 11.1 24.1 27.3 33.7 28.6 16,3 11.9 16.6 23.5 29.8 10o9 16.1 11.3 'ij 2a 7 29.2 10.0 87.1 94.8 57.8 82.6 9a 0 .45.5 14.2 11.3 9.6 19.4 26.8 8.6 92.1 96.6 66.9 85.9 93.3 49,1 35,8 10.8 40.4 9.6 46.5 a,i 30.2 23.8 19.6 51.1 15.8 57.8 15.6 65.7 15.4 30.9 27.0 23.4 33.1 38.2 11.9 17.6 38.7 44.7 9.9 13.3 48.9 52.9 8.1 11.3 36.0 46.1 25.6 29.8 16.6 21.4 28.1 10,6 36.9 ia6 47.4 7.4 37c 7 28.7 15.6 81.4 15.7 83.0 11.9 86,3 10.4 19.3 14.3 12.1 59.7 89.3 42.3 75.9 86.7 35.0 ia5 9.9 10.2 : iao 16.6 24o 4 16.7 60,2 1ÍU) 68.2 18.0 75*1 15.9 31.6 26.4 21„2 Dirección de Estadística y Censos. XII Censo General de PoblaclófTy I de Vivienda Í952, Tomo Resumen del País, Santiago de Chile, 1959. " i i ) Dirección de Estadística y Censos. X I I I Censo de Población 1960, Tomo Resumen del País. Santiago ds Chile, 1969. i i i ) I n s t i t u t o Nacional de Estadística (Ex-UirecciCn 3e t s t a d f s t í c a y Censos). XIV Censo Nacional de Población y I I I de de Vivienda 1970. Muestra de Adelanto de Cifras Censales, Santiago de Chile, iy/'¿* I n s t i t u t o Nacional de Estadística, Entidades de Población 1970 (Tarapacá a Magallanes). Santiago de Chile, 1972. ) 19 ( census years (1970), based on the Indices of "urbanization" and "industrialization", means a delay in the secondary production and a asymmetrical enlargement In the urbanization of the population. Furthermore, the small development of the secondary branch of the economy, in conjunction with the large increase of the urban population, means an enlargement in the rest of the branches of the economy and, specifically, the sector of "services"i.e. the tertiary sector of the economy. Indeed, the tertiary sector of the econoity, from the point of view of persons enployed, increased by 56.9 per cent during the period 1952-70, while the secondary sector of the economy, from the same point of view, was increased 32/ by 1.6 per cent during the same as above period (1952-70).—• Finally, the symmetry coefficient in accordance with the urbanization index, shows that the urbanization of the population of the country realized, which during the year 1970 reached 73.1 per cent, was not only the result of the increase of employment of the industrial sector of the economy, but it was attributable to the other reasons mainly social and those of unemployment in the primary production, the lower per capita income of rural areas, etc. As a conclusion, we can say that urbanization of the population in Chile does not correspond to industrial development, without of course, overlooking that a part of the urbanization realized is attributable to the industrialization of the country« The inflow of internal migrants to the urban centres, which in the case of Chile precedes the industrial development of the country, apart frcm the enlargement of the tertiary sector of the economy, covers unemployment, the so-called "invisible unemployment" of the population of the urban areas. This type of unemployment causes unfavourable consequences on the social and economic growth of the country. Finally, worthy of particular note, is the fact that it comes from the above table, as regards the development of the symmetry, coefficient, that it followed a downwards trend during the sample period 1952-70. Thus, in the year 1952 it was 31.6 per cent and it fell to 26.4 per cent and to 21.2 per cent during the years 1960 and 1970 respectively. 327 Chart 2 shows the development of the labour force by branch of economy, from the point of view of persons employed, during the period 1952-70. The above chart clearly shows the enlargement of the "services" branch of the economy. ) 19 ( Chart 2 THE'DEVELOPMENT'OF THE LABOUR FORCE IN CHILE BY BRANCH OF ECONOMY iW THE SAMPLE PERIOD 1952 - 1970 (Percentage) Labour force x 10 3 £ 2 500 Services 2 000 60.0 50.5 46.2 1 500 4.7 JJL—L. 1 000 19.0 -—-Mine Industry 18.0 15.9 500 Agriculture )l . 27.7 1952 Sources: ! 21.2 G O ! 70 Years Dirección de Estadística y Censos, XII Censo General de población y | de Vivienda 1952. Santiago de Chile, 1959 Dirección de Estadística y Censos, XI | | Censo de Población 1960. Santiago de,Chile, 1969. I n s t i t u t o Nacional de Estadísticas, XIV Censo de Población y I I I de Vivienda. Maestra de Adelanto de Cifras Censales. Santiago de~Chile71972. ) 19 ( This fact can be attributed, on the one hand to more rapid increase of the labour fores than the corresponding increase of employment opportunities in the secondary branch of the econorry and, on the other hand, to the decrease of enployment in the aforementioned branch of the economy. Indeed, it was previously said that secondary production increased only by 1.6 per cent during the period 1952-70, while during the semi-period 196C-70 it was reduced, from the point of view of the number of employed, from 428.9 thousand persons in 1960, to 415.4 33/ thousand persons in 1970, that is by 3.2 per cent.— However, during the same semi-period (1960-70) the percentage of the urban population of the country, frem 62.2 per cent in 1960 amounted to 75.1 per cent in the year 1970. These ascertainments verify the conclusion drawn previously, concerning the asymmetrical development of the urbanization and industrialization of the country. From the application of the symmetry model by region, the following were ascertained. The numerical, limit of the symmetry coefficient, in all the prov- inces of the country, is less than half of 100, as same occurs at a national level. Therefore, the ascertainment at a national level as regards the asym- metrical evolution of urbanization and industrial development is also extended to a regional level. In other words, the general conclusion, in the case of the regional level can be interpreted as follows: that the urbanization of the provinces of Chile precedes their industrial development. In fact, in the province of Santiago, capital of the country, which shows the largest degree of urbanization of the populationr which during 1970 was 93.3 per cent, the index of industrialisation of this province during the same year (1970) was at the level of 22.8 per cent, that is four times smaller than the urbanization index, the same can be said for the provinces of Aconcagua and Valparaiso, where the index of urbanization during the last census year (1970) was 85.9 on the average, while the proportion of those employed in the secondary branch of the economy, in relation to the labour force of these provinces, was 14.3 per cent on the average. The region of Santiago, of course, shows the largest numerical limit of the symmetry coefficient (24.4 per cent) in comparison to 337 The sector of the economy, mines, etc. from the point of view of labour force, showed also a drop during the period 1952-70. Thus, the number of enployed in this sector of the economy decreased from 101.4 thousand persons in 1952 to 75.3 thousand persons in 1970 (25.7 per cent). )19( the other provinces of the country. This is because the majority of the indus- tries are established in the capital of the country. After the province of Santiago, from the point of view of development of urbanization and industrialization, there follows the group of provinces of f3uble, Concepciôn, Arauoo, Bio-Bio and Malleco, whose symmetry coefficient in the year 1970 was 2 3 . 4 per cent on the average. There foliar/ the provinces of Valdivia and Osorno, with a symmetry coefficient of 21.4 per cent, while in the remaining eighteen provinces, the coefficient is at low level. Finally, the symmetry coefficient at a regional level, during the three census years of the period 1952-70, followed the same trend as at a national level, that is, a downwards trend for all the provinces. Furthermore, we can say that the same factors, as at a national level, caused the downwards trend of the symmetry coefficient in the ssme period (1952-70). Of course, the numerical value of the coefficient varies in each province and this is attributable to the differentiation of the urbanization and industrialization indices. We do not go into more detailed analysis of the results obtained from the application of the symmetry model by region, because both the conclusions drawn and also their causes* show a similarity with those for the country. We can, however, in the general conclusions, add the following as regards the urbanization of the population by province? that the provinces which have the „ largest numerical value of the symmetry coefficient, as for instance Santiago, etc. ceteris paribus, will be poles of attraction for the inhabitants of the rural regions, who desire to migrate to the urban areas of the country. * 2. The development of urbanization and industrialization frcm the symmetrical point of view in the remaining Latin American countries at the national level. Their classification based on the model applied The application of the symmetry model in latin American countries allows us, in addition to the conclusions drawn for each country, to classify them .into two main groups. This classification is based on the estimated numerical value of the symmetry coefficient, s^. In the first group are included those countries where, during the last decade (1960-70) the development of the urbanization of their population and industrialization, from the point of view of the number of employed in the secondary brandi of the economy, were not symmetrical. In other words, the degree of urbanization of these countries )19( was high during the past decade (1960-70), while the industrialization index did not experience any fundamental change during the census years 1960 and 1970. In the second grotip, the rest of the Latin American countries are included, whose characteristic is the fact that both the urbanization of their population as also industrial development, showed a normal development during the period under examination, 1960-70. Of the eighteen countries of Latin America where the symmetry model was applied, eleven belong to the first group while the remaining seven belong to that group in which countries show symmetry as regards the development of urbanization and industrial development during the period surveyed (1960-70). In the Table 2.1 and in the Chart 2.1 we give the aforementioned classification of the countries, with their indices of urbanization and industrialization.—/ The countries of the second group are in a considerably advantageous position compared to the countries of the first group, as to the development of urbanization and industrialisation, because the development of these two indices shows a certain symmetry, that is, urbanization does not precede industrial development. In the countries of the first group, we observe a large increase of tlis urban population, while industrial development follows a small upwards trend. This means that the development of the two magnitudes of urban- ization and industrializations is asymmetrical. Furthermore, in certain coun- tries, of both groups, we observe that the index of industrialization remained at the same level in the years 1960 and 1970. This, of course, does not mean that the labour force of the country, or the number of those employed in industry, remained unchanged between the two censuses. On the contrary, as same appears from the statistical data of the Table 3 of Appendix 1, in these countries, with an unchanging industrialization index, we have an increase in the labour force of the country in general and an increase in the number of employed in the secondary production, on account of the increase of the population of the countries and the industrial development taking place. 34/ Bolivia and Haiti are not included in this table, because of the lack of adequate statistical data as regards the labour force in the secondary branch cf the economy. The same can be said for the countries of Uruguay and Cuba, in the year 1970. Table 2.1 INDICES OF URBANIZATION AND INDUSTRIALIZATION, AS ALSO THEIR RATIO, IN VARIOUS LATIN AMERICAN COUNTRIES DURING THE CENSUS YEARS OF THE PERIOD 1960-1970 Country n/n census Year Index of urbanization z 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 1. 2. 3. 4. 5. 6. 7. Argentina Chile Colombia Ecuador Honduras Mexico Panarra Dominican Republic Venezuela Uruguay Cuba Brazil Costa Rica El Salvador Guatemala Nicaragua Paraguay Peru Sourcess i Index Symmetry of Coefficient industrialization S. I. i Census Year i Index of urbanization Z. i Index Synsnetry of Coefficient industrialization S. X J i A Group 33.9 1970 1970 26.8 24.2 1970 1970 40.6 1970 33.2 1970 27.0 16.8 1970 78.9 75.1 57.7 45.7 27.7 62.3 50.2 25.0 15.9 12.8 14.6 7.7 13.7 7.5 31.7 21.2 22.2 31.9 27.8 22.0 14.9 1960 1960 1964 1962 1961 1960 1960 73.7 68,2 52.8 36.0 23.2 50.7 44.0 25.0 18.0 12.8 14.6 7.7 13.7 7.4 1960 30.3 8.1 26.7 1970 36.9 8.1 22.0 1961 1963 1953 67.4 32.2 57.0 12.2 21.2 16.6 18.1 25.8 29.1 1970 1960 1963 1961 1964 1960 1962 1961 46.3 . 34.5 38.5 34.0 40.0 35.8 47.4 8.6 11.6 12.8 11.5 11.7 15.0 13.2 B Group 18.6 1970 1970 33.6 33.2 1970 33.8 1970 1970 29.3 41.9 1970 1970 27.8 71.9 12.2 17.0 - - - - - -- - - 47.6 33.6 37.9 30.8 40.0 36.0 49.2 8.6 11.6 12.8 11.5 11.7 15.0 13.2 18.1 34.5 33.8 37.3 29.3 41.7 26.8 Programa de Intercambio ELAS/CELADE. Juan Carlos Lerda, Adolfo Aldunate. Distribución de la Población Económicamente activa en los países de América Latina, 1940-1960. Centro Latinoamericano de Demografía, Boletín Demográficos N°2 y 3. Proyecciones Población Urbana 1970. ) 82 ( Chart 2 1 . THE DEGREE OF INDUSTRIALIZATION AND URBANIZATION OF THE LATIN AMERICAN COUNTRIES IN THE YEAR 1970. (Percentage) 13,7 fs '.n^.-N. V\ v £ 62J 81 . / \ 1.' 16. U. s. A. \ !„ r ' V \ 115 / J*-/ / t"' ' „ " / " ' % Q '"¿Î-a- . . „ ,,. " • / "36.9 ominican Republic ~ < P , t o . Rico, ' V' fi < 1 «1 T / I« ' — V r " 30.8 Honduras ^ -'/Haiti ! f 71.9" , / Z y - Trinidad and Tobago / "T Guyana • • y I Venezuela w / Surinam V '-/french Guiana ¡Colombia ' s / 57.7 Ì Wi " "i /T^V. S y S Brazil Peri ( u •^1:11. 1 v „.. s < . Chile Bolivia It T5.9~Ti?b75.Ì " ^ ^ V ' i •V' Source: Based on the data of the Table 2.1 \ • V .O , .f --rfh? • l'-ct .. : . • 70.9 Note: /K • N p^tt——7 _,> Index of i n d u s t r i a l i z a t i o n : persons employed i n the industrial sector of the economy to the total labour force of the country 180° » 100 Index of urbanization: urban population to the total population of the country 180° » 100 ) 83 ( In general, we can say, for all the Latin American countries, on the basis of the numerical limits of the symmetry coefficient, S., which are less than half of 100, that the development of urbanization of their population and industrial development has not been symmetrical up to now. Based on the above general conclusion, it can be said that the countries of latin America, from this point of view of development of urbanization and industrialization, can be considered as being less-developed countries and probably the socio-economic growth, from this point of view would follow an upwards trend in the countries of the second group for the caning years. Of course, the larger numerical value of the symmetry coefficient in certain countries of the second group of Table 2.1, as for instance Paraguay and Guatemala -their coefficient is 41.9 per cent and 33.8 per cent respectively- means the better development of urbanization of their population and industrialization, compared with the other countries of the samp group, whose numerical value of the coefficient is at a lower level. How- ever, the higher rate of the symmetry coefficient of the countries, even though these countries belong to the second group, is an index and not a proof of their socio-economic growth. The criterion of the economic growth of the country, as we mentioned in the previous chapters, is the per capita income and, consequently, the symmetry coefficient in the case of its higher value of certain countries, must be related with the main dimension of economic development -the per capita income- for the determination of the socio-economic level of the countries. Indeed, in the aforementioned countries, Paraguay and Guatemala, the per capita income amounted to US$ 149 and US$ 258 in the year 1960 -Table 4.2 of the present chapter- vdiich is lower than the per capita income of certain countries which belong both to the second and the first group»—^ 35/ For instance, the countries of Venezuela and Uruguay on the one hand, belong to that group of countries -first group of Table 2.1- whose development of the urbanization of their population was not symmetrical in the period under study (1960-70) and, on the other hand, their symmetry coefficient is at a lower level, 18.1 per cent and 25.8 per cent respectively in the year 1960, while the per capita income of these countries can be considered to be at higher level among the Latin American countries (US$ 859 and US$ 593 respectively in the same year, 1960). ) 84 ( Finally, frem Table 2.1, it appears that there ,is deviation in the symmetry coefficient, S^, among the various countries of Latin America. The deviation between the two extreme numerical values of the coefficient, 41.9 per cent (Paraguay) and 16.8 per cent (Panama) is 25.1 per cent in the year 1960. This deviation can be considered significant and shews the "gap" that there is in Latin American countries from this point of view of development of urbanization and industrialization. In the Chart 2.2 we present graphically the position of each country, from the point of view of the symmetry coefficient, as also the direction of the development of this coefficient, in the period 1960-70, based on the clas- sification of the countries in two groups and the data of Table 2.1. 3. Reliability of the model The reliability of the symmetry model is tested by the symmetry of the existing population distribution of each country. By the symmetry of popula36/ tion distribution, we mean that the development of the metropolitan areas— must be in proportion to the development of the cities of small population size, of each country. Therefore, as a criterion of the normal population distrib- ution we take the population size of each city, as also the existing relationship between cities of different population size. More specifically, for formulation of the aforementioned demographic criterion, we classify the cities of each country in accordance with their population size, as follows? a) b) Cities of c) Cities of 100 000 - 500 000 inhabitants d) 36/ Cities of less than 20 000 inhabitants Cities of more than 500 000 inhabitants 20 000 - 100 000 inhabitants The concept of. a metropolitan area, according to an international classification, is defined as follows s "An area with 100 000 or more inhabitants, containing at least one city with 50 000 or more inhabitants and those administrative divisions contiguous to the city that meet certain requirements as to metropolitan character". (The World's Metropolitan Areas. Edited by International Urban Research, University of California, Berkeley, U.S.A.). Chart 2.2 THE SYMMETRY COEFFICIENT OF URBANIZATION AND INDUSTRIAL DEVELOPMENT IN LATIN AMERICAN COUNTRIES, IN THE DECADE 1960 - 1970. Symmetry coefficient (Pfircentago) Symmetry coefficient (Percentage) 45 45 35 Argelina " 35 r Honduras El Salvador \ Costalea ^ U a W 3 Cbil^ ltxc?7 'éio* Peri Nicaragua Uruguay ' C \ Colombia 55 hR. Dominicana 25 20 20 Panamá 15 B: GROUP „Paraguay 40 \ 30 A: GROUP Ecuador 40 A c \ Venezuela : ^Brasil Sa 15 10 —i 1960 65 1 > 70 Years 10 1950 65 70 Years Source: Programa de Intercambio ELAS/GELADE. Larda, Juan Carlos, Al duna t e , Adolfo. D i s t r i b u c i ó n de l a población o c o n f i m c ^ n t a activa en l o s países de K America L a t i n a . 1940-1960. Note: 1) The a r r o w s ^ V i n d i c a t e the development of the symmetry c o e f f i c i e n t during the decade 1960-1970 2) For the c l a s s i f i c a t i o n of the countries in A and B group, see table 2.1 of the present section. ) 86 ( Based on the above classification of the cities, the following indices, as regards the symmetrical population distribution of each country, have been accepted: i) If the total sum of tne population of the two first groups of cities (a and b) is approximately equal to the population sum of the last two groups of cities (c and d), the country is characterized by a symmetrical population distribution. The countries, furthermore, which have this characteristic, that is, a symmetrical population distribution, are the developed countries, ii) The less-developed countries, are characterized by an asymmetric population distribution, in this case, it is. observed that, the sum of the population of the two last groups of cities (c and d) is essentially larger than that of the first two groups of cities (a and b). The aforementioned thoughts can be written in mathematical form as follows: a + b - c + d Symmetrical Population Distribution, Developed Countries c + d > a + b Asymmetrical Population Distribution. Less-developed Countries where: a : Population summation of cities less than 20 000 inhabitants b s Population summation of cities 20 000 - 100 000 inhabitants c : Population summation of cities 100 000 - 500 000 inhabitants d ? Population summation of cities more than 500 000 inhabitants In the case of Chile, the symmetry coefficient was estimated at 26.8 per cent in the year I960.—^ On the basis of this numerical value of the synmetry coefficient and what we have accepted as regards its interpretation, we were 37/ The lack of statistical data as regards the number of cities according to population size, of the Latin American countries in the year 1970, with the exception of Chile, for the classification of the cities on the basis of the adopted criterion of reliability of the model, imposes on us the use of the data of the 1960 census, for all Latin American countries, to apply the aforementioned criterion. This, of course, does not alter the importance of the conclusions drawn- ) 87 ( led to the conclusion that the realized urbanization of the population of the country was asymmetrical, in relation to the achieved industrialization of the country and, the country, based on this ascertainment, belongs to the lessdeveloped countries. The verification of the aforementioned result is achieved by the application of the criterion of reliability of the symmetry model which we described at the beginning of the present section. Thus, on the basis of the population data of I960, which are given in the Table 4.3.1., of Chapter III, Section 4.3, the number of cities with less than 100 000 inhabitants, amounted to 85 cities with a total population of 1 833 thousand inhabitants. The population of the cities of a population size of more than 100 000 inhabitants, amounted to 2 424 thousand persons during the same year (1960). Based on these data, it is seen that the population of the third and fourth population groups (c and J) is larger than the population sum of the first and second groups (a *nd b) by 590 thousand persons. The percentage increase in question is 32.2 38/ per cent.-Consequently, we can say, based on the inequality, c+d > a+b, that the population distribution in Chile is asymmetrical and the country belongs to 39/ the less-developed c o u n t r i e s , — a s a conclusion drawn also during the application of the syrrmetry model. Finally, the index of the population distribution of the country, which was estimated at 32.2 per cent, is in agreement with the calculated symmetry coefficient, 26.8 per cent, because the existing difference, 5.4 per cent can be considered very small. Therefore, the symmetry model becomes the most reliable statistical element for measuring the symmetrical development of urbanization of the population of the country and industrialization. 38? 39/ The application of the criterion of reliability of the model, on the population data of the 1970 census, for the case of Chile also, leads us to the same conclusions as above. Thus, the population sum of the last two groups of cities (c and d = 3 722 thousand parsons) is larger than the population sum of the first two groups of cities ( a +b = 2 049 thousand persons) by 673 thousand persons, that ir. by 38.8 per cent. This criterion is an indication as regards the determination of the countries in less-developed and in developed countries. In this case, the per capita income must be taken into account for the classification of a country as a less-developed or a developed country. ) 88 ( The classification of the cities by population size, and in accordance with the requirements of the criterion of the model was made possible for certain latin American countries, which are included in the following table. Table 3.1 THE NUMBER OF CITIES BY POPULATION SIZE AND THEIR POPULATION FOR CEKEAlN COUNTRIES OF LATIN AMERICA Number % Country Census Year of cities 20-100 Population Number of cities more than Population 100 000 thousand inhabitants inhabitants 1. Argentina 1960 51 2 039 334 12 9 512 665 2. Bolivia 1950 5 272 402 1 321 073 3. Brazil 1960 159 6 612 727 31 13 309 463 4, Colombia 1964 35 1 591 020 12 4 802 612 5. Costa Rica 1963 - 1 320 431 6. Cuba 1963 21 869 400 5 2 211 400 7. Chil© 1960 31 1 353 424 4 2 423 788 8. Ecuador 1962 11 385 058 2 865 550 9. El Salvador 1961 5 187 053 1 255 744 Guatemala 1964 3 92 269 1 572 937 11. Haiti 1950 1 24 423 1 134 117 12. Honduras 1961 2 83 495 1 134 075 13. i'fcxico 1960 87 3 612 414 17 8 496 492 14. Nicaragua 1963 4 118 371 1 234 580 15. Panama 1960 2 .. . 1 . 273 440 16. Paraguay 1962 - 1 309 660 17. Peru 1961 27 995 313 4 1 827 672 18. Dominican Republic 1960 6 198 720 1 369 980 19. Uruguay 1963 Ì4 456 232 1 1 159 085 20. Venezuela 1961 31 1 304 261 5 2 256 892 10. Source: 32 522 • - Centro Latinoamericano de Demografia (CELADE). N°9, Santiago de Chile, diciembre de 1971. Boletin Demografico ) 89 ( From Table 3.1 it appears that in all the Latin American covmtries, the population of the cities of the1 second group (c and d) is larger than the population of the first group (a and b). This means that we have an unequal dis- tribution of the population of the countries. Furthermore, the fact of the inequality of the population of the two groups indicates that the countries belong to the less-developed countries, because the relationship cfd>a+b exists, In the case of Guatemala, the population difference of the two groups, arfb and c+d, is 152 thousand persons, that is, 36.4 per cent. This result is in agreement with the index of the syrrmetry model, which is at the level of 34.0 per cent approximately. Finally, from the data of the above table, the reader can be easily led to the relevant conclusions as regards the population distribution of each country, and their classification at the level of economic growth, on the basis of the criterion of reliability, by which, as we said pre**\ously, we can verify the obtained results for the application of the symnetry model. 4. International comparison based on the symmetry model. Conclusions drawn In the present section, we provide the estimation of the symmetry coefficient of urbanization and industrialization, for certain countries which do not belong to the Southern hemisphere of America (Latin American countries). The symmetry coefficient was evaluated on the basis of the data of census years, or on the basis of the population projections in the period 1960-70. These countries are mainly European countries where the urbanization and industrial- ization indices differ in each country. In this manner, we wish to make com- parisons of the data, as regards the increase of the urban population and the industrial development in the European countries, with those concerning the countries of Latin America, and to classify the countries as developed or lessdeveloped countries, based on the numerical limit of the symmetry coefficient. From tte Table 4.1 in which are included sixteen countries with their indices of urbanization and industrialization, we are led to the following conclusions.^ 40/ The lack of statistical data as regards the economically active population by economic branch for the countries which are included in the table, in the year 1970, did not allow us the estimation of the industrialization index and, consequently, the syrrmetry coefficient of these countries, for the same year (1970). ) 90 ( Table 4.1 INDICES OF URBANIZATION AND. INDUSTRIALISATION, AS WELL AS THE SYMMETRY Î-DDEL FOR CERTAIN COUNTRIES, DURING THE PERIOD 1960-70 n/ Country i Census Year t Urban. vy ization index Z. Industriellr: ization index I. „ . Symmetry model i x S. i North America 1. Canada 1961 69.6 22.9 32.9 2. United States 1960 69.9 26.5 37.9 1965 68.1 24.2 35.5 Asia 3. Japan Europe 4. Austria 1951 50.0 28.6 57.2 5. Belgium 1961 66.4 34.6 52.1 6. Denmark 1965 45.5 29.4 64.6 7. France 1968 70.0 27.0 38.6 8. Germany, Federal 1969-^ 38.4 37.9 98.7 9. Greece 1961 43.3 13.4 30.9 10. Netherlands 1960 80.0 29.9 37.4 11. Norway 1960 32.1 25.5 79.4 12. Sweden 1965 77.4 31.9 41.2 13. Switzerland 1960 51.3 39.8 77.6 1966^ 77.9 . 34.7 44.5 14. United ¥ Kingdom^ Oceania 15. Australia 1966 83.3 27.0 32.4 16. New Zealand 1966 68.3 26.6 38.9 Sources; i) ii) a/ b/ United Nations, Demographic Yearbook, 1970. United Nations, Yearbook of Labour Statistics, 1970. International Labour Office, Geneva, Includes Scotland and Northern Ireland, Estimates. ) 91 ( In certain countries, the symmetry coefficient exceeds 75 per cent, or approaches the numerical limit of 100. These countries are characterized by a symmetrical development of urbanization of their population and their industrialization. In the case where, in certain countries, the symmetry coef- ficient approaches the limit of 100, this means, as we explained in the relevant chapter, that these countries require labour force to cover the needs mainly of the secondary branch of the economy, because of exhaustion of their domestic resources in manpower. These needs -as is known- are covered by the immigration of workers from other countries which have a surplus labour force. The coun- tries which experienced a symmetry coefficient of more than 75 per cent, in the period 1960-70 are West Germany, Switzerland and Norway. Indeed, in these countries we had an inflow of workers from other countries, mainly Mediterranean countries where industrialization was at a lower level than in the aforemen- ticr«d three countries and, consequently, they had a surplus of labour force, a problem of unemployment, etc. Of the sixteen countries surveyed, six coun- tries show a symmetry coefficient of more than half of 100, and ten countries have a coefficient of less than 50 par cent. In general we can say that the urbanization index of all the countries is at a high level, with the exception of W. Germany, Norway and Switzerland, where the development of the urbanization of their population and industrial development has been symmetrical in the period 1960-70. symmetry coefficient in these countries The numerical value of the approaches the optimum level and from this point of view, we can say that these countries will cover the needs in labour force by immigration. Regarding the Latin American countries, and based on the symmetry coefficient, Table 2.1 of the present chapter, we ascertain the following s Of the 41/ eighteen countries surveyed, in the countries of Ecuador and Paraguay,— the symmetry coefficient is more than 40 per cent, in the year 1960, and 1962 respectively, while there are five countries, Argentina, Honduras, Costa Rica and El Salvador where this coefficient is 33 per cent approximately. In the re- maining eleven countries, the symmetry coefficients are at a level lower than 30 per cent. As a .conclusion, on the basis of the symmetry coefficient which 41/ In these countries, Ecuador and Paraguay, the secondary sector of economy is mainly craftsmanship. ) 92 ( reflects only the development of trie urbanization in conjunction with industrial development» vie can say that all the countries of Latin America belong to the less-developed countries. • Of course, as v?e said in the previous sections, the normal development of these two dimensions, urbanisation and industrialization, is an index and hot proof, of the economic growth of the country in general. The symmetry coefficient must Lie related with the per capita income, which is the main dimension of tie degree of the economic growth of a country, for a determination of its social-economic level. TO prove this fact, in the following Table 4.2., t r give the per capita- income of certain countries and the degree ie of urbanization of their population and industrial development. Table 4.2 THE SH-WETRST «^EFFICIENT AND HIE PER CAPITA INCCM3 OF CERTAIN COUNTRIES DURIEG THE PERIOD 1960 - 1970 Symmetry i Census Year t Per capita income in market prices in US$ i Census Year t 1. United States 1960 37.9 1960 2 559 2. Japan 1965 35.5 1970 1 650 3. t J. Germany 1969 98.. 7. 1970 2 690 4. Switzerland 1960 77.6 1960 1 429 5. Korway 1960 79.4 1960 1 093 6. Argentina; I960 33,3 1960 529 7. Chile 1960 28.9 1960 245 3. Guatemala 1960 33.8 1960 258 9. léxico 1360 27.6 I960 315 10, Panama 1360 16.8 1960 349 11. Paraguay 1962 41.9 1960 149 12. Uruguay 1963 25.1 19SQ 593 13. Venezuela 1951 18.1 1960 059' n Country , /n Sources; i) ii) coefficient S Symmetry coefficient; J.C. Lerda, A. Aldanate, Programa de Intercambio ELAS/CELADE, Distribución de la población econiSmicaraante activa en los países de América latina» 194Q-19GQ. Per capita income in market prices in US? dollars s United nations,yearbook o.f qational Accounts Statistics 1971, Vol. III. ) 93 ( Frcm the above table, we see that In certain countries, as for instance Wo Germany, Switzerland, etc., the main dimension of economic growth, which is the per capita income, is in agreement with the numerical value of the symmetry coefficient. In other countries, and particularly, in Latin American countries, the value of the index of urbanization and industrialization differs frcm their per capita income, a fact that means that in many cases, the value of the symmetry model is not in agreement with the main indicator of the economic growth of the countries. 5. Population movements anticipated in Latin American countries and probable redistribution of their labour force In the previous section, we saw that in certain European countries, such as West Germany, Norway, etc., a scarcity of labour force was observed, particularly in the secondary branch of the econary, on account of exhaustion of the domestic resources in labour force. These countries economically developed, permitted the entry of workers frcm other European countries, to cover their needs in labour hands. The symmetry coefficient in these countries was form- ulated in numerical limits of more than 75 per cent. The "migratory" countries •were the less-developed countries, where the numerical value of the symmetry coefficient was at a lew level. On the basis of the results obtained from the application of the symmetry model, in Latin American countries, such situations as those which appeared in the European countries, that is a scarcity of labour force, in the industrial sector of the economy, and coverage of these needs by immigration of workers from other countries, are not expected to take place in the catling years.—^ Biis is because the increase of employment opportunities in the secondary production of the economy can be covered by domestic resources of manpower of each country. It was ascertained, in previous sections, that urbanization consid- erably precedes industrialization in all countries of Latin America and, consequently, there is a surplus of urban labour force which can cover the needs in 42/ International migration will probably take place between the latin American countries for covering services needs in the coming years, as this happens in Argentina today. ) 94 ( labour force created in the industrial sector of the economy. Furthermore, there is a surplus of rural labour force, which can be transferred to the industrial areas of the country. As a conclusion, on the basis of what we have mentioned above, we can say that in each Latin American country, under the present conditions, or a probable redistribution of the labour force by branch of economic activity on account of an increase of industrial development, the needs in labour force c r e a t e d in the secondary production, can be covered by the domestic resources in manpower for the coming years. Therefore, population movements (external migrations) between the countries of Latin America, on account of the created needs for labour force, and in general, of the economic growth of each country, are not anticipated for the coming years. manpower will remain unchanged. The existing balance of their Of course, a redistribution of the labour force, by branch of economic activity (agricultural, industrial, etc.) and by region, will take place in each Latin American country. This is because all the countries are at the stage of economic development and, as a consequence, the demands for labour force by branch of economy will be changing. The degree of redistribution, the period of time that a such redistribution of economically active population will take place, would depend on the rate of economic growth of each country. Such changes in the labour force of each country were noted in the past, 1952 - 1970, as one could ascertain from the investigation of the applied syrrmetry model and, particularly, from the development of the "urbanization" and "industrialization" indices, whose values changed during the census years of the sample period, 1952-70. ) 95 ( VI.; FINDINGS AND CONCLUSIONS I» A review of findings .and conclusions 1.1. From the analysis of development of the urbanization of the Chilean population we ascertained that the urban population increased by 36.4 per cent during the sample period 1952-70 and the degree of urbanization amounted to 75.1 per cent in 1970» which can be considered high. Furthermore, we ascer- tained that a redistribution took place of the urban and rural population in the country's provinces. Finally, we were led to the conclusion, as regards the population movements that they took place from rural areas and small population size-urban centres towards the large urban areas and particularly to the capital of the country. 1.2. From the statistical investigation of the influence of the industrial development on the urbanization of the population, we evaluated that the main daterninative factors of the increase of the urban population is the per capita income of the urban areas, the demand for labour force in the secondary branch of the economy, unemployment in agricultural production, and the wage of the industry. Based on these determinants, we made forescasts as regards the develop- ment of urbanization in the coming years and the results obtained were considered satisfactory. 1.3. As regards the regional problem in Chile, we ascertained that the main factors prevailing in it are the unequal distribution of income and the differentiation of industrial development by province. We were also led to the conclusion that there exists a correlation between the per capita income by province and the number of migrants towards the metropolitan area of Santiago. Furthermore, the main factors of increase of the population of Santiago, which is considered the largest urban centre of the country, is the higher level of per capita income and the demand for workers in the industry because, as we said in the relevant chapter, we have a concentration of many industrial units in this metropolitan area. Finally, we arrived at the co: elusion that an equilibrium in the economic development of the province could possibly reduce the migration flow towards the capital of the country, because as it was proved, the movement of population in that direction is attributable, mainly, to economic reasons and secondarily, to other reasons, such as social, cultural, etc. ) 96 ( 1.4. Frcm the application of the symmetry model, in.the case of Chile, and at a national level, the numerical value of this coefficient vas equal to 21.2 per cent in the year 1970. Following this, we were led to the conclusion that the development of the urbanization and industrialization of the country is asymmetrical, that is, in other words, that urbanization precedes industrial development of the country and from this point of view the country can be considered as belonging to the less-developed countries. Furthermore, this coefficient in the period 1952-70 followed a downwards trend. The asymmetrical development of the urbanization of the population and industrialization, is observed in all the geographical regions, even though the syrrmetry coefficient varies in each province. The area of Santiago has the highest limit of this index. This is natural because of the fact that the majority of industries are established in this area and, as a consequence, the industrialization index is at a higher leve^ than other areas. 1.5. From the estimation of the urbanization and industrialization indices, as also the syrrmetry coefficient for nineteen countries of Latin America, we were led to the following conclusions s i) in seven Latin American countries, urbanization and industrialization showed a symmetrical development, while in thé remaining twelve the evolution of these two dimensions -urbanization and industrialization- was asymmetrical, as same was said in the case of Chile at a national and regional level, ii) Hbe value of the syrrmetry coefficient in most of the Latin American countries is less than 50 per cent, a fact which shows that these countries are less-developed frcm the point of view of the development of the urbanization of their population and their industrialization. If the previously mentioned index is related with the main dimension of economic growth which, as we said, is the per capita income, then the fact is verified that these countries are economically less-developed as having a per capita income less than 1/4 of the per capita income of the United States. 1.6. Frcm the application of the reliability criterion of the symmetry model, we ascertained the relationship existing between cities of various population sizes, both as regards the case of Chile, as also in certain other countries of Latin America. On the basis of the results achieved, from the application of the above criterion, we were led to the verification of the ascertainment which emerged from the application of the syrrmetry model. Finally, we accepted ) 97 ( that the symmetry model is the most reliable statistical technique for measuring symmetrical or asymmetrical development of urbanization and industrialization» 1.7. The lack of labour force, on account of exhaustion of the domestic manpower of the country, and the covering of these needs by external migration, events which took place in certain European countries, are not foreseen to appear in the Latin American countries for the coming years and this because, on the one hand there is a surplus of rural population and, on the other hand, the increase in labour force is much larger than the rate of growth of industrial development, whose rate, in certain countries of Latin America can be considered at a low level. 1.8. The comparison of the results obtained from the application of the symmetry model in the countries of Latin America and the European countries, has led us to certain conclusions as regards the degree of urbanization and the industrial development, whose repetition is not considered advisable. 1.9. Finally, all the ascertainments as regards the urbanization, the influence of industrial development on the urbanization of the population, regional problems, the relationship of urbanization and industrialization, are in agreement with what has been said during the theoretical consideration of the problem under study. 2. Epilogue The econometric investigation of the problem of urbanization and industrial development for Latin American countries is the first effort to be made by the author of the present work, in the analysis of the general problem of urbanization and economic growth by constructing a model in which demographic and economic factors will be included. industrial development and The estimation of the influence of the urbanization, the correlation of the main factors prevailing in the regional problem and the degree of development of urbanization and industrialization, lead us to useful conclusions as regards the relationships existing between demographic and economic variables. Today, in all the developed countries, an effort is made to relate all the factors, economic, demographic and social, which contribute to the increase of the national income. Furthermore, the economic and social structure of a ) 98 ( country mainly depends on the structure of its manpower. It is known that the age-composition of the population influences the age-composition of the labour force, productivity Jepends on the labour force, its structure, educational qualification, etc. Therefore, the ..-„¿elation of all of these factors, in the frame of national accounts of the country will provide the data needed for allocation of the economic and human resources of a country for a rapid socioeconomic growth. Furthermore all the countries of Latin America are at the stage of economic development and prepare long-term economic plans and consequently, they have need of such data referring to the existing relationships between demographic and economic factors. Finally, from the results obtained of the present work, we can say that the first effort of connecting demographic and economic factors can be considered successful. APPENDIX A ) 101 ( A. List of variables used fear urbanization of the population In Chile 1. N s 2. N3 s 3 Rural population of a country (in thousands) s = Total population of a country (in thousands) Urban population of a country (in thousands) Gross national income at constant 1965 prices and in million Escudos (E°) 3. N-N 4. Y % 5» Y3 % Agricultural income at constant 1965 prices and in million Escudos (E°) 6. Y-Y a = Y2" % 7o Y*, W % Per capita urban income at constant 1965 prices and in Escudos (E°) 8. 0s ? Unenplqyment in the agricultural branch of the economy (in thousands) 9» D % Demand for labour force in secondary production (in thousands) 10. ^ s Wages in secondary branch of the economy (index: base year = 1962) 11. N** 8 s Urban population of the metropolitan area of Santiago (in thousands) 12. Y14^/!^3 s Per capita urban income of the area of Santiago at constant 1965 prices and in Escudos 13. vH*s D s Demand for labour force in the secondary branch of the econorty of the area of Santiago (in thousands) w B. Urban income at constant 1965 prices and in million Escudos (E°) List of variables used for the estimation of the symmetry model in Latin American countries 1. P .. s Total population of a country or of a region (in thousands) 2. P s Urban population of a country or of a region (in thousands) s Rural population of a country or of a region (in thousands) ij 3. P . ij 4. 2.. = 5. P 13 a i-i ^ ioo % o .. s ij 6. p Ps e ij Index of urbanization of a country or of a region (in percentage) Total economically active population of a country or of a region (in thousands) s Economically active population employed in secondary production of a country or of a region (in thousands) } 102 ( p ioo s e.y L ij 100 s Index of industrialization of a country or of a region (in percentage) Synraetry Model (in percentage) Zij as in relationships 4 and 7. i s refers to country j s refers to province of a country« • - ; . Table 1 • ' ' OATA OF.VARIABLES USED IN FUNCTIONAL RELATIONSHIPS FOR URBANIZATION OF.THE POPULATION IN CHILE, DURING THE SAMPLE PERIOD 1960-71 . n/ n (3) (2) (4) Y N»NasNr Na N ( i n thou- ( i n thou- ( i n thou- (1n m i l l i o n sands) sands ) Escudos) sands) Year t 1 2 3 4 5 6 7 8 9 10 11 12 (1) 1960 1961 1962 1963 1964 1965 1965 1967 1968 1969 1970 1971 7 7 7 7 7 8 8 8 8 8 8 9 Sources: ' 2* • ,3. ^ & f i ó l e s : , .1, • 2. . 3. 374 510 649 790 934 080 229 381 535 693 853 016 2 346 2 334 2 320 2 303 ' 2' 285 2 265 2 242 2 217 2 189 2 160 2 127 2 092 5 5 5 5 5 5 5 6 6 6 6 6 028 176 329 487 649 815 987 164 346 533 726 924 14 14 15 16 17 17 19 19 20 20 21 22 099 941 698 434 096 970 280 765 241 915 572 586 (6) (5) (7) Y r /N r Y-Y a =Y r Ya (in (in million (in million Escudos) Escudos) Escudos) 1 1 1 1 1 1 1 2 2 1 1 2 580 639 598 677 769 728 867 004 037 857 949 041 12 1Í 14 14 15 16 17 17 18 19 19 20 519 302 100 757 327 242 413 761 204 058 623 545 2 2 2 2 2 2 2 2 2 2 2 2 489 570 646 689 713 793 909 881 869 917 917 967 (8) (9) • (10) Ua . DW ( i n thou- ( i n thouIndex sands) (1962=100) sands) _ - _ - - - - - 9o8 10,0 10.5 10.9 9.5 10.2 10.4 - 650.6 663.9 663,2 67a 3 693.3 719.8 734«. 2 88.4 88.1 94.6 96» 8 95.8 ' 92.2 90,3 - (11) (12) (13) 0ws r s Y r + S /M r + S ( i n thou- ( i n ( i n thousands) Escudos) sands) 2 2 2 2 2 2 2 2 2 2 3 194 268 344 423 505 589 677 767 860 956 056 - 2 448' ' 2 541 2 675 . 2 762 2 801 2 934 3 107 3 085 3 014 3 043 3' 054 221.5 ^ 206.5 210.2 210.8 225.2 248.6 252.5 w 261.4 M 265.5 o 274.1 276.8 ~ • •»' 5flV Censo Nacional de Población y i l l de Vivienda, 1970, -Muestra de Adelanto da Cifras Censales» I n s t i t u t o Nacional de Estadística. Santiago de Chile, T$7T. • '.-•.. X I I I Censo, de Población, 196a I n s t i t u t o Nacional de Estadística, Santiago de Chile, 1969. El Empioo y e l Procoso do Oosarrollo on Chile 1960 - 1970.. (PREALC). V o l . í - V , Santiago de Chile 1973. ^ a n "ffijj "Economía Nado nal' 1Ó71 -76> Antecedentes sobre el Gasarrollo Chileno. Oficina de P l a n i f i c a c i ó n Nacional (OOEPLAN). Santiago oe Chile, 1971. Producto GeosrSfico Bruto 1960-70, por rogionos. OOEPLAN. Santiago de C h i l e , 1972 (rairaeo). ; For inierpretation..of the symbols of the variables used, see the corresponding l i s t of variables. Wages i n the secondary branch of the economy deflected by the index of consumption prices. The urban population i n 1970 d i f f e r s from the urban population i n the same year of Table 2.-2.1 „ e Chapter 111. But the lack of statistical data regarding the urban population by province requires the using of both two f i g u r e s . Tabla 2 ¡HILE: THE DISTRIBUTION OF URBAN AND RURAL POPULATION BY PROVINCE DURING THE CENSUS YEARS 195 2, 1960 AUD 1970 A 789 824 113 169 378 254 954 593 531 432 693 181 257 342 566 289 292 419 072 647 059 986 687 J _ 102 184 80 262 128 498 754 224 13 9 89 173 72 146 251 411 72 138 159 365 23 2 123 13 9 100 a 26 262 55 206 932 995 Distribution (percentage) Distribution (percentage) 61 165 41 103 51 424 521 90 36 33 68 26 44 87 313 20 383 005 441 230 032 799 831 827 501 332 515 486 270 941 297 699 59.72 89.28 51.73 44 61 120 86 49 45 18 11 44 146 647 923 157 773 193 096 677 921 37.27 39.45 36,69 3 a 27 34,99 76.12 28.63 31.92 38,67 33,12 37« 03 4 a 45' 32.28 17.97 44.46 81.37 573 122 6a 22 33.38 39.75 85.26 86,72 4 a 44 26.16 41 19 38 158 77 73 233 133 103 56 105 45 101 163 98 51 94 97 244 146 73 94 2 14 10 2 359 873 3! ;a y Cei 5 ( I n s t i t u t o Nacional de E s t a d í s t i c a ) Tomo: 40,28 10.72 48.27 ro j a ) 62.97 59,55 67.72 82.03 55.54 18.63 123 215 116 308 140 617 2 437 259 158 .105 ' 206 79 171 285 539 89 158 174 394 259 1& ' 167 99 37 73 39.78 406 819 672 939 346 455 123 766 030 100 178 695 987 401 269 590 146 772 V'9 490 286 793 591 585 285 ¡as y Cei "Censos de Población". 1970 1960 1952 7 374 115 5 028 060 6a 62 6 a 25 14.74 13.28 59 56 73.84 62.73 6 a 55 63.31 69,73 65.01 23.88 71,37 68.08 61,33 66.88 070 107 211 219 203 997 235 85 459 991 160 148 543 78 001 510 548 481 425 2 193 966 470 138 315 509 51 823 802 43 359 154 89 822 736 31 768 350 62 257 639 113 159 521 440 770 460 31 947 718 62 .630 300 78 350 654 152 896 794 114 146 005 66 417 671 70 347 211 21 956 770 19 966 358 ' 60 869 "Si'siribution (percentage) 87.11 94.79 73,52 51.83 55.50 88.82 9a 0 1 53.31 32.69 4 a 98 43.57 39.84 36.33 39,62 81.70 35.71 37.12 44.95 38,74 43.94 46.12 41.96 22.13 52.86 82.98 68a 19 Distribution (percentage) 15 859 12.89 5.21 26.48 48.17 44.50 P • j • J 950 758 648 588 324 255 804 489 56.06 53.88 58.04 77.87 47.14 17.02 175 069 .161 320 .251 555 243 098 152 748 127 .246 339 403 202 043 161 .451 97 720 735 162 672 210 233 945 3 018 152 306 460 169 491 168.580 63 915 114 737 52 448 23 0 885 121 275 82 "700 37 015 189 053 76 077 306 220 147 944 645 385 549 746 98 664 48 947 193 539 95 053 91 814 176 963 205 919 420 850 145 906 277 768 86 017 160 237 103 758 198 803 35T01 111 150 31 163 48 423 77 151 89 396 2 346 055 31.81 8 869 166 6 660 529 11 222 30 148 62 69 243 121 106 62 116 47 109 172 98 57 776 843 542 029 459 155 686 443 332 968 093 480 751 513 106 088 95 241 145 77 97 77 17 12 Resuaen del Pafs (para 1952 y 1960); Dirección de E s t a d í s t i c a "Entidades de Población", 1970 (Tarapacá a A g a l l a r l e s ) . 11.18 • 9,99 .46,69 67.31 59,02 56.43 60,16 63,67 6 a 38 18,30 64,29 62.88 55.05 61.26 13 8 25 137 63 62 215 136 104 62 109 45 112 T58 95 49 98 85 214 131 74 95 76 17 12 769 457 502 360 731 952 793 969 665 289 610 685 976 276 639 717 486 149 931 862 220 045 049 260 245 2 208 637 / Censos, Santiago de Chile, 1959 y 1969. Dirección de E s t a d í s t i c a y Censos, Santiago de Chile 1972. i1 o Table 3 THE ECONOMICALLY ACTIVE POPULATION OF THE LATIN AMERICAN COUNTRIES IN THE CENSUS YEAR OF THE PERIOD 1950-70 Country Census year Gì i t 1947 1952 1951 1950 1950 1950 1950 9 10 11 Argentina Chile Colombia Ecuador Honduras ttexi co Panama República Domini cana Venezuela Uruguay Cuba 1 2 3 4 5 6 7 Brasil Costa Rica El Salvador Guatemala Nicaragua Paraguay Peru 1950 1950 1950 1950 1950 1950 1 2 3 4 5 6 7 8 Sources: a/ Census year P 6 2 3 1 447 155 755 205 647 8 272 241 6i t 631 293 609 269 393 093 104 „ 424 388 134 442 567 11 332 299 7 2 5 1 Year^ P8 e, i P 524 667 125 591 988 016 386 1 855 428 655 210 44 1 556 22 1970 1970 1970 1970 1970 1970 1970 2 294 415 837 276 61 2 192 33 082 «0 754 087 910 423 186 1960 820 710 66 850 1970 1 225 000 99 781 1970 1970 1970 3 372 000 1 111 000 2 968 000 412 082 235 703 1 555 828 167 726 - - 1961 1963 2 351 291 995 500 28 7 344 211 200 1 972 266 327 208 - - - 718 984 409 814 976 344 2 231 205 29 870 74 424 - 000 360 000 000 000 000 000 _ 1950 - 177 607 557 895 799 15 966 450 408 713 460 907 233 292 _ 972 542 18 018 825 607 16 604 271 653 967 329 437 9 2 6 1 1960 1960 1964 1962 1961 1960 1960 1950 1953 ei 0i 't 994 862 961 174 010 091 079 Ps P - 1960 1963 1961 1964 1963 1962 1961 22 651 395 807 1 317 474 586 3 124 263 273 092 140 960 415 579 2 005 45 103 149 55 88 410 775 332 476 460 631 486 980 1970 1970 1970 1970 1970 1970 1970 30 419 535 1 081 1 623 659 784 4 285 000 000 000 000 000 000 000 - 2 621 62 138 18 6 77 117 563 315 081 594 633 187 979 610 1 ) Programa de intercambio ELAS/CELADE. Distribucitfn de l a Poblacitfn econòmicamente activa en los pafses de America Latina 1940-1961 2) Centro Latinoamericano de Demografía. Boletín Demográfico 2 (para 1970). Población Económicamente Activa, Estimations. ) 107 ( BIBLIOGRAPHY 1. Athanassiadis, C., Mathematical Model in Economics, Papazissis Publishing Co., Athens, 1959. 2. Athanassiadis, C., Statistics, Papazissis Publishing Co., Athens, 1957, Vol. II. 3. Athanassiadis, C., Business Statistics, Papazissis Publishing Co., Athens, 1963. 4. Athanassiou, S., "Statistics in Economic Research", in Modern Technology (review), February, 1972, Athens, Greece. 5o Athanassiou, S., Statistical Forecasting Model for Transportations, Rand Corporation, USA, 1972. 6. Athanassiou, S., "Techniques of Forecasting of Economic Magnitudes", in Modern technology (review), June, 1972, Athens, Greece. 7. Athanassiou, S., "Statistical Model for Teleconraunication Services™, in Modern Technology, October 1972, Athens, Greece. 8. Athanassiou, S., Manpower Planning in Greece, NATO Conference, Cambridge, England, 1971. 9. Athanassiou, S., "The Demographic Problem of Greece from an Economic Point of View", in Local Self Government (review), September, 1970, Athens, Greece. 10. Athanassiou, S., "Emigrations Reasons-Development Policy" in M o d e m Technology, August, 1971, Athens, Greece. 11. Athanassiou, S., "Emigration and Population Growth", in Social Research Review, April, 1970, Athens, Greece. 12. Allen, C., Elementary Mathematics of Price theory, Wadsworth, Publication Co., California, 1968. 13. Baxevanis, J., Economy and Population Movements in the Peloponnese, Greece, National Centre of Social Research, Athens, Greece, 1972. 14. Centro Latinoamericano ds Demografía (CELADE) y Corporación de Fomento de la Producción (GOHFQ), Perspectivas de Crecimiento de la Población Chilena, 1970-1985, Santiago, Chile, 1970. 15. Centro Latinoamericano de Demografía (CELADE), Boletín Demográfico, Nos. l - H , Santiago, Chile. 16. Cassels, L., "Emigration", in Estia (newspaper), 26 April, 1970, Athens, Greece (Republication). 17. Chenery, A., "Patterns of Industrial Growth", in American Economic Review, September, 1960. ' 18. Clark, C., The Conditions of Economic Progress, Mac Millan Co., London, 1957. ) 108 ( 19. Conning, Arthur., Estimación de la Migración Interna Neta, Clasificada por Edad y por Sexo, en las Provincias y Regiones de Chile durante los Años 1930-1940, 1940-1952 y 1952-1960, Serie D, N°36, CELADE, Santiago, Chile, 1968. 20. Elizaga, J.C., Migraciones a las Areas Metropolitanas de América Latina, CELADE, Serie E, N° 6, Santiago, Chile, 1970. 21. Ezequiel, M., Methods of Correlation and Regression, John Wiley and Sons, New York, 1967. 22. Hetzoglou, S., "Internal Migration", in Spoudé (magazine), Vol. 1 1966, Athens, Greece. 23. Kerrick, B., Urban Migration and Economic Development in Chile, The M.I.T. Massachusetts, USA, 1965. 24. Hughes, H., "Industrialization, Employment and Urbanization", in Finance and Development, Vol. 7. Bank for Reconstruction and Development, Washington, D.C., USA. 25. Instituto Nacional de Estadísticas, IKE., XII Censo General de Población y I de Vivienda 1952. Tomo Resumen del país"Santiago, Chile, 1959. 26. Instituto Nacional de Estadísticas, XIII Censo de Población i960. Resumen del país. Santiago, Chile, 1969. 27. Instituto Nacional de Estadísticas, Entidades de Población (by province), i960,Santiago, Chile, 1963. 28. Instituto Nacional de Estadísticas, Entidades de Población (by province), 1970. Santiago, Chile, 1972. 29. Instituto Nacional de Estadísticas, XIV Censo nacional de Población y III de Vivienda 1970, Muestra de adelanto de Cifras Censales (by regions) Santiago, Chile, 1972. 30. Instituto Nacional de Estadísticas, Ligia Herrera Jurado, Tendencias de Poblamiento en Chile desde 1940 a 1960. T-4, Tomo Santiago, Chile, junio de 1969. 31. Kanelopoulos, A., Economic Development, Vol. I., Athens, Greece, 1968. 32. Kevork, C., A model for Consumption in Urban Population, Bank of Greece, Athens, 1962. 33. Lampsidis, C., "Internal Migration", in Tahidroroos, (iragazine), 29 May, 1968, Athens, Greece. 34. Lewis, J., An Introduction to Mathematics for Students of Economics, Mac Millan and Co., New York, 1965. 35. Lipsey and Steiner P., Economic. 1969. 36. Lowry, S., Migration and Metropolitan Areas, Chardel Co., San Francisco, USA, 19661 37. Miró, Carmen, "Hie population of Latin America" in Demography Review, Vol. 1. The Population Association of America, USA, 1964. A Harper International Edition, New York, ) 109 ( 38«, Miró, Carñien, Aspectos oer.ográxicos de América Latina, CELADE, Serie A, M* N°88, Santiago, Chile, 1968. 39. í¡iróf Carmen, The Influence of Population Changes in Latin America, CELADE, Serie A, N°69, Santiago, Chile, 1967. 40. líandelbaum, K., The Industrialization of Dank I3ard Areas, tollman Publishing Co., Hew York, 1945. 41. Mac Mahon, C., Techniques of Economic Forecasting, O.E.C.D. Paris, 1965. 42. Myrdal, G., Economic Theory and Underdeveloped Regions, Duckworth, London, 1957. 43. national Statistical Service of Greece, Results of the Population Census, ÍI.S.S.G., Athens, Greece, 1963. 44. Odmann, E., Urbanization, A.F. Government Publishing Co., StockhoL-n, Sweden, 1970. 45. O.E.C.D., I Manpower Policy and Problems in Greece, Paris, 1965. 46. Qrcutt, G., Mj.cro-economic Analysis for Production of National Accounts, Institute of Sciences, Lisbon, 1967. 47. Oficina de Planificación Nacional (ODEPLAN), Plan de la Economía Nacional. Antecedentes sobre el Desarrollo Chileno 1960-1970, Santiago, Chile, 1971. 43. Oficina de Planificación Nacional (ODBPLAN), SI Desarrollo Económico y Social de Chile en la Década 1970-80, Santiago, Chile, September, 1970, Tomo I - IV. 49. Ofic-ina de Planificación Hacional (ODEPmj), Producto Geográfico Bruto 1960-70, Santiago, Chile, 1972 (Mimeo). 50. Papageorgiou, C., "The Regional Problem of Greece", in Review of Social Research, (magazine), Center of Social Research, September, 1972. 51. Poulopulos, S., "Urbanization and Astifilia", in Economicos Taliidroraos, 15, April, 1971, Athens, Greece. 52. Programa Regional del Empleo para América Latina y el Caribe (PREALC). Oficina Internacional del Trabajo (OIT), El Empleo y el Proceso de Desarrollo en Chile 1960-70, Tomos I - V, Santiago, Chile, agosto, 1973. 53. Schultz, H,, The Theory and Ijeasurements of Demand, The University of Chicago Press, Chicago," 1938. 54. Suits, D., The Theory and Application of Econometric models, Center of Planning and Economic Research, Athens, 1963. 55. Thomas, D., Social and Economic Aspects of Internal I'iigration, Mac M i l a n Co., Londonr 1941. 56. United nations, Processes and Problems of Industrialization in Underdeveloped Countries, U.N., New York, 1955. 57. United Nations, Economic Development of Underdeveloped Countries, U.U., New York, 1959. ~ ) 110 ( 58. United Nations, Industrialization in Latin American Countries. U.N., New York, 1965. 59. United Nations, International Research on Migration, U.N., New York, 1953 60. Westeble, R., "UrbanizationVol. 7, N°4, December, 1970. Finance and Development, (magazine), Problems and Prospects", in 61. Wolf. H., "Causal Inference from Observation", in Journal of Royal Statistical Society, Vol. 119, 1956.. 62. Wold, H., Forecasting by the Chain Principle, University of Uppsala, Sweden, 1963. 63. Wold. H., Time as the Realm of Forecasting, University of Uppsala. Uppsala, Sweden, 1966. 64. Wold, Ho, The Approach of Model Building and the Possibilities of its Utilization in Human Sciences, University of Uppsala, Uppsala, Sweden, 1964. 65. Wold. H., Demand Analysis, John Wiley and Sons, New York, 1952. 66. toulf, P., Macro-Economic Forecasting, Institute of Sciences, Lisbon, 1967. 67. Zolotas, X., International labour Migration and Economic Development, Bank of Greece, Athens, Greece, 1966.