83286 ·l ' THE URBANIZATION PROBLEM IN THE LESS DEVELOPED COUNTRIES: Csuses, Impacts and.Policy Implications for the Eank Group. R.M. Westebbe April 1970 The views expressed in this report are those of the author and may not be attributed to the International Bank for Reconstruction and Development o~ its affiliates • . -I • TABLE OF CONTENTS Page No. Drl'RC)liJC'l'IOH • • • • • • • • • • • • • • • • • • • • • •. • • • • • • • • • • • • • • • • • • • • • • • • • • • • • iv -; • PART ONE: THE· SE'rl'ING ... ~- I. THE HISTORIC PROCESS OF URBANIZATION ••••••••••••••••••• 1 ' A. Why Cities ••••••••••••••••••••••••••••••••••••••••• 1 B. Characteristics of Urbanization •••••••••••••••••••• 3 1. Development Patterns ••••••••••••••••••••••••••• 3 2. Urban Growth: 1920-1960 ••••••••••••••••••••••• S a. Analysis by Level of Development and Regional Density........................ 6 b. Analysis by Recency of Urbanization • • • • • • • • 8 o. Parameters ot Urban Growth • • • • • • • • • • • • • • • • • 10 3. Concentration in the Big Cities ••••• • ••• •....... 11 4. Urban Pull and Rural Push •••••••••••••••••••••• 13 ~. ·urban Population Composition • • • • • • • • • ••• • • • • • • • 15 6. Urban Employment Pattems •••••••••••••••••••••• 17 c. A Comparison ot the Developed and Leas Developed Countries •••••••••••••••••••••••••••••~•••••••••••• 20 1. Technology and Labor ••••••••••••••••••••••••••• 20 2. Rate ot Natural lnoreaae ••••••••••••••••••••••• 20 ·3. Social and Ebonomio Progress ••••••••••••••••••• 21 4. The Case at India •••••••••.••••••••••••••• ., • • • • • 21 II. IMPLICATIONS FOR THE FUTURE •••••••••••••••••••••••••••• 22 A. Urban and Rural Population Projected: 1960-2000 • • • 22 B. Employment and Output •• ~...... •.• • • • • • • • • • • • • • • • • • • • 27 PART TWOa ISSUES AND POLICIES • I. CONTINUING URBANIZATION: MIGRATION TO THE CITIES •••••• 33 A. Why Cities Grow: Urban Pull and Rural Push • • • • • • •• 33 ~ B. Can Cities Absorb the Migrants?..................... 36 c. Bringing City Lights to the Country •• ~............. 38 D. Do Migrants Cause Instability? ••••••••••••••••••••• 39 I. Is Overurbanization the Issue? ••••••••••••••••••••• 42 - 11 - TABLE OF CONTENTS (Continued) Page No. II. URBAN SCALE FACTORS 1N NATIONAL GROWTH • •••••••••••••••• 45 A. The Costs of Urbanization: Can They Be Managed? •••• 45 1. The Case of Latin AMerica •••••••••••••••••••••• 45 2. The Case of India •••••••••••••••••••••••••••••• so 3. Do Big Cities Cost More? ••••••• ~••••·•••••••••• 52 4. Concentratio:f\ Decentralization and New Towns ••• 511 S. Can Cities Keep Growing? ••••••••••••••••••••••• 55 6. Can the Costs Be Managed? •••••••••••••••••••••• 57 B. The Economic Development Benefits of Urbanization ••• 5R 1. The High Correlation Between Urbanization and Econom.c GrO'Vth •••••••••••••••••••••••••••••••• S8 2. How Urban Econo11liea Work in the Wation ••••••••• 61 a. Export Base, Service Base and Growth ••••••• 61 b. Innovation, Incomes and Industry ••••••••••• 62 c. Management as a Li11liting Factor •••••••• ·•••• 64 3. The Regional EconoliW': National Growth Versus Equal Growth ••••••••••••••••••••••••••••••~•••• 65 a. The Strategy of Concentrated Development ••• 67 b. .Growth Poles and the Spread of' Innovation •• 69 U. The Rural Beano~ •••••••••••••••••••••••••••••• 70 a. A ·"Holding" Strategy •••••••••••• • •••••••••• n III. THE METROPOLITAN POLICY ISSUE •••••••••••••••••••••••••• 73 A. Urban Public Goods and Services..................... 73 1. The Relevance ot Scale Economies ••••••••••••••• 74 2. M&naging Urban Systems •••••••••• ·•••••••••••••••• 74 3• Benet'its and. Costa •.••••••••• ·• •••••••••••••• ,.... 75 h. The Scope of' Urban Planning••••••••••••••••••••• 77 B. Distribution Policy for Urban Stability •••••••••••• 78 c. Urban Housing ?olicy ••••••••••••••••••••••••••••••• 79 PART THREE: POLICY IMPLICATIONS I. GENERAL STRATEGY FOR URBAN-REGIONAL DEVELOPMENT •••••••• 82 A. For the NatiOn ••••••••••••••••••••••••••••••••••••• 82 B. For the Metropolitan Region......................... 84 c. Resource Mobilization tor Urban Development • • • • • • • • 85 D. Population Policy •••••••••••••••••••••••••••••••••• ·86 - iii - TABLE OF CONTENTS (Continued) Page No. • II. POLICIES POR BANK CONSIDERATION • •••••••••••••••••••••••• 86 "'· " ·A. Policy tor the Short Run •••••••••••••••••••••••••••• 88 ' B. Policy at the National Level •••••••••••••••••••••••• 90 C. Policies at the Metropolitan-Regional Level.......... 90 1. The Metropolitan Plan............................ 91 2. Urban Management................................. 92 3. The Bank and Urban Infrastructure Systems........ 93 D. Possible New.Forms of Bank Lending •••••••••••••• ~.... 9h 1. Housing - Sites and Services ••••••••··•·•••••••• 94 2. Land Development••••••••••••••••••••••••••••••••• 95 3. . Mobilizing Housing Savings Through Savings and Loan Mechanis~ ••••••••••••••••••••••••••••••••• 96 ANNEX: I •. DEVELOPING POLICIES FOR THE 70'S THHOUGH AN OPERATIONAL RESEARCH PROGRAM II. A NOTE ON INTERNAL MIGRATION STATISTICS IN SELECTED COUNTRIES . III. A NOTE ON URBANIZATION AND ECONOMIC AND SOCIAL PROGRESS INDICATORS TABLES 1 - 10 • ,. - iv- • INTROOTICTION!/ ' The developinr, nRtions are urbanizinr. far more rapidly than they can create or borrow resources to accommodate their urban population. This is the heart of whRt is generallY termed an acute and universal urban crisis. The pace of economic development is not sufficient to cope with the problem. The economic growth of the cities does not provide sufficient modern employment opportunities for the growing urban labor forces. The demand for resources for urban overhead and public service systems is far greater than the supply of internal resources and transfers .from higher levels or government. A "critical" situation is thus created as the sharply rising demand for urban services cannot be met and the quality of urban life declines for the lower income groups who are forced to live in the mushrooming slums and shanty towns. An examinRtion of several case histories and the relevant literature as well as visits to a number of the world's urban trouble S?Ots lead one to question whether the wor4s "urgent" and "critical" properly describe the urbani- zation phenomena. It is the thesis of this paper th~t the urban problem cannot be dealt with by urgent measures and that it is not critical in the sense that • the social structures of the le.ss develope~ world will soon collapse without quick remedial action. The urbanization process involves the economic and • social transformation of the less developed world. It is an unbalanced 1( I would like to acknowledge the assistance of Miss Pamela Brigg who made an important contribution by collecting and organizing the statistics on which this study is based, by analyzing the sources for the basic figures and by calculating the labor force projection& - v - transformation. This unbalance will probably worsen given the present demo- graphic trends, economic growth rates and constraints on resources in the less developed world. Yet urbanization cannot be stopped; it is at the core ot the development process. The economic growth of urban areas, Where most industry is centered, determines that of the nation. The task then is to develop a long-run plan for the most efficient spati~ and economic organiza- tion of the system of cities in the less developed world in order to maXimize growth. Related issues involve minimizing the. cost of urbanization, improving the mobilization or urban resources and greatly improving the management of urban centers. . 'nJis paper is divided into three main parts. The first sets the stage for discussion and analysis of the problem. It reviews the historic processor urbanization and urban demographic characteristics from 1Q20 to 1960 as well as possibilities for the future. It contains projections of urban population in the less developed countries through the year 2000 and implications of these projections for the growth of urban labor forces and the growth of non-agricultural employment and outp~t. The second section analyzes the problem. Urban-rural relationships are explored, as well as the implications of migration for stability and for distribution policy. The relationship between urbanization and economic development is analyzed. Relevant theoretical and empirical findings are discussed. The costs associated with urbanization are examined. The urban economy, regional development, and such issues as equity and efficiency in resource allocation, optimal city size and growth poles are explored in terms of strategies for development. Finally, section three anA-lyzes the main policy implications of urbanization for the Bank Group in terms of the national economy, the metropolitan regional economy, and resource mobilization. PART ONE: THE SETTING I. THE HISTORIC PROCESS OF URBANIZATION A. Why Cities? We may view the process of urbanization and changes in the nature and size of urban centers over time as part of the response of society to technological advance. This view is not all inclusive, rather it is intended '. to focus on the role or the urban center as a mode or organization tor in- creasing efficiency in economic activity.!! The essential factor in creating urban settlements was the production ot agricultural surpluses. Cities of antiquity {pre 1000 B.C.) were of small size. They were supported by agricultural surpluses stemming from such in- ventions as animal drawn plows, domesticated plants and water control. Sjoberg stresses the key role of urban social organizations in managing surpluses and labor forces. An ideology, usually religious in character, and a favorable environment were required. Literacy was a critical factor in making possible complex systems of administration and thought.~ The main cities were religious and administrative centers. "He who controlled the annual agricultural surplus exercised the powers or lite and death-over his neighbors. That artificial creation or scarcity in the midst of increasing natural abundance was one of the first !( For a fuller discussion ot the cultural determinants of technological change in the growth of cities, see Eric Lampard, "The History of Cities in the Economically Advanced Areas," Economic Development and Cultural Change, Vol. III, No. 2, January 1955, pp. 81-136. V Gideon Sjoberg, "The Origin and Evolution ot Cities," Scientific American, September 1965, pp. 55-62. - 2 - characteristic triumphs of the new econo~ of civilized exploitation."l/ Large urban populations could not be maintained because of the high cost of transporting aoods, particularly agricultural products. Du.ring the Roman Empire agricultural advances enabled an estimated 2 percent ot the Empire •s population to live in urban areas. This in tum led to the develop- ment of many secondary and tertiary activities including a great increase in land and sea transport. A bureaucracy developed which was able to control population over a wide area. Rome itself probably contained UJ"f&rda of 300,000 people at the peak of the Empire. Arter the 14th century, European oities became cat'lters ot co~~~~~~erce and indust27. They were, in eesenoe, city states not dependent on large interior hinterlands but rather linked with each other by a growing movement of goods and by increased specialization ot production. By the 16th and 17th centuries the industrial revolution had become the main impetus to urbaniza- tion.Y Whereas the towns ot pre-industrial Europe were largely ·local in function and constrained by stratified societies, the ~ advances in industrial technology led to a concentration: ot production and capital in towns and cities, as well as to the factory syatem. In tum a host of secondary industrial activities developed in the same urban centers. The history or industrialization and urban growth in the West is one or increasing specialization. This changed the technical-organizational !I Lewis Mumford, The City in History, Baroourt, Bruce and Wald, Ino., New York, 1961, p. 36. Y Bert Hoselits, "The Role ~ the City in Hi~torical Perspeotin," !!!!. Urban Explosion in Latin Allerioa, ed. c. !eyer, Co:mell Uni118N1t,. Press, Ithaca, 1967, pp. 17-33. - 3 base ot society and with it the spatial ordering or economic activity. "Areal differentiation is, in f'aot, the. spatial oorolla.ry or tunotional specialisation and logically serves the sa. end -- the eoonoJII1' :'!/ B. Characteristics of Urbanization 1. Development Patterns The pattern and rate of city growth differ greatly among countries, depending on culture, historical circumstances and economic structure. Indus- trialisation in France made much less use of historic administrative and commercial centers than was the case in England. France's industrial growth took place more slowly in specia1ized centers and was strongly resisted by the guilds; the econoJV remained heavily under the influence of Paris and a few large regional cities. America's pattem of' urban centers in the early 19th century still reflected colonial economic dependence and therefore was mainly based on a f'ew port cities. Later with the opening of' the continent through the railroad, the port cities turned inward and beC&JilEI industrial and service centers tor the growing interior eoono~ 'Which in turn developed its own major urban centers along transport nodal points.Y Africa is experiencing the most rapid rate of urban growth today. Urbanization is a very recent phenomenon here. A good part of the urban growth in Africa occurred after World Wlr II with an intlux of western capital and western experts, who provided the necessary organizing and managerial talent. Administrative capitals and port cities were of' relatively !( Lampard, op. cit., pp. 88-92. y ~-, pp. 114-121. - u- small size during the colonial period. The migrant comes to the city from a tribal society with little education and no knowledge or the discipline needed to live and to work ett'ectively in large cities. Many ot India's great cities are toreign creations, responses to external econondc forces rather than to India's own social structure. The Indian elite, which dominates these cities today, is western in outlook, training and langUage, whereas the elite in the still predominantly rural areas speaks a vemacu1ar language and participates in the reli~ious and social cuato1118 ot one ot the ·n~rous disparate groups existing in India. CoJIIIIUDica- tion between the ~llte urbanite and the rural migrant is dif'ticult. When rural migrants bec0111e urbanized, they bring the culture of' the village w1 th them. In eftect, they reconstitute village society in the city. Institutionally, Iridian cities are quite ditterent trom their European counter- parts. The tradition ot the tree citizen enjoy.lng a large degree of' urban self-government baa never existed.!! Urban political and adlld.nistrative institutions are thus less well equipped to cope with the complicated problem of modern city growth. In Latin Allerica, towns and cities were established in a highly organized way by conquerors. First, regions and their resources were defined. Then the main cities and systems ot towns were established for each region. There 'Was no frontier."Y A city began as a center ot administration, religion, !/ Bert Hoselitz 1 "The Role of Urb8nization in Economic Development t Some Intemational Comparisons," India's Urban P\l.ture, ed. Ro7 'l'uriler, 1962, · pp. 271-276. - 2/ Ralph GakenheiJier, "The Peravian City in the Sixteenth . ~tury, n - in The Urban Explosion in Latin America, ed. Gl.erm Beyer, Cornell University Press, 1967, P· 3u. - 5 - military control and marketing. It was essentially a political creation rather than a tunation of indigenous econo11Jic or social forces. Rights to land were derived from the colonial urban adJII:l.nistrators. The dominance of the town by conservative landowner classes discouraged the immigration of \. others. Financial resources were limited and municipal works were slow to be built.JI A syatem of port cities was built in which each city looked toward Spain. Contacts b.etween cities were discouraged. The interior of the con- tinent with its wealth and natural resouroes remained largelY unexploited. ~er independence there was little change in the social and economic structure. Greater international exposure made the cities less provincial as the;y imitated Europe and the u.s. Immigration from Europe was a major tactor tor change. It led to the opening ot the interior, the beginning or 1110dem technology in agriculture, the creation or urban industries, the construction or interior lines ot comrrnmication, and the emergence or a new professional class.Y Yet still today the large, dominating, central city is more a center tor the preservation of traditional society even on the part of the middle class, than a torce tor national integration. 2. Urban Growth: 1920-60 Around 1800 an estimated 2 .!a% ot the wc;>rld 1 s population lived in cities of 20,000 and over. By 1850 this percentage had risen to 4.3%, by • 1900 to 9.2%, and by 1920 to 11&%. By 1960 almost a quarter ot the world's • population lived in suoh cities. !I Ibid., pp. 36-56. Y See J. Hardoy, "The Role of the City in Historical Perspective," !!!!_ Urban Explosion in Latin America, p. 61. - 6 - The rate or urbanization seems to depend on a region's leVel of development, density of habitation and existing level of urbanization. In the laO years from 1920 through 1960 the less developed regions or the world experienced higher population growth rates than those of the more developed regions and considerably higher urbanization rates. The big city population of the less developed world showed the most dramatic increases or all. a. Analysis by Level or Development and Regioftal Density From 1920 to 1960 the world's population rose from 1. ) billion to some 3 .o billion, a rise or 61%. The population ot the mere developed regions rose by laS% and that or the less developed regions by 70%. In this period the world 1 s urban population (defined as 20,000 and over) rOse from 270 million to 750 111illion, a rise of 185%. The urban population ot the more developed countries rose by 125% and that ot the less developed regions by 330%. The world's rural population rose from 1.6 billion to 2.2 billion or by laO% in the same period largely in the less developed regions .!1 !I The figures used in this section are derived from UN sources and are subject to a number or qualifications with respect to definition, accuracy and interpretation. For a full statement of the sources and qualifications see: United Nations, !rt Working Group on Problems or Internal Migration and Urbanization, and selected papers, Bangkok, 1967, p. 32. l/. J. c. Caldwell, "The ~vement to Ghana's Towns," Af'rican Rural-Urban Migration, Australian National University Press, Canberra and London, 1969, pp. 87-119. This is based on survey done between 1962 and 1961& ot 582 urban households and 1, 78? rural ones by the Demography Unit or tbe Univ. of Ghana. !!I Beyer, The Urban Explosion in Latin America, p. 97. - 15 - or Sao Paulo. Many were small landowners and sharecroppers or cotfee and cotton lands which had become depleted and then abandone:i as new areas opened to the West. The rest or the migrants were mainly trom the Northeast where they had been engaged in marginal subsistence agrioul ture.. They were usually predisposed to move although they retained strong ties to family remaining behind.k' A prevalent theor.y supported by case studiee is that a good deal or migration occurs in a series or steps. Rural migrants go first to small towns and then proceed to larger ones, making a relatively short journey in any given phase. There may be exceptions such as Bogota, Colombia where the majori1;y ot migrants. have moved directly .Y 5. Urban Population Composition The sex and age distribution or urban population varies according to level or development and volume or migration. Census data tor individual countries var.y in reliability and completeness. From studies done ·so far it appears that there were 40 million more females than males in the more developed regions and 40 million more males than females in the less developed regions. There is some indication that the differences may be explainable in terms ·or male and female mortality rates and in some oases war losses. Urban areas in the less developed urban areas exert a greater attraction for male migrants 1/ James Rubens B:-andon Lopez "Aspects of the Adjustment ot Rural Migrants to Urban-Industrial Conditions in Sao Paulo, Brazil," Urbanization in Latin America, pp. 214-238. Y W. L. Flinn, Rural to Urban Migration, A Colombia Case, University of Wisconsin, Land Tenure Center, July 1966. - 16 - than do the urban areas of the developed regions. A measure of the relative number of males to females can be calculated by the "not excess masculinity rate." In latin America with a relatively high level of urbanization of 49%, females exceed mA.les in the urban population. The masculinity rate is highest in Africa and South Asia where urbanization levels are about 18%. Males per 100 ~e~~es ~n_the Totat, U~ban,,Rural=!egulation Excess Total Urban Rural Urban Population Population Population Masculinity World Total 100 99 101 - 2 MOre developed regions 92 91 93 - 2 Less developed regions lOU 111 103 +6 Of which: East Asia 1o6 114 104. +10 South Asia 105. 114 103 +11 Ia.tin America 100 93 lo6 -13 Africa 96 110 96 +14 Source: A number of factors may account tor this phenomena. In the least developed countries, males may be attracted in larger proportions to urban centers where some form of work opportunities can be found. Females in the least developed countries are less likely to find work in urban areas. Also it is more difficult to t1nd adequate shelter for.families. The attraction of urban amenities may be less in a society with poorly developed services. - 17 - In general, urban areas are growillg more from migration than from natural increase. Since migrants to cities in the less developed countries .. · tend to be of wrking age, the distribution of urban population by age is generally more concentrated in the working age category. , 'While urban male participation rates do not differ significantly among countries, there are cultural factors inhibiting the participation of females in economi~ activities in urban areas, particularly in Asia.!! The more developed a country, the higher its female participation rates, parti- cularly in office and other service occupations. The sex composition ot the labor force varies greatly with the level of urbanization. Proportionately many more women are working in urban areas in Latin America than in other less developed areas. · 6. Urban Employment Pattems The distribution of non-agricultural employment (Annex Table 7) may be regarded as a rough measure or the distribution of urban employment. It ~ be useful to point out that the definition of non-agricultural employ- ment is not the same tor all countries. In most cases it includes the mar- ginal service occupation categor,y which absorbs so many recent migrants to the cities. Urban labor force figures and estimates or non-agricultural employment cannot be compared tor the .Purposes of analysis (Annex Tables 7 and 10). First, because the figures include different g~oups of countries • for different time periods and are based on different national definitions. • Second, because non-agricultural employment includes employment in localities smaller than 20,000 while urban labor force, as here defined, only includes 1/ See "Urbanization in Asia and the Far East," Joint UN/UNESCO Seminar, ed. P. Hauser, Bangkok, 1956, p. 108. - 18 - localities or 20,000 and over. The urban labor force as defined by different nations approximates non-agricultural employment more closely, though still tends to be smaller for many nations whe~ mining, manufacturing, transporta- tion, etc. also take place in rural areas. For example, in India in 19Sl, 6% or the total rural labor force was employed in processing and manufacturing. In absolute terms, India's working force in processing and manufacturing was almost evenly divided between urban and rural areas. Rural areas contained predominantly cottage and handicraft industries while urban areas contained the bulk ot factory employment. It is believed that the Indian pattern or employment is generally valid tor most of Asia.!/ · In Annex Table 7 non-agricultural employment is di vi.ded by main sectors for the main regions or Atrica, Asia and Latin America. The less developed regions show considerable variations in the proportion of manu- facturing and public utilities employment to total non-agricultural employment. Manufacturing and public utilities constitute trom 19% to 2~% of non-agricultural employment in Africa, from 23% to 27% in other East Asia, South-East Asia and South-West Asia, 37% in Middle South Asia (including India, Pakistan and Iran), and 31% to 35% in Continental South America. Latin America with relatively higher per capi.ta income levels also has generally higher percentages employed in manufacturing and public utilities.5/ !I See Hauser, ed., "Urbanization in Asia and the Far East~' pp. 1112-luu. 5I More detailed an8lysis suggests that the industrial composition or the male labor force in Latin American cities is closer to that of the cities of the industrialized countries than to that of the cities of less developed Asia. See Urbanization in Latin America, p. 115. - 19 - Middle South Asia, including India and Pakistan, has a low pro- . portion or non-agricultural employment and a relatively high proportion or people employed in manutacturing and public utilities. As manufacturing includes handicrafts and cottage industries, the higher Indian proportion • in this category may simply refiect a higher proportion ot these manual occupations in the non-agricultural labor force. South Africa which has a high proportion ot non-agricultural employment shows a low proportion in manufacturing and public utilities. This is explained by the high proportion employed in mining. Most or the variation in the distribution ot non-agricultural employment takes place between manufacturing, public utilities and collllll8rce and services. '!be latter category is high tor all regions, varying between 4.5% and 62% ot _total non-~icult_ural emplo~nt. A good deal or employment in this category is or a marginal service nature. For this reason the category exhibits a high absorptive capacity in most less developed countries. As countries advance to the higher levels or development, the service category also absorbs a high proportion or the non-agricultural labor toroe but then in professional and other highly paid occupations. In the less developed countries the service occupations contain many underemployed people, which is one reason that orticial rates or unemployment are usually low in these countries."!! # • !I Annex Table 8 shows non-agricultural employment indices in selected countries tor the last ten years. - 20 - C. A Comparison of the Developed and Less Developed Countries 1.. Technology and Labor The urban growth ot the less developed nations differs fundamentally from that. experienced b,y the industrializing nations ot the world during the 19th century. The technology being introduced in the less developed nations now is far more advanced than that which transformed Europe and North America during the 18th an4 19th centuries. The technology of industrialization evolved gradually in the West. It was relati ve]Jr labor intensive in the early stages and advanced gradually to its present high degree of capital intensiveness. Thus there was a long- run demand for labor in the industrializing cities of the ~at corresponding to the· supply of labor freed from the agricuitural sector, which was simul- taneous]Jr increasing in efficiency. In contrast, the less developed countries are industrializing with a relatively advanced capital intensive technology under surplus ·labor conditiona far more severe than those experienced by the Western countries during their phase of rapid urbanization. 2. Rate of Natural Increase · In addition, mortality rates in the developing countries of today have declined --rar more- rapidly, because of modem enVironmental sanitation and medicine, than they did in the advanced countries of today when they were at comparable stages of urbanization. For example, infant mortality fell by about &;)% in Sweden between 1880 and 1920 and by 130% in Ceylon between 1920 and 1960, while lite expectancy rose by about 10 years in Sweden between 1880 and 1920 and by s01118 25 years in Ceylon from 1920 and 1954. - 21 - As a result urban areas of the developing countries of today are growing both trom high rates of natural increase and migration. In contrast, the urban areas of the now highly developed urbanized countries grew pri- marily trom migration as their rate of urban natural increase was low during • their period of rapid urbanization. 3. Social and Fconomic Progress Countries w1 th similar urbanization trends and levels may be com- pared with respect to selected social and economic indicators. (See Annex III. SUch a comparison indicates that in the developing countries economic and social progress is not keeping up with the pace of urbahization. At similar levels ot urbanization, the developed countries had progressed more economic- ally and socially, with the exception ot health conditions. 4. The Case or India Let us consider the case ot India. Between 1891 and 1951 India's urban population, in towns or 20,000 and over, rose from 5% to 12%, a level reached by most &lropean countries by 1900. Germany, Britain, the low countries, and to some extent France, urbanized more rapidly than India during comparable periods of growth. Scandinavia, Switzerland and Russia were about the same as India. Although comparable to Europe in urbanization, India has lagged far behind in industrialization and in income per capita. At the same level of urbanization (about 11%) slightly more than half or the population or .. European countries derived their livelihood from agriculture compared to two-thirds ot India's population. About one-fourth of Europe's labor force was already engaged in manufacturing at this early stage of urbanization - 22 - compared to only about 10% in India in 1950. About a fourth of India's urban population was engaged in manufacturing in 1950 compared to over halt in mining and manufacturing in Germany in 1882 at a comparable level ot urbanization. India also suf'tering from high population densities in rural areas bas ex- perienced little change in agricultural productivity. A good deal of India's manufacturing and non-agricultural employment is or the cottage variety in rural areas. Industry in developing Europe benefited trom the external eco- nomies of location in large urban centers and from internal. economies or scale. This led to rapid capital formation in urban areas. In India the generally smaller scale and dispersion or manufacturing have limited capital formation, technological change and the growth or employment in rapidly growing centers. The increasing urban population accordingly must tind employment in miscellaneous low productivity service occupations.!! II. IMPLICATIONS FOR THE FUTURE A. Urban and Rural Population Projected: 1960-2000 The increasing urbanization or the ·less developed world has profound implications tor economic and social policy in the countries affected. Urbani- zation relates intimately to the location or modern sector activity, to the concehtration or unemployed and underemployed labor and to the demand for social overhead facilities and social services. Thus development planning, th~ assessment or investment priorities, the formulation or investment programs and the location ot and return on projects depend greatly on what demographic patterns are expected. In this section the. main trends or regions will be assessed. !/ Hoselitz, India's Urban·Future, pp. 164...176. - 23 - _,. • !/ For a fuller explanation, see Growth or the World's Urban and Rural Population, 1920-2000, pp. 55-57. The "medium" variant of the UN total population projections was chosen. Future urban population was based on the weighted average results or projections based on tour patterns ot past growth. See p.·lO tor a discussion of these patterns. g! This assumption may. not be realistic tor highly urbanized and developed countries like the u.s. A recent estimate indicates the urban popula- tion ot the U.S. may reach 85% of the total by 2000. See wall Street Joumal, March 2, 1970, p. 12. See also UN Working Paper No. 15, pp. 50-64 tor an earlier version or the projections with somewhat different assumptions. See ISDR No. 1, pp. 17-19 tor a summary covering .. 1960-1980. - 24 - At present the UN estimates show that the world's po~ation 111&7 be rising b7 6; million annuall7 or which 30 million are in urban areas. The urban, population or the less developed areas ma7 be rising b7 20 million annuall7. The estimated breakdown or the annual urban and rural population increases in the less developed major areas is as follows: Urban Rural & Small Towns (in millions) South Asia 7S 20.0 · Bast Asia 6.0 6.S Africa 2.; 6.0 Latin America 4.5 3.5 Source: Growth . or the World's Urban and Rural Population, 1920-2000," pp. ;6-;7. It should be noted that Israel, Japan, ~entina, Uru,gua7 and Chile all have passed the SCY,C urbanisation level and most have relativel.y higb per capita incomes even though the7 are part or the less developed major areas. - 2S - In brief1 the longer term projections show that the urban population of the 1110re developed countries 1118.1 double by the year 2000 as it did from -- 1920-1960 to a total of some 900 million. The urban population of the less 9 developed regions may quadrUple by the year 2000 as it did from 1920-1960 to some 1 1 400 million. The rural and small town population of the more developed regions is expected to grow slightly to some 540 million by 2000. The rural and small town population ot the less developed regions may almost double to so11l8 31 200 million by 2000. While the urban population ot the developed world 11111.7 rise at a decreasing rate 1 the momentUJil ot urban population increase in the less develop areas is expected to be .maintained• .As the follow:l.ng table shows 1 all JD&jor areas of the less developed world J1U17 teel the etrects or massive increases in both urban and rural popu- lation, some to a .greater degree than others. Absolute P (In millions) 1920-40 1940-60 1960-80 1980-2000 Mai;r Areas Urban at .Aaia 34 73 120 158 So".lth Asia 24 67 148 302 Latin America 12 44 94 179 Africa 7 23 53 128 Major Areas Rural & Small Towns "' East Asia 47 87 127 88 South .Asia 116 181 402 44~ Latin America 28 38 72 81 Atrica 42 58 123 191 - 26 - The variations in the rate o£ increase of urban and rural population by area depend on population growth rates and the level or urbanization in each region. Africa may emerge through the rest or the century as the region with the highest· rates of growth of both urban and rural population. ATerue Annual Rates of Growth Urban & Rural in the Less DeTeloped Major Areas 1920-2000 Urban 1920-hO - 1940-60 1960-8~ 1980-2000 liit·~ia 3.1 3.S .).0 2.4 South Asia 3.2 4.3 4.1 3.9 Latin .AtaEn-ica 3.5 5.2 4.4 3.8 Atrica .3.5 s.o 4.6 4.5 Rural !&it Asia· o.s 0.7 0.9 0.5 South Asia 1.2 1.4 2.2 1.7 Latin America 1.6 . 1.6 2.1 1.4 Atrica 1.3 1.4 2.1 2.2 Source: Growth of the World's Urban and Rural Population, 1920-2000, Table 34, p. 64. Ellrope 1 s share of the world's big city population is expected to decline from 49% in 1920 to 16% by 1980. In the same period the less developed countries may increase their proportion or the world's big city population from 13% to 48% •. The less developed regions may be called upon to accommodate almst 200 million people in big citi~s between 1960 and 1980. - 27 - . The prejections are on the whole probably cautious. Earlier UN projections of the urban population of the less developed regions show a .-· low figure of l.S billion by the year. 2000 and a high figure or 1.9 billion by 2000, based on high and low assumptions of total and urban population , growth.!/ Even the lower projection is slightly higher than the projections discussed here. The precision of the estimate is, of course, not the issue. What is important is the implication of the figures, even i f roughly accurate, for the size and pattern of fUture resource allocation •. · Underlying this issue &re questions concerning the organization and quality ot lite in cities of the less developed world which in many cases are doubling in . size over 10 to .lS years. The issue is really the economic development implications or tuture urban growth. B. Emplo;yment and Output--- The urban labor force projections for Africa, Asia and Latin America from 1960-2000 in Annu Table 10 indicate the extent of the surplus labor problem in the less developed countries. These projections are based on the lower ~ urban population projections and on national census data on urban-age specific participation rates tor the period 19S0-66 for 3 countries in Af'rica, 10 in Asia and 13 11 UN Working Paper No. 15, also Annex Table 9. ' - 28 - in Latin America.!/ According to them the urban labor foroe may increase through the year 2000 by sixfold in Africa and by fi ve.fold in Asia and Latin America. In Africa and Latin America urban population and labor force were projected to increase at approximately the s&Jile rate while in Asia, particularly in East Asia, the urban labor .force was projected to increase :taster than urban population. ln the period 1955-1964 the developing countries increased their output or 1110dem sector manufactures by an annual rate. ot 7% but their direct demand for manpower in this sector by an annual rate of only 4%; employment in light manufacturing rose by 3~ annually in this P'riod and that in heavy industry by 6~.'U Urban industries are peculiarly well placed to benefit from increasing labor productivity. Apart from changes in the technique of production, firms in growing urban centers can usually count on achieving productivity increases through changes in labor .force quality, better Management, expanding markets, economies of scale and fuller use of capacity.lf !/ Ettore Denti 1 "Sex-Age Patterns of Labor Fol"Ce Participation by Urban and Rural Populations," International Labor Review, December. 1968. y UNIIX>, "Industrial Developn~ent Survey," International Symposium on Industrial DevelopMent, Athans, 1967 1 p. 307. 'J! c. R. Frank, Jr., "Urban Unemployment and Economic Orowth in .Africa," Center Paper No. 120, Yale University, 1968, p. 257. - 29 - Assuming about one-fourth of the urban labor force in Asia and Africa will remain in ~nufacturing and public utilities and about one-third in Latin America as at present, the following number of jobs will have to be created in these sectors from 1960-2000. (millions) 1960 Estimated 1960-2000 Pro- Urban Labor jected Increase 196o Force in Manu- in Jobs in Manu- Estimated facturing & facturing and Total Urban Public Utili- Public Utili- I.abor Force ties ties (1} (2) (J} Africa 14.4 J.J 18.6 Asia 45.9 14.6 70.5 Latin America 22.2 7.4 36.8 The prospects that industrial grovth will absorb more than a fraction . of the rapidlf growinp, urban labor forces forecast for the less developed world is not promising. ]n order to achieve the employment levels shown above in manufacturing and public utilities by the year 2000, emplo,rment in these sectors would have to rise between five and six told, the same as the projected increase in labor torce.As we have seen, recent industrial grovth in the less developed areas produces only about ~ percentage point p,ain in employment for each percentage point r.rovth in industrial output. Thus, in order to achieve the relatively - JO- • modest goal ass~d for the increase in industrial and public utilities empl~nt, and ·allowing tor a wide margin of uncertainty- in the figures, a doubling in the rate of industrial investment and output vi th respect . to the employment projected would appear to be necessary, givan present technology and_ the recent trenda in industrializatio:q. According to one sup,gestion, the highly paid modern sector workers in Africa are becoming an increasingly smaller percentage or vhe urban work force at the same time that the wage differential is widening.!! An. analysis ot three East African countries shows not only a consistent lag between indus- trial growth and employment for the late fifties and early sixties but a decline 2/ in manufacturing employment while output expanded.- However, t.he indirect demand tor manpower due to greater industriali- zation grew undoubtedlY tar more than the 4% for direct demand. we need to know a go~ deal more about the indirect empla,yment impact of industrial growth in developing countries. In particular we need to know how the growth or manufacturing industry leads to a rise in demand for transport and commercial services. Further there may be important linkages between modern sector emplo,y- ment and ~o-called traditional service employment. There appears to be a !! Frank, op. cit. Y J. Harris and K. Todaro, "Urban Unemployment 1n East Africa: An Economic Analy'sis of Policy Alternatives," East African Economic Review, December 1968, p. 25. - 31 - high income elasticity of demand on the part of those with modern sector incomes for the services of those in the low wage traditional sector. This . is a subject where further research could well profoundly change our under- standing of the economics of service sector employme~t in the less developed countries. One study analyzes the growth of labor forces in relation .to growth in real output for tha recent past and through 1980. It concludes that surplus labor will rise by some 12% of the labor force by 1980 on the basis of present output-emplo,yment ratios adjusted. This implies an overall rate of surplus labor of over 20% when the estimated surplus labor force in 1965 is added. In Latin America given present trends in non-agricultural employment,· overall rate of growth of output of 5%, and a growth of labor force of 2.8% per annum, productivity in agriculture and services could not rise by more than 1. 3% per annum from 1965 to 1980 in order to avoid the emergence of a higher proportion of surplus labor to labor force.!/ In selected African countries the government sector accounts for close to half of total non-agricultural employment. Government employment cannot grow much more rapidly than current government expenditures. Growth of current government expenditure in turn relies on the expanslon of total budgetary ±/ OECD, "The Employment Problem in the Less Developed Countries: A Review", OECD Second WOrking Conference on Research Into Emplgrment Problems in Developine Countries, WOrking Document, December ?, 1969, pp. 98 _ 100. - 32 - revenues and t,he distribution of such resouroes between c:urrAnt and investment. needs. It is reasonable to assume liha.t commerce, trade and misoellaneouE; services probably grow at roughly the same rate as tot.al national inc:ome .1/ The c:onstruction industry, as will he shown, absorbs less than 10% of non- agricultural employment and is hip,hJy unstable, sub,ject to high rat.es of unemployment. A recent study shows that in Latin America well ov~r half the increase in the labor f'orQe is still being absorbed in traditional low productivity activities.~ '!he lesson or the recent past would seem to be that a large pro- portion of the less developed world's rising urban labor force will haw to be absorbed in low productivity service occupations given the distribution or employment in present urban centers and allowing for reasonable growth ra~es in national product in the future. 1/ Frank, op. cit., p. 250. Y G. Jones, "Underutilization of Mmpower and Demographic Trends in T.atin America," ILO Review,Nov. 1968, pp. u~.3, u5u. - 33 - PART TWO: ISSUES AND POLICISS I. CONTINUlNG URBANIZATION: MIGRATION TO THE crriEs In the less developed countries, the urban labor force is rising more rapidly than urban employment opportunities despite fairly high rates of indus- trial output growth. Nevertheless, migrants continue to flood to the cities. Migration is essentiRllY unpredictable and unlike population growth, is not greatlY affected by birth and mortality rates. Economic incentives in particular, but also social factors play a large part in decisions to migrate. Under con- ditions of rapid population growth and urbanization, prediction is especially hazardous. Migration is a mRjor factor leading to policies to control primate city growth in ma~ countries and to corresponding policies to build up new regional centers. It also concerns the issue of the best allocation of national resources for urban and rural growth. A. lily Cities Grow: Urban Pull and Rural Push The motivation tor people to migrate to cities is based on some combination of perceived welfare and economic incentives; in the preceding part some of the results of case studies on migration motivation and satisfaction was discussed. This motivation is by and large rational in that people who migrate believe they will be better off as a consequence ot the decision. Assum- ing homogenous labor supply and perfect factor mobility amongst other assumptions the migration of labor could be ascribed to difference in the marginal producti- vity of labor in rural and urban areas. Migration would thus cause workers to raise their marginal products and would cease once urbBn-rural wage difterenoes had disappeared. the real world - 34- is a good deal more complicated, as labor is not homogenous and factors are not perfectl7 mobile. Innovation and the process of economic development exert considerable influence on the forces producing economic incentives to migrants. In tact, people· do not move only because of real wage difference. A number o£ tactors are responsible which may be categorized as urban pull anQ rural push. The movement to the cities is not without cost. The migrant and his 1"amil7 pay the psychic cost of the disutility o£ l~aving a i"amiliar environment. MOney costs include the direct additional cost of making the mone and the income foregone until the wage earner becomes reestablished. All these factors must be weighed-against the advantages cited i"or the urban center. This suggests that major dii"ferences between expected or actual wages in urban and rural areas may not be the only important element in stimulating migration and that migration can be substantial even in the absence of large wage dii"i"erentials. Y ~ge differentials are not always the same as real income difi"erentials in view of the cost of migration and the high urban cost of living. Availability o£ public services would tend to increase real wages relative to money wages. FUrther wage rates for central city emplo,yed do not take into account the large percentages of people unemployed particularl7 the masses in slums on the fringes 2/ of the cities who may not be counted on the wage i"igures at all.- The impact of urban pull and rural push factors differs greatly among different regions in the less developed areas. Pressure of population on limited !( Fbr a discussion of push-pull and cost-benefit factors associated with ndgration, see Bruce H. Herrick, Urban Migration and Economic Development in Chile, M.I.T., 1965, pp. 10-22. £( Resources for. the FUture, Design for a WOrldwide Stugy ot Regional Developmen • 1966, p. 50. • - 35 - land resources is an important factor. Countries in JWrope in the mid 19th centur,y had from 3 to 5 times as much cultivable land per household as do Indian .tarm households toda)". The JWropean farmer produced surpluses and entered the market econ~. The Indian farmer, except tor selected recent changes. in productivity, has remained closer to subsistence. Latin Am&rica represents an intermediate case between Europe and Asia, insofar as generalization about such a varied conti..'lent can have any validity. Pressure on the land in Latin America is not so great as in Asia. Substantial capital accumulation has taken place in urban industry and the modern sector is. growing more rapidl)". Nevertheless, .culturally the poorer migrants of latin America resemble those ot India in their persistence of rural behavior patterns and social groupings, a phenomena which led Oscar Iswis to speak of a "Culture of Poverty. n'!/ In Africa one must distinguish between areas north and south of the Sahara. In the north cultivable land is limited. High rates of rural popu- lation growth and stagnant prodUctivity levels are at the root of the high migration rates. In the south, population and agricultural resources are some- what better balanced although rigid economic and social structures may be in- hibiting a gre&.ter absorption of the labor force on the l~nd.Y Rates of urbanization are amongst the highest in the world in the south. The socio-cultural settlement pattern in the cities resembles that in Asia.JI 1( Cited in Bort F. Hoselitz, "The Role of Urbanization in Economic Development: Some International Comparisons", India's Urban Future, pp. 157-181. E/ Economic Commission for Africa, Localisation Du Peuplement, Migrations Internes et Urbanisation en Afrigue, Cairo, 23 Cctober 1962. J( Dr. Prothero, "ChRracteristics of Rural/Urban Mi~ration in SUb-Saharan Africa," UN,.Wbrld Population Conference, Belgrade, 1965. - j6 - B. Can Cities Absorb the Migrants? Policies designed to increase urban emplo,yment, if successfUl, will cause a corresponding reaction on the supply side by inducing those who might have withdrawn from the labor force to stay and by encouraging those who might not have come to the city to come. SUccessful industrial growth policies in addition to creating more employment opportunities may also increase urban un- employment.!/ In short, within wide margins the supply of urban labor may be perfect:cy elastic. We do not lmow at what level of unemploymen·:,, of under- employment or of deterioration in the quality of urban life migrants are per- suaded not to come. Certainly' the widespread prevalence of "chain migration" implies the existence of an effective communication link between the urban and rural poor over wide areas. In Latin America high levels of internal migration are consistent with urban unemplo,yment rates of 10% to 15% particularly among the young, and even higher rates if underemployment equivalents are added. In India urban unemploy- ment in 1956 was estimated to be 10% and generally underemployed 11%. Big city un~mployment in India was reported to be four times the rate in urban areas of less than 15,000 people.£! In Chile an unemployment rate of 6-7% was found in Santiago and one of 8-10% in provinci&l cities. If low productivity emplo,yment 1/ See Frank, op. cit., and J. Harris and M. Todaro "Urban Unemployment in East Africa: An Econon;ic Analysis of Policy Alternatives," East African Economic Review, December 1968, p. 25. 2/ OECD, ''The Employment ProblelllS in less Developed Countries: A Review". Also, "Unemployment and Underemployment in India, Indonesia, Pakistan and the Philippines," ILO Review, October 1962, pp. 4-12. - 37 - 1/ or uncleremploymen.t is included a f"igure o£ 20% £or "marginality" may be used.- Urban 1lnempl01111ent may not always be a matter of" lack o£ job oppor- tunities. In the case o£ those with some education there ~ be resistance to accepting jobs they regard as inf"erior. For example in the Ivory Coast rural population, with some education and high expectations, migrate to the cities where they £ace high unemployment rates in the unskilled job category. This movement has caused a rural labor f"orce shortage which has in turn induced migration o£ even less skilled people £rom neighboring countries. In Abidjan there is the seeming paradox of" surplus labor in unskilled job categories and labor shortages in more skilled job categories creating a demand £or expatriate labor.&' In general, the urban emplo,yment structure is characterized by wide- spread underemployment and marginal labor f"orces. This usually concerns workers at the lowest income levels. Virtually everyone has something to do and manages to earn something. In part this phenomenon results !'rom oversta££ing o£ public c services and £rom public policies aimed at I'IIBXilllizing employment in simple con- struction and maintenance work. Howaver, an important part o£ the ability o£ these marginal labor f"orces and their dependents to survive relates to ver;y f"lexible labor absorption capacity o£ the numerous service, trade and small industrial activities of the !/ J. Friedmann, "Urban ani Regional Development in Chile: A Case Study of Innovative Planning," Santiago, June 1, 1969. According to G. Jones, ~· ~., p. 458, Chile in 1960 reportedly had an unempla,yment rate of 6.7% and an urban underemployment rate o£ 28%, Panama 11.2% and 20%, respectively, and Peru (1961) 2.7% and 25%. '!./ See H. Mulder, "Urban Planning an:! Development in the Ivory Coe.st," IBRD Economic Report, February 1970 (Draf"t ). - ~~ - 1/ nonmodern sector of major urban centers which we do not tully understand.- we do not know at what level the growing labor surpluses of cities can no longer be concealed in underemployment by the flexible absorptive mechanism thRt is the traditional economic and social sector of the major cities of the less developed world. C. Bringing City Lights to the Country In view of this situation, attention has been given to policies affecting the supply of labor in rural areas. An analysis of labor migration and unemployment in less developed countries concludes by suggesting that "the net benefits of bringing 'city lights' to the countr,yside might exceed whatever net benefit might be de~ived from luring more peasants.to the city by increasing the attractiveness of urban living conditions." The object, of course, is not to reproduce urban amenities on the farm but rather to keep the urban-rural real income differential within tolerable limits and not depend on the "long run stabilizing effect of a lower probability of successfUlly finding modern sector employment."'!! A 'suggested means of accomplishing this would be to encourage small-scale producers in the traditional sector as well as increasing agricultura: real incomes by reducing taxes on agricultural output.J! At the margin and giTen the choice of policies which would in- crease urban-rural wage differentials and encourage migration, it might well be sound policy to choose the rural O})tion depending on what can be said about the margina1. social product or increased employment in rural vs. urban areas. In this respect, increased urbanization inTolTee higher social costs or infrastructure, services and o.tten housing. Certainly nothing should be done which wuld deliberately worsen the urban problem ±( For an analysis see ECLA, "Creation of Empla.yment Opportunities RelAtive to Supply," Urbanization in Latin America, pp. 118-132. - 2/ M.P. Todaro, "A Model of Labor Migration and Urban Unemployment in Isss . Developed Countries, " American Economic Review, M!t.rch 1969, p. ·147. )/ Frank, op. cit., pp. 270-2n. - 39 - without a clear corresponding benefit. In short, the problem will be to achieve growth and employment objectives with a 1111tually consistent urban- rural development strategy. Such a strategy 1111st recognize that the trend is tor modemization and rising productivity in agriculture. This 1118ans that iileritably fewer people w:l.ll be needed in the direct process ot pro- duction on the tal"lll. The l'UJ"al eoonom,y is discussed later in the context of an urban-rural strategy tpr cievelopment. D. Do Migrants Cause Instability? The presence of a large and growing mass of Underemployed, uprooted migrants living in miserable conditions on the fringes of a relatively efficient urban society, baa o£ten been regarded as a serious potential source of violent protest and revolution. Yet according to a number of surveys, slums such as the bustees of Calcutta, the barriadas of Lima and the favellas of Rio may be a greater source of stability than would be expected. Their numbers have been increasing dramatically. For example, in Lima the barriada population rose from 10% to.20% of the population between 1958 and 1964, and Rio's favellas increased from 8.5% of the population in 1950 to 16% in 1964.Y In Calcutta the bustees constitute one-third of the land area. Joan Nelson and others point out that there is no evidence that migrants are particularly politically disruptive. Migrants, as we have mentioned usually come from nearby areas and follow a steplike procedure from small to large town~. The so-called tight structure of rural society has been partially eroded already in most areas through contact with modern influences so the transition to urban life is not that severe. Further, migrants are usually young and adjustable and receive help from an extended family circle. They also find their first jobs relatively quickly. Living conditions are probably not worse than those in rural areas and being politically inexperienced, it is not surprising that they 1/ Harley L. Browning, "The Demography of the City, 11 The Urban Explosion in Latin America, p. 101. - 40- are not sources of' political action.ll In Africa, Chtkind concludes "so long as the unemployed are dependent on their close and far kin and frierxls, and as long as they can, i£ they wish, return to rural areas to (tey to) obtain a living from the land, they remain immune to the intensive social and economic pressures which might make them a more active political community."Y Germani in a study of Buenos Aires confirms that emplo,yment expecta- tions were usually fulfilled although housing and social disorganization were problems.JV Resentment, without violence, was shown towards recent arrivals in Buenos Aires on the part of established low income groups. In Bombay, resent- ment of newcomers of different cultural backgrounds reportedly is a serious cause ·or strikes and violence on the part of employed workers. ~other version of the instability argument claims that migrants become frustrated by conditions over time and are thus potential sources of instability. In Calcutta, demonstrations. by the middle class are reportedly more violent than those b.1 the predominantlY working class. It seems that as long as people are in the marginal categor,y without access to urban institutions and processes, they are unlikely to be sources of open discontent no matter how long they have been in &n urban center. People at the lowest level of skills and income are concerned with survival and do not aspire for the impossible. 1/ Joan Nelson, "Migrants, Poverty and Urban Politics in Developing Nations,'' December 9, 1968, PP• 1-13, (Unpublished) and "Urban Growth and Politics in Developing Nations: Prospects for the 1970's, "Pearson Conference, Doc. No. 31, Williamsburg, Va., February.l970. y Peter Gutkind, "The Poor in Urban Africa: A Prologue to Modemization, Conflict, and the Unfinished Revolution1 11 No. 12, Center tor Developing Area Studies, McGill University, Montreal, p. 890. V Germani, "Inquiry Into the Social Effects of' Urbanization in a Working Class Sector of Greater Buenos Aires," Urbanization in Latin America, pp.228-2~0. - 41 - MB~ do make progress and others find other ways than political action to express frustration such as crime. Finally, the urban poor are particularly vulnerable to police control as they lack the necessary contacts and influence to avoid it.1! Turner and Mangin argue persuasively that successful squatters who have acquired effective tenure of land tend to be mainly interested in consoli- dating their housing investment and in identit,ring themselves as respectable, property owners with a stake .in society.!! Admittedly the conditions for pro- .moting self-improving squatter settlements are not present everywhere, nor has the process been thoroughly studied in all regions. Nevertheless, numerous cases exist to provide a good deal of information to those concerned with the orderly integration of masses of migrants into urban life. The fact that the large and growing marginal population of cities are less politically volatile than had been feared does not necessarily mean that the process of urbanization will be without strife and disruption. It simply means that policies to prevent the rural marginal populations from migrating should not be based on the potential instability argument. 1/ Nelson, "Migrants, Poverty and Urban Politics in Developing Nations," p. 11. .See als~ Oino Oermani, ''The City as an Integratinr, Mechanism," The Urban Explosion in Latin America, pp. 185-187. Y J. Turner Rnd W. Mangin, ''Progressive Architecture," May 1968. Also J. Turner, "Uncontrolled Urban Settlement: Problems and Policies," Inter- Regional Seminar on Development Policies and Planning in RelRtion to Urbanization, United Nations, University of Pittsburgh, 24 October - 7 November, 1966. - 42 - High levels of urbanization in relation to development, or '~yper- urbanization," can be shown to be a constructive process for accelerating political development and transferring power from traditional elites to the emerging middle and working class groups. This may indeed involve severe dis- ruption. In effect, the large city is the mechanism which best trAnsforms societies from traditional to modern ways. In the process, a large unorganized urban proletariat may be excluded !rom effective participation in economic and political life until it can shed its marginal status.11 In shor~, the growing marginal population is not a critical problem requiring urgent solution, but one which can be dealt with by sound long-run programs of urban economic and social development. E. Is Overurbanization t.he Issue? A number of writers who would influence national migration policies resort to what may be termed the overurbanization argument. TYpical of these is Barbara Ward's article entitled, "The Cities That Came Too Soon." She expresses the anguish of one who is appalled at the miser, of life in the great cities of the developing world, and examines the symptoms of their groWth with- out however suggesting a remecty. "Urban growth in the developing world today is not so much a measure of healthy, inevitable processes o! modernization as a pathological acceleration of urban 'cell creation' which could put whole societies into a terminal crisis of sooial and economic disintegration. ,y Y A case in point is Chile. See J. Friedmann and T. Iacld.ngton, "Hyper- urbanization and National Development in Chile: Some HYpotheses," Urban Affairs Quarterly, pp. 3-29. Sf Barbara fthrd, "The Cities That Came Too Soon," The Economist, December 6, 1969 - 4.3 - The use. of the word "pathological" suggests deviation from the normal. Yet i f Barbara Ward regards the city slums as unbearable, what would she s~ of their rural counterparts, whiCh as we suggest may be little better in ~erms of living conditions)/ In fact, the argument against overurbanization must be seen in the context of overall population growth rates which exceed the growth of new resources through their process of economic development. The problems must be anAlyzed in national terms and not in terms of the urban and rural symptoms. The idea that there is a norm and that less developed countries are overurbanized is essentially a conclusion drawn from the industrial histor,y of the developed nations. There is, however, no theor,y or credible body of analysis which demonstrates that all countries must follow the same path of urbani1ation and industrialization. In fact, as Sovani has shown, even the developed countries did not all follow the same path. In 1695, Sweden had ~ppro.ximately 6% of its population in cities over 100,000, with almost 45% of its labor force in non- agricultural occupations. SWitzerland had a far higher percentage of non- agricultural employment at that time with a lower level of urbanization.!/ 1/ See Lowdon Wingo's conunents in this respect on Philip Hauser's paper, "Urbanization Problems of High Density Living," World Population - The View Ahead, ed. R. F.Rrmer, J. I~ng and C. Stolwitz, Joint Development Research Center, Indiana University 1968, p. 226. y N.V. Sovani, "The Analysis of Overurbanization," Economic Development and Cultural ChanJI!, Vol. III, No. 2, January 1964, pp. 117-122. - hL - Miss Ward fUrther makes the statement that in the large tertiary sectors of major cities thia "type of employment ••• keeps the man from absolute starvation but contributes all but nothing either to the ~cono~'s development or to human acquisition of skills and confidence." Yet, the marginal product of migrants in urban areas must be higher than their marginal product in rural areas, as measured by wages received and satisfaction realized, as otherwise the migrants would return to their former homes. As we hAve seen, the lure of higher incomes is the factor most responsible for migration, and migrants who come to the cities are in the main satisfied. There is also the well established fact that urban per capita incomes are some multiple of average incomes in most countries particularly those of the less developed world. In the absence of more proof than has been presented, we must assume that the marginal productivity of labor in urban areas is higher than in rural areas. Finally, the overurbanization argument is narrowly based on observatioru; drawn only from one branch of the discipline of economics. It omits the idea that urban centers may play the role of dynamic center$ of social and cultural change in the absence of large-scale industrialization. The less developed world may well adopt different patterns of urbanization. There is no reason to believe, for example, that the primate cities of Asia are not playing a dynamic . 1/ role in bringing about cultural change and modernization.- . In brief, what is called as overurbanization in one context could just v as well be termed as overruralization in another.- Further, experience with programs to steer urban migration is limited. Even it' we wanted to, it is unlikely that we would be able to control the pace of urbanization to any great extent as we do not know what is involved in the way of managing such a social mechanism. Finally, there are substantial economic forces creating long-run urbanization trends with which we must come to terms in development policy. y Ibid, pp. 120-121. 'Y See The Urban E!cplosion in latin America, p. 74. · - 45 - II. URBlN SCALE FACTORS IN NATIONAL GRO'..tl'H A. The Costs of Urbanization~ Can they be ManAged? The process of urbanizing substantial portions of the population of the less developed countries will require corresponding investments in urban social overhead and service systems. These investments will be larger than those required if the same population were to remain in rural areas. RoAds, transport~tion, sanitation and water are different in form and cost far more per capita in urban than in rural areas. In practice, rural areas also receive much less in education and health expenditures per capita. FUrthermore, large cities usually have higher public outl.,s per capita than small ones~ The trend, as we have shown, is for urban growth to concentrate in the big cities. At the broadest level of aggregation (localities of 20,000 and over) the urban population of the less developed world may increase from 409 million in 1960 to close to 800 million by 1980 and to 1.4 billion by 2000. This implies that the investment in urban infrastructure as well as the use of land will probably have to double by 1980 and more than triple by 2000, depending on what is assumed with respect to the efficiency of capital, the standards of public services per capita and the densities of settlements. For urban economic policy, problems arise i~ determining the size of the existing capital stock in urban public investment, the capital investment required for further urban growth, as well as the impacts of social capital on growth and prodUctivity. 1. The Case of Latin America Some understanding can be acquired !ron what happened under conditions - 46 - .1/ of rapid urbanization in Latin America in the four decades after 1920.- From 1920 to 196o the population in latin America rose from 90 to 213 million and the urban population ( 20, 000 and over ) from 1$ to SS million. The percentage urbanized rose from 17% to 26%; resulting in 19 million more peoPle living in urban areas than would have occurred had the same percentage urbanized prevailed throughout this period. In the period 192$-1956 the fixed reproducible capital of Latin America per capita remained virtually stable at some $6o0 in 19$0 prices while CIDP per capita rose from $168 to $285, A rise of some 60%. Rapid urbani- zation· took place during this periOd with constant capital investment per capita and rising income per capita levels. The gross product per unit of capital invested rose by 40% in this 2/ . . period.- From 1940 to 195$ employment in manufacturing and construction rose b,y 6 million from 13% to 18% of total employment and the gross product per worker rose b,y about one-third. Government and other services emplo,yed an additional 7.5 million in this period with an estimated lO!C rise in product per worker. The share of manufacturing and construction rose from 19% to 23% ot gross product, while that of agriculture fell. Thus the shift of capital and labor resources to industry increased the overall productivity of both capital and· labor. l/ See Alexander Oanz, "Problems and Uses of National Wealth Estimates in La tin America," The Measurement of National Wealth, ed. R. Goldsmith and C. Saunders, Income and Wealth, Series VIII, Bower & Bower, London, 1959, pp. 217-273. Y Based on data from Argentina, Brazil, Colombia and Mexico which contain ?a!o ot the gross product and stock of capital of all of Latin America. - 47- Certain stages and basic factors can be identified in this 40-year period which will enable us to gauge the impact or these broad developments on investment requirements for basic social capital. In the 1920's Latin American economies were dominated by export sectors and imports were relatively free. There was little protection for domestic industry and much excess capacity. The situation changed in the 1910's when demand for domestic-manufactured products grew under autarkic conditions. This required expanded investments in road, power, health, education and sanitation in growing urban areas. During World war II the unavailability of foreign manufactured consumer goods, capital goods and food gave a strong st~mulus to domestic manufacturing industry and agriculture. Production beyond capacity took place and machinery and equipment were used beyond their estimated normal useful life. After the War, huge reserves of foreign exchange supported a major expansion or domestic industry. Domestic demand grew as a result of rising per capita income and better income dis- tribution. Investment in basic social capital, particularly in transport and energy, lagged because of the poor condition of government finances. This condition was related to the need to subsidize exports, to the high cost of poorly planned social secur1ty programs and to rising defense outlays. The shortage of basic social overhead .facilities became a U.miting factor to industrial e">Cpansion. Private f"oreign investment we~t into manufacturing for the domestic market rather than public utilities where public policies caused lower rates of return to prevail. The deficiency in social capital was aggravated by much misdirection of investment and an underestimation of the demand for such -48- investment. Further, the rapid growth of urban centers with exPandi.ng per capita incomes led to an increasing demand tor urban amenities and services such as roads, hospitals, schools and recreation. Thus a substantial diver- -., __ co.:;: -· sion or investment frOm more directly productive purposes took place. Two count.ries illustrate the problem to different degrees. Colombia experienced an annual rate or growth in per capita gross product or 1.9% from 1929 to .19~~' compared to about 0.5% ror Argentina. Argentina started the period with a relatively high quality or urban amenities and service systems. . . . Argentina's structure of social overheads required a large proportion of gross fixed investment to simply replace the existing stock of obsolete capital. Colombia, which developed later than Argentina, concentrated a greater pro- portion of its investment in sectors, such as iron and steel, with relatively .. • high output to capital ratios. Further, it enjoyed the benefit of higher prc;>ductivity through the adoption or the latest technology. Strategic errors in the allocation of resources also contributed to Argentina's relatively poorer economic performance. In the postwar period ;o% of investment was allocated to government administrative, social and military services compared to half this percentage in 1944.1/ Allocations to directly productive investment decl~.ned correspondingly, particularly to transport, power and heavy industry. In contrast, Colombia's allocation of investment in manufacturing, mining, transport. and electricity rose by threefold from 1925 to 1953 and that in agriculture, housing and government administrative and social services by only 7CYI,. 1( Ganz, op. cit., p. 250. - 49 - A comparison of capital invested per worker in the main sectors as a percentage of capital invested per worker in the whole econ~ for Argentina, Colombia and Venezuela showa that Colombia arid Venezuela invested relatively more per worker in basic manufacturing, transport, communication, electricity, and enjoy'ed relatively higher returns per worker in directly productive sectors such as manufacturing and mining)/ The difference in the capital stock structure of the three countries is compared to the average percentage allocation of resources in the following table.· Percentase Distribution of CaEital Stock and Investment Arsentina . Co~bia Venezuela I !mO (1950-54 > 122j ~945-53) 12M (1956-58) Agriculture 14.7 (9) 35.6 (25) 13.5 (9) Manufacturing, minerals, 15.3 (19) 16.5 (23) 23.4 (39) petroleum & cons·truction .._ Transport, communications 12.7 (16) 14.1 (19) 13.7 (19) and electricity Housing, commerce, services 57.3 (55) 33.8 (33) 49.4 (36) and government Total 100 (100) 100 (100) 100 (100) -- --- Sources Banco Central de Venezuela, Memoria, 1959, p. 417, 421. The figures show thAt Venezuela and Colombia spent only about a third of current investment in housing and services compared to over a half in Argentina ~ devoting a higher proportion of capital invested to higher productivity uses, !/ Banco Central de Venezuela, Memoria, 1959, pp. 423,424. - so - Colombia and Venezuela were able to achieve much higher rates of growth of per capita income. Such an effort required a tighter control of less productive investments. Venezuela and Colombia experienced rates of urbanization from two to two and a half times faster, respectivelY, than Argentina. For the continent as a whole the current shortages or basic social capital tor productive purposes will tend to raise the per capita requirements of fUture urbanization. To this must be added the fact that large cities are more costly' to buUd and operate per capita i f only because star1ards per capita are higher and that political pressures operate to satisf,y the demand for urban amenities and services. The differences in allocation may be explained in part l;ty the higher per cap;Lta income level in Argentina. This would normally be expected to lead to higher standards of urban services. 2. The Case of India A rough computation for India gives some idea of the order of magni- tude of resources required to cope with the investment needs of urbanization over the short run. The calculations were originallY done by Britton Harris in the earlY 1960's and refiect price levels of that period. The urban ·growth figures ~ve been adjusted upwards by one-third. Y India's urban population (20,000 and over) is estimated to increase currentlY by some 60 million a decade. In terms of a five-year plan period, 30 million more urban dwellers will have to be accommodated. This implies some 12 million more households and about 20 million new jobs. On the basis of a reasonable average cost of housing, including both self-help and publie, work places and urban services, a cost of some 28 billion rupees may be estii'IBted. 1/ - Britton Harris, ''Urban CA!ntral jzation and Planned Development," . India's Urban Future, p. 263. - Sl - The 1969-74 five-year plan calls tor total investments, public and private, ot Rs. 2hl billion ot which some Rs. 31 billion il!i allocated to housing and urban development, water supply and aallitation.!l B;y updating tor estimated price changes of some SO% trom 1963 to 1969, the percentage of investment devoted to urban investments, as defined by the five-year plan, would rise trom 1.3% to 16% of total public and private sector outlays. .About BS% ot the Rs. 31 billion planned investment in urbani- zation is in the private housing sector, consisting to a lRrge extent of self- built type housing. Thus in the Indian context the cost ot urbanization appears to be manageable at present st-andards of public services at least in. the near term. For the private sector the large proportion of low standard housing and work places wUl probe.bly continue. Gi.ven the fact that selt-buUt construction involves substantial elements of new capital mobilization, we would not expect urbanization financed in this way to create severe competition with more direct- ly productive investments. This tentative conclusion is almost the opposite of that reached by Harris with respect to the implications of urban investment needs in terms of the second Plan which, however, involved a tar lower level of total investment and when his calculations showed urban investments, as defined, constituted a third of total outlays. 1! The 3,100 crores does not include such basic social capital as transport and communications nor such services as health, education and social welfare. Government of India, Fourth-Five Year Plan 1969-74, p. 48, (Draft). - 52- In an estimate covering the period 1960-80, P. Pant concludes that a feasible investment program to accommodate the costs of urbanization can be projected and would require some 2$% of total investment assuming an annual $.)% growth r~te of national income. This investment would cover · existing quality housing, utilities, education, heal~h and recreation and ''bring about some measure of improvement in urban facilities." Pant argues for lower cost urbanization. His estimates are based on the addition of 110 million people to the urban population in towns over 5,000 in the period (1960-80 ), which is somewhat lower than our own.Y 3. Do Bis Cities Cost More? A study being done in the Econanics of Urbanization Division on public current .and private production costs per capita in 26 Pakistan towns - . - 2/ - . shows wide variations.- ?ublic current costs per capita in west Pakistan ra~e from 20 rupees in Peshawar to 41 rupees in Karachi. In East Pakistan costs ranged from 12 rupees for Dacca to 31 rupees for Chittagong, a rising industrial 1( Pitambar Pant, rurbanization and the Long-Range strategy of Economic Development," India's Urban Future, pp. 186-190. y Samir K. Bhatia, ''The Public Cost o,f Urbanization in Pakistan," IBRD, 1970, (Draft). - 53 - center. In general, the bigger towns incur higher per capita costs. Centers beinp,: promoted by the Government. to achieve better re~ional balance may require exceptionally high current per capita expenditure. For example, Quetta in the far ·weste:rn part of the country has a per capita cost of 52 rupees mainly due to buildOng of excess capacity in social capital invest- ments in orcier to stimulate industrial growth. In this case the stimulaUon has not worked and industrial growth is lagging. Our~udy also shows that pri va.te production costs PEl' -worker in key industries 'lll'e often lower in big towns with high public costs per capita. Karachi's cost of cement is far lower t.han in an internal city li.ke Rawalpindi.. Karachi 1 s cost of vegetable oil is at least 10'/. lower than Hyderabad and Peshawar.. The cost o.r jute processing is higher in Dacca than in Chittagong although Dacca's public cost per capita is far lower. These figures can begin to provide the framework for developing location criteria and policies for industry which will lead to a marginal least socjal cost location for new industrial plants and thereby contibute to a more efficient pattern of tndustrialization in terms of returns on national capit.al and the employment of urban labor force. The pattern in Pakistan shows that the towns generating hieher per capita incomes also require higher per capita outlays. In part this is due to the ract that the demand .for het.t.er quality services is higher where hi~her :i ncomAs are earned and in p:::trt due to the complexity of publj c systems in large rather than small towns. In Karachi per capita public costs rose from ?.~ to It? rupees between 1 Q67 and 196l3 almost entirely due to a Government decision to improve urban services following aeriou• student-worker riots and strikes. - 54 - 4. Concentration, Decentralization and New Towns The cost figures as such do not tell us much about the relative merits of decentralization versus concentration. The evidence available suggests thnt the larger towns enjoy economies of seale for public investments, and generate external economies tor the urban eeono~ in sufficient amounts to yield r,reater net returns per unit of investment than do small urban centers. In India although per capita public costs per worker were higher in Bombay than in Orissa, this was more than eompensa ted for by the higher per worker product i.1 Bombay. In Japan an econometric stu~ shows that GNP is higher under conditions of con- centration than deconcentration. FUrther, the most densely populated urban areas produce lower social overhead costs per capita and higher incomes than less dense urban centers.!/ New tcMts which have been attempted in India and Pakistan show, upon preliminary examination, fairly high per capita investment costs.~ Such costs can usually only be borne. for showcase cities with substan- tial public administrative fUnctions. A study of new towns in India found that the decentralization policy was based on the anti-city sentiment of decision- makers and that capital costs associated with these towns were excessive in . . ~ relation to their financial means to cover current and capital costs. !/ Koichi Mera, "On the Concentration of Urbanization and Economic Efficiency," IBRD, October 13, 1969, (Draft). g( See Catherine Bauer WUrster, "Urban Living Conditions, Overhead Costs, and Development Patterns~ 1 ' India's Urban FUture, pp. 290,291. Research is needed on comparative urban overhead costs for various scales of urban size and for different standards of services. Until better cost information and·investment location criteria are available, forecasting the cost of urbanization will remain a hazardous undertaking. 21 Ved Prakash, Financing New Towns in India, Cornell : .&':\: Unive~sity, Doctoral Thesis, June 1966·;~~. 1-~35. - 55 - Alonso criticizes new towns as a solution for U.S. urban problems in that they would not permit the necessar,y complementar,y a~ specialization of .• fUnctions required of modern cities and would not result in a~ discernible .. 1/ cost advantages.- 5. Can Cities Keep Growing? The issue relates to whether we can define an.optimal sized city and whether it is possible to devise economic alternatives to continued central city growth. The next section will discuss this issue in connection with external benefits. Empirically it is difficult to demonstrate th~t cities do reach optimum aize. Urban concentration must achieve a certain "critical mass" for external economies and internal economies of scale to raise productivity and foster more growth. There are probably increasing returns to urban scale over a considerable range as.marginal productivity per capita probably rises faster than marg~al costs. It will take carefUl study of individual situations to determine at what point the external diseconomies of city growth in the form of pollution, traffic congestion, and crime, for example, become sufficiently internalized by the private sector to cause new investments to leave.· Even where this appears to be.happening, as in Sao Paulo and New York, the center may simply acquire new functions while industrial activities seek new locations on the periphery. The new peripheral location would still enable an enterprise . to enjoy most of the economic advantages of big city location • An empirical study or this issue in the context of the regional-urban development of Northeast India concludes that "the c3se is strong for deferring 1/ William Alonso, "What are New Towns For?", University of California, Berkeley, August, 1969, (Drat't). - 56 - the· idea of findinr. an optimal city si.~e in favor of determininp, merely the minimum efficient si?.e for new developments.'' The study found that t.he unit costs of an increment of public infrastructure for a lnrr.e industrial expan- sion are virtually the same for cities of population sizes 130,000 and 300,000. The study cautions ar,ainst the relevance of the size t1ndinr. for other countrie3 1/ where conditions are quite different.- The-. grQwth of .::!ties should be promoted within a size ranP-e that is efficient in terms of-unit infrastructure costs and that is abl,_ to achieve positive agglomeration economies in terms of the size of industrial structure. This will require a carefUl examination of the location of existing cities, the cost of equipping them with suitable infrastructure, the linkages between firms in the industr/ in question and the complementarities among cities of different sizes in the region. The developtnent· issue cannot be put in terms of centralization versus decentralization or urban versus rural. In referring to the high incomes and high proportion of industry, transport, commerce and administrative jobs in urban centers in India, Britton Harris writes, "to a very considerable extent, the price which must be paid for urban development is a price paid for conduct- ing these activities at all." Y The real issue is what kind of urban rrowth and where· it will be located. !/ R. Morse et al., "Costs of Urb:m I:ritrastructure for Industry as Related to City Size in Developing Countries: India Case Study," Stanford Research Institute, 1967, pp. 10-11, 18. Y Harris, India 1 s Urban FUture, p. 266. - 57 - 6. Can the Costs be Managed? It appears that big concentrated cities are more efficient than smaller less concentrated cities in that the cost of social overhead per capita is lower and income per capita is higher in the bigger cities. Nevertheless, fUture urbanization is likelY to require a larger share of the nation's capital supply because a larger proportion of the population will be living in cities. Most countries will find it necessar,y to devote more capital resources to pro- viding at least minimum levels of settlement infrastructure and social services as standards cannot be allowed to decline continuously. The private returns of migrants should th~s be weighed against the social cost in terms of investments which are diverted from other more productive social purposes (excluding income ~roups who pay for the services they demand). ~ile the evidence of the past may seem to be inconclusive with respect to the fUture cost of urbAnization, it does suggest that urbanization can be managed, indeed, it must be i f development itself is to proceed. _The critical variables in th·iR respect are the rate of growth, the rate of capital formation 1 and the standards of urban overhead and service systems which are to be provided.- !/. Barbara Wlrd, "The Cities That Came Too Soon," The Economist, December 6, 1969, p. 57. This conclusion may represent A more balanced approach than that of Barbara ward who noted that urban population projections for the period 1960-1980 may require "the equivalent of all the cities built in the developed w~rld over centuries of urban growth. The gigantic scale of such a task, its virtual impossibility of fUlfillment, are enough to sugr,est that urban growth ••• " is " ••• a patholop,ical acceleration of urban cell creation." - 58 - B. The Economic Development Benefits of Urbanization This discussion will focus on the process through which urban centers o! various sizes and patterns contribute to the ~rowth of output and employment in a nationa~ econ~. The analysis will outline a conceptual framework for the urban area, the regions and growth poles. The primary purpose is to explore a number of important policy issues in urban-regional economics sue~ as ef- . 1/ ficiency and equity in the allocation of resources and economic returns to city size.- 1. The High Correlation Between Urbani7.a tion and Ecor 11Tiic Qrowth The high degree of correlation between urban growth and national growth has been commented on by a number of writers. Their empirical findings suggest some of the dynamic factors relating to urban growth and economic development and some of the relationships which may be irrelevant. Russell, for example, demonstrated the relation between urbanization and growth by re- greasing the percent of urbanization in 72 countries with per capita incomes using data of the mid 50's. This study produced an index of correlation co- efficient of .69 at a high level o! significance.ll 1/ Urban economics has ~nly recently come to be recognized as a field of a.cademic specialization. Formerly, it vas considered a part of rerional economics or regional science which itself was described as a "new field of specialization" in 1963. The theoretical roots of both fields may be found in location theory, multiplier theory, input-output analysis and mathematical programming. J. R. Meyer, "Regional Economics: A Survey," Surveys of Economic Theory: Growth and Development, Vol. II, st. Martin's Press, N.Y., 1967, pp. 250-257 and I. Hoch, "Prorress in Urban Economics," Resources for the Future, 1969. y Bruce M. Russell, "World Handbook of Political and Social Indicators," Yale University Press 1964, Part B, cited in l"riedmann and Iacldngton, op. cit. , p. 6. - 59 - The degree of concentration of per capita income can be related to the stage of economic growth. In examining historic income per capita dif- ' ferentials in Italy, Brazil, the u.s., Canada, Germany, Sweden and France, ldlliamson found that "increasing inequality is generated during the early development stage, while mature growth has produced regional convergence or a reduction in differentials.1f Alonso attributed the excessive concentration of activity in countries at the earliest stage of develop~ent to the fact that their industries are new and rely heavily on the externalities of the city.g/ In a study of city size distribution and economic development in 95 countries Berry noted two patterns of urban structure. One in which urban centers are of regular graduated size and another in which one or more primate · centers dominate with a deficiency of intermediate centers.JI He concludes that there are no relationships between "city size distribution and either relative economic development or the degree of urbanization of countries, although there is a strong association between urbanization and economic de- velopment." The primate distribution is associated in small countries with political or economic dependence on an outside country, dual or multiple ]/ Williamson, "Regional Inequality and the Process of National Development: A Description of Patterns," Regional Analysis, ed. Needleman, 1968, Penguin, p. 155. Y William Alonso, "Industrial Location and Regional Policy in Economic Development," Working Paper No. 24, Center for Plannin~ and Development Research, University of California, Berkeley, 1968, p. 27. J/ Brian Berry, "City Size Distribution and Economic Development," Economic Development and Cultural Change, Vol. IX, No. 4, Part 1, July 1961, pp. 573-588. - 60 - colonial economies, centers of culture and administration and the capitals of empires. Primate centers may occur where economies of scale make it un- economical to have more than one major city. Countries with more than one large city m~ have city regions like New Zealand or specialized complementar.y cities like Sweden and Japan. Where complex forces affect the urban structure a more normal distribution arises. "Countries with the lovest degrees of primacy include many with considerable industrialization and also those with .long urban traditions and histories of urbanization." Belgium and the U.S. are in the former category and India and China in the latter. Higgins notes the strong correlation between industrialization, urbani- zation and economic development. He ascribes differences in productivity and incomes to occupational differences between regions, even with a given country. He cites Schnore's rank correlation for 69 countries of urbanization with ten indices of economic development. The correlation coefficient for urbanization and emplo.yment in nonextractive industry is .77. It is high for the other in- dicators as we11.1/ Higgins, however, stresses the complexity and diversity of the development process and concludes that there is no proven causal relation- ship between the variables. Y Benjamin Higgins, "The City and Economic Development," The Urban E!cplosion in Latin America, pp. 126, 127. - 61 - 2. How Urban Economies Wbrk in the Nation a. Export Base 1 Service Base and Orowth 1/ The city may be viewed analytically as a product of its export base;- smaller metropolitan areas produce as much as half their goods for sale to areas outside their borders. In the short run the industrial composition of an urban economy's export sector is unlikely to undergo much structural change. Thus demand for export products is paramount in determining urban growth and employ- ment. In the longer run however, the local service sector may be seen as the critical input for the.export sector. Its efficiency is the key to the survival of export·firms. A growing econam7 presents a tar greater threat to a small single export industry toWn than to a larger more diversified urban center. Technological change often fosters inter-industr,y linkages, clustering and complementar,y labor demands, fUrther enhancing the attractiveness of the larger manufacturing center. As growth proceeds export specialization in small towns is replaced by the establishment of an export complex. With it the productive process widens to other products and deepens by completing forward and backward linkages in the production chain and leads to import substitution. At each stage the local economy acquires skills, services, social overheads and entrepreneurial talent. Whe~ the stage of the regional metropolis is reached, a "node" is created which both connects and controls former rival clties. '!be regional 1/ This analysis is largely based on the conceptual framework pioneered by Wilbur Thompson. See Wilbur Thompson, "A Preface to Urban Economics," Resources for the FUture, 1965 ______. "Internal and External Factors in the Develop- ment of Urban Economics," Issues in Urban Economics, ed. H. Perloff and L. Wingo, Resources for the Future, 1968, pp. 43-80. - 62 - metropolis exports services as a major part of its activity. Whatever the cause of one city gaininp, dominance over others, such dominAnce is usuplly maintained,as the larger, dominating city generAtes external economies in a number ot sectors particularlY in services such as finance and administration. The large city is a diversified productive machine made up of a blend of growing and declining industries. The linkages between firms in such ~ center make it difficult for any of them to break out of the system. The big ·cities represent enormous st.ocks of fixed capital in social and ?rivate overhead. Even i f these cities' productive enterprises are obsolete, their current costs of public utilities and services are likely to be low. '~o n~tion is so affluent that it can afford to throw away a major city." 1/ The great city is in itself an important market which attracts productive activities and being a transport hub becomes the focal point for activities throughoUt the nation. b. Innovation, Incomes am Industry The innovative fUnction of the l~rge city has important income effects. New, more sophisticated products require more skilled workers who can command higher wages. In turn the innovating large city spins off the traditional low wage ·industries. The smaller to~ receiving such in!iustries may well suffer from built-in growth limitations, as a result of low wage rates, limited skills and slow growth indu~tries. The fiscal incomes of such lag~ing centers are in- sufficient to fin~ce the better schools and training facilities needed to match the quality of the labor force in the large innovating centers. }/ Thompson, "A Preface to Urban Economics," p. 23. - 63 - This theory makes the distribution and stability of income in an urban area a function of industry mix. For industries such as durables, clustering takes place reflecting external economies; in some consumer r,oods 1/ industries such as textile single industry towns can survive.- The growing importance of the multiproduct, multiplant firm in an advano.ed country reinforces this trend, as· the least cost location for such organizations is a large city. In one sense the urban economy may be viewed as packages of industries in space. The nation's system of cities is a function of its system of indus- tries. Large cities and industrial complexes result from ~conomizing on the hiP..h cost of transportation of intermediate products in heavy industry. The external economies of large cities in services and overheads attract new in- dustries. With its higher income levels and demand for labor the city attracts -workers from outside. It also attracts the poor migrant whose marginal product may be low thus accounting for the paradox of poverty amongst plenty even in the affluent cities of the west. 1/ Not all analysts agree that we understand agglomeration fUlly. A path breakin study, after reviewing the literature, concludes: "not enough is known about agglomeration to indicate in what degree it represents a response to economic and noneconomic forces; whether there are alternative ways of achieving the economies that ngr,lomeration may brin~; or what ~oats or economic losses mi~ht result from measures diminishing the tendency of industries to agglomerate." See R. Morse et al., "Costs of Urban Infrastructure for Industry as Related to City Size in Developing Countries: India Case study." - 64- The innovating large city has no choice but to becOme bigger in size as city size and industrial growth are closely related. The large city has great advantages in transportation and utility systems, academic research complexes, skilled labor and training facUities. These economies ot scale are internalized by private firms in the form of lower costs, while the social costs of congestion and pollution are borne by society as a whole. These external diseconomies are an increasing fUnction of city size but are not usually reflected in factor market prices. lilen such diseconom:..es are .tUlly reflected in the marginal cost or production in urban areas they may be a limitation on city size. c. Manasement as a Limiting Factor It would appaar that the ~jor scarce factor in large metropolitan centers is effective management. As land, labor and capital do not conceptually pose limits on the size of the- city perhaps the real limitation or disecono~ of scale is managerial. This may be defined as the point where the unit cost of public services starts rising as the quality of services starts declining~ If these marginal costs to local export firms are rising more rapidly than marginal returns, a fact limiting growth will have appeared. This probably depends in large part on how successful mOdern management techniques and the latest tech- nology in such fields as transportation are. It is a central issue in public policy in the less developed world and concerns not only technical management efficiency, but also the effectiveness ot the political process in coping with the changing times as well as the ability of major grdups and classes to co- operate in an urban environment. - 65 - .·. ). The Regional Economy: National Growth Versus !qual Growth A city is not an isolated economic unit. It is a sub-system of a ' national econ~. A nation may be said to conaist of a system of cities in which each performs tunctions tor its surrounding regions. Cities may also be regarded in a hierarchal framework with certain primate cities performing functions tor the entire economy and other cities dependent on the primate city in varying degrees. The regional development issue may be discussed most simply in terms of efficiency or national growth and equity or a more equal distribution of income. In many countries policies designed to limit the size of primate cities on economic grounds complement policies designed to promote development in backward regions. Alonso points out that such policies may otten be detri- mental to economic growth, that efficiency and equity policies may therefore be in conflict.!! The ambiguity of a region with respect to scale is a problem. A region in one country may be as big as a country elsewhere; Chile's population is halt that of the New York Metropolitan area. Growth pole theory stresses the key role ot regional urban centers in equalizing income between regions under a strategy termed "concentrated decentralization" by Rodwin. Yet the inequalities in income within regions may be increased so that the average 1( William Alonso, "Urban and Regional Imbalances in Economic Development," Economic Development and Cultural Change, Vol. 17, No. 1, October 1968, pp. 1-14. See Part II AS for a discussion of the-empirical diffi- culty of defining an optimal sized city. - 66- distribution of a nation's income ~ be little improved. "In the absence of empirically determined scales of relevant sizes ot regions and growth poles, these concepts have limited relevance as a guide to poli~ and action.11 Alonso questions whether there is any evidence that big cities necessarilY experience rising marginal costs to scale. Even if marginal costs are rising, the net marginal product per capita may still be increasing, thus yielding increasing returns to urban seale. On the other hand, regional equalization policies have not been empirically tested for consistency with maximizing national growth. Location theory holds that investment proportional to resources and population in the presence of unexploited regional resources will yield the fastest rate of national economic growth. Yet the conditions assumed b,y classical theor.y are not present in the less developed countries. Transport, tor exalliple, is not fully developed but usually converges on the port cities. Entrepreneurs and managers are reluctant to move to .the bush. Investment in remote regions in- volves great uncertainty and there is good reason tor businessmen to locate close to the government decision-making apparatus. In the short run a poli~ of deciding on a project-b,r-project basis may yield· one location strategy on efticien~ grounds. In the longer run national growth plans may require a grouping of projects which will create a "critical" mass sufficient to start a process of self-sustaining growth through induced external economies, although the human and technical conditions for !! Ibid, p. ). - 67 - •. achieving this mass are not well lmown. In the long run, policY' 11Dlst allow for tundamental transformations in population, transportation and virtuallY" all the fiXed paraMters or incremental project analysis. Studies show that primacY" or concentration rises in the earlY" stages ot growth. As develoPment proceeds, a trickling down effect takes place. Interior areas are opened and transport and education become available, "making externalities more nearly equal everywhere." Al.onso mentions that . "an invisible hand 1D8Y be at work, and, given world enough and time, it ma7 reconcile the etticieno7 and equitY" goals." He nevertheless concludes that dispersal theories have yet to be proved and that the ambiguous definitions ( equity •ke· dit.ficult the .measurement of tradeoffs with eftic.iency or income.!! a. The Strategy or Concentrated Development In an earlier work, Il'riedmann and Alonso consider regional develop- ment in a more pragmatic vein as a longer run policy issue. They reject the choice between social equity and growth in production as "the real options are not bracketed by extreme formulations in terms of either welfare or efficiency, present time or future." The choice of regions for development should be based on their potential development contribution. Most transitional societies have as their basic goal the economic and social integration or their l!locieties in space. A first criterion tor policy concems national integration stated as follows: "One of the funda- mental purposes of economic develop11'1Bnt is the creation of a national economic !I ~~ pp. 9,10. -68 - space articulated b.Y an interdependent system or cities, areal functional specialization and national markets." Second, with respect to efficiency, projects should not be located where private and social interests diverge. 'nlird, regional investment allocation should be designed to maxilldze oppor- tunities tor future national growth, as regions cannot progress it the whole nation stagnates. Fourth, a reasonable balance in standards ·or living JIIUSt be maintained, "at least sufficient to preserve political stability in support or the drive tor ~tional. growth."!/ They suggest a fourfold classification which dist1Dguishes metro- politan growth regions, development axes between such centers, frontier development regions and finally depressed regions with limited prospects. ·~thin the framework ot a national policy tor development, ditterent problems are faced in each ot these regional-urban types and dit.rerent methods or development are called tor. nY In the end even Friedmann stresses the core region 8trate17 tor an industrializing society. Core regions are areas or concentrated interdependent activity which can decisively intluence the economic developJDent or a wider area. 'nley can be activated through a "plarmed, coordinated, long range and massive investment ettort'~ll In short, it may tor political or soci~ reasons be necessary to divert resources to re·gions which have little development V John Friedmann and Wllliam Alonso, "Introductionl R~onal Development as a Policy Issue" 1 Regional Development and Pl!!!lins, M.I.T. Press, 1964, pp. 4-7. y Ibid, pp • .3, 4. 'JI John Friedmann, "Regional Planning and Nation Building: . An Agenda tor Internatio~ Research," Economic Development and Cultural Change, Vol. 16, No. l, Octo~er 1967, p. 120. - 69 - potential. Growth, however, can only take place in salected regions where a number of economic conditions are tultilled. Resources should be concentrated on such regions in the interests of achieving satisfactory national growth rates and thus having some resources left for distributianal purposes. To a great extent the analysis and selection or critical growth regions in most countries will relate to the opport\inities for industrial growth. Policies to promote such regions will have to recognize the essential industrial supPly and market factors which lead to agglomeration in urban centers. b.. Orowth Poles and the Spread or Innovations Considerable literature exists on this subject which cannot be covered in this paper. A major problem in using growth pole concepts or theory as a guide to policy is raised by Lasuen in his research on the regional flow or innovations and the role of business organizations in creating growth poles. He views development as a process by which firms adapt successive sets of innovations on a successively larger scale. Both transport innovation and the increasing scale of innovating firms have operated to reduce the number of locations or centers where innovation can occur, leading to a relative decay in other places. The crucial question concerns the evolution of business organizations from the one product, one plant model to the multiproduct, llllllti- plant, multicity model. In Europe the multiproduct one plant system led to widespread vertical integration of firms, but Europe lagged behind the management revolution in the u.s. which enabled firms to produce a variety or products for different markets. In Britain the success or industrial parks resulted from the diversi- fied management or firms which could operate new factories anywhere. In con- trast the Northern Italiar.astill operate multiproduct one plant firms and are - 70 - unable to produce in the less developed south. In order to develop regions through the establishment or growth poles the locational limitations inherent in a .f'irm only able to operate on a one product one plant basis must be re- cognized. Wide regional wage di.f'f'erentials or tax and infrastructure incentives cannot be exploited under these conditions. At best a selective policy en- c·ouraging f'irms organizationally capabl~ of moving and stimulating the re- organization of others on a multinational basis is needed. Lasuen concludes that in less developed countries services such as banking and CJmm&rce rather than the typical manufacturing tirms can best exploit final markets in under- developed regions.!! 4. The Rural Econo!!Jl Little · attention has been paid so tar to regional imbalances and the rural econom,y. The tendl!!1cy towards concentration in primate centers at certain stages or the development process may "inhibit the transformations necessary to succeeding stages or national development. n'!,/ The dominating primate city not only draws in rural migrants, but also the hilher quality human resources .from smaller towns. Private investment institutions favor the big city, while public investment is also allocated to the big city because of' its heavy concentration of' political power. A self'- perpetuating mechanism is set in motion in which capital, productive labor, managerial talent and ent~reneurship concentrate in one urban center. A V J. R. Lasuen, "On Growth Poles," 28th Conf'erence of' the Southem Economic Association, Washington, D. c., November 8, 1968 (unpublished). Y Lowdon Wingo, "Implications of' Regional Development& Programs f'or the Patterns of' Settlement," Design .f'or a Wlrldwide Study of' Regional Develop- ment, Resources f'or the Future, 1966, p. 31. - 71 - sort ot "developmental dualism" is created which divides the country into a modern and a traditional sector. This is a barrier to the economic inte- gration required tor new development thrusts J/ WilliaJIISon ana.l.)'ses histori•al cases ot this kind ot "perverse" development but concludes that the elements leading to divergence diminish in.time allowing equilibrating ettects to make themselves. telt.Y · The rural areas or such countries may tail to show any productive response to the growing demand or the urban center tor agricultural products. ThUs the rural areas remain poor and. export their surplus popul.at ions to the big city, as even the small towns otter 11ttle opportunity. Under these cir- euastances, regional policy might attempt to increase the interdependence or the rural and urban economies and thus gradually close the gap between them• The object would be to integrate rural areas with the nation's market economy, involving possiblY subs'bantial structural reforms such •s with respect to land tenure. By modernizing agriculture, however, the number ot people supported by the land will be reduced and the growth or rural populations will have to be reduced. Migration to the city may simply increase and the resultant social costs may ottset the productivity gains in agriculture. a. A "Holding" Strate g;y As we have seen rural push can be a powerfUl £actor in inducing migration to the city. A short-run policy might call tor an expansion ot !I Ibid, p. 31. 'lj J:. G. Williamson, "Regional Inequality and the Process ot National Develop- ment: A Description ot Patterns, II Regional Analysis' pp. 101-106. - 72- output penl'litting a veater proportion of population to be retained on the land. In essence, this wouldbe a holding strategy designed to limit the outt'low until urban areas could absorb more people. Policies designed to increase !arm incomes and improve rural living conditions vCDl.d be elements in such a strategy J/ !be holding strategy may prove to be feasible particularly in situations where expanding markets for agricultural products and technology permit a greater absorption or manpower in labor intensive production. The impact or urban settlements in rural areas as centers or commerce, distribu- tion and rural based industry needs to be examined empirically. Such an approach may contribute to a better pattem or rural development~ Such centers could provide the alll8Di ties and semoee delu.nded by higher income farmers, supplY modem inputs and export rural output. A strategy or this type would have to be care.tully designed to accord with the minimnm scale economies or urban centers. For this reason only a few such centers could be promoted in a given region and then only atter a caretul study. Such study would 8xamine the impact or major trans- port investments in promoting such centers in rural areas. Transport iftvest- ments can improve access or the primate center thereby eliminating tbe economic functions of rural centers. Study is needed or the responsiveness .or the rural econo~ to urban demand changes, ot farm incomes and employment pattems and of the institutional structure or agriculture in dit~erent countries. V Wingo, "IIIIpl.1oations ot lerional. Development: Programs ror the Patterns of Settlement," pp. 34-JS. - 73 - III •. THE METROPOLITAN POLICY ISSUE A. Urban Public Goods and Services The principal issues ot urban-regional econom1:c development policy are easier to bring into focus on a broad scale than at the level ot urban project evaluation or systems ot projects in urban areas. The subject will not be treated exhaustivelY, given the purpose and space limitations of this paper. It may:~ however, be use:f'ul to discuss some or the characteristics or investments in an urban context as well as the complexities . in dealing . w1 th them and what this suggests tor establishing criteria tor investment polie)" in metropolitan regions. In a larger sense market mechanis1118 may be said to determine the organization ot functions in a city and the ~ in which behavioral processes& are worked out. Accordingly, we would be well advised to understand these mechanisms and perhaps their imperfections in attempting to gauge the develop- ment impact or public investmsnts. Some ot the evidence ot diseconomies or scale or· cities may result trom the failure or the prioe system to discipline behavior. Urban space will be used more freely i! the means ot access to it, i.e. transporation or public utilities, are priced too low. To the extent users do not pay tor . such services, society must pay. The use ot price as a control and equili- brating mechanism for the solution or all urban problems in the less developed world is not recommended. Its acceptability and applicability is probably greater in the presently more developed regions. Congestion may be more ac- ceptable in less developed societies as it involves trading time, with low opportunity cost, tor money with higher opportunity costs in alternative public - 74 - uses. Nevertheless, the use or price as a management tool !or large urban areas should be overlooked, despite formidable problems in determining demand, costs and user charges. 1. The Relevance of Scale Economies The size of the city to be governed is relevant in a~ discussion of the suppl7 or public services. Scale economies and area-wide externalities exist !or such integrated services as air pollution control, hospitals, public health services, power, water and sewerage, public transportation and planning. FOr aqy given size city, some services will be operating under conditions ot . v overfUll use of capacity and others will h3ve excess capacity.- However, tor a substantial. rmmber of "horizontal" government supplied services such as libraries, schools and police protection long-run cost curves are flat and little in the way of scale economies can be achieved as unit size is restricted by the necessity to locate geographica~ close to service recipients. Thus, the scale economy argument has to be qualified. 2. Managing Urban Sratems It would seem that urban government consolidation would offer sub- stantial advantages in terms of rational planning or the urban econom, and the least cost supplf or packages of urban public services. However, the possibility of better coordinated and s,ystematic planning tor growth must be weighed against the likelihood of increasingly complex communications between sub-units and 1/ TemporarilY unbalanced capacities can be shown to represent optimal long- run planning among components or an urban s78tem. R. !i)rse et al., "Costs of Urban Infrastructure for lndustr,y as Related to City Size in the Develop- ing Countries: India Case Study," p. 36. - 75 - remoteness from considerations of consumer choice of l~ge government units. Urban governments do not operate on profit motives. · Even where possible, least cost combinations of inputs may not be chosen because of con- flicting goals. Urban services are offered in a monopolistic market where marginal cost pricing is not usually considered. Hirsch concludes that in the United States local governments with jurisdictions of So, 000 to 100,000 people can b.est provide most services, while district administrations can best provide those services with substantial scale economies, and income redistribution objectives can best be accomplished by state and federal authorities.!/ Quite different solutions may be appropriate for countries with .different economic structures and administrative and legal traditions. J. Benefits and Costs In considering the demand for public services there is no way that benefits or preferences can be assessed when the good concerned is given away. A case in point is that of so-called "merit" goods, of which education, low- cost housing and hospitals are examples. The individual receives more of the good than be would purchase. The supply of these goods is seen as a legitimate function of society. For goods supplied at zero price, demand always exceeds supply. Rationing takes place through congestion, administrative action and an active market in other assets which give one access to the public serv1ce.- . 2/ 1/ Werner s. Hirsch, "The &lpply of Urban Public Services," Issues in Urban Economics, pp. 477-524. gl Julius Margolis, "The Demand for Public Services," Issues in Urban Economics, pp. 527-565. - 76 - The essential point is that the supply or such joods is usually not related to any measure of benefits but rather to a definition ot "need" ad- vanced b7 the publio supplier. In the case of. collective goods, such as street lights, health inspection, streets and justice, users cannot be directly charged for benefits received. Similarly where external economies ot con- Sumption are present the beneficiaries cannot be made to pay. An example would .. be the dissuasion ot a boy not to leave school in order to reduce potential .1 delinquency.J .. Merit goods 1118.7 be analytically distinguished from public goods. The latter include such publicly supplied services as urban trL~sportation and water supply. Measuring costs and benefits produce familiar problems or which discount rate to use, uncertainty and the determination ot optimal prices. Further, analysis takes place in the absence of ~~~arkets and often in the presence of increasing retUrns, indivisibUities aDd extensive externalities. In brief, benefit-coat anely8is.Usually attempts to relate the service supplied to some 11J8rket behavior: This ~thod assumes thAt public goods are not supplied accor- ding to community criteria as is often the case or merit goods. In the case of transport, the savings in costs of supplying the . .. . ~ service in question is-~he basis for evaluating the return of investments. In education .the service is regarded as an intermediate good and the increased productivity o~ the individuals receiving education is put in the form of !/ Robert Dorfman, ed. "Measuring Benefits of Government Investments," The Brookings Institution, 196.3, p. 5. gj See H.O. van der Tak and A. Ray, "The Economic Benefits of R-- - ~ double in population and probably in size in the foreseeable tuture. An opportunity exists to provide tor this growth in a far more efficient manner . - than vas done in the past, precisely because so much of urban growth is yet to come. · The less developed countries can plan tor urban growtn ·patterns : ~ . ~ . ~ .. Which allow tor a more rational growth of cities. The.y can conserve re- ' . sourcies through improved management and resource allocation procedures. . ' . . '!bey can greatly iaprove the mobilization ot·resoUrces trOm local sources. The. problem is one_ ot econ~c and social development at the sub-national urban levei as well ·aa at the na~ional level.· Objectives must be specified, means identified, choices made and a •chanism devised tor the efficient c&1T1inl out or decisions. A. For the Nation At the level ot nation&l poliQy lllllld.nl, and· resource allocation a new di.nsion in thinking about growth should be added. The cities are not just absorbers ot resources which must be rationed i t other national goals are to be tinanced; rather, they are the instruments ot economic and social transformation ot the nation. The choice must be made over a wide range between investment in commodity production and what may be termed settlement intrastructure and urban services. In this the quality ot urban services should be considered a variable 1n detendning investment allocations so that adequate resources will be available to achieve urban growth and employment objectives. - 83 - A related is8ue concerns the problem or regional imbalance which frequently leads to policies designed to limit the size or big cities. The ditticul ty is that such policies may lead to a more equitable distribution ot good& by assisting the backward regions, but may limit overall growth. or course, national policy must consider the longer run objectives or inte- grating the nation both socially and economically. In carrying out a national strategyt however, consideration should be given to selecting regions where concentrated activity and investments can have a decisive impact on the economic development ot a wide area. One preliminar,v conclusion is that urbanization solutions which require the establishment of numerous widely distributed "growth poles" with attendant industrial parks, infrastructure and other incentives should be avoided. The lessons of experience so far appear to be that the pursuance or distributional goals in the absence of efficiency criteria can lead to wasted investmen~s and little growth in resources either for distribution or for new investment. Rather, a good guide to the selection or a1ternate viable centers is to determine which centers are already growing and have the potential tor further growth. To a great extent the analysis and selection of critical growth regions will relate to the opportunities for industrial growth and, in some oases, to the development of untapped resources. Publicly provided infrastructure and fiscal policies can be a po1orerful force for the promotion or such centers. FUrther, given the state or knowledge, and in view of past experience, new town strategy, at least as carried out so tar, does not appear to be promising as an urban solution in the less developed countries. The entire question of alternative growth poles and new towns requires fUrther study in the course of our work in the field and at head- quarters. - 84 - Given the economic factors favoring the continued expansion of existing large urban centers and desirability of concentrating growth in a few selected centers it follows that national policy will not be able to change the distribution of urban population decisively. Rather, national policy can aim at diverting part or the now or migrants towards the few viable alternatives. National strategy should reconcile the goals or raising agricultural productivity with the impacts or such policies on migration to urban areas. In rormulating national strategy, thP criteria for a· system or settlements should be to achieve the growth pa\tems most efficient in terms or the cost or overhead and services as well as the retums in the form or incomes and employment. B. For the MetroPolitan Resion The creation of coherent national strategies for urban-regional development will require careful reappraisal of the economy of the 1118tro- politan region. In many cities of the world there are problems or frag- mented administration, ineffective municipal or regional utility service systems and a general lack of coordination in planning and in the allo- cation or investments. Municipalities are often under-financed, do not adequately mobilize local resources, and have little conception ·or bud- geting or of what constitutes an effective allocation or resources. ~at is needed is a development framework within which investment programs can be evaluated and carried out. The metropolitan ·development strategy should be aimed at optimizing output and employment in a given area and providing for a least cost overhead and service syateu investment consistent with the society's distributional objectives. - 85 - 'nle main limiting Actor in the urban economy may well be manage- ment aa the other factors such as land, labor and capital do not conceptually pose limits on city size and growth. As the large cities provide the bulk of modern sector jobs and incomes and a good part of a nation's traditional service employment, effectiveness of management may be a critical determinant in the ability or a society to achieye its development goals. c. Resource MObilization tor Urban Development For a good many countries effective regional and urban development programs cannot be carried out because or inadequate d1 vision ·or resources between national and local authorities. In some countries changes in the decision-making process may be required so that sub-national growth require- ments are renected at the national level. The· system tor decision making should be one which coordinates urban, regional and national development strategies and resource allocation. It is important that .the proVincial and metropolitan agencies requesting an allocation or resources be able to do so in terms of an economic justification which relates urban growth and investments to national objectives. Allocations justified by distributional, political or welfare criteria often fail to maximize growth. An important problem facing most urban administrations is lack of sufficient funds for capital requirements and often lack of current revenues to maintain existing services. It can be shown that a substantial part or the difference between the performance levels of comparable cities in a given economy lies in the efficiency with Which resources are mobilized locally through tax systems, rate structures, assessments and borrowing on local markets. In particular, the rate policies ot public utility systems need to be reviewed in relation to the distribution or incomes in urban areas and their implication tor financing public investments from internally generated sources. - R6 - D. Population Policy The high rates of population growth in most less developed countries contributes greatly to the pressure on urban areas and to the demand for resources tor current consumption instead ot tor growth. 'nle Bank's policy towards encouraging family planning is thus relevant to the ability ot devel- oping countries to raise their average per capita incomes in the context of rapid urban growth. In general, birth control programs will not provide any re11et tor about two decades to the problem ot surplus labor wh~ah bas a tendency to increasingly gravitate towards the cities: the numbers entering the labor force for. the next fifteen years are alreaey bom. Nevertheless, successful birth control programs can ~ead to itiiJ)Ortant short-term savings in national resources required tor urban settlement overheads, soctal services, housing and food. Even i t we were able to sharply reduce population growth rates in developing nations, their urban a_reas would continue to grow rapiclly. As we haw shown, this migration is based on powerful eoono~c and social toroes. As the lowest income areas ot the world including 'lllOst of Asia and Tropical Africa have 1*t to reach 2S% level ot urbanization, a vast number ot potential migrants are available to come to the city. II. POLICIES FOR BANK CONSIDERATION The Bank should seek to be a constructive torct! in helping to channel future urban growth into patterns 11lore consistent with national growth and distributional objectives. Bank policies may be considered at two levels, the metropolitan and the national. ibeir implementation will involve the Bank's country economic work, its project lending and its opera- tional research. It will require coordination wi tb others, particularly in - 87 - technical assistance. We will have to consider how best to adapt existing types of Bank lending to take account of new criteria related to urban- regional development as well as what new forms of lending may be needed to enable us to play a more effective role in an urbanizing world. Section I of the Annex proposes an operational research program designed to assist the Bank in projects lending in urban areas as well as in formulating general policies for making the Bank as effective as possible in promoting healthy urban development in the less developed countries. A successfUl Bank policy for urban-regional development will require coordination of the Bank's economic, financial, industrial, public utilities, transportation and agricultural work at the very least. The financing of investments in these sectors either directly or indirectly has an important impact on the efficiency of urban growth, on the distribu- tion of benefits and on the employment of surplus labor. The lines of approach suggested for Bank consideration in the field or urban-regional -development can be subdivided as follows: 1. For the short run - improving the effectiveness of urban management, resource use and resource mobilization. 2. At the national level - integrating the Bank's macro- economic and sector approach to development performance with investment programs at the urban-regional level which promote a balanced growth of basic urban services. 3. At the metropolitan-regional level - metropolitan develop- ment strategies, management or urban systems, increasing the returns on the Bank's urban infrastructure investments. 4. New forms of Bank lending - sites and services, savings and loan mechanisms, land development. - 88 - A. Policy for the Short Run - How the Bank Can be Effective in Urban Development It is otten said by those familiar with both the complexities ot urban-regional problems and the paucity ot empirical studies of them that we do not lmow enough to do effective work in urban-regional economic development. Tbe fact is, or course, that we know a good deal. We are also aware of what we do not know and how and where we must seek answers through operational research and field studies. In the short run, we can analyze the performance ot selected metropolitan-regional economies in three areas. First, we can assess the quality and the scope of the management of the urban complex by studying the various bodies msponsible for managing a particular urban area. Matters ot concern here are the ability to make and coordinate decisions in the relevant areas and the degree of effectiveness in allocating resources and in building and maintaining urban service systems. Tbe role ot public utility institutions needsto be analyzed. Within a metropolitan region the jurisdiction of public utility institutions is often partial or over- lapping. As well, the in~eraction ot utility investments with other major investments and services needs to be assessed. This leads to a second area or concem, that of the mobilization of resources. MOst urban administrations lack sufficient fUnds to provide the new services required. As well, they otten lack sufficient current revenues to maintain existing services. Much could be done by improving local tax systems, rate structUres, assessments, and by opening opportunities tor - 89 - borrowing on local markets.ll Finally, we come to the issue ot metropolitan planning and plan implementation. The urban area should be approached as a sub-national eoonom,y in which the use ot resources will be related to the achievement ot development objectives in' a given metropolitan area. We Might assist countries in formulating the criteria for studies leading to such plans and we might make the existence of such plans a condition for further investments in urban areas. This is partly a matter for Bank econoMic Missions and their economists to deal with and may in some oases lead to recommendations tor technical assistance. This will be discussed turther under our third policy suggestion. Plans are, hoW&ver, only a guide to what. to do. They DIUst be continuously revised in the light of experience and new taots and there must be a mechanism for implementing them it they are to have any practical me~. This requires institutional mechanisms capable ot advising management currently ot the best strategy tor achieving objeotives. '!be appropriate torm ot coordinating investment policy will vary in different places and brings us back to our first point. The Bank may have an important role in bringing about such coordination. As we are new at urban development and as statts are small, it would appear prudent to deal only w1 th a few urban areas, particularly those which show the greatest proMise or meeting the criteria laid down. We have already begun to look at selected countries and metropolitan areas (Brazil, Colombia, India, Pakistan and Kenya) and this may be usetul as a basis tor deciding which one or two major cities are most promising as places to begin more operationally oriented work. !/ For an example of how these three objectives have been analyzed in a metropolitan context, see R. Westebbe and K. Mera, "A Note on Calcutta's Urbanization Problems and Prospects," Economics or Urbanization Division, Economics Department, January )0, 1970. - 90 - B. Policy at the National Level In evaluating the performance of national economies and assessing their requirements for assistance the Bank usually looks at the national econoJV both as a whole and by particular sectors. This approach needs to be modified by taking into account metropolitan regional considerations which would relate the locational efficiency of investments and the spatial linkages between investment programs in different sectors. National planners and decision makers would thus also be concerned with carrying out national plans and achieving national goals at the metropolitan-regional level. They' would have to be concerned with associated policies regarding internal migration and the creation of appropriate sub-national administrative and planning mechanisms to integrate sectoral programs at the regional level. The Bank might assist in this process of introducing urban-regional investment and locational criteria at the national level through the work of its economic missions. Already urban-regional consulting economists have been included in three economic missions, Venezuela, Argentina and Ivory Coast. An attempt bas been made to draft preliminary guidelines for mission economic work and reports produced by the Bank staff in· selected countries. _The_ work of~ introducing urban-regional analysis in our own economic reports should be continued and progressively expanded in key countries. Countries should be encouraged to do the necessary studies and acquire staff for national and sub-national agencies in the field of urban-regional development. c. Policies at the Metropolitan-Regional Level Coherent national strategies for metropolitan-regional development are most desirable. Yet here knowledge is limited and it will take time to achieve results. - 91 - 1. The Metropolitan Plan At the outset we face in many cities the problem of fragmented municipal administrations, ineffective municipal or regional utility service aystems and a general laok of coordination in planning or_ the allocation of investments. Municipalities are under-financed and have little· conception ot budgeting or ot what constitutes an effective allocation of resources. The Bank's present programs in public utility and urban transport investments may provide part of the leverage for achieving improvements. Bank policy should encourage the planning and evaluation or invest- manta with a view to their areawide developi'llent impact. 'ntis would require the establishment ot regional development plans tor the metropolitan areas in question and the adoption ot strategies and programs to carry them out. We would in etteot ·request a deVelopment framework within which we could assess projects we are considering financing. As urban projects have inter- related impacts and traditional Bank beneti t-cost analysis is based mainly on the returns which may be attributed to the project itself, we will have to improve our methodology and analytical techniques. The operational research program is partly designed to do this. In general, we should attempt to in- duce a major shift in our approach towards urban infrastructure investments away from just meeting deficiencies in the supply of certain services which appear as a resW. t of the incremental growth or demand in urban s.reas towards anticipating and planning for investments on a coordinated basis over time. It we recommend a UNDP financed development study and plan tor a specific metropolitan area we should ensure that the terms ot reference and criteria used evaluate urban-regional development in a national framework. - 92 - This will require that the study be fully coordinated with and backed, by national plan author~ties, who usually have decisive control of regional resource allocation and are responsible for a coherent national policy tor urban-regional development. . We should also do more to see that technical assistance in the form or study and foreign expertise leaves behind a group able to continue, adapt and improve on the initial strategy tor growth set out in the plans. It should be emphasized that what is being discussed nere is not the traditional city master plan which usually takes little or no account of economic criteria and provides no basis for making choices between alter- native uses of resources. As one senior Indian official J>ut it, "The highest mortality rate in India is of town master plana. 11 Mbat is wanted is a metropolitan development strategy aimed at optimizing output and employ~ ment in a given area and providing tor least cost overhead and service systems investments. The appropriate regions tor such planning should encompass the areas which interact with the metropolitan urban center. 2. Urban Management The appropriate management structure for an urban area cannot be precisely defined. Local traditions, politics and the inherited adminis- trative structures will determine what is possible. Super metropolitan governments have ~awbacks and many functions can most economically be carried out by smaller units of government. The required system should coordinate major metropolitan invest- ments, make them consistent with national development strategy, establish a rational system for d~stributing benefits and assessing public utility - 9.3 - rates and attain an acceptable level or local resource mobilization. It should also be coordinated with national government agencies concerned with urban-regional development and should be abie to claim an appropriate share ot national resources tor local development. 3. The Bank and Urban Infrastructure Systems Utilities are usually organized as integrated s.ystems tor supplying a single service such as water or electricity. This is necessary in order to provide a rational organizational structur~ but in the future there should be more emphasis on coordination with the region or which they are a part. lor the optimal use or space, utility investments need to be coordinated with area- wide investment progra.c:s which cover other areas as well. The impact or utility rate structures on the distribution or benefits throughout the area should be considered. Standards should relate to what· a country can attord and should not be based on arbitrary intem.ation&l criteria (as are suggested at times by some or our sister international organizations). Further, the distribution ot benefits from urban utility systems should be examined tor their relation- ship with the social requirements of particular metropolitan areas. Thus the distribution of benefits to lower income workers and others who have established themselves in the urban structure may be more urgently needed than a distribution to upper income groups or to the most recent group of relatively inarticulate migrants. While the need for basic water and sani- tation facilities for the most recent migrants is recognized, the provision or somewhat better standards or services tor those low income workers at the next stage of development (whose political power is disproportionately larger) may be more pressing. - 94 - The analY:tical framework for evaluating Bank projects in urban areas shou~d be widened to include the ·interactions and tradeoff's with other related projects. For example, economizing on road investments may cause a dispersion ot population and a ribbon type development ot settlements. This could lead to higher costs of supplying water, sewerage and electricity.!/ Further, Bank investments might, if' cozmected with other infrastructure investments, be an important means or providing the basic structure which can determine the density and pattem of urban growth. The ~'{ will need to study what kind or Bank investments in urban areas will, singly or in combination, increase the productivity of resources the most, whether they be in transport, public health or education. D. Possible New Forms ot Bank Lending 1. Housing - Sites and Services The issue of' ·housing is a central one in the problem or urbaniza- tion. The basic problem is to facilitate the process or building shelter for the masses of' migrants and iow income people who are becoming a larger proportion of' existing urban centers. At this level institution building is difficult as financial savings are not large. Yet the lesson of' experience is that people who are offered . the incentive of' a housing site with perhaps minimum services will build their own dwelling trom intemittent savings. Self-built housing does not saddle the owner with large fixed debts and provides more housing more quickly; in the cases studied, than does standard - 1/ I aa indebted to Mr. David Knox tor pointing this relationship out in the case of' Ceylon in his memorandum to 118 ot March 16, 1970. - 95 - public housing. Public policy should find ways to mobilize this large savings potential by channelling squatter building into land uses consistent · with metropolitan development strategy and without a large diversion or public funds. The Bank • s existing programs for water, sewer, roads and electricity in urban areas ild.ght be coordinated with site and service schemes tor squatter building and may in some cases make such schemes feasible. By providing a rational settlement pattern tor migrants such schemes can assist in avoiding much or the overcrowding associated with the rapidly growing squatter settle- menta in most or the leas developed regiona. 2. Land Development In a number of countries consideration is being given to various institutional mechanisms for rational land development in growing urban centers. Such schemes may achieve more economic uses or land trom a national point or view. As well, they may permit increases in land values resulting from development to benefit the public sector primarily1 thus providing more resources for urban needs. They are intended to be self-liquidating over a period or time. One form of such an institutional mechanism is the improve- ment trust found tor example in India, Pakistan and other former British overseas territories. The Bank already has been associated with such organi- zations in carrying out urban water and sewer systems. In addition to this 1 we may want to consider providing part of the necessary seed capital through direct loans to a revolving tund as well as making our infrastructure invest- ments part or the scheme. - 96 - Our investments could in this way become the basis for assuring that physical planning would lead to the creation or Viable communites including provision for the lowest income levels who ar,e now squatters living on the fringes or society. Our purpose in this should be to coMbine economically and financially feasible projects with the social goal or integrating the poorest groups into modern urban society. ~ are investi- gating the prospects tor promoting such schemes through our tield research efforts. 3. Mobi1izins Housing Savings Through Savings and Loan Mechanisms At higher income levels a number or countries have had a good deal ot success in creating associations tor home buildir~ through a savings and loan mechaniSm. When tied to the possibility ot acquiring a long-term loan tor a house suoh savings may in large part be new. A variety or forms ranging trom tull public to mixed and private associations have been used. The element or public subsidy need only be sufficient to supplement what should be largely a savings ettort on the part or the private sector. We are presently examining the desirability or doing research on the need tor outside seed capital to promote such institutions in selected countries, particularly on whether more outside capital than that now provided by other external aid agencies is warranted. It seems that external aid for these purposes can be most effective when given to countries' central agencies which in turn sponsor the growth or local housing finance institutions. ANNEX I.· DEVELOPING ·poLICIES .l"OR ·THE JO'S THRODGR ·AN ·oPERATIONAL RESEARCH PROGRAM 1. Case Studies in Urban Development 2. Review and Research Studies a) National Economic Development and Urbanization (i) A Review of Case Studies on Migration r~· (ii) Research Paper: The Interrelations Between Urban and ---- b) ~ural Development in Developing Countries Metropolitan Regional Development (i) . Review Paper: Th~ Management of Resources for Urban Development· ' (ii) Research Paper: Urbm': Land Markets in Developing Countries c) Urban Transportation (i) A Review of Mass Transit Technology and Coat Charac- teriatica (ii) Research Paper: The Economics of Managing and Developing Integrated Urban Transport Systems d) Public Utilities (i) A Review of the Techniques for Analyzing Public Utility Investments in Urban Economies (ii) A Review of Technologies for Environmental Control e) Housing (i) A Review of the Economics of Housing in the Leas Developed Countries Annex Page 2 I. DEVELOl'tNG POL!Cl.~ l"OR ·nm 70'S THROUGH . AN OPERATIONAL . RESEARCH PROGRAM The development of a full set of Bank Group policies and programs for urban development in the 1970's will require a systematic effort to ad- vance the state of our knowledge and experience. We will want to simulta- neously and over time go deeply into the problems of important selected countries, learn what we can of the best practice and thinking with respect to key areas of urban public policy, and finally improve our ur1erstanding and ability to operate in the field of urban-regional development by means of carefully designed operational research studies to be done by our own staff and by experts outside the Bank. In this policy oriented research effort we will have the benefit of the considerable urban-regional analy- tical talent of Resources For the Future which is willing to enter a col- laborative arrangement with us to share part of the expense of doing these studies and to supervise them. · A proposal has been drafted to this effect. We have also established links with the research activities of the Urban Institute and a number of research and academic institutions in various parts of the world. These contacts have put us in an advantageous position with respect to the current state of the art and enable us to draw on the best talent for doing studies of particular interest to us. 1. Case Studies in Urban Development Current work involves urban-regional development studies of selected countries. Such studies are under way for India, Pakistan, Iran and Jamaica, ~nd are proposed for Brazil, Colombia and Kenya. They are Annex Page 3 intended to have genex-al 'Value foJ: the Bank. t s emerging policies in the field of urbanization as well as to provide guidance for the work of the Bank. in specific countries. Exploratory missions have already been to the first three countries in order to acquire data. Other missions are planned, partly in conjunction with Area economic missions, to carry out the rest of the program. In some cases we are supervising the work of consuitant urban economists who accompany economic m~ssions· such as in Venezuela and the Ivory Coast. The scope of our country work is neces- aarily constrained by the availability of staff. In addition to the above, we have gained knowledge of conditions in other places by responding to Area and Projects Depar~~ents' requests for analytical assistance as in Chile, Niger, Sao Paulo and Singapore. 2. RevieW and Research Studies The program of operational research concerns work to be done by the staff as well as studies to be commissioned outside. A number of rela- tively inexpensive review papers designed to be done in a few months' time, are proposed for topics about which we need to. know the present state of the art for the purposes of current operational analysis and in order to provide a sound basis for defining further research efforts. The entire program may be considered in terms of a series of topics covering broadly the impact of urbanization or national development, and pr~blems of urban metropolitan policy. A number of specific areas of high operational importance have been singled out such as urban transportation and public utilities. The following is a summary of the research program by topic. a. National Economic Development and Urbanization The key issue for national development planning and strategy is the interaction of the urban and rural economies which are characterized Annex Page 4 by large scale migration to the cities. A study is und~ way in the Division entitled, uopt:bDal Location of Economic Acttvity in Pakistan" based on an analysis of the public and private costs of locations in selected urban centers. A draft bas been completed of a gene~al paper entitled, "On the Concentration of Urbanization and Economic Efficiency." A review paper on migration and a related full study on urban and rural development are proposed. (i) A Review of Case Studies on Migration Migration of population from rural areas to urban centers bas been a significant factor for expansion of urban population in developing countries. Several hypotheses argue against migration such as that migrants decisions to migrate to urban areas are not motivated by economic reasons but rather by irrational expectations, that.urban migrants are not so productive as they would have been in rural areas and that individual decisions to migrate do not lead to the socially optimal resource allocation. The proposed survey is intended to summarize the findings of existing studies with respect to the motivation and expectations of migrants, the degree the expectation is fulfilled, the process of searching for first jobs, the time required to obtain jobs, their initial income relative to the income which they were getting before migration and the income non- migrants are getting in urban areas and the process and_ speed of assimilation into the urban life. Such summaries will not refute or confirm the hypo- thesis conclusively, but at least would provide a better basis for judgment with respect to them. (ii) Research Paper: The Interrelations Between Urban and Rural Develop- ment in Developing Countries Very little research bas been focussed on the ways in which rural Annex Page 5 and urban eco'nOII)iea :r:elate to each other in developing countries or on the illapacts on national e.conomtc development which diff~rent kinds of rural-urban linkages imply. A good part of the economic dualism in these countries can be described in terms of the impact of the growth of the major metropolitan center on the rural hinterland of the nation, Under generally competitive conditions and within the market context which those conditions suggest. one would expect that urban growth and develop- ment would be accompanied by increasing productivity and prosperity in the agricultural region tributary to the urban market. One could also pose the hypothesis that the market institutions concerned with rural-urban economic relations t~nd to work perversely in developing countries, and in doing so, distort the allocation of capital between rural and urban development opportunities. Urban growth should be creating demands for capital investment in agriculture not only in terms of directly pro- ductive capital on the land, but also in terms of basic infrastructure essential to more productive agriculture farm-to-market roads, produce markets, irrigation and drainage . works, and the like. This distortion of the allocation of capital should be of considerable interest to policy makers. When this problem is related to the rural-urban migration potential of agricultural development and its implications for the demand for urban infrastructure, the importance of the problem becomes even more pronounced. This research proposed would explore and appraise the imperfections in the market mechanisms by which the economic impulses resulting from urban development are imperfectly transmitted to the tributary agricultural sector in developing countries and the extent to which substantial misallocations of capital result therefrom. It is hoped that the research would culminate Anna Page 6 in a be.ttQ' .understandtng ot the. ~t1..tut1onal, legal and other sources of market :t:mpe.rfecttona and tbus provide a basis for recommending measures to reduce ~uch imperfections. Furthermore, such research might well evaluate same of the policy opportunities that have been suggested to deal with such problema as rural-urban migration, the provision of rural public services, and the like. In brief, it is hoped that the research will produce some guidelines for national strategies of develop- ment which consider the urban and rural sectors as interrelate~. b. Metropolitan Regional Development A review paper is proposed on the management of resources in urban areas. A research paper is proposed on land markets. (i) Review Paper: The Management of Resources for Urban Development !/ An important factor in the performance of urban economies concerns the effectiveness of the institutional structure and management in using and mobilizing resources. A paper is needed which would provide a survey of the best practice in selected countries with differing institutional structures and ezperience in allocating and managing resources for urban development. The paper would deal with the issue of resource allocation between national and subnational governmental levels, and would appraise the distribution of revenues in selected areas from the point of view of their contribution to achieving development objectives. At the level of the urban center the review would describe the existing patterns of mobilizing tax and other resources as well as the institutional and other constraints to and prospects for further resource mobilization. The review would 1/ This should be carried out in close cooperation with the Domestic - Finance of Developing Countries Division. Annex Page 7 analyze selected local spend:blg patterns :t.n relation to development priorities. Specific cons1deratton would be given to the techniques for financing urban development, such as betterment levies, self..assessment, changes in state structures, borrowing, etc. The paper would discuss the economics of various methods of mobilizing resources, i.e. the effects on incentives, the distribution of income, on investment. (ii) Research Paper: Urban Land Markets in Developing Countries It has been observed that the value of land in and around the major metropolitan centers of developing countries generally equals or exceeds that of parcels stmilarly situated in comparable metropolitan regions in the more developed nations, even in situations where urban land markets are severely constrained. This situation has a number of serious implications for _development of the country as a. whole, especially where the metropolitan region contains a major fraction of the nation's productive plant. Clearly land development schemes in major metropolitan areas are likely to be sub- stantially shap•d by land coats: new low income housing developments are frequently located at a considerable distance from job centers because land development agencies try to maximize the amount of land that can be bought by limited budgets. Land values have a substantial influence on the costs of land-intensive urban infrastructure; a major proportion o£ the costs of new freeways, schools and other public facilities in inner metropolitan areas is chargeable to land costs. On the other hand, the mechanics of the land market, reflecting as they may the benefits of public investment, suggest the possibility of betterment charges on land value increments as a source of finance for the development of public infrastructure in metropolitan regions. In short, the metropolitan land market has a substantial influence Annex Page 8 on the costs of production, and hence, on the eff~c~ency of the metropolitan economy. It would be useful to know how exogenous changes in the national economy are likely to affect the level and distribution of land costs, as well as bow the distribution of land value is likely to be altered by policies and projects of the government. The research would attempt to identify and analyze the impact of imperfections and speculations on land markets. -The conclusions would be expected to throw some light on the ap- plicability of U.S. land market models for the developing count-ries. c.· Urban Transportation From an operational viewpoint, investments in urban transportation are likely to be of primary operational interest for the near future. A draft review paper entitled, "The Framework for Urban Transport Studies," has.been completed in the Division. A survey paper on mass transit tech- nology and cost characteristics is proposed as well as a research paper on the economics of managing and developing integrated urban transport systems. (i) ARevieW of Mass Transit Technology and Cost Characteristics It. is o·ften argued that rail mass transit is more desirable than road mass transit in high density, large scale urban areas. In terms of line haul, rai1 transit tends to be faster than road transit, but rail transit tends to be more capital intensive and less adaptable to changing land use patterns. However, for fast growing urban centers, rail transit may economize by reducing the land area used for transportation because of high capacity and probably by reducing automobile ownership. Some of the basic questions posed here can be answered by surveying the basic technological properties of rail mass transit (elevated, subway, opencut, monorail) and road mass transit (bus with or without exclusive lanes) and costs associated to each mode. Here, costs must be evaluated with the Annex Page 9 domestic facto~ price proportions p~evailing in developing countries ~ather than by merely couverttng cur·rencies of developed countries to those of developing countries. (ii) Research Paper: The Economics of Managing and Developing Integrated Urban Transport Systems Urban transport involves the use of public policy and control measures in economizing in investments in roads, public road transport and mass transit. From the point of view of Bank policy formulation towards .the urban development, it would be operationally highly useful to _have an analysis of the economics of, first, improving the performance of existing transport systems in metropolitan centers through better policies and management and, second, different modes of producing transportation, the J:'elationship between them and their impacts on the long-run development of urban areas including land values and use patterns. This longer run analysis would also deal with the whole range of public policies and planning including fare structures, subsidies, and other control measures. The study would be expected to consider the purposes of transport investment in an urban setting. It would examine the various categories of costs and benefits and in particular the distribution of benefits by classes of the population. Emphasis would be placed on maximizing rates of return through a coordinated approach to urban transport planning. d. Public Utilities A survey paper is proposed to improve our understanding of the present state of the art in analyzing public utility investments in an urban environment. Further research will be proposed as a result of ongoing collaboration between the Public Utilities Department and the Economics of Urbanization Division. ~n~ Page 10 (i) A• Raytew . of the c Techntguea • 4 for Analyztns Public Utility Investments . 4 -I • • . in Urban Economies ~~ A review is needed of the available analytical techniques for evaluating the costs and benefits of public utilities including water supply, sewerage, drainage, electricity, gas and telephone. The traditional tech- niques used by the Bank need to be broadened to include the developmental impacts in urban areas of public utility investments. In particular, methods of determining the costs and benefits of alternative utility i~vestments and packages of utility investments would be examined in this paper. The cost and scale of public utility systems will be considered in relation to popu- lation density and the quality of service. In this respect, so-called "Standards" of service would be critically examined in terms of the oppor- tunity cost of resources and. the supply of related services •. Finally, the "tree" character of utility systems would be analyzed as well as 4iurnal and seasonal peaking and spatial structure. (ii) A Review of Technologies for Environmental Control It has been amply demonstrated in developed countries that indus- trialization usually brings about environmental degradation through factory - ~- - -· - -- - -- smoke and liquid and solid waste. Airpollution in major metropolitan areas in many develpping countries has achieved levels of intensity c~arable to or greater than those in most polluted of American cities. The issue of e~ironmental quality is likely to become more pressing in many of these countries because of increasing density and size of urban centers. Such environmental problems are due to externalities of production and consumption processes. They can be internalize4 to the producers of 1/ This paper is to be done in collaboration with the Sector and Projects Studies Division. Annex Page 11 e:xte.rnaU. ttea, but thi.a has ;ra:r:ely been done becauae of add:ltional economic costs involved for tnternaltzation, If costs aTe suffictently·low, the government should be encouraged to prohibit production of negative e:xter- nalities or to dispose pollutants by itself. As a first step in answering this question, the proposed paper would survey available technologies and their costs for internalizing or disposing of wastes and pollutants. These basic technological data will became a basis for economic analysis of environ- mental control which may be planned in a later research study. e. Housing An initial review paper is proposed in the field. Discussions are presently being carried on with a nUmber of experts concerning longer term research on savings mobilization and savings institutions with respect to housing in less developed countries. (i) A Review of the Economics of Housing in the Less Developed Countries This is a subject about which a good deal has been written, but which requires a systematic review of the sources in order to provide a basis for understanding the economics of the subject as well as its relation to national economic and social development. The importance of housing stems from the multiplying requests from less developed countries for Bank assistance in meeting the growing problem o~ slums and squatter housing in areas under- going rapid urbanization and because of the large sums being expended for public housing in a number of countries. The proposed survey would deal with the impact of housing on employment and on the construction and building materials industries as well as the demand side from squatters to middle and upper income groups. The financing of housing would be considered also from the point of view of the efficiency of financial institutions in mobilizing underutilized resources as well as from one of efficiency of existin~sti- Annex Page 12 tutions. l"tnally, the. re.sou:rce. .mobi.li.zation aape.ct of self-built housing would be evaluated, together with the social implications of such housing in selected countries, The study paper would indicate the gaps in existing knowledge with respect to various economic impacts of housing and the institutional structures needed to manage housing. Annex Page 13 II. A NOTE ON INTERNAL MIGRATION STATISTICS IN SELECTED COUNTRIES • Migration statistics should be interpreted with caution.ll The best procedure is to place reliance, wherever possible, on individual country statistics where appropriate adjustments have been made. Studies of inter- censal natural increase and net migration for urban areas ot 20,000 and over in Latin America indicate that migration accounted tor trom some 40% of urban growth in the case ot Mexico ·to some 7~ in the case of Venesuela. Percentage Population Growth Due to Natural Increase and Immigration for Selected Latin American Countr~es Percentase due to: Intercensal Period Natural Increase Migration Venezuela 1941-50 29 71 Colombia 1938-51 32 68 Brazil 1940-50 51 49 Chile 1940-52 53 47 Mexico 1940-50 58 42 Source: Urbanisation in latin America, OR· cit., p. 110. - 1/ It is di.ffioult w:it.h . t.ha available data . to measure w1.th precidon t.he . effects of internal m:tgration on urban growth. The figures on past urban growth patterns are at best estimates. They must be qualified for changes in urban boundaries and for changes in the number of settlements classified as urban as towns reach the minimum size necessary to be included in the urban category. In fact, the growth rate of urban centers o.f all different sizes m :innuenced by the graduation of smaller centers to larger ones. Using a somewh~t different method for the period l950-60 in which the difference between rural and urban rates of natural increase is used to compute the net migrating balance, Lowdon Wingo found that migrants consti- tuted. as low as 27% of urban change in Argentina, ~1:: in Hexico and CM.le, over 4'5% in Brazil, Colombia, Panama, Paraguay and over '50% in Peru and Haiti."!! In India net migration from 1941 to 1951 to towns over 5,000 amounted to 20% or the 1941 population over 5,000. Individual cities sUch as Madras experienced Jlligration rates of 2/3 their 1941 population.Y "};/ Lowdon Wingo, Jr. "Recent Pattems of UrbanizatiDD Among Latin American Countries," Urban Affairs Quarterly, Vol. II, No. 3, March 1967, pp. 81- 109.. The method used consists of assuming that urban and rural net natural increase .rates are similar and that deviation of urban and rural growth rates from the national can ~e ascribed to net rural-urban migra- tion. Wingo makes the point that internal migration is a complex process of human .mobility which includes seasonal and temporary movement as well as permanent relocation. Further net migration may give no clue to gross migration which can be several times greater. Y D. J. Bogue and K. c. Zachariah, "Urbanization and Migration in India," India's Urban Future, ed. by Roy Turner, University of California, Berkeley 1962, pp. 31 and 42. Annex Page 15 III. A NOTE ON URBANIZATION AND ECONOMIC AND SOCIAL PRoGRESS INDICATORS .. Classification of countries according to when they reach 25~ urbanization levels can be used to compare the developed countries with the developing countries at the same level of urbanization, i.e. Groups I and II 1 and to compare the developing countries at different levels or urbanization.!! The .first group consisting of the developed countries or England and Wales, Australia, the Netherlands, the United States, Sweden and Finland attained a 2S~ urbanisation level by 1920 and a S2~ level of urbanization by 1960. The next group consisting of the developing countries or Uruguay1 Argentina and Chile reached a 25~ urbanization level by 1920 and a 56~ level by 1960. Compared to the first group in 1960 they bad 1/3 the per capita income level, a 10~. lower expectation of lite, an infant mortality rate over 3 times higher, 1~ lower calorie consumption, 1/4 the energy consump- tion, 17~ lower school enroll.Jnent and over 1/3 more persons per room. This second group bad virtually the same percentage of GDP derived from non- agricultural activities and 20~ higher dependency ratio. Hence, at similar levels of urbanisation the developing countries had not attained the same level of economic and social progress. "' V United Nations, Department of Economic and Social Affairs, "Urbanization and Economic Social Change," International Social Development Review, No. 1, New York, 1968. Azmex Page 16 A third group consisting of Venezuela, Ja~, RJuku Is., the U.A.R., Mexico, Brazil, Peru, Iran, Ma.lqa and Algeria U~Cm~•t others reached a 2S% urbanisation lnel between 192S and 1960. Tb1s sroup had an average urbanization level of 32% in 1960, 2/3 of that ot the developed countries and an average per capita income level almost 4 times lower. The percentage of GDP derived from non-agricultural activities at 80% was about 10% lese, lite expectancy was some 20% lees, literacy about 1/2, caloriQ conswnption about 2/3,· and persons per room and the de]:Mmdency ratio about 1/3 higher. The fourth ~oup consisting ot 30 of the least developed countries with an average level ot urbanization ot m iD 1960 has 78t to attain the 2S% level. Per capita income in 1960 was about 1/7 that of the developed countries, non-agricultural eaplo)'Mnt about 2/3, oaloric consumption a little over 2/3, the inhabitant/physical ratio 1/12, the cinema attendance rate 1/8, the literacy level 1/4, lite expectancy 2/3 and the dependency ratio 1/3 higher. Rene- among t)le developing countries those at higher levels of urbanization have achieved a higher level of social and economic progress. • .tr•.,ex " Tab::.~ l: Sn!!'..L'l! OF !IAT::S OF IJ?.B.,!l .,:0 ~\LUL IB:X•"lH, 1920-1960 A..tfl) 1960-2000, AND t!UU.J POPiliATIO!: AS .l PmC?:.'fr.\33: OF TOTAL POPUL\TION, 1920-1960, AID 2000 Urban Population u • ot ~e!l'ion Rate c! !ctal ~l"C'rl.h Rate c! !tural :lro1o-th Rate ct Urban G:-ovth Bjte ot Bis:-City Gro,.-th Total Iil.O:O~ t;otS-2~ I1'21S~ i~-2lR!Ci il'20-60 i9lli-2C5ll!S ~20~ 19bc-2il00 i92lS i~ 2~ (l) {2) m (L) (S) (6) (7) {6) (9) bo) (u) :ioro!c! 'ror.al ft 59-::.18 ~ SJ.-62 m ~ ~ 216-)10 !11 l1 L.O-W, 1. ~o:-e develooed regior~ or high densi~ Total o! P::"O':l'!) J§. ~ 11 !l u S::li 1l S7-B1 Jl 111 ~ ".ieste:-:t !111'0?8 So:-t!ler:t !::!"'Ope 3!& 22 2$-)) l:!-21 12 3 0 0 61 L1 S0-66 22-3$ . S2 29 So-66 22-3S b1 S2 la9 S9 $9-62 64-66 Sc...:t!-'.~:1 Zw.:rcjl8 · la2 2b-33 12 10 s 12S 60 !J-76 283 . 22 78-111 2S 39 )2 bB-$2 !:&s~e:-:: ::u:-"?8 22 32-!:2 12-lli 76-100 114-lS1 24 b2-L.S JapL"' 6S 31-h9 19 1 2)0 67-10S 206 S9-92 .21a la6 S9-64 2. ~t':-e c!eve!croed reslons of lover densit;r: fota! or ~r:-Olr.) 21 se-!!;. ! i !?1 116-170 ihl!. ~ l1 !I! 6$-70 ~:e:-t!te:-r. ~-:er!ca 72 1:8-76 23 0 11a1 a.3-13S 169 66-!07 b1 sa 72-76 So~et ~!o:t )!! 6$-SS : -2 8 )8d 164-227 11aS2 237-3:28 10 )6 S8-63 T~~e:-ate Scuth A..~~:-lca 122 66-Sl& ss 0 260 129-160 296 100-l2S 32 SJ 73-74 1~st.:-a:1~ a::.d !hra ZealL"lc! 92 71-92 29 0 16S llG-lla) 206 8$-112 la7 6$ 79-82 ). less c!evelO'!)ed regioos o! high density !e~'!. e!' ~'.!o ~ 9!!-130 12 6$-81 m )03-L.32 21! J39-Sl2 i lh 29-32 ~a!n::.~c! East Asia 37 6C-106 2$ 41&-71 2$2 1SS-323 7SO 137-.307 s lh 22-28 ~1~!:8 South Asia 7la llB-138 60 79-BS )21 377-h67 644 Sla0-7h3 s 13 29-31 So-::~!l.-!:ast Asia Othe:- ::.as t .bla 102 117 151-176 13S-1S6 as 61 97-106 21 496 11S5 S30-6S9 4c5-h79 66) $)) ~ 3 295-741 493-$91 Ia s 1J 29 31-35 6la-67 Caribb~a."' 109 lOS-liaS 78 39 380 319-h88 133 a S!U-9.36 10 24 b8-S7 b. Lesa c!e?e:!.c::ed rnions or 1011ero de:~li~ ':eta! o! C"'".!O 100 1S9-19B .!U. 69-78 bbO S2;-6SS 1l!! 69$-9::.9 ! ~ !!!.:2! Tro?!ea:!. South ~er'ica 11:7 167-222 96 20 SS3 )16-702 768 521-70!! 11 )0 68-74 ::ertherr. A!r!ca 88 121-191 sa )2 312 la97-652 689 292-746 12 26 61-66 So-.:th.:.·:est As:!.a 106 127-1!!2 a5 Ia) 229 S43-667 21aO 1076-1471 lla 2) $7-61 11&1 191-25'4 87 17 61$ S87-79b 786 681-919 10 )1 72-77 ~idC:!.e A."'erlun 1'.a1nland T:-op1cs:!. Arrlca Scut!:ern Ar:'ica 92 126 1S0-177 129-187 a1 80 120-139 13 769 37S 225-668 la63-ssla -IJ!. 333 ))27-b227 SOB-608 16 2 7 32 18-20 70-73 Othe:- Oceania · sa 107-lla7 S3 100-l)la - 300-soo - - 0 3 1-9 a ro:- 19)0-60. 1J2. ·Far 19b0-60. Sou.-ce: Con;lutat1olll 'bued on data traa Ul Popalation Dl'fislon, •Urban aDd aaJ.a]. l"opptlation Orovth, 1920-1960 111th P!'Ojectlons,• WorldJII Papeio Jlq~ 1$, Septelliler 1967. · . · · Annex Table 2: URBAN POPULATION (20,000 AND OVER) AS A PERCENTAGE OF TOTAL POPULATION, ESTDIITES 1920-1960 Region 1920 1930 191i> 19.50 •1960 World Total !!! 16 1!! 21 g2. - - 1. More develO]:'!!d resions of higE densi tz Total of groUP Western Europe ll 41 2§. 43 !?. - uo 45 !!2. 49 47 Northem Europe 52 56 58 ~8 ~9 Southern Europe 25 26 ~0 34 39 Eastern Europe 2u 26 29 2Q )2 Japan 24 29 )9 . )R 46 2. More developed regions of lower density ga Total of group 28 - - 34 ~ ~ Northern America 41 46 46 5i .58 Soviet Union 10 13 24 28 36 Temperate South America 32 35 38 h6 .53 Australia and New Zealand 47 48 52 57 65 ). Less develoE!d regions ot hi&h densitl Total of group Mainland East Asia 2. 5 1 1 - 8 8 - 11 10 - 14 14 Middle South Asia 5 6 8 11 13 South East Asia 4 6 7 10 13 Other East Asia 5 8 14 21 29 Caribbean 10 13 16 20 24 u. Less developed regions of lower density Total of group 8 - - 8 10 - 11! 20 Tropical South America 11 13 16 21 30 Northern Africa 12 14 17 21 26 South-West Asia lh 12 13 16 2) Middle American Jltlinland 10 1) 16 21 31 Tropical Af'rioa 2 2 3 4 7 Southern Africa 16 18 22 28 32 Other Oceania 0 0 0 1 ) Source: UN Working Paper No. 15. . Annex Table 3: SUMMARY OF CHANGES IN TOTAL, RURAL, URBAN AND BIG-CITY POPUJ..ATIONS IN ~RL: REGimJS CLASSIFIED BY RECENCY OF URBANIZATION ~, 1920-1960 Group ot Regions 1920 1930 1940 1950 1960 • TOTAL POPULATION (MILLIONS) World Total 1,860.0 2,068.6 2,295.1 2,515.5 2, 093.2 Group I 3B0.7 h2h.h 456.4 564.3 569.8 Group II 421.5 484.9 544.5 577.0 696.4 Group III 1,057.8 1,159.3 1,294.2 1,434.2 1,727.0 RURAL AND SMALL TOWN (MILLIONS) 1vor1d Total 1.593.6 1, 730.4 1,863.6 1,982.5 2,232.9 Group I 231.8 243.8 255.6 268.2 272.7 Group II 359.2 399.3 412.8 421•8 459.9 Group III 1,002.6 1,087.3. 1,195.8 1,292.5 1,500.3 URBAN POPULATION (20,000 AND OVER) (MILLIONS) World Total 266.4 338.2 431.5 533.0 760.3 Group I 148.9 1fk) .6 201.4 236.1 297.1 Group II 62.3 85.6 131• 7 155.2 236.5 Group III 55.2 72.0 98.4 141.7 226.7 BIG-CITY POPULATION (500,000 AND OVER) (MILLIONS) World Total 106.6 142.3 181.1 227.4 352.2· Group I 74.6 93.6 102.9 122.2 159.7 Group II 21.8 33.6 52~5 58.7 100.0 Group III 10.2 15.1 25.7 46.5 92.$ URBAN POPULATION AS A PERCENTAGE OF TOTAL POPULATIOl 't\br1d Total 14 16 19 21 25 Group I 39 43 t4 T?f 5'2 Group II 15 18 24 27 34 Group III 5 6 8 10 13 BIG-CITY POPUJ.ATION AS A PERCENTAGE OF URBAN POPULA' World Total 40 42 42 43 46 Group I 5'0 5'2 ~ 5'2 ~ Group II 35 39 40 38 42 Group III 18 21 26 33 41 BIG-CITY POPULATION AS A PERCENTAGE OF TOTAL POPULA~ World Tot:ll 6 7 8 9 12 Gro~I ~ ~ ~ ~ H Group II 5 7 ·10 10 14 Group III 1 1 2 3 5 (Footnote on next page) .Annex Table 3: (Continued) - 2 - Footnote regions at least 25 percent urbanized by 1920: Western Europe, - /a Group I: Northern Europe, Southern Europe, Nort.h America, Temperate South America, and Australia and New ZealRnd. Group II: regions at least 25 percent urbanized by 1960 but not by 1920: Eastern Europe, Japan, Other East Asia, the Soviet Union, Tropical South America, the Middle American MainlllJld, Northern Africa, and Southern Africa. Group III: regions not yet 25 percent urbanized by 1960: Mainland East Asia, Middle South Asia, South-East Asia, South-West Asia, the Caribbean, Tropical Africa and other Oceania. Source: tJH Working Paper lo. lS. Annex Table 4a DECENNIAL INCREASES IN TOTAL POPULATION, ESTIMATES 1920-1960 I (Peroent) • Region 1920-.30 19.30-40 1940-50 1950-60 World Total ll. ll. 1:2 12 1. More developed regions or high density Total ot groUP . - lO B - ~ 2. 10 ~stern Europe 7 4 8 ' 6 Northem Europe Southern J!Wtope Eastern Europe Japan 5 12 11 15 ' ll 8 12 -7 16 6 8 9 12 2. More developed resi~::::s ot lower denaity Total ot sroup Northern Amerioa - 16 16 - 9 B l lS 20 - 20 Soviet Union 15 9 -8 19 Temperate South •rioa 27 18 20 22 Australia and New Zealand 20 9 16 2$ 3. Lese developed resione ot high denai ty Total ot croup - 9 - 11 6 - 10 6 - 20 16 Mainland East Asia Middle South Asia South-kat Aa:la Other Eaet .Aaia ' 11 l? 19 14 19 17 14 15 2? 2~ 27 22 Caribbean 20 20 20 22 4. Leaa developed reciona ot lower denaitl Total ot croup Tropioal South ~rioa 1:2, 19 - 19 2) !2 26 - 28 3u Northern Atrioa 12 14 17 2$ South-Welt Alia 9 19 21 31 ·Middle .A.merioan Mainland 16 20 29 3$ 'I'Z'OJ)ioal .A.trioa 1$ 17 1$ 24 Southem Atrioa 24 20 19 26 Other Oceania 9 14 9 19 Sou:raea Ul Workiq Paper lo, l$, Annex Table 5: DECENNIAL INCREASES IN URBAN POPULATION, ESTIMATES 1920-1960 (Percent) • Region 1920-30 1930-40 1940-50 1950-60 1vorld Total 28 g1: 1!1 ll - 1. More develoEed resions of high densitz Total of group 19 18 - u 21 Western Europe 12 13 6 16 Northern Europe 13 8 7 8 Southern Europe 27 20 18 24 Eastern Europe 21 17 -8 22 Japan 42 49 14 38 2. More deve1oE!d rel!one or lower densitl Total or group li 20 1±1 li - Northern America 30 7 41 37 SoViet Union 50 96 6 56 Temperate South America 37 29 47 39 Australia and New Zealand 22 19 29 40 ~- Less develoE!d resions of hiSh densitl Total or srouE Mainland East Asia - 34 30 - 39 25 12 30 57 - 65 Middle South Asia 31 46 52 48 South-East Asia 57 53 57 59 Other East Asia 86 110 90 67 Caribbean 55 41 51 46 4. Less develoE!d resions or lower densitl Total of group Tropical South America - 29 35 ~ 51 60 82 88 70 Northern Africa 34 38 47 53 South-West Asia -9 53 57 59 Middle American Mainland 53 44 70 97 Tropical Africa 60 h9 77 112 Southern Africa 49 48 liB 47 other Oceania - - - 350 Source 1 UN Working Paper No. 15. Amg Table 6:Numbt'r of hlg cities nud multlmllliun tltlc:sln the world and mnjor areas, 1910-1960 /tltl/ur .,.,. 1910 19JO IHQ IPSO 196t1 ----------·----- Bit: ~ith-., ( 500,000 il•lmMtums ond over) World total • • • • • • • • 83 102 126 JS8 2.14 More developed m:l,)or at c::1s • 62 73 8.5 97 126 Europe ••••• 40 47 50 52 56 Northern America • • • • 18 20 21 29 41 Soviet Union • • • • • • 2. 4 12 14 25 Oceania • • • • • • • • • 2 2 2 2 4 Less developed major an:as • • 21 29 41 61 108 East Asi:\ •••••• 11 13 J6 22 50 South Asia • • • • 4 7 14 22 29 Latin Amcric:n • • • • • 5 7 8 l1 19 Africa • • • • • • • • • 1 2 3 6 10 More developed rcaions • 69 8J 94 lOS 139 Europe ••••••.• 40 47 50 52 56 Other"' • • • • • .' •• 29 34 44 53 83 Leu developed regions • • 14 21 32 53 95 Multimillion clll~s (2,500,000 ln- . lulbitontstllld over) world totnl • • • • • • • • 7 II 15 20 26 More developed rruijor areas • 6 8 10 12 12 Europe • • • • . Northern ~rica·. • • • • 4 2 4 3 4. 4 4 6' ,. 4 Soviet Union • • • • • • • 1 2 2 2 ~1nia • • • • • • • • • ..5 Les.~ developed nuijor IU'C3S • East Asia • • • • • • • • J I 3 2 3 8 3 14 ,. South Asia • • . . • • • 1 2 3 Latin America • • • • • 1 1 3 4 Africa • • • • • • • • • More developed regions • • • 7 II .. 13 ... J5 15 1 Europe ••••••• 4 4 4 4 4 Other• • • • • • • • 3 7 9 11 11 Leu developed regions • • 2 5 11 • Nortbcro Amcrlc:a, Soviet Union, Japan, Tcn1perato South America, Australia and New 7A:Aiancl. Souroe1 Annex Table 7: PSRCmTAOE OF Noti-A. LATIN AMERICA Tropical South America: Bolivia (5-9-50)*, Brazil (l-9-60)c, Colombia (15-7-64)*• Ecuador (25-u-62)*, GUyana (20-3-65)a, Peru (2-7-61)*, Surinam (31-3-64)*, Venezuela (26-2-61)*. Til!fate South America: Argentina (30-9-60)*, Chile (29-11-60)*, Paragua;y ( . - o-62)*, Uruguay (16-l0-63)c. · Middle American Kainland: Briiish Honduras (7-4-60)b, Costa Rica (1-4-63 )*, •Guatemala (1B-4-64)c, Honduras (17-4-61)*, Mexico (8-6-60)*, Nicaragua (25-4-63)*, Panama (11-12-60}*, El Salvador (2-5-61)*. · Caribbearu Netherlands Antilles (31-12-60)*-t Bahamas (15-11-63}*, Barbados (7-4-60)1, Cuba (28-1-53)*, Dominica (7-4-60J*1 Dominican Republic (7-8-60)C, Grenada (7-4-60}*, Haiti (7-8-50)*, Jamaica (7-4-60)*, Martinique (9-10-61)*, Annex Table 8 1 NOH-AGRICUL'l'ORAL BMPLO!MENT DmiCES IN SILIC'rlm COORTRIES {1963 • 100) JUPRICA !!:!mical Africa '.; - Year Zambia0 Cameroons0 Gabonb Ghana0 KenyaC 109.6 1958 .7 115- 109.9 82.3 75.8 19S9 109.4 104.1 92.4 80.7 110.1 . 1960 i07.9 100.0 . 93 •.4 8b.O 111.7 1961 .104.7 103.0 92.0 92.1 107.7 1962 101.5 79.7 96.5 96.8 106.6 1963 . ioo~oe 100.0 100.0 100.0 . '100.0 : 1964 .10S.8~. 101.6 ' 100.oe 102.6 100.oe 196$ I 119.•$ 108.8 100.1 103.6 -, 100• 8t . 1966 n.a. . ll2 ..6 . 10$.·3 9$.3 95.$ l 1967 n.a. n.a. n.a. n.a. n.a. 1968 Rate ot RrGVth* n.a. . <-3.o> a n.a. (-6.3) (8.1) ZI n.a. (2.4) n.a. (3.6) n.a. (-1.8) . u.- - - - ~ _ - - • ;. -·' ;?' : • < Siena !!!!: - ,, :.L~ .. Malav1° · ... . lfilter1&0 . Leone0 S. Rhodesia0 Tanzania0 '· Uganda0 19S8 ·- • i.. ~~ .,. 114.8 . 106.7 . 81.3 . - 110.~ 109.4 112.b 19$9 . ' ~ 113.3 . 10$.2 83.3 il0.6 10$.0 111.2 1960 lo0.1. _, 109.8 113.3 84.6 .. m.6 · 106.6 111.6 1961 l :: ._ - 107.3 9$.0 89.1& 108.1& 113.8 109.8 1962 'o- . c, 4 1,. 99.9 120.1 · 94.1 10$.0 uo.s 1~.0 196) \ -~ _ 1oo.oe 100.0 100.0 __1, 1oo.oe 100.0 100.0 .__: - 196h 92.0 129.7 ' . 104.2 99.0 104.3 100.8 1965 - -- ----- - _.., ~ " n.a. - 12$.6 ': - . 113.1& n.a. 106.9. 111.1 1966 !,- "-· n.a. n.a. 11).8 ~ n.a. 114.3 . 111.b 1967 c I. ~~ n.a. : n.~! . : 110.1 n.a. n.a. 117.8 ' 1968 n.a. - n.a. n.a~ n.a. n.a. n.a. Rate ot growth* (-1.7) (2.1) (3.7) (-2.1) . {0.9) ' (-1.7)/U . (4.4)2[ · (Continued on next page) - 2 - . Annex TRb1e 8 : (Continued) AFRICA South Atrioac - Year 19S8 Total Population 9S.O White Poe!J.ation 9S.S 19S9 9$.6 9S.u 1960 ~.1 9S.3 1961 96.9 95.8 1962 97.6 97.1 1963 100.0 100.0 1964 108.1 99.S 196; us.s 104.S 1966 120.8 110.2 1967 123.4 112.1 1968 n.a. 11.&. Rat~ ot growth* (2.8) (1.6) - ASIA Other East Asia China& -Year 19S8 Korea• n.a. (Taiwan) n.a. Ryukyu Is.a 72.9 1959 n.a. n.a. 79.2 1960 n.a. n.a. 86.7 1961 n.a. n.a. 92.5 1962 n.a. n.a. 9S.8 1963 100.0 . 95.'18 100.0 1964 10'7 ..0 100.0 101.) 1965 120.S 101&.7 105.4 1966 124.8 . 112.0 111.7 1967 136.6 123.; 118.3 1968 11&9.0 n.a. n.a. Rate ot growth* (6.)) (6.4) {4.3) (Continued on next page) - 3 - Annex Table 8 : (conti~ed) • - ASIA - South-West Asia - Year 1958 Turkeyb Syria& Israelb 83.6 2l:2rus< 97.0 86.3 n.a. 1959 86.7 n.a. 87.8 95.6 1960 87.0 n.a. 91.2 96.2E 1961 96.3 85.2 85.oe 96.7 1962 95.2 84.3 92.8 98.4 1963 100.0 100.0 100.0 100.0 1964 107.7 94.9 107.0 96.6 1965 128.2 94.7 110.8 99.3 1966 141.9 95.7 110.8 102.7 . 1967 153.0 n.a. 107.6• n.a. 1968 n.a. n.a. 114.9 n.a. Rate or growth* (5.9) (1.9) (4.4) !.§. (o.6: South-East Asia !!!!: Pbi1ippineso Singaporec 1958 87.4 107.3 1959 90.2 103.1 1960 92.5 93.0 1961 94.8 94.5 1962 97.3 95.1 1963 100.0 100.0 1964 102.5 104.6 1965 105.9 110.6 1966 105.1 121.;8 1967 105.7 122.0 1968 n.a. n.a. - Rate or growth* (2.0) (3. 7) t:L (Continued on next page) - 4- \rmex Table 81· ( Contimled) LATIN AMERICA TroEical South America - Year 1958 - P81'"Ud n.a. 1959 n.a. 1960 n.a. 1961 92.5 1962 96.1 196.3 100.0 1964 104.1 1965 108.4 1966 n~a. 1967 n.a. 1968 n.a • Rate ot trowtb* . (4.1) Middle American Mainland - Year 1958 ElSalvadorO n.a. Br. Hondurasc 6.3.3 1959 n.a. 68.2 1960 n.a. 7.3.0 1961 9$.7 97.8 1962 99.3 100.0 196.3 100.0 100.oe 1964 109.6 n.a. 1965 128.·8 82.5 1966 126.6 90.1 1967 144.0 94.1 1968 n.a. n.a. Rate of growth* {7.3) {1.3.6) L!! Caribbean Year Puerto Rico• - 1958 85.2 1959 87 • .3 196o 89.08 1961 91 • .3 1962 94.2 196.3 100.0 1964 108 • .3 1965 ll4.5 1966 120.$ 19~ 19 12g.1 Rate ot growth* .7) 1?4.7 {Continued on next page) - 5 -. Annex Table 8 .1 (Continued ) ~ . ·• Hotel Generally, the employment indices reter to the number ot employees. , In some oases, other status groups are also included. In addition,. tor certain series "services" are not tully represented. In a few cases some sectors or occupations have been excluded. In some cases only persons over a minimum age are included. Some of the time series are based on one month ot the year. a labor force sample surve,ys b compulsor,y social insurance statistics c establishments statistics d otticial estimates e new or revised series t excludes rural areas * Rates ot growth oal~ulated· by fitting a logarithmic time trend. /1 1958-1962 ~ 1958-1962 1962-1966 1958-1963 ~ 1963-1967 1961-1966 1960-1967 1958-1962 Source: International Labour Ottice, 1968 Year Book of Labour Statistics, Tat pp. 306-308. . 1 ~ -·- - ._ _ ,...,.. ... .· -· -·-- -· ·----- - - -·- - ·- .•. ·-· · - -··· .... --- . ... _ 1Mf!'X Table q a PROJBCTtONS OJI' URBAll POPULATIOtl, 1960-2000, ASSUMED .'1'0 DlCB&SE AT MCII: 'fHI RA.TIS Or . . . . TOr.U. POPULATI~ d (MILLIONS) • 1960 Lower As8u~iona · Bl~er AsSUMPtions Region lstiute i97~ 1980 ~ ~~ i9'~ 9~ 19~ 2ooo World Total 760.3 1.027.1 1.~8L.6 1.~11!_.~ 2,292~1 1,061.1J 1..521..2 ?Ll_47.L 2.R97 .2 1. More deve1ol!!!! region~t ot h1f'J1 densi tz: TotAl of rrouop · !l2.& 263.L ~ '24.6 lSO.la . 268.5' lli.:.2 ~ h02.8 W.etem !uror- 6S.4 7L.4 82.4 90.2 98.0 7S.6 86.0 96.7 108.6 NoJ'theJ"ft Europe 4S.o L8.l S0.6 5'2.4 SL.7 49~0 5'2.6 S6.2 60.8 South•m Eut'OJ18 45'.9 . S2~S· ;n~t.· 6U.4 70.~ S3.S 61.7 70.7 8o.9 . Eutern Europe [! · . )1.6 )7.) 43.S 1.9.9 SS.6 )8.0 hS.7 5'4.) 63.3 Japan 42.9 5'0.9 60 •.) 67.7 n.e S2.b 66.0 77.6 88.1 2. More neveloE!d regions or lov81" dendtz: Total or RI"OU'P ill& m.:2 ill.& ~ ~ 287.0 ~ h7:).3 . 5'90.8 Northern ~rica 115'.3 1)8.8 . . l6U.8 190.8 210.9 1b).4 178.2 222.5' 270.6 Soviet Union 78.o· 102.) 1)0.8 168.6 205'.6 109.3 lb8.7 198.S 2SS.3 Temperate South Amorica 17.3 2).4 28.6 . )b.o )9.6 2).8 )0.7 . 37.7 16.0 .lUitNl.ia and Nov .Zealand 8.2 10.4 . 12.8 . 15'.1 17.2 10.5' 13.2. 16.6 19.9 3. Lees develoi!d rs1ons or hie!! densitz: . ' Total or group 213.0 JlLl ~ ~ ~ -~. ~ 801.? l.B3.~ Mainland East Aaia 90.t. 118.2 1S2.6 190.6 2)0.6 1)2.2 199.) 287.9 )82.1. Middle South Aaia 76.6 122.9 1R8.9 269.3 36S. 7 . 12).9 201.~8 )07.8 434.7 South-Eaet Asia 27.b . 16.3 72;.8 115'.4 172.6 . . 16.8 75'.9 127.4 208.1 Other last .laia 13.8 23.4 )6,6 52.2 69.8 23.5' 1;0.) 5'9,6 . 19.9 Caribbean 4.8 ?.S 11.2 15'.4 . 20.1 7.6 • 12.2 19.0 28.2 - ---- ··-- Ia. Leas deve12i!!d 1'flgiona of lover denaitz ·--- -· . . ... .. . - ·· - ·-·· ... Total o t tn•Oiiop .. t. !Ll. ~- !22:.1 1&38.8 6i0.3 •172.9 311.9 ru.:.! ~ tropical South •rtca )),) 61.8 lOS.? 15'4.9 205'.0 62.2 Ula.S 184.6 267~0 Jlort.hern Africa 16.9 28.8 48.2 73.S 100.9 29.2 52.6 87.6 127.1 8outh-W.et Asia 13.S 22.6 )8.2 . 60.4 · 86.8 22.9 b0.3 68.;L 10),6 Middle Allerioan Mainland lb.) 27.0 48.b ·'12.7 98.2 27.2 S2.h 86.7 127.9 Tropical Atrica 13.9 21.6 . 33.5' 5'3.2 86.9 . 21.7 35'.6 60.S 106.8 Soutbem Jilrica S.. 7 9.5 15.9 23.8 )2.1 9.5' 16.2 25'.7 )7.3 Other Ooeania 0.1 0.2 0.2 o.l o.l. 0.2 0.) . 0.4 0.6 ~ Until one-halt ot the 'POJ'U].ation ia u~. thereafter ni"all'OJNlation 11 ua\aMCI to r.aiD oonatant (abaolutel7) 1 vhile urban population a'baorba all the 1Aoreue in popq.ation. I &, Inaludee Berlin. Soui'Oel UN Po~lat.ion Diviaion,•Urban and air&l Population Orovt.h, 1920.1960, with PJ"ojeotiona,•Workinl Paper lfo. 1~, • 6I Sept.OIIber l967, ~ 134·135. . · · l .. ... ~ •• Annex Table · 10 : PROJECTIONS OP URBAN LAB>R FORCE [!, 1960-2000 · ( lllilllons ) 1960 Lower Assumptions Higher As!'!DI!J)tions Estimate 1970 1980 1990 2000 1970 1900 1990 · 2000 - MALE Atrica 10.0 16.3 26.6 41 ..0 S9.8 16.S 28.4 47.3 73.8 Asia 38.2 62.4 98.0 144.7 202.2 62.9 104.3 163.9 240.S Latin Aaerica 1S.9 27.3 44.2 63.1 82.7 27.5 47.8 74.7 106.6 nMALE Africa 4.4 7.0 11.4 17•6 26.2 7.2 12.0 20.0 32.0 Asia 7.7 12.6 19.8 29.2 44.9 12.7 21.1 33.1 48.6 Latin J.erica 6.3 10.8 17.S 24.9 32.7 10.9 18.9 29.S 42.2 TOTAL Africa 14.4 23.3 38.0 S8.6 86.0 23.7 40.4 67.3 10S.8 Asia 4S.9 7S.o 117.8 173.9 247.1 7S.6 12S.4 197.0 289.1 Latin America 22.2 38.1 61.7 88.0 11S.4 38.4 66.7 104.2 148.8 (Continued on next page) - 2- Annex Table 101 (Continued) 1JRBAH BASIC ACTIVITY lU.TES - MALE ·Africa so.o Asia so.? Latin America IJ9.1 I'IMALB Africa 13.2 Asia 13.8 Latin Allerica 16.8 JJRBA! pr U:IO (•1•• per 1,000 fenaales) Africa 1,088 Asia 1,147 Latin America 928 l! "Urban" refers to localities or 20,000 population or more, "labor force" to persons 15 years and older. Sources: Urban population projections: UN Population Division, Urban and Rural Population Growth, 1920-1960, with Projections, lbrking Paper No. 15, September 1967, pp. 134-135. Age-sex specific labor force participation rates: Ettore Denti, "Sex-Age Patterns of Labor Force Participation by Urban and Rural Populations," International Labor Review, December 1968. Sex-age structure of urban populations at