Urban development and housing in the Indian draft Sixth Plan : a comment

DRAFT u,p-7LI THE WORLD BANK DEVELOPMENT ECONOMICS DEPARTMENT URBAN AND REGIONAL ECONOMICS DIVISION URBAN AND REGIONAL REPORT No. 79-14 URBAN DEVELOPMENT AND HOUSING IN THE INDIAN DRAFT SIXTH PLAN: A COMMENT RAKESH MOHAN May 1979 These materials are for internal use only and are circulated to stimulate discussion and critical comment. Views are those of the author and should not be interpreted as reflecting the views of the World Bank. References in publications to Reports should be cleared with the author to protect the tentative characters of these papers. URBAN DEVELOPMENT AND HOUSING IN THE INDIAN DRAFT SIXTH PLAN: A COMMENT Rakesh Mohan * Urban development is a field that has traditionally been neglected in Indian planning. With the new emphasis on the importance of agriculture and rural development in the nation's economy, the Draft Sixth Plan document pays much greater attention to rural areas than earlier plans, as indeed it should. It is regrettable, however, that urban .development has been neglected almost completely: a scant six pages being devoted to the problems of India's towns and cities which among them now contain more than 130 million people. Even if it is true that urban areas have received a greater proportion of public investments in the post-independence period than, perhaps, was optimal, urban development and planning, per se, have received scant attention. However, an argument suggesting that greater care should be taken in the formulation of the urban development portion of the plan does not necessarily imply the need for higher investment in urban areas. Rather, the argument is that explicit attention should be given to urban development and planning so that the development of the country's urban areas (and, by implication, rural areas) is consistent with the location of activities implied by the sectoral plans. Regional and sectoral planning cannot be done separately. This comment first reviews the recommendations and policy actions suggested by the plan. An alternative approach is then suggested in light of the pattern of urbanization, the size distribution of cities and location of activities that are commonly observed in less developed countries and, in particular, in India. * This comment reflects solely the opinions of the author and not the World Bank, the organization with which he is employed. -2- The Draft Plan Two premises can be inferred to underlie the view of urbanization taken in the document: (1) that India has been urbanizing too-.rapidly in recent decades, and (2) that urbanization is undesirable in itself. From these premises follow the recommendations within the plan context: . the rate of urbanization should be slowed down . the rate of growth of metropolitan cities should be slowed down or reversed o investment should be directed toward infrastructure in small towns . industries should be banned within limits of large cities . programs for small industries should be restricted to small towns of less than 25,000 population . the setting up of offices, commercial establishments and centers of entertainment in cities should be banned * central and state government offices, banks and other institutions should be moved to less congested areas of the city. The actual programs then suggested for the plan are not all consistent with these general guidelines. The main components of these programs are: . slum improvement . investment in urban infrastructure, i.e., roads, pavements, bus sheds, market places, etc. 0 development of state capitals: Calcutta, Gandhinagar, Chandigarh and Bhopal * water supply, sewerage, roads and road transport investment in small and medium size cities 6 development of growth centers in the national capital (Delhi) region . investment in research in order to find measures to arrest growth of large cities and measures to decongest them. 3- Finally, an assertion is made that the pace of urbanization will come down to "desired" levels when the income levels and quality of life of the bulk of the rural population reaches "acceptable" levels. Under the heading of housing a similar recommendation is made to bring down the growth of large cities in order to reduce the cost of housing. The specific recommendations for action include: . decrease the cost of urban land . encourage private housing . encourage sites and services projects . implement the urban land ceiling act and release undeveloped land . amend rent control acts in order to remove impediments to investment in private housing. There are therefore no specific measures recommended to arrest the growth of large cities nor of urbanization in general. It should be noted that the guidelines outlined earlier are essentially negative in nature and that little constzuctive is offered. While the guidelines suggest that the setting up of offices should be discouraged in cities, 20 percent of planned investments in urban development are then earmarked for state capital projects. The other measures suggested are unexceptionable but have little relationship with the guidelines. It is clear, therefore, that the urban development sections of the plan being based neither on sound premises nor on clear objectives, are internally inconsistent. In any case, the total urban development investment of Rs752 crores (a little over one percent of total planned public sector outlay) spread over the five-year plan period is negligible. 4- An Alternative Approach A more rational and practicable approaTh to urban development and urbanization would view it as an integral component of national development. When viewed in,this manner, the phenomenon of urbanization should be imbued with neither intrinsic desirability nor undesirability. The urban develop- ment plan should then be subject only to the overall objectives of the plan and the existing resource constraints. The overall objectives of the plan are stated to be: (1) the removal of unemployment and significant underemployment (2) an appreciable rise in the standard of living of the poorest sections of the population (3) provision by the state of some of the basic needs of the people in these income.groups, like clean drinking water, adult literacy, elementary education, health care, rural roads, rural housing for the landless and minimum services for the urban slums. As long as these objectives are satisfied, it should be a matter of indifference whether people live in urban or rural areas. Indeed, if rapid urbanization promotes the removal of poverty with lower resource costs it should be encouraged rather than discouraged; if it is easier to create employment in urban areas that is what should be done; if basic needs can be provided more economically in population agglomerations rather than in dispersed settlements in rural areas, migration to towns and cities should be encouraged. The point is that neither rural nor urban living is intrinsically desirable so the plan should merely seek to attain its overall objectives regardless of the implied rate of urbanization. One basic problem is the plan document.relates to the confusion between policies affecting the spatial distribution of population nationally and those within cities. Even if it is the case that large cities are 5- undesirable, the arrangement of activities within the large cities should receive adequate attention. Thus the approach to urban development should consist of two distinct components: measures which aim to attain the desirable spatial distribution of population across different sizes of cities and measures which aim to achieve an efficient spatial structure within cities. Both sets of measures necessitate an understanding of the functions of cities: the first set needs appreciation of the different functions of different sizes of cities and the latter how these functions can be efficiently performed within cities. Urbanization Urbanization should be seen as a phenomenon which is part and parcel of development in general. Thus urban development does not take place at the expense of rural development nor vice versa. Certain activities are better performed, indeed require, agglomerations of people while others do not. The location of activities has therefore to be seen in the total context of the pattern of activities existing in a country and its likely development in the future. The rate of urbanization in India has been one of the slowest in the world. The level,that is, the proportioh of total population residing in urban areas, rose from 17 percent in 1951 to less than 20 percent in 1971. This slow growth reflects both the low absolute level of income in the country as well as the low rate of growth of the economy as a whole. There is therefore little reason to believe that urbanization has been too rapid in India. Contrary to the suggestion in the document that the rate of growth of urbanization in the country will decrease when income levels increase, it is likely that this rate will actually increase if the rate of growth actually 6- increases to 4.7 percent per annum as planned. It is important to recognize this so adequate and timely investments are made in urban infrastructure before it is too late. The reason for such an expectation is simple enough. As incomes increase, a smaller proportion of people's budgets will be spent on food (the income elasticity 9f demand for food being less than unity and decreasing with growth in income) while that spent on urban-type goods will progressively increase. Thus the demand for labor in urban areas will also increase and consequently the level of urbanization as well. Concurrently, productivity growth in agriculture (caused both by biological as well as mechanical innovations) will make possible the release of labor to respond to urban demands. Urban type goods are taken to include both manufactured articles as well as services. It may be argued, of course, that such goods need not necessarily be produced in urban areas. The manufacture of goods needs a multitude of inputs as well as networks of distribution to reach their markets. They can then be based in rural areas when transportation and communication is efficient and swift enough such that proximity to inputs and markets is not longer a factor in the location-making decision. That can be expected to happen only at the highest level of development and the available evidence suggests that even then the large majority of manufacturing and service activities choose to locate in urban areas. The crux of the matter comes down to locating industries where it is socially most economical. Similarly, services such as education,health, entertainment, trade activities, characteristically need large pools of population to be viable. A university or a hospital needs to be a certain minimum size to be able to afford a Variety of functions as well as specialization in some. Lower level services must -7- naturally be brought to rural areas: investment in primary schools, health centers, mobile markets, theaters, etc. must now receive priority. But they should form part of a planned system where they have an adequate relationship with higher level institutions which service them as necessary. The higher level institutions should constantly innovate to improve the servipes offerred by the primary and secondary centers. A concrete example of such a system is offerred by the network of health centers in Cuba: a country which has devoted explicit attention to the appropriate rate and pattern of urbanization. In Cuba, each rural community or small town is provided with a primary health center which refers difficult cases to a secondary health center located in a larger town which, in turn, refers the most complex cases to research university-type hospitals. The analogy can clearly be extended to other services such as education. The practice of agriculture itself requires innovation which can only be carried out in a reasonably advanced system of research institutions such as the one that India possesses. The Indian Agricultural Research Institute in New Delhi is of little use without its far flung research stations and the research stations would lose their usefulness without the IARI. Thus cities are as necessary to agriculture as food is to cities. The lack of investment in one area slows the growth of the other. Indeed, it may be said that the relative neglect of agriculture in the past has also slowed the growth of our cities. An urban development plan therefore must not seek to slow down the rate of urbanization nor attempt to reverse it. Higher rates of investment in agriculture and the resulting higher overall growth rates will speed up urbanization and therefore cities should receive an appropriate level of investment adequate for their growth. As such the main constituents of an urban development plan should be similar to those of rural development plans. The plan should provide for: (1) activities which promote employment of the poorest sections of the population; and (2) promotion of schemes to provide drinking water, sewerage, education, and health care for the poor. The urban poor i are as much in need of attention as the rural poor and a plan aimed at relieving poverty should pay compazable attention to both. The Size Distribution of Cities There is, as yet, little agreement on the optimal size distribution of cities. Systems of cities are often found to follow certain mathematical distribution functions but there are seldom related to any understood economic rationale. There is thus no prima facie case for discouraging the growth of large cities nor of encouraging them. The case against large cities is often based on the existence of diseconomies of scale in the provision of urban services. There is as yet little evidence to support this claim. What evidence there is suggests that costs of urban services per capita decrease from small towns to intermedite size cities (with population three hundred to eight hundred thousand). Thereafter, the evidence is mixed: some find that they decrease further, others find that they increase or remain largely constant, depending on the particular concept of cost used. This is an empirical question which can, in principle, be solved; though practically the appropriate methodology and cost concepts that ought to be used is not at all clear.A 9.- Therefore before large cities are actively discouraged in India, there should be some attempt to study the comparative costs of services in different size cities. This question though has the benefit side too. Do cities produce productivity effects which also need to be accounted for? These effects are difficult to quantify but are normally adduced to agglomeration effects. The higher per capita income levels in large cities (even price adjusted) indicate that productivity increases with size of city. This productivity effect tends to compensate for higher service costs even if they exist. Higher transport costs are often cited as one component of diseconomies of scale in large cities. It must be pointed out that the opposite might be the case: transport costs necessary for similar activities could be still higher if the same population were dispersed over a wide area. Thus there is little economic rationale for limiting the size of cities. It is also worth pointing out that even planned systems (such as China) have not been able to limit much the size of their cities nor the rate of urbanization. Indeed, the rate of urbanization in China has been higher than that of India over the past three decades. What is understandable is that large cities appear to be more difficult to manage from the public management viewpoint--in particular that of the city administrator. It is not clear, however, if this is so from the national point of view. It might, in fact, be easier as well as more economical to manage large numbers of people in a few cities rather than dispersed over a greater number of towns. Management costs per capita would probably be lower.,for large cities. While it is understandable why city level public officials might favour limiting the size of their cities, it is surprising why state or national level officials often do the same. - 10 - The preceding is not meant to imply that large cities should necessarily be encouraged. But it does indicate that they should not be -actively discouraged. That larger cities have grown faster in India than the smaller ones in recent decades itself indicates that there are some economies of scale pushing the pattern of the spatial distribution of population in this direction. The rationale should basically depend on the net benefit/cost calculation. Would the poor, for example; be serviced more economically in smaller or larger towns? Are costs per job lower in smaller towns? Can large city activities be located at lower overall costs in smaller settlements? Location of Activities within Cities Irrespective of the overall distribution of population across cities, it is important that the structure of cities be adequately understood so that urban investment programs are properly directed. Where should different types of industries be located within cities? Where should commerce be located? Where should offices be located? What are the functions these activities are serving? 1. Location of Industry The current plan seeks to ban industries within limits of large citi6s and to encourage programs for small industry in small towns only. It is a misunderstanding of the function of cities which leads to such policy recommendations. Cities exist because of the realization of agglomeration economies. Such evidence as there is suggests that small industry requires large cities for success. Small industries live in a greater atmosphere of uncertainty in both the input and output sides. Thus the existence of small industry requires a significant pool of labor - 11 - to draw from and the ready availability of a variety of raw materials at short notice on the one hand and market proximity on the other, Large industries can locate outside large urban centers much more easily than small industry"since their product lines and input requirements are more stable and therefore do not necessitate frequent face to face contact. For large industries, raw materials and products are characteristically transported in large amounts and therefore do not need to share transport with the products of others. It may therefore be more rational to encourage the location of large industry in or near small towns or the periphery of large cities. Small industry is more likely to succeed if it is located in or near central cities as it often does. The congested areas of old Delhi are a case in point. Despite official policies to the contrary, small industry has continued to expand in these congested areas.- The experience of other large cities is similar. While larger firms characteristically move out of central cities to the periphery or to other towns small firms continue to locate near the center. Thus garment manufacturers, printing and publishing firms, repair shops, etc., are the type of activities that can be encouraged in central citiesi 2. Location of Commerce The plan suggests that central and state government offices, banks, and other institutions should be shifted to less congested areas of cities. This would merely result in increasing the cost of transactions both in terms of time as well as money costs. The function of a city is to facilitate commerce by locating it in congested areas such that business can be carried out with a minimum of transport and other costs. Service activities such as government offices, banks, insurance companies, schools, - 12 - hospitals, other financial institutions, retail and wholesale activities, require congestion for success. It would be negating the functions of cities to encourage the dispersal of such service activities. 3. Provision of Infrastructure The provision of infrastructure such as roads, sewerage, water supply, electricity, housing, telephones is often regarded as a drain on the productive side of the economy. This is not necessarily so and such activities should be regarded as at least as productive as most manufacturing industries. The technology for such activities is characteristically labor-intensive. In addition, they have a multiplicity of backward and forward linkages. The demand for materials created by such activities promotes industrial production. They have clear income and emplolment- generating effects with their own second order effects on general demand. Further, the health effects of such basic need provision themselves can be expected to have significant productivity effects. 4. Housing The subject of urban land and housing must be one of the more misunderstood subjects in economics. The plan suggests that the price of urban land should be brought down in order to bring down the cost of urban housing. That the price of urban land has been rising very rapidly is a popular idea but not backed with evidence. There are little data to support this view since data on the price of urban land are typically difficult to collectw However, even if the price of urban land has risen compared with the general price index it indicates its rising scarcity value. That should be taken as an indication of a scarce resource requiring more efficient husbanding. It indicates that housing should be more dense rather than less. Lowering the price of land would induce opposite measures. - 13 - The provision of housing is an activity that is typically labor- intensive. Housing construction--both private as well as public--should therefore be encouraged. Given the resource constraints and priorities of the public sector, it is clear that investment in private housing should'be'increased. This can only happen if there is appropriate financial intermediation which is lacking in India. Thus housing banks, etc. need to be established with governmental initiative in order to channel .private savings towards housing. Incentives in such financial intermediation should also be directed to low income housing. The poorest of the population cannot afford housing but need space for shelter. It must be recognized that there is little solution to this problem because of their lack of purchasing power. Unsightly shanty settlements will long remain part of our cities' landscapes. Public policy can only provide some security of tenure or rental innovations to these lowest income groups along with essential services like water and sewerage which also improve the health of the population. Housing standards have therefore to be brought down to induce increased private investments in housing and shelter. A Suggestive Plan for Urban Development The above analysis suggests a plan on the following lines: 1. Urban Public Works Along with rural public works, the key element of urban development should be r2assive investment in urban public works to build urban infra- .8tructure for the provision of water supply, electricity, housing and shelter, sewerage, etc. Such a program would generate significant induced demand for a host of raw materials and provide employment to the lowest unskilled sections of the population. The income generated will, incidentally, - 14 - also raise the demand for agricultural goods as well as industrial goods. These public works should be located in all sizes of cities, according to need. Such a plan is eminently practicable as long as there are significant food reserves which can form the base of the required resources. 2. Industry Much more information is required on the relative job creation abilities of small and large industries. Indeed, the accent, perhaps, should be on achieving the right product mix rather than on industry size per se. The size of establishments and their location as between small and large towns and between the core and periphery of cities should then be dependent on the pattern of input and marketing requirements of the product concerned. Thus there is little reason for specific encouragement or otherwise of the location of industries in large or small cities. Adequate attention should be paid, however, to pollution and anti-pollution measures in the shape of regulations as well as fiscal devices. The expansion of industrial production essentially requires a substantial increase in urban as well as rural incomes. If an investment oriented massive urban and rural public works program is successful, its induced demand effects will increase the demand for consumer goods required by low income sections of the population. Thus industrial strategy, in addition to providing key capital and intermediate goods, should be aimed at encouraging low income type consumer goods. The accent should then be on producing low cost goods whether they are from the handicraft or handloom sector or from the factory sector. Linked with the industrial strategy should be attention to systems to aid rural as well as urban marketing. Thus there - 15 - are no simple guidelines which advise the banning of large industries and which limit the creation of small industries in small towns only. The rationale should be much more product and market related. 3. Services Cities exist as much for the provision of services as for industry. Some marketing studies tend to show that beyond a certain income even rural people tend to look for services in urban areas rather than their own village. There should be explicit recognition of this to aid better planning of the distribution of services. Banking and financial services, wholesale and retail. trade, personal and administrative services all thrive on agglomeration, proximity and even congestion. Thus much more thought should be given to appropriate location before administrative offices and other services are moved to city peripheries in'the interest of decentralization. Investment in health and education services in cities is not necessarily detrimental to rural areas. Indeed, what is needed is a hierarchy of services in different areas. Low level health centers and primary schools should clearly be located near the population they serve. But higher levels of services can be located in agglomerations aloing with adequate provision for transport and referrals from the rural areas. The urban poor are as much in need of attention as the rural poor. Special efforts to provide them with employment, along with nutrition, health and education services are as necessary as they are for the rural poor. The provision of water services and sewerage through public works can also be seen as a means to improve public health leading to higher productivity. - 16 - 4. An Agenda for Research It should be obvious from the foregoing that urban development policy needs to be improved considerably in India and applied research on urban related problems is therefore very necessary. No longer can spatial implications of other policies be left to take care 6f themselves. An economic plan should be accompanied by a spatial physical plan which spellls out the consequences of the economic plan. Much concern has been expressed about city size without adequate understanding of the functions of different types of cities. Further research should focus in the desirable distribution of activities and services with regard to the ultimate objectives: e.g. removal of poverty; rather than on the desirable distribution of city size. The latter should be a by-product, or outcome, of economic policy, rather than an explicit objective. Research is needed under the following broad themes: (i) Employment Creation and Activity Location. The plan suggest that certain types of activities and industries should be restricted to small towns. Before such a policy is put into effect work should be done in (a) the existing pattern of activities within cities as between small, intermediate and large size cities; (b) the comparative job creation aspects of small and large establishments for similar products, as well as total costs; (c) comparative costs of industrial production in certain selected items in different size cities as well as between urban and rural areas and their job creation aspects. Much more needs to be done than the mere cataloguing of these comparative costs. The differences need to be analyzed. Are they because of city size per se or other associated causes? Further, the concept of - 17 - costs should be a broad one: account must be taken of marketing costs, access to information, access to capital, labour, etc. It should be emphasized once again that the issue is how to obtain maximum output and employment along with a reasonable distribution of income rather than city size but the latter is important if only because it has become an explicit policy concern. It is important that, at a minimum, blatantly wrong policy choices are not made. (ii) Provision of Services, Much importance has been given to the provision of services in urban areas and comments were made on comparative costs of different size cities. The latter issue is difficult to tackle because different types of cities and towns provide different levels of services. What we need to find out is how a basic minimum of services can be delivered to the majority of, if not all, the population. Only then it becomes a meaningful question to ask if a similar level or bundle of sdrvices are more or less economical to provide in rural and urban areas on the one hand and in different sized towns on the other. Thus compilation of information on costs of existing services in different towns and cities, while an essential first step, will not be adequate. Costs need to be calculated for specific delivery systems and the desirable sizes of agglomerations then deduced. (iii) Future Patterns of Urbanization: The variety of development experience within the country provides an instructive window into the future of urbanization in the country. The green revolution in Punjab, Haryana, and Western U.P. has been analyzed extensively in terms of its effects on agricultural production, productivity, employment and income distribution in rural areas. But what have been its effects on urbanization? What have been the linkages in rural areas on the demand for urban type -18- goods and services? How have medical, educational, governmental services responded to this demand? Where have they located? How has urban employment responded and in what areas? What types of small scale (and large scale) manufacturing activities have appeared? Such work would provide useful lessons for future rural as well as urban policy for other parts of the country which have yet to reach the income levels of Punjab and Heryana. This is clearly a tall order for a realistic research program: But the essential directions are clear as well as their importance. FOOTNOTES 1/ The Urban Land (Ceiling and Regulation) Act came into force on February 17, 1976. Its key provision is that it limits the extent of urban land that can be owned by each individual or other entity. The limits vary with city size: 500 sq. meters in the four metropolitan areas and up to 2000 sq. meters in the smallest urban areas. 2/ For a demonstration of this point and a more detailed discussion of urbanization in the context of development see Rakesh Mohan, "Development, Structural Change and Urbanization: Explorations with a Three Sector Dynamic General Equilibrium Model Applied to India, 1951-1984". Unpublished Ph.D. Dissertation, Princeton University. Princeton, 1977. 3/ There is now some evidence that the urban poor may suffer more from malnutrition. See Samir Basta, "Nutrition and Health in Low Income Urban Areas of the Third World", Ecology of Food and Nutrition, Vol. 6, 1977, pp. 113-124. 4/ Should we, for example, merely consider water, sewerage and electricity type urban service costs, or also transport, road, pollution and other costs. What are urban costs? 5/ For some calculations for Japan, see Koichi Mera, "On The Agglomeration and Economic Efficiency", Economic Development and Cultural Change, January 1973. 6/ See Ashish Bose, "India's Urbanization 1901-2001" Tata McGraw Hill, New Delhi, 1978. 7/ Gangadhar Jha, "Changing Structure of Delhi's Walled City: Problems and Prospects" Nagarlok, Vol. IX, No. 2 (April-June 1977)., pp. 78-88.