Child Mortality in New Industrial Localities: an Indian Steel Town

As Asia becomes increasingly urbanized the effect of new industrial development on child mortality becomes of increasing interest. In India, considerable investment has been made in the social infrastructure of industrial new towns. This survey of Durgapur steel town in West Bengal shows that although the average level of child mortality in the working class population is favourable in comparison with other Indian cities, considerable differentials, that can be related to social, economic and environmental differences within the population, have arisen since the creation of the city in the late 1950s. The paper argues that the undertaking of selective sanitary interventions to improve access to drinking water (in particular) would be administratively feasible in these industrial new towns, of immediate impact, and indeed necessary if the differentials in mortality are to be eliminated. Introduction and objectives of the study As Asia becomes increasingly urbanized, demographers and social scientists interested in policy need to focus their attention on the mortality characteristics of populations living in towns and cities, and the impact of the latter on their citizens’ health. Although some studies have been undertaken on health and mortality in the metropolitan cities in the region, such as Hongkong (Goldman 1980), Shanghai (Jiang and Karkal 1981), Manila (Basta 1982) and Bombay (Ramasubban and Crook 1985), very little work has been done on the towns and cities of the next rank in population size, say between 100,000 and 1,000,000. Yet since the 1950s there has been a very substantial growth of new industrial complexes often located outside the metropolitan regions, giving rise to rapidly growing and by now very large urban settlements, sometimes styled ‘new-towns’. An example of such growth can be seen in the ‘steel-towns’ of India, three of which date from the late 1950s when the country was embarking on a strategy of establishing heavy industry largely sponsored by the public sector. These towns are Bhilai in Madhya Pradesh, Raurkela in Orissa, and Durgapur in West Bengal; the last is the subject of this paper. All three were new-towns, built on green-field sites, and have now experienced about 35 years of growth, some of which has been very rapid. Durgapur, the least of the three in size, had a population of 415,986 at the last Census count in *The authors are grateful to the International Development Research Centre and the Leverhulme Foundation for supporting their research, and to the Indian Statistical Institute, Calcutta, and the School of Oriental and African Studies, University of London, for facilitating the preparation of this paper. 166 CHILD MORTALITY IN NEW INDUSTRIAL LOCALITIES: AN INDIAN STEEL TOWN CROOK AND MALAKER HEALTH TRANSITION REVIEW VOL. 2 NO. 2 1992 1991. Elsewhere in the developing world similar cities have mushroomed from the establishment of a single heavy industry, where, as in the case of steel, plant size is subject to huge economies of scale; the steel-towns of Maanshan in China (Qichang n.d.) and Chimbote in Peru (Pryer and Crook 1988) are examples geographically far apart, but sharing certain demographic characteristics, such as a period of initially very rapid demographic growth and a concentration of young adults in their populations, most of whom are migrants. In India the creation of new industrial towns is still taking place: the port and steel-city of Vizakhapatnam in Andhra Pradesh is the most recent example, with a 1991 population of 1,051,915. It is an appropriate time to ask certain demographic questions regarding industrial cities of this kind. Is fertility falling faster here than in other cities? Is health improving more rapidly? Are the differentials in fertility and mortality lower than elsewhere? Are the social and physical environments of industrial new-towns associated with better health and lower fertility than in other towns, or not?1 This paper will focus on aspects of mortality. We take as a working hypothesis that the state is able to intervene to lower mortality levels in cities in that the provision of clean water and good sanitation can help towards that end; so can clean air and adequate shelter. In theory the State can have considerable control over the supply of these amenities, but not, of course, without cost. In mature urban environments much of the population is usually without such amenities: 30 to 40 per cent of the population in India’s metropolitan cities is designated as living in slums on this account. This is because in the past the State was either not willing to meet the cost, or not able to keep up with the increasing need, or a combination of both. To make up the backlog today in the older metropolitan cities is obviously a mammoth task which would use a prohibitive quantity of resources if attempted in a short space of time. But in newly built urban environments a more successful attempt could have been made to keep up with the need for the provision of amenities, especially when much land had been obtained in advance, as in the Indian steel towns. Insofar as mortality differentials depend on differential access to these amenities, it should have been possible to achieve relatively low mortality differentials in these towns. One aim of this study is to see whether this has indeed been the case. Another is to ascertain the potential for this to be achieved in the future. India’s crude death rate was estimated at 7.7 per thousand in the urban areas for 1980-1982 (Government of India 1988b). The corresponding infant mortality rate (IMR) was 64 per thousand live births. In the State of West Bengal, where Durgapur is situated, mortality is believed to be rather lower than the national average, with a crude urban death rate of 6.7 for 1980–1982, and infant mortality rate of 48 for 1983, the closest available date to the 1981 Census. There are a few estimates for some of the metropolitan cities alone in India, for instance Bombay, where the registered crude death rate was 8.6 (IMR of 61) in 1981, and Calcutta, where an infant mortality rate of 52 was estimated from mortality questions asked at the 1981 Census (Government of India 1988a); the latter is probably an underestimate, resulting from the use of an inappropriate model life-table.2 One of our objectives in this study is to provide an estimate of mortality for a non-metropolitan city with a heavy industrial economic base, so that we may begin to obtain a picture of mortality 1 One of the authors has examined various other demographic characteristics of such towns from census materials (Crook 1992). 2 This critical observation was made to us by Tim Dyson. It should be noted, for example, that the estimated infant mortality from this source for Bombay is about 15 per cent lower than the registered rate, itself likely to be an undercount. CHILD MORTALITY IN NEW INDUSTRIAL LOCALITIES: AN INDIAN STEEL TOWN 167 CROOK AND MALAKER HEALTH TRANSITION REVIEW VOL. 2 NO. 2 1992 differences in Asia across urban areas of differing economic character.3 A second and more important objective is to indicate the differences in mortality that exist within such cities, and their social or environmental determinants, to add to the still modest literature on intra-urban mortality differences in developing countries: for example, Behm (1980); Tekce and Shorter (1984); Bisharat and Tewfik (1985); de Lima Guimaraes and Fischmann (1985); and Basu and Basu (1991). For the determinants we are particularly concerned to obtain and use measures of physical environmental quality (as indicated in access to sanitary facilities, for example), as well as household socioeconomic measures, and, insofar as sample size allows, to model their independent effects on child survival. It will be particularly important to see how far differentials of this kind have been contained in cities that were planned from scratch, so to speak, with large investment outlays on infrastructure and housing. It has been documented that such new-towns have accumulated a substantially ‘unplanned’ population over time, partly because the local resources expected to be generated for ongoing social infrastructural development failed to materialize, and partly because the speed of demographic growth was grossly underestimated (Sivaramakrishnan 1982; Jagannathan 1987; Crook 1992). Nonetheless, such cities are still sometimes regarded as being economically privileged, as indeed they once were, and most development aid is still focused on the older metropolitan areas ostensibly because their need is greater. In this study we seek to establish whether or not there are mortality differences that can be attributed to the unequal provision of social infrastructure for different social strata in new-towns, despite the initial commitment to social equality in this respect.