The United Nations urban growth projection and its impact on urban development and regional disparities in India

Résumé du document

The United Nations Population Division (UNPD) has been publishing and revising its World Urbanization Prospects since 1991, the latest being the 2002 revision, and this has become a popular source of data and analysis of the past, current and future proportion urban in each country, region or continent of the world. As urban issues get more attention, notably in the Millennium Development Goals (MDG), it is increasingly used by other international agencies such as UNDP, World Bank, EIA, UN-Habitat as an instrument for projections of some other global trends, like poverty, energy consumption, environment and resources. Projections and even estimations, for recent years, of other global trends cannot indeed afford to do without urbanization projections, as they are often a key indicator of global integration. No other organization than the UN has been successful in compiling a database on urbanization that equals the UN database in scope and quality. An early attempt to offer alternative to the UN database is the GEOPOLIS database, which is using a common agglomeration and population criteria for all countries.

Extraits

[...] The Mean Percentage Error (MPE) for the projection of the world urban growth excluding China during the period 1980- 2000 was indeed 19%[6]. It was for the 20-years length of forecast as high as for South Asian countries: indeed China and India particularly contribute to the difference between the two estimations with a respective MPE of and But even the MPE for the 5-years length of forecast exceeded in countries with low and lower middle level of development. In the wake of these observations, Professor A. [...]

[...] In the wake of these observations, this paper proposes to focus on the specific use of United Nations urbanization estimates in India. I argue that overestimated UN projections regarding India allow national policy- makers to give the illusion of an allegedly widespread, fast and homogeneous urbanization within India's regional states. In reality, these figures hide a much more uneven situation and in particular a great disparity in urbanization speed between backward and advanced regional states, which widens the development discrepancy. [...]

[...] It is therefore not easy to demonstrate what should be the form of the relation between urban-rural growth difference and the percentage urban during the early stage of the urbanization process, especially outside Europe. The reason why a linear model was used when the UN started its projection exercise in the 1980s was that at the time, very few countries had more than three available data between the 1950s to the 1970s. In absence of deeper historical trends, a sensible solution was to model the trend across countries at a given time. [...]

[...] To produce these estimates, the United Nations uses equation to interpolate between censuses and to provide reasonable guesses of urban proportions between 1950 and the first national census available after that date. The weak point of the United Nations method emerges when it is employed to generate forward projections of urbanization levels. The method assumes constancy in the URGD, and this is an untenable assumption in medium- and long-term projections. Indeed, it has been shown in other studies that with other things held constant, the URGD should decline with urbanization. Noting this, the United Nations sought and found in the empirical record supporting evidence for URGD decline. [...]

[...] The United Nations urban growth projection and its impact on urban development and regional disparities in India Table of contents I. The UNPD projection methodology: assessing the Urban Rural Growth Differential (URGD) 4 A. The evolution of the URGD over time The initial formulation and its 1974 extrapolation The 1980 refinements 5 B. URGD: a method that has proven disputable both theoretically and empirically A questionable method which produces debatable projections 6 II. The impact of an overestimated urban growth on political choices: widening regional disparities 7 A. [...]