IS THERE A METROPOLITAN BIAS? URBAN POVERTY AND ACCESS TO SERVICES BY CITY SIZE IN SIX DEVELOPING COUNTRIES.
with Francisco Ferreira and Peter Lanjouw.
Abstract: The spatial heterogeneity of poverty – in its many dimensions – is not restricted to the rural-urban dichotomy. There is considerable spatial heterogeneity among urban areas, and one important dimension of that heterogeneity is across city sizes. A greater understanding of how poverty – both in terms of incomes or consumption expenditures and in terms of access to public services – varies across different types of cities should help inform the discussion of appropriate poverty reduction strategies in most countries. Yet, the evidence base needed for this disaggregated analysis is seldom available. In this paper, we draw upon the considerable additional insights generated by small area poverty estimation (based on the combination of welfare estimates from household surveys with “sample” sizes from National Censuses) to investigate the relationship between poverty and city size in six developing countries, namely Albania, Brazil, Kazakhstan, Kenya, Morocco and Sri Lanka. We find substantial variation in the incidence and depth of consumption poverty across city sizes in five of the six countries. For all five countries where the data permits a disaggregation of the incidence of public service access, there is also considerable variation across city sizes. In all cases, poverty is lowest and service availability is greatest in the largest cities – precisely those where governments, the middle-classes, opinion-makers and airports are disproportionately located. This leads us to ask whether, in addition to Lipton’s original urban bias, there might also exist a “metropolitan bias” in the allocation of resources (including policy attention) to larger cities, at the expense of smaller towns, where most of the poor are located.
RURAL POVERTY AND GEOGRAPHY: TOWARDS SOME STYLIZED FACTS IN THE DEVELOPING WORLD.
with Peter Lanjouw, Piet Buys and Timothy Thomas.
Abstract: We combine detailed estimates of rural poverty at a spatially disaggregated level, with geographically referenced information on agricultural potential in those localities and their and proximity to urban centers, in five developing countries. We explore the association between rural poverty and “marginality” which we define in terms of low agro-potential and remoteness. We present evidence that rural poverty rates are often, but not .everywhere, higher in marginal areas than in better endowed localities. However we also show that overall poverty numbers tend to be substantially higher in the non-marginal areas; a consequence of greater population densities in such places. We illustrate with respect to one country, Ecuador, that during a macro-crisis population movements within the country have been such that rural poverty rates (and numbers of rural poor people) have risen most substantially in the relatively well-endowed areas. Finally, we document for Brazil that rural poverty rates are closely (and inversely) associated with proximity even to small towns – not just large cities. We illustrate, further, that where urban poverty rates in small towns are low, rural poverty rates tend to be lower. We suggest that a possible mediating role is played by rural-non farm employment: rural non-farm employment rates tend to be higher in localities that are located close to urban areas (including small towns) and are also higher where poverty rates in small towns and cities are low. We caution that these patterns are simple correlations and do not demonstrate direction of causality. We argue, however, that they point to possibly fruitful directions for further research.
POVERTY DYNAMICS IN RURAL MOROCCAN COMMUNES: 1994-2004.
with Peter Lanjouw, Abdeljaouad Ezzrari and Mohamed Douidiche.
Abstract: Using two sets of poverty maps available for Morocco in 1994 and 2004, we look at the relationship between initial characteristics of the communes, their change over the ten-year period and the dynamics of poverty. The purpose of this paper is to assist Moroccan policymakers by providing detailed information on the relationship between some policy measures (proxied by the change in commune characteristics) and poverty reduction. The work presented here is a joint publication with the Department of Statistics, Observatoire des Conditions de Vie, Rabat. The results of the analysis presented in this study generally accord with our basic understanding of poverty decline and economic development in Morocco. Poverty has fallen most rapidly in those communes with high and growing employment rates and where dependency ratios (children per adults) are low and are falling. Improvements in education, particularly increases in the share of the adult population with lower secondary education are associated with falling poverty. There is some indication that poverty has been rising in those communes where people have shifted out of agriculture towards certain non-farm activities such as construction. An important and robust set of findings in this paper concern the relationship between poverty and infrastructure access: t commune-level poverty has fallen with expansion of infrastructure services, notably electricity and also water access. The results of the analysis do point to one or two puzzles however: poverty is higher, and has increased over time, in communes in which the share of adults with higher, post secondary, education is greater, and where this share has increased over time. A corollary to this finding is that poverty has also been observed to have risen in those communes where the share of working population employed in administrative services has increased. It should be noted however that we have not been able, in this study, to establish clear causality from our various variables of interest and poverty.
POVERTY MAPS OF DJIBOUTI CITY: A MULTI CRITERIA ANALYSIS [
in French]Abstract: In the absence of good quality data on household consumption or income, how can we advise policy makers and international donors as to where poverty lies? The present document shows a set of poverty maps for Djibouti city, disaggregated at the neigborhood level. While we present three maps based on monetary poverty, those are complemented with infrastructure access and child health measures. We conclude that all indicators of poverty are highly correlated, especially in the poorest and richest areas. neighborhoods that experience medium levels of monetary poverty simultaneously experience higher volatility of non-monetary measures.