Brazil within Brazil : Testing the Poverty Map Methodology in Minas Gerais

Elbers, Chris ; Lanjouw, Peter ; Leite, Phillippe George

World Bank, Washington, DC

关键词: ADULT POPULATION;    ANALYSIS OF POVERTY;    AUTOCORRELATION;    BOOTSTRAP;    CALCULATION;   

DOI  :  10.1596/1813-9450-4513
RP-ID  :  WPS4513

学科分类：社会科学、人文和艺术（综合）

来源: World Bank Open Knowledge Repository

【 摘 要 】

The small-area estimation techniquedeveloped for producing poverty maps has been applied in alarge number of developing countries. Opportunities toformally test the validity of this approach remain rare dueto lack of appropriately detailed data. This paper comparesa set of predicted welfare estimates based on thismethodology against their true values, in a setting wherethese true values are known. A recent study draws on MonteCarlo evidence to warn that the small-area estimationmethodology could significantly over-state the precision oflocal-level estimates of poverty, if underlying assumptionsof spatial homogeneity do not hold.Despite these concerns,the findings in this paper for the state of Minas Gerais,Brazil, indicate that the small-area estimation approach isable to produce estimates of welfare that line up quiteclosely to their true values. Although the settingconsidered here would seem, a priori, unlikely to meet thehomogeneity conditions that have been argued to be essentialfor the method, confidence intervals for the povertyestimates also appear to be appropriate.However, thislatter conclusion holds only after carefully controlling forcommunity-level factors that are correlated with householdlevel welfare.