Corral, Paul2025-05-012025-05-012025-05-01https://hdl.handle.net/10986/43150This paper explores why small area poverty estimates from models at the household level that only use aggregate data as covariates, exhibit systematic bias. The analysis demonstrates that this bias stems from the model’s inability to capture the complete between-household variation in welfare, as they rely solely on covariates aggregated at geographic levels. Through model-based simulations, the paper shows that the bias in these models is minimized when the empirical variability of simulated welfare based on the model is closest to the true empirical variance of welfare at the area level. This finding also has implications for bias in unit-level models.en-USCC BY 3.0 IGOSMALL AREA ESTIMATIONPOVERTY MAPPINGSATELLITE IMAGERYCENSUSOFFICIAL STATISTICSWorking PaperWorld Bankhttps://doi.org/10.1596/1813-9450-11110