Revisiting the Neighborhood Effects on Disadvantaged Households in Indonesia

Abstract

The latest statistics of 2022 record 9,57 percent of the Indonesian population living in poverty. As most of the study in this field merely focuses on the internal factors of poverty, this study contributes to providing more perspectives for poverty alleviation by revisiting spatial linkages on poor households’ income in East Java, the second most populated province in the economy. This study employs spatial econometrics analysis in a three-step procedure: (i) designing spatial weight matrix within nearest districts, (ii) doing Moran's I Test to identify spatial effects on the variables, lastly (iii) doing LM Test to find best regression model. Estimates on the SUSENAS 2017 database show that both Ordinary Least Square (OLS) and General Nesting Spatial (GNS) models provide similar results. Nonetheless, when compared with Akaike’s Information Criterion (AIC) model, the GNS model provides better accuracy. This means accommodating spatial effect is imperative. This study also finds a positive spillover of poor households’ income within neighboring districts. This implies the existence of multiplier effects in a household’s economic activities that later might form clusters of poverty. These suggest that the Sustainable Livelihood Approach interventions in the province should take into account differences in social institutions and other socio-cultural characteristics.