A spatial model of community-level poverty: evidence from Indonesian family life survey

Abstract

This study is aimed at examining how spatially-related factors affect poverty in the context a developing nation. We explore the following question: Does space matter in affecting poverty in a developing country and does social capital reduce poverty levels? We employed a dependent variable that is spatially lagged. To account for the spatial effect on community-level poverty, spatially-lagged variables are incorporated in this empirical model. Social capital was used as the independent variable. Although there might be endogeneity of poverty rates  and social capital level, however, assuming the fact that in the context of a developing nation social capital is inherited rather than affected by the level of poverty, we omitted the effect of poverty on social capital and only considers the effect of social capital on poverty. Five proxies to measure community. The data were sourced from the fifth wave of the Indonesian Family Life Survey (IFLS5), designed and implemented by the RAND Corporation. The results of the study show that, regardless of the proxies used to measure community level poverty, space significantly positively affect poverty. However, the effect of social capital on community level poverty varies in terms of significance and their signs. This implies that any poverty alleviation policy should take into account the importance of bringing marketplaces and related market opportunities closer to poor people at the community level. Positive spatial spillover effects could reduce poverty and therefore granting village-level public investments should be directed toward creating physical market connectivity at the village level.  This has policy implication that governments should pay greater attention to providing closer linkage among villages through road infrastructure and providing incentives to building social capital to village communities.