In this paper, we use a spatial Durbin model (SDM) to analyze poverty in South Carolina - a predominantly rural state with large pockets of persistent poverty counties. This allows for identifying specific factors in surrounding areas that influence local poverty rates, something missing from the existing literature. Earlier research relied on the general spatial model that we now know is not appropriate in applied research. We find poverty rates increased with more retirees and kids, more single moms, and higher average TANF payments in a county. There were no significant effects from income and job growth. Population density was the only variable with a negative effect during the study period 2003 to 2014.