| Abstract: | Abstract Efforts to explain declines in welfare caseloads in the 1990s focus on three factors (economic opportunity, welfare policy, and demographic characteristics) and produce mixed results. Although several recent studies consider state differences, few consider local differences within states. Failure to control for caseload persistence and within‐state variation in program implementation may be one reason why studies produce contradictory results regarding the effect of economic conditions on caseload decline. Using data for Louisiana parishes from 1991 to 1998, we analyze local caseload decline. Our results support other researchers' finding of a “threat effect” before 1996 welfare reform; they indicate that, all else being equal, reducing caseloads will be most difficult in nonadjacent nonmetropolitan parishes and easiest in adjacent nonmetropolitan parishes. |
