Considerable research attention has been devoted to the question of whether and to what extent changes in welfare policy legislated in the 1990s might have deterred immigrant participation in welfare programs, although only post-1996 immigrants were explicitly targeted by most of the changes. Some analysts have argued that such so-called chilling effects have lowered immigrant participation, and others have argued that this is true only in California. This paper analyzes the role of local labor market conditions in explaining declines in the welfare participation trends of immigrants and reductions in the nativity participation gap for the period 1994 to 1999. The data, extracted from the March Current Population Survey, indicate that immigrants’ participation in welfare declined more rapidly than natives’ during the latter half of the decade. Our results show that variation in the unemployment and employment rates across MSAs and states explain the observed relative post-welfare reform decrease among immigrants, with immigrant welfare utilization being sensitive to changes in both employment and unemployment rates. The inclusion of state fixed effects in probability models suggests that the relative decline among immigrants is not due to unobservable heterogeneity across states, but rather to differences in local labor market conditions. The policy implications of the findings are discussed.