Local Labor Markets and Local Area Effects on Welfare Duration

The Survey of Income and Program Participation (SIPP) has become an important tool for studying how long people stay on welfare programs because it has monthly data on a variety of welfare programs. This article presents estimates of duration models for unmarried women with children who are on the Aid to Families with Dependent Children program (AFDC) using the 1984 and 1985 panels of SIPP. A weakness in previous welfare duration studies is that they do not include local labor market conditions or other local area effects; this omission may bias the estimated effects of policy variables (such as benefit levels) and labor market variables. This article incorporates relevant local area information from the City County Data Book and links this to SIPP welfare recipients based on county of residence. I find that local variables such as unemployment rates or per capita sales affect welfare exit rates, especially for blacks. Living in an urban area lengthens welfare spells for both whites and blacks.