View article

An experiment in Columbus, Ohio, randomly assigned clients in a mandatory welfare‐to‐work program to one of two case management models. A traditional model required clients to interact with two staff members: an income maintenance worker who processed welfare benefits, and an employment services worker who enrolled clients in work activities. An integrated model required clients to interact with one worker for income maintenance and employment services. The study examined effects on program participation, welfare receipt, employment rates, and earnings. Over 2 years, clients in the integrated group had significantly higher rates of participation in program activities and significantly lower welfare payments but had similar employment rates and earnings to clients in the traditional group.