Over the past two decades, federal and state policymakers have dramatically reshaped the nation’s system of cash welfare assistance for low-income families. During this period, there has been considerable variation from state to state in approaches to welfare reform, which are often collectively referred to as “welfare-to-work programs.” To help states assess various program approaches in an informed way, this report draws on an extraordinary body of evidence: results from 28 benefit-cost studies of welfare-to-work programs based on random assignment evaluation designs. The synthesis addresses such questions as: Which welfare reform program approaches result in positive payoffs on the investments made in them? Which approaches make participants better off financially? Which approaches improve the government’s budgetary position? The report thus presents findings that can aid policymakers and program developers in assessing the often complex trade-offs associated with balancing the desire to ensure the poor of adequate incomes and yet encourage self-sufficiency.

Different types of welfare-to-work programs emphasize different goals. Whether a program is judged as a success in terms of its benefit-cost performance depends on what policymakers were attempting to accomplish in that program. The following policy conclusions are suggested by this synthesis:

• If a chief goal is to increase participants’ income, then programs that provide individuals with financial incentives or earnings supplements intended to encourage work appear to best achieve this goal. While beneficial for participants, earnings supplement programs tended to result in a net cost for the government. Participants, however, often gained more than a dollar for every dollar the government spent, making this type of program an efficient mechanism for transferring income to poor families.



• If a chief goal is to reduce government expenditures, then programs that require individuals to look for jobs immediately and that assign other activities if work is not found are relevant strategies. These programs tended to be beneficial for the government budget (and to be less expensive than the type of program described next) but to result either in small benefits or in net costs for participants.



• If a chief goal is to balance reducing welfare expenditures with increasing participants’ income, then programs that require individuals to participate initially either in an education or training activity or in a job search activity can meet this goal. This type of program, when targeted to both short-term and long-term welfare recipients, was beneficial for both participants and the government’s budget.



• Mandatory programs that require individuals to participate in General Educational Development (GED) completion and Adult Basic Education prior to job search do not appear to achieve the goal they emphasize: increasing the income of participants. Nor do they achieve the goal of saving government money.



• Some mandatory work experience programs — which assign individuals to unpaid jobs, often following a period of job search — resulted in limited benefits for participants but did provide valuable goods and services for the general public. They did not consistently reduce government costs, however.

These studies measured only benefits and costs that are reliably expressed in dollars. Other benefits and costs that are not easily expressed in dollars — for example, changes in children’s school performance or well-being — were not estimated and, thus, are not incorporated into the benefit-cost analyses. Overall assessments of program types, however, also should take into account noneconomic considerations when determining whether a program achieves policymakers’ and society’s goals.