The blog of the Urban Institute
October 22, 2012

How Do We Decide Who Gets the Cadillac?

October 22, 2012

In our present era of constrained federal and state budgets, the tired mantra of doing more with less has been trotted out again and again. While doing more with less is seldom possible, there are ways that we can better target the resources we have.

Figuring out how to best target resources is one of the major challenges in serving vulnerable families. Not every family needs intensive services, but determining who does and who may benefit from a less expensive approach can be difficult. In an article published last week in Social Service Review, we demonstrate a way to identify groups of residents that informs the targeting of effective, streamlined service delivery. This work was developed during our evaluation of the Chicago Family Case Management Demonstration, which provided intensive case management services to residents of two distressed public developments on the south side of Chicago. These lessons don’t just apply to public housing, however, but also to social services for the elderly, ex-offenders, low-income dads, homeless individuals, and youth.

Deciding which families received intensive, expensive services (like inpatient drug treatment, transitional jobs, or mental health counseling and medication provided by psychologists) and which families required light-touch services (like job-search assistance or work-related child care support) was a critical question that emerged for the case managers in the demonstration. Service providers—particularly those in the homelessness, child welfare, mental health, and criminal justice systems—have encountered similar questions and typically target services on the basis of clinical assessment questionnaires, structured interviews, self-report questionnaires, or actuarial assessments.

While some of these tools are long-established, few have ever been empirically tested for their predictive value. Given this lack of rigorous research, practitioners have little besides their intuition to guide them.

The public sector, philanthropic community, service providers, and researchers must do more to develop and critically test assessment and analysis tools to aid in targeting. This process must be coordinated across different policy areas, as many offer, at their core, variants on the same social services. Creating evidence-based targeting tools will require effort—for example, requiring a commitment to integrate administrative data across different social service systems. But the dividends will be better processes for ensuring that families in need of a Cadillac get one, and that others, only in need of a scooter, get that instead.


As an organization, the Urban Institute does not take positions on issues. Experts are independent and empowered to share their evidence-based views and recommendations shaped by research.