Conventional child support systems rely heavily on a blanket set of punitive measures to help ensure that families receive the financial support to which they are legally entitled. Los Angeles County is taking a different approach, one that seeks to tailor child support interventions to the circumstances of each case and reduce the use of punitive measures, such as wage garnishment, driver’s license confiscation, and, in extreme cases, incarceration. FUSE executive fellow Carlos Thomas worked with the Child Support Services Department to refine a predictive model used to forecast payments and allocate staff resources. 

Carlos tested the model’s accuracy and identified changes that could improve its predictive power. The proposed changes led to a roughly 5 percent increase in the model’s ability to correctly predict payment. Carlos also developed training modules to help staff make better use of data and statistics, and he created the prototype for a data dashboard that enables department leaders to make decisions based on comprehensive, real-time data. These efforts have helped the department target resources more effectively and reduce the use of punitive interventions, while also improving payment outcomes for families.