Child protection and child outcomes: Measuring the e ects of foster care', The American Economic Review

The child welfare system investigates over two million children each year for parental abuse or neglect, yet little is known about the effect of placing children in foster care. Outcomes are rarely observed, and children removed from home likely have worse family backgrounds than children not in foster care, which makes comparisons difficult. This paper uses the removal tendency of investigators as an instrumental variable to identify a causal effect of foster care placement on a range of long-term outcomes, including juvenile delinquency, teen motherhood, employment and earnings. Investigator removal tendencies are found to predict foster care placement in subsequent investigations, despite a rotational assignment process that results in similar child characteristics across investigators. The results suggest that children on the margin of foster care placement have better employment, delinquency, and teen motherhood outcomes when they remain at home