Foster Care Kills

If you are on the margin of remaining in your home with your family and being removed and placed in foster care, then foster care will cause the following outcomes: increase the likelihood of child teenage pregnancy, lower your earnings, lower your schooling attainment, and increase the likelihood of engaging in criminal activity. So finds two new papers by Joseph Doyle, an economist in the business school at MIT. The two papers are here (AER 2007) and here (R&R and JPE). His identification (we're economists, so of course, we have to talk about his identification) is pretty cool. He uses as source of identification not simply placements in foster care, but the assignment of case workers to the child's case. Specifically, some of the case workers have very high removal rates and some have very low removal rates, and the idea is that there is something about the ones with the very high removal rates (call them the "hard asses" if you will) that causes a child to go to foster care, independent of their family situation. That is, if the easy case worker had gotten that case, the child wouldn't have gone to foster care, but because the hard ass did, the child did. Instrumenting for foster care placement with the type of case worker, Doyle finds worsened outcomes in the child using a dataset from Illinois that basically only University of Chicago graduates or faculty members can touch (Doyle did his PhD at Chicago). Very cool. Check out his entire research agenda here. Lots of cool papers there.