The review has wise words about the difficulties of non-experimental evaluation of active labor market programs like WIA, about the lack of policy response to evaluation results, and about performance management. I particularly liked this bit on the latter topic:
A very closely related problem turns on the merits of using performance measures to proxy for rigorous impact estimates. Since these measures were first conceived during the CETA program, attempts to refine them so they actually “work” have amounted to the workforce development field's equivalent of the quest for the Holy Grail. Like its predecessor quest, so far this effort has been futile. There is no convincing evidence that using performance measures as a proxy is a good idea and lots of evidence against it. As explained by Burt Barnow in his chapter here, “Lessons from the WIA Performance Measures,” workforce performance measures do not correlate well with program impacts. That really should not be a surprise, because coming up with reliable performance measures requires that we be able to confidently and consistently solve the evaluation problem.Though nominally aimed at the Europeans, there is much that US policymakers could learn from the book as well. They could also learn from the Europeans (at least some of them) about how to increase the quality of non-experimental program evaluations via better administrative data.
You can order the book from Upjohn (or Amazon) and you can read the final draft of my chapter for free. They made me take out the bit about Farrell's from the book version, so I actually prefer the final draft.