Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We model complexity in social programs as a by-product of the screening process. While a more rigorous screening process may improve targeting efficiency, the associated complexity is costly to applicants and induces incomplete take-up. We integrate the study of take-up with the study of rejection (Type I) and award (Type II) errors, and characterize optimal programs when policy makers choose screening intensity (and complexity), an eligibility rule, and a benefit level. Consistent with many real-world programs, optimal programs feature high complexity, incomplete take-up, classification errors of both Type I and II and, in some cases, "excessive" benefits. (JEL D04, D82, H23, I18, I38)