Many of the society’s most consequential allocation problems are mediated by systems we design. Families seek places in schools, transplant programs allocate scarce organs, and organizations divide tasks and decision-making authority between humans and artificial intelligence.
Although these settings appear different, they share a common structure: the relevant information is dispersed across people, participants respond to the rules they face, and reasonable social objectives may conflict.
This lecture presents an empirical approach to economic engineering—combining economic theory, causal evidence, and behavioral models to understand how allocation systems work and how they might be improved. The examples show why finding a better algorithm is rarely enough. Successful design must consider what information to collect and reveal, how to encourage participation, and who should make which decisions. Data cannot determine society’s objectives, but it can reveal tradeoffs, anticipate responses to new rules, and help us build systems that serve those objectives more effectively.