Simplicity and Portability in Mechanism Design: A Case for (and Against) the Worst Case

A central issue for mechanism design is how to identify theoretical environments that will lead to practically useful insights about optimal mechanisms. For a mechanism to be practically useful, it must be simple enough so that agents can optimize their behavior. The mechanism must also be portable, in that it also performs well in environments other than the one it was optimized for. We argue that by focusing on the informational environments that are the most challenging for designer, will necessarily identify optimal mechanisms that are both simple and portable. We survey a recent literature that operationalizes this idea, and we compare it to other approaches in robust mechanism design. Finally, we offer an (adversarial) critique of worst-case analysis in mechanism design.