Bundling Complements

Abstract

I develop a duality-based multi-dimensional screening framework with a geometric characterization of combinatorial preferences. For a mechanism to be optimal, the type distribution pins down \emph{required} directions of binding feasibility constraints, while the complementarity among bundles determines the \emph{covered} directions; optimality reduces to full coverage of required directions. I apply the framework to a one-parameter family in which every bundle containing a fixed \emph{core} of items earns a complementarity premium. Two thresholds organize the optimum: above a lower threshold the grand bundle must be offered; above a higher threshold a \emph{core-peripheral} menu -- a bundled core with optional add-ons that are not sold standalone -- is optimal. The tight distributional condition for finiteness of the higher threshold is \emph{inclusivity}, that the menu exclude no near-top buyer.