Class action litigation presents a common problem that has received little discussion in the academic literature. In almost every case, the plaintiff class's recovery is not fully distributed. For example, all possible plaintiffs may not come forward with their claims, the plaintiffs may not be ascertainable, or claims may not be timely submitted. Administrators are regularly posed with the problem of what to do with these residual funds. Currently, courts are free to do virtually anything with such funds. The system is ad hoc, unpredictable, and unguided by any normative principle. In these cases, I propose that the funds should escheat to the state. On a regular basis, the state could then distribute these funds to the citizenry. Ex ante, all citizens are equally likely to find themselves exposed to the risk that led to the class recovery in the first place. Escheat to, and distribution through, the state thus disburses money on average equally to all of those who are potential victims. I generalize from the context of residual funds to argue that recovery in all class actions and other collective adjudication should escheat to the state for distribution (assuming that losses are small relative to wealth or that pecuniary loss is covered by first-party insurance). Individuals would not be compensated ex post in close tailoring to their actual loss as under the current tort system, but with a view to the average risk borne ex ante. Ex ante, individuals could find themselves in any of several worlds (wealthy, poor, injured, not injured, &c.), and since all individuals expect to share in an increase in the state's treasury, money in state coffers would accrue to the benefit of all individuals. Distributed this way, tort damages would solve several problems: the link between harm and compensation, uncertainty, distorted consumption and production incentives, free-riding, and high administrative costs. This argument dramatically reshapes the way damage awards are viewed in our tort system.