The third chapter addresses ethical issues in decision-making about SRM, primarily addressing the matter of procedural justice. Whereas the previous chapter considered the justice (or lack thereof) of certain distributions of harms and benefits, procedural justice concerns how decisions ought to be made. I argue that a necessary (but not sufficient) condition for procedural justice is that all parties that stand to be substantially burdened by some SRM technique be afforded an opportunity to contribute to decision-making about whether to deploy it. This creates some problems for would-be proponents of SRM. Given that certain SRM techniques are relatively inexpensive and (unlike emissions mitigation) technically feasible without large-scale cooperation, SRM is prone to unilateral and small-scale multilateral decisions to deploy. Such deployments fail to meet the foregoing necessary condition of procedural justice, for they would exclude from decision-making many parties that stand to be substantially burdened by SRM. However, taking procedural justice seriously here threatens to undermine SRM’s political feasibility, for better or worse. Although procedural justice does not require consensus among all relevant parties, it does require that they all have an opportunity (e.g., through representatives) for meaningful influence on decisions. Their divergent interests could make it difficult to secure sufficient agreement for deployment of SRM, even a variety of SRM that stands to deliver a non-ideally just distribution of harms and benefits. There is thus a plausible concern that a procedurally just decision-making framework for SRM could prevent policies that are ethically attractive in other respects. To deal with this problem, I argue for a non-ideal-theoretic approach to procedural justice, where the requirements of procedural justice may be relaxed if doing so is necessary for some more important ethical aim, such as securing relatively just distributions of harms and benefits.