Abstract

The precautionary principle justifies postponing the implementation of development projects to await better information about their environmental impacts. But if implementation capacity is congestible, as is often the case in practical settings, a postponed project may have to vie for implementation priority with projects that arrive later. Limitations of implementation capacity create two risks. First, it may sometimes not make sense to go back to a postponed project, even if it is later revealed to be a good one. Second, the planner may find it worthwhile to go back to it, but at the expense of undesirable delay of subsequent projects. We consider a planner facing a sequence of projects that vary stochastically in their (1) importance and (2) improvability, but knowing that implementation capacity is congestible. The scope for congestion implies a ‘bonus’ for earlier-than-otherwise decisions, in common parlance ‘keeping the desk clear’, which works against the well-understood option value that encourages postponement. The optimal decision rule depends upon the stochastic environment whereby future projects are generated, in ways that are not obvious. The value of the bonus is increasing in the expected importance of future projects but decreasing in their expected improvability. Higher variability of the importance of projects, in the sense of mean-preserving spread, increases the size of the bonus, but variability in their improvability has a generally ambiguous impact. We characterize the adjusted decision rule and note its implications for the conduct of cost-benefit informed policy.