Managing Multiple Research Projects

In standard experimentation problems, decision makers are allowed to experiment on at most one project at a time. Would it ever be beneficial to experiment on more than one project simultaneously if we let them? We show that the answer is Yes, even if it is known that only one of the projects is fruitful ex-post. We consider two projects which are conflicting in the sense that one and only one of them can produce successes; but there is uncertainty as to which project is the profitable one. The decision maker only observes the outcomes of the projects undertaken, and undertaking each project entails a cost. Thus, simultaneous experimentation involves higher costs, but can produce more data. At the same time, the outcomes of either project are informative about the other: Due to their conflicting nature, bad news about one is good news about the other. Provided the costs are not too high or she is sufficiently patient, if the decision maker starts working on a single project that produces no successes, she becomes gradually pessimistic and eventually takes on the other project while keeping the first one — despite the higher costs and the fact that one of them is doomed to fail. Thus, there is a rationale for simultaneous research even on costly, conflicting projects