Clinical trials rarely occur in vacuum. Generally historical information, perhaps extensive, is available on the comparator arms in a study. When applicable, this information presents the opportunity to design more efficient trials, achieving superior power, type I error, and mean square error of estimation properties that trials which do not utilize this information. However, the potential for drift in the comparator effect also gives the potential for bias, reduced power, or inflated type I error. In this talk we review some common methods for borrowing historical information within a trial and illustrate when and how the information provides benefit to a trial as opposed to reducing efficiently. Unfortunately, knowing whether or not historical information will help a trial depends on unknown parameters (the discrepancy between the historical and current studies). We then focus on what kinds of long term distributions of that discrepancy would result in a drug development process that is on average more efficient than ignoring the historical information.