Two studies investigated young adults' expectations about long-term contraceptive effectiveness. Subjects were told about five hypothetical methods of contraception varying in reported effectiveness, which was expressed in terms of the likelihood of avoiding pregnancy for base periods of 1 year (Experiment 1), 5 years, or 10 years (Experiment 2) of use. For each method, subjects estimated the likelihood that a woman would avoid pregnancy while using it for periods ranging from 1 month to 15 years, and then rated how satisfied they would be with it. For nearly half of the subjects, estimates of cumulative effectiveness did not decline as time period increased. Those subjects who did realize that cumulative effectiveness declined over time estimated rates that declined too slowly for methods of modest and low reliability, and at rates that were too similar for methods differing in effectiveness. Subjects were overly optimistic about effectiveness for time periods longer than the base period, and overly pessimistic about effectiveness for shorter time periods. Not surprisingly given these results, subjects expressed more satisfaction when a method's effectiveness was expressed in shorter base periods. Such faulty understanding of the long-term implications of contraceptive effectiveness information may undermine people's abilities to make informed contraceptive choices.