This paper discusses results of a two-year collaborative research project between the authors and the Demand Response Research Center focused on behavioral response to a voluntary time-of-use pilot rate offered by the Sacramento Municipal Utilities District (SMUD) under the PowerChoice label. The project had two purposes: one was to assess the potential for increasing demand response through the introduction of enhanced information and real-time consumption feedback; the second was to better understand behavioral response to a TOU rate. Three successive waves of telephone surveys collected details about reasons for participation, actions taken, capacities and constraints to altering behavior, and a range of salient conditions, such as demographics and dwelling characteristics. Pre- and post-program interval meter data for participants and a comparison sample of households were also collected and analyzed to consider initial and season-change price effects of the rate and the effect of supplemental information treatments on response. Over half of surveyed participating households reported that they had made a great deal of effort to adjust their electricity consumption to the rate. Despite this, load data analysis revealed only minimal price effects; and, though households subjected to information treatments seemed to have learned from these treatments, load data analysis again detected only minimal effects on load. Given the currently high hopes for behavioral intervention and residential TOU rates, these unexpected results require explanation. We suggest a number of possibilities and discuss some implications for TOU programs, and for understanding demand response behavior and approaches to experiments with TOU rates.