In recent years, policymakers have shown increasing interest in dynamic pricing as a means to encourage demand response (DR). Proponents of real-time pricing (RTP) and critical peak pricing (CPP) rates rely heavily on conceptual portrayals of the DR benefits that would inure from their wide-scale adoption. A few empirical studies of voluntary RTP tariffs implemented by vertically integrated utilities confirm that at least some of the largest customers can adjust their usage in response to high prices. Recent CPP and RTP pilot programs involving residential customers suggest that some of these customers too might adjust usage to price signals. However, the overall experience with RTP in vertically integrated settings has been less than encouraging, with several notable exceptions. Furthermore, these results do not necessarily translate to situations in which RTP is implemented as the default service in the context of retail competition. During the past five years, regulatory commissions in several states (Maryland, New Jersey, New York, and Pennsylvania) have adopted default-service pricing for large customers that is indexed to day-ahead or real-time energy markets operated by Independent System Operators (ISOs). These initiatives were driven primarily by retail market development goals with DR being a secondary objective in a few cases. While there are indications default service RTP has resulted in more customers facing, and possibly responding to, high prices, no study has been undertaken to fully characterize or measure this response. Despite its promise, critical questions must be resolved before policymakers will be comfortable implementing dynamic pricing on a large scale. To what extent does default service RTP actually encourage price response? Does RTP-type default service also satisfy the goal of switching customers to competitive suppliers? To what extent do retail market choices, such as hedges, complicate RTP's capacity to elicit price response? What are the relative roles of default-service RTP and ISO-sponsored emergency programs in eliciting DR? Which factors and characteristics account for differences in customers' willingness and ability to respond to hourly varying prices? We contribute to the resolution of these issues, for day-ahead RTP, with a study of the largest customers at Niagara Mohawk, a National Grid Company (NMPC), that have been exposed to default-service electricity pricing indexed to the New York Independent System Operator (NYISO) day-ahead market since retail competition was introduced in 1998. The study, conducted in two phases, included customer surveys and interviews (in summer 2003 and fall 2004) to collect information on customers' facility and business characteristics, their perceived ability and willingness to respond to prices or other signals to curtail, and their electricity supply and hedging choices. These data were combined with NMPC billing records to estimate price response. The study focused on the summers of 2000-2004, to characterize how customers adapted to changing market circumstances, alternative supply choices, and, after 2001, opportunities to participate in NYISO demand response programs. The 149 subject customers, served under the "SC-3A" service classification, range in peak demand from 2 MW (the threshold for inclusion) to over 20 MW and include manufacturers (32%), government/education facilities (30%), commercial and retail businesses (11%), and health care (11%) and public works (16%) facilities.