| Existing procedures used to develop marginal costs for electric utilities were not designed for applications in an increasingly competitive market for electric power. The utility's value of receiving power, or the costs of selling power, however, depend on the exact location of the buyer or seller, the magnitude of the power and the period of time over which the power is used. Yet no electric utility in the United States has disaggregate marginal costs that reflect differences in costs due to the time, size or location of the load associated with their power or energy transactions.;The existing marginal costing methods used by electric utilities were developed in response to the Public Utilities Regulatory Policy Act (PURPA) in 1978. The "ratemaking standards" (Title 1) established by PURPA were primarily concerned with the appropriate segmentation of total revenues to various classes-of-service, designing time-of-use rating periods, and the promotion of efficient long-term resource planning. By design, the methods were very simple and inexpensive to implement. Now, more than a decade later, the costing issues facing electric utilities are becoming increasingly complex, and the benefits of developing more specific marginal costs will outweigh the costs of developing this information in many cases.;This research develops a framework for estimating total marginal costs that vary by the size, timing, and the location of changes in loads within an electric distribution system. To complement the existing work at the Electric Power Research Institute (EPRI) and Pacific Gas and Electric Company (PGandE) on estimating disaggregate generation and transmission capacity costs, this dissertation focuses on the estimation of distribution capacity costs. While the costing procedure is suitable for the estimation of total (generation, transmission and distribution) marginal costs, the empirical work focuses on the geographic disaggregation of marginal costs related to electric utility distribution investment.;The study makes use of data from an actual distribution planning area, located within PGandE's service territory, to demonstrate the important characteristics of this new costing approach. The most significant result of this empirical work is that geographic differences in the cost of capacity in distribution systems can be as much as four times larger than the current system average utility estimates. Furthermore, lumpy capital investment patterns can lead to significant cost differences over time. |
