“ Optimal Market Grain Over Space and Time ”

Markets facilitate the exchange of precise quantities of goods and services at particular locations and times; yet for convenience and to reduce transactions costs, these markets customarily specify the terms (e.g. the price) uniformly over a substantial interval of space and time. If customers’ tastes and producers’ costs vary continuously over these dimensions, then assessing uniform prices within these market intervals leads to economic inefficiencies (marginal value unequal to marginal cost) for many participants, and it affords incentives for hedging at the boundaries of these market segments where prices differ. The determination of an optimal “grain” for market design, therefore, requires the balancing of four different categories of costs: (1) the inefficiency from not equating marginal benefit and marginal cost for every transaction, (2) the increasing transactions (decision) costs of having a larger number of market segments, (3) the increased hedging costs across the boundaries if there are fewer market segments and therefore greater price disparities, and (4) the increased opportunity for participants to exercise market power as fewer competitors are aligned with each segment of a more finely-grained market structure. One practical application for these concepts is in the emerging markets for electricity where the spatial grain, locational marginal pricing (LMP) based on congested points in the transportation system, provides incentives for the geographic investment in new generation facilities that could ultimately alter line flows, and therefore congestion -the rationale for the original spatial structure. These concepts also apply to the structure of forward markets, particularly if they can influence capacity investment and therefore, subsequently, spot market prices. As an example, it is shown that in markets with a limited number of suppliers, a forward market conducted before the required lead-time for investment in new capacity results in greater investment and lower spot market prices. Conversely, forward markets conducted only after the capacity-expansion is committed imply less capacity and higher spot market prices, even without risk-aversion. * This work was sponsored by the Power Systems Engineering Research Center (PSERC) and the Consortium for Electricity Reliability Technology Solutions (CERTS). We express our appreciation for the support provided by PSERC's industrial members, by the National Science Foundation's Industry/University Cooperative Research Program under grant NSF EEC-0118300, and by CERTS, supported by the United States Department of Energy.