Pricing the use of capital-intensive infrastructure over time and efficient capacity expansion: illustrations for electric transmission investment

Traditional economic theory provides a conundrum for pricing large, lumpy infrastructure investments: very different short- and long-run pricing prescriptions. Unless the facility is congested, efficient short run prices should only cover operating costs (short-run marginal cost, SMC); any higher price designed to also recover capital costs would risk inefficient under-utilization. However, if the facility becomes crowded, capital costs should be included in the calculation of user-fees since that burgeoning demand is likely to cause the construction of more capacity, and users should be confronted with the cost-consequences of their decisions. Once additional capacity is completed, however, and if because of the large size of the addition the facility is no longer congested, then price should once again fall to SMC. The resulting jagged pattern of prices offers little assurance to investors of capital cost-recovery without a government guarantee, and it may lead to schizophrenic behavior by both customers and potential suppliers. Just because the physical investment is lumpy, should the price pattern also be dichotomous or can a smoother transition be employed? By integrating the use of congestion fees that are based upon the external costs imposed by one user on all others prior to the construction of added capacity, and then by using the same congestion charge to gauge the “willingness-to-pay” for new capacity and to set an “opportunity-cost”-based benchmark for capital cost recovery afterward, a smoother sequence of prices can evolve. The capital cost recovery portion of these prices, whose magnitude is based upon the congestion eliminated, is premised on a long-run, dynamic view of markets and the transitions they can facilitate, and these cost-recovery adders can be combined with “peak-load-pricing” and the “inverse-elasticity” rule, for example, to improve efficiency and fairness over both space and time. The resulting price patterns can provide compatible incentives for all parties, and they complement several existing electricity system planning processes in those regions where congestion rents are already assessed for the use of transmission. The net effect could be similar to a sequential “real-options” analysis of efficient capacity expansion.