Home > Electricity > Electricity Publications > Electricity Prices in a Competitive Environment


Electricity Prices in a Competitive Environment

Available Format

PDF (900 k)


Executive Summary

The emergence of competitive markets for electricity generation services is changing the way that electricity is and will be priced in the United States. This report presents the results of an analysis that focuses on two questions: (1) How are prices for competitive generation services likely to differ from regulated prices if competitive prices are based on marginal costs rather than regulated "cost-of-service" pricing? (2) What impacts will the competitive pricing of generation services (based on marginal costs) have on electricity consumption patterns, production costs, and the financial integrity of electricity suppliers? This study is not intended to be a cost-benefit analysis of wholesale or retail competition, nor does this report include an analysis of the macroeconomic impacts of competitive electricity prices.

Assumptions of the Analysis

Prices for generation services are based on marginal costs under competition

Economic theory states that competition drives price to marginal costs if there are many producers and consumers. 1 For electricity, this means that competitive prices for generation services would be based on the costs of producing the last kilowatthour of electricity. This method of pricing is different from the cost-of-service regulatory practice, which uses average costs (total costs divided by total sales) as the basis of prices. The application of marginal costs as the basis of prices assumes that no supplier or consumer exercises market power. Market power exists when a supplier or consumer influences prices by virtue of size or control over important aspects of the market, such as access to transmission lines. If suppliers exercise market power, prices could be higher than marginal costs. If a consumer segment exercises market power, then that segment could have a price advantage over other customers.

In this analysis, regulated prices for generation services are based on average costs, and competitive prices for generation services are based on marginal costs. Average costs are defined as the total costs of production, including a return on investment equal to the producer’s cost of capital, divided by sales to ultimate consumers. Marginal costs are defined as the operations and maintenance (O&M) costs of the most expensive generating plant needed to supply the immediate demand for electricity (the marginal cost of generation).2 During periods of high demand, when demand approaches the limits of generating capacity, prices may rise above the marginal cost of generation. In this analysis, during those high demand periods a “reliability price adjustment” is added, which represents the value that consumers place on the reliability of electricity service. The marginal cost of generation and the reliability price adjustment are added to unit taxes (per kilowatthour) to estimate the competitive price for generation services. In this report, the competitive price for generation services and the regulated price of transmission and distribution 3 equal the retail competitive price of electricity.

Time-of-use prices could affect the timing of electricity usage

Because electricity cannot be stored, as its demand rises and falls by season or during the portions of a day, additional generating plants in reserve must be immediately brought on line (“dispatched”) to serve increases in demand. As a result, the costs of electricity production—and thus the marginal cost and competitive price—rise and fall with changes in the demand for power, as generating plants are dispatched or taken off line. Competitive prices that vary with demand may lead to changes in consumer electricity usage patterns. Time-of-use prices give consumers an incentive to reduce usage during high demand ("peak") periods and increase usage during low demand ("off-peak") periods. It is uncertain what effects time-of-use prices will have on demand patterns. Therefore, several estimates of short-term consumer responsiveness to changes in price are used in this analysis to examine the effects of time-of-use prices.

The costs of new generating capacity are recovered through competitive prices

For most of the competitive cases presented in this analysis, generating capacity is added only if it provides an increase in system reliability 4 sufficient to recover the cost of building the capacity. Investments in new capacity used only during peak demand periods (“peaking” capacity) are recovered through the reliability price adjustment. Investment costs for other new capacity (non-peaking capacity) are recovered through the premiums received during periods when more expensive plants are running and setting the price. In this way, the analysis is conducted so that the costs of all new generating capacity are recovered through competitive prices. Other than the 109 gigawatts of fossil fuel and nuclear capacity assumed to be retired for economic and other reasons in the Annual Energy Outlook 1997 (AEO97) Reference Case,5 there is no additional economic replacement of generating capacity in any case in this analysis. That is, new generating plants that are cheaper to operate are not built solely to replace the existing stock of generating capacity.

Competition is assumed to begin in 1998

Finally, the analysis assumes that regional competitive markets for generation services are in place on January 1, 1998. In reality, it is not clear how fast the restructuring of the U.S. electric power industry will proceed. Some States will move more quickly than others to establish rules and institutions to facilitate a competitive market for generation services. The speed at which competitive markets for generation services will emerge is a political issue and an economic issue. Political forces will establish the ground rules, and market forces will respond according to those ground rules. Consequently, the rate of change is very difficult to estimate. This simplifying assumption is used to facilitate the purpose of the analysis—i.e., to compare competitive electricity prices based on marginal costs with regulated electricity prices based on average costs. The estimates of future prices in this analysis are not forecasts, but they represent a range of possible outcomes consistent with the variations in the assumptions for the cases.

Results of the Analysis

Table ES1 summarizes the major findings of this analysis. Average annual electricity prices are displayed for two regulation cases and five competitive cases. The regulation cases are the AEO97 Reference Case and the No Competition Case. The No Competition Case is identical to the AEO97 Reference Case, except that no reductions in nonfuel O&M and general and administrative (G&A) costs over time are assumed. In the AEO97 Reference Case, cost reductions are assumed to result from competitive pressures in the wholesale market for electric power, as well as from supplier preparations for retail competition, and result in a 7-percent reduction in the price of electricity in 2005.

The competitive cases are the Flat Rates Case (similar to the AEO97 Reference Case, but with prices based on marginal costs instead of average costs, and no time-of-use rates); the Moderate Consumer Response Case (testing the effects of moderate consumer responsiveness to competitive time-of-use electricity prices); the High Consumer Response Case (same as the previous case but testing a higher assumed level of consumer responsiveness); the High Efficiency Case (illustrating the price impact of a set of higher competition-induced cost savings and efficiency improvements); and the Intense Competition Case (showing the effects of severe competitive pressures, which drive prices for generation services down to a level that is only slightly higher than the cost of the fuel used to generate power).

Short-term competitive prices could be lower due to uneconomic costs in regulated prices

Over the short term (2 to 3 years), if stranded costs are not recovered through prices,6 average electricity prices nationwide under competition could be 6 to 13 percent lower than regulated prices, as indicated by the results for the year 2000 from the competitive cases (excluding the Intense Competition Case), compared with those from the AEO97 Reference Case. As shown in Table ES1, the AEO97 Reference Case represents the effects of price reductions under regulation that result from the limited competition that currently exists in the wholesale market for electric power. These reductions are shown in the relative prices for the AEO97 Reference Case and the No Competition Case. Hence, the 6- to 13-percent reduction in electricity prices due to full competition in the generation services market is in addition to the price reductions that are occurring due to the current level of competitive pressures reflected in the AEO97 Reference Case.7

The Intense Competition Case represents conditions under which competition becomes so fierce that producers price their power at virtually the cost of the fuel used for generating it. End-use price projections in this case range from 5.3 to 5.5 cents per kilowatthour (1995 cents), compared with 6.1 to 6.3 cents per kilowatthour in the Moderate Consumer Response Case. This result is a 22-percent price reduction relative to the AEO97 Reference Case and a 24-percent reduction relative to the No Competition Case in the short term (the year 2000)—a price reduction greater than that which results from the 40-percent cost reductions assumed in the High Efficiency Case.

Low-cost producers may recover their fixed costs through the premiums they receive during high demand periods when more expensive producers are setting the price, but it is possible that the low average annual price in the Intense Competition Case would not be high enough to provide full cost recovery for any but the most efficient generating technologies (combined cycle, for example), operated as efficiently as possible (as baseload capacity). It is not evident that this low price would provide cost recovery for less efficient technologies operated under less than optimal conditions, such as combustion turbine technologies generating electricity only during peak periods. Consequently, it is likely that generating plants with higher operating costs would be forced to retire. The consequences of such economic retirements are not included in this analysis.

As the need for new capacity increases, competitive prices will rise until capacity expansion becomes profitable. Therefore, the prices projected in the Intense Competition Case are not considered to be sustainable over the long term.

Figure ES1 shows the competitive regional markets for electric power that were analyzed in this study. Regional markets based on the North American Electric Reliability Council (NERC) regions and subregions formed the basis for the analysis of prices. Some regions of the country could experience larger price reductions, some smaller, and some could see prices rise under competition (Figure ES2). The 6- to 13-percent price reduction in the short term is the result of the assumption of no stranded cost recovery through prices. Most current Federal and State legislative initiatives, however, including the California restructuring bill (AB 1890), and recent rulings from the Federal Energy Regulatory Commission (FERC), allow at least some level of stranded cost recovery. Therefore, over the short term, the degree to which electricity prices may fall under competition depends on how much regulatory and legislative relief producers receive for stranded costs. In cases where policymakers grant suppliers 100 percent stranded cost recovery, average competitive prices will closely resemble regulated prices in the short term.

If there is no stranded cost recovery, the price projections presented in this analysis suggest that the reduction in market value of current generating assets (stranded assets) that results from the inability to recover stranded costs could range between about $72 and $169 billion (1995 dollars). In the Intense Competition Case, estimates of net stranded assets are as high as $408 billion.8 However, this estimate of stranded assets assumes that there will be no reduction in costs as a result of competitive pressures (beyond those assumed in the AEO97 Reference Case), whereas it is likely that price reductions of 20 to 25 percent under competition would result in intensive efforts on the part of suppliers to reduce costs. A significant portion of the $25 to $30 billion that electricity suppliers incur each year in nonfuel, non-capital-related costs would not be recovered through prices in the Intense Competition Case. These costs would become stranded unless suppliers became more efficient. As mentioned above, the prices projected in the Intense Competition Case are not considered to be sustainable over the long term.

If the O&M and G&A costs that are removed from competitive prices in the Intense Competition Case were eliminated completely from generating costs as well, then net stranded assets would be much lower— $110 billion as opposed to $408 billion. In other words, the O&M and G&A costs that are potentially fixed, and therefore not included for pricing purposes, could con tribute roughly $298 billion to net stranded assets through 2015 if the costs were incurred but not recovered through prices. As producers find ways to reduce costs under the intense competitive pressures assumed in this case, it is likely that these O&M and G&A costs would be reduced, and that the realized level of stranded assets would be between the $110 and $408 billion cited above. Also, as mentioned previously, it is likely that the low prices illustrated in the Intense Competition Case could not be maintained without substantial cost reductions.

The stranded asset estimates are "netted out" within the regions used in this analysis. In other words, this calculation assumes that the revenue gains of winners (with average costs lower than competitive prices) offset the losses of other suppliers in the same region. Absent this netting out effect, stranded assets could be as much as 20 percent higher than the calculations presented in this analysis. Also, in the absence of stranded cost recovery, Federal income tax receipts could fall by as much as $2.5 billion per year on average under the range of most likely cases in this report, as a result of reductions in the taxable income of electric utilities. It is likely, however, that these first-order revenue reductions would be partially offset by reduced Government outlays for electric power.9 Additionally, it is possible that the macroeconomic effects of competitive pricing could result in higher tax revenues.

Long-term competitive prices will be lower if there are efficiency gains and cost reductions

Over the long term (10 to 20 years), prices under competition will likely be lower than prices under regulation would have been for a number of reasons. New uneconomic costs that producers incur under competition will not be recoverable through prices as they have been under regulation. Either suppliers will avoid the uneconomic costs that they incurred under regulation in the past (high-cost generating plants, high-cost contractual agreements), or their shareholders, not their customers, will bear the consequences. Additionally, prices under competition will be lower than regulated prices to the extent that suppliers reduce costs and improve operating efficiencies in response to competitive pressures.

Future cost reductions and efficiency improvements under competition are very difficult to estimate. The reductions in nonfuel O&M costs, reductions in the costs of new generating capacity, and improvements in operating efficiencies (heat rates) assumed in the High Efficiency Case lead to estimated reductions in competitive electricity prices of 11 percent relative to the AEO97 Reference Case by 2015 (Table ES1).10

Customer response to time-of-use prices affects average prices

If time-of-use prices for electricity do reduce electricity consumption during high usage (peak) periods and increase consumption during low usage (off-peak) periods, the result will be a more constant level of demand for electricity across seasons and portions of the day. The extent to which consumers might change their electricity usage habits when faced with prices that change over time, or when offered the availability of lower rates for interruptible service, is unknown. Nevertheless, this analysis suggests that if consumers respond, competitive prices will be affected. Average annual competitive prices could be pushed lower if consumers respond to time-of-use prices by reducing peak period consumption, thus reducing marginal costs and competitive prices as the most expensive generating plants are used less frequently.

The price effects achieved through a reduction in peak period demand could be at least partially offset if total demand for the year increases. In other words, while prices could be reduced by the effects of lower peak period electricity usage, lower prices during off-peak periods could increase off-peak usage, and off-peak prices could be higher than they would have been with a lower level of consumer price responsiveness.

If consumers display a high level of responsiveness to changes in prices, the increase in demand during off-peak periods (due to lower off-peak prices) could be greater than the decrease in demand during peak periods (due to higher peak period prices). Higher off-peak demand would push competitive prices higher than they otherwise would have been. As a result, competitive prices could be higher given a higher level of consumer responsiveness, although they would not rise to the level of regulated prices. Average annual competitive electricity prices in 2000 vary by about 0.3 cent (a 5-percent range) over the range of consumer response cases used in this analysis (Flat Rates, Moderate Consumer Response, and High Consumer Response cases) (Table ES1).

On average, utility profits will decrease during the transition to competition in the absence of stranded cost recovery, but the financial effects will vary from utility to utility

On average, the move to competition without any stranded cost recovery would cause prices to fall by about 6 to 13 percent over the range of most likely cases presented in this report. In the aggregate, utility profits and the taxes directly associated with them are about 20 percent of revenues. Thus, the revenue reductions caused by a 10 percent decrease in prices would result in substantial reductions in profits (and taxes). Common stock dividends and common stock prices would fall, but the industry as a whole would remain solvent.

There are, however, very large variations in prices (and average costs) within and between regions, and the financial effects will vary depending upon the utility and its costs. In the very short term (e.g., 1998), the revenues of roughly 25 percent of the privately-owned utilities supplying financial data to the FERC would actually increase. Conversely, about 20 percent of privately owned utilities would observe revenue reductions of over 30 percent, and these relatively high cost utilities would be financially distressed. Without substantial cost reductions or stranded cost recovery, some of these utilities may not be able to remain solvent and would have to declare bankruptcy.


Contact:
Robert T. Eynon
robert.eynon@eia.doe.gov
Phone: (202) 586-2315