This chapter presents a quantitative analysis of the likely impacts that competitive electricity generation markets could have on fuel supply industries. The primary tool used for the analysis is the National Energy Modeling System (NEMS), a comprehensive model of energy markets that projects energy supply, demand, and prices. NEMS is an integrated model that represents the supply, conversion, and end-use demand sectors in domestic energy markets. By balancing energy supply and demand, NEMS projects production, imports, consumption, and prices of energy in the mid-term forecast horizon (in this analysis, through 2015). Because restructuring affects all energy consumers and producers, all the demand and supply modules within NEMS were used in the analysis.

Case Descriptions and Assumptions

In order to explore the potential impacts of a competitive electricity market on fuel markets, several cases were constructed. The regulatory, legislative, and environmental policies that will eventually emerge are currently being debated in a number of different forums. Therefore, there is considerable uncertainty about the conditions under which a competitive electricity market will operate. In order to capture this uncertainty, a range of possible outcomes was prepared, each based on different assumptions about key electricity and energy variables. Although these cases are not forecasts, they do represent potential outcomes that could occur under the range of assumptions analyzed. Two full competition cases in addition to a partial competition case (the AEO98 reference case) are compared with a no competition case in order to illustrate possible impacts of competition.

The first case (no competition) represents a market in which there are no further competitive initiatives and in which participants assume that no further move toward competition will occur. This case was developed to provide a base against which the competition cases and the AEO98 reference case could be compared. While the AEO98 reference case assumes that only three regions (California, New England, and New York) will move to full competition over the next decade, it also assumes that electricity market participants will anticipate the onset of full competition.²⁰² To develop the no competition case, EIA modified the following assumptions from the AEO98 reference case:

• Heat rates for new plants are assumed to improve less over the forecast horizon than in the AEO98 reference case, because there would be less incentive for vendors to improve them if electricity markets remained regulated. For example, while heat rates for new advanced combined-cycle plants were assumed to be 6,350 British thermal units (Btu) per kilowatthour in the AEO98 reference case, the no competition case assumes that they would be only 6,668 Btu per kilowatthour by 2015, an efficiency that is 5 percent lower (Table 18).
  
  
• The capital costs of new generating plants are assumed to be 15 percent higher than those assumed in the AEO98 reference case. In regulated electricity markets with full cost passthrough, plant equipment manufacturers are assumed to be less aggressive in lowering costs to maintain market share. In addition, it is assumed that equipment would be tailored to meet individual customer needs, thus reducing cost savings that could be realized if more factory construction and modular design were employed.
  
  
• Capital costs for new construction are assumed to be based on the regulated utility cost of capital, rather than on the project cost of capital used in the AEO98 reference case. In a regulated environment, utilities are allowed to recover their capital costs over 30 years. The AEO98 reference case assumes higher costs of capital based on project financing by unregulated investors. In a competitive market, new capacity additions are riskier and investors are assumed to plan for a 20-year recovery for capital costs.
  
  
• Both general and administrative costs, as well as operation and maintenance costs, are assumed to decline by 5 percent, compared with the 25-percent decline assumed for the AEO98 reference case. Much of the incentive to cut staff and reduce costs comes from the anticipation of competitive electricity markets. In a regulated market, these costs are paid by consumers, dampening the incentive to reduce them.

Table 18. Comparison of Selected NEMS Assumptions

The competition cases described below contain varying assumptions on how a deregulated electricity market may evolve. Two full competition cases are considered, combining assumptions about low fossil fuel use with the AEO98 reference case electricity demand and about high fossil fuel use coupled with higher electricity demand. While both cases assume full competition, they differ from each other in assumptions about consumer responses to prices, technological progress for oil and natural gas production, legislation promoting generation from renewable sources, and retirement decisions for fossil and nuclear generators. These competition cases are designed to characterize the effects of competition that is more intense than is assumed in the AEO98 reference case. While the cases may overstate the intensity of competition, they provide an outer boundary on the effects on electricity markets. Assumptions common to all the cases are as follows:

• Both the reference case and the competition cases assume that California, New York, and New England will become fully competitive within the next decade. Electricity prices for commercial and industrial customers in California are assumed to remain at 1996 levels between 1998 and 2001, with residential customers receiving a 10-percent reduction from 1996 prices during the same period. After a transition period between 2002 and 2007, California markets are assumed to be fully competitive by 2008. This transition period reflects the time needed to establish the institutions for a competitive market and to allow for recovery of stranded costs to the extent permitted by the State. New York and New England have a similar transition period between 1998 and 2007. In the competition cases (unlike the AEO98 reference case), all other regions are assumed to move to competitive markets beginning in 1998 with the same transition period and to become fully competitive beginning in 2008. Full competition, in addition to the cost and efficiency gains assumed, means that electricity prices will be driven by competition among electricity generators rather than by regulatory proceedings.²⁰³
  
  
• Limits on power transmission are relaxed in three regions from those assumed in the AEO98 reference case. For the competition cases, it is assumed that Texas, New York, and New England can transmit more power to adjacent regions than they could in the AEO98 reference case. Texas is assumed to have an incentive to build new transmission connections to neighboring States in order to allow its low-cost fossil plants to sell electricity outside the State. In New York and New England it is assumed that new transmission connections to Canada will be built, allowing additional sales of electricity from Canada to the United States.
  
  
• Investments in new generating capacity are assumed to exceed the levels that would be expected on the basis of optimal economic efficiency alone. This could occur if suppliers invest in new capacity in order to increase their market share. The level of overbuilding to reflect this investment behavior is assumed to be 10 percent above that which would occur under assumptions of economic efficiency.
  
  
• Because of competitive pressures to maintain market share, a higher rate of improvement in coal mining productivity is assumed in the competition cases—2.5 percent annually compared with 2 percent in the AEO98 reference case.

In order to represent outcomes from restructuring that result in higher or lower fossil fuel consumption, additional assumptions were made in the competition cases. The following assumptions were made for the high fossil case:

• Optimistic technological progress rates that lower costs for oil and natural gas supply are assumed because of competition. Compared with the 1.3-percent annual reduction in the AEO98 reference case, technological improvements are assumed to reduce onshore drilling costs by 1.6 percent per year. The impact of technology on costs is offset by other factors, including rig availability and drilling levels. Improvements in technology are assumed to result from pressure exerted by electricity markets on oil and gas producers to lower their costs to maintain (or to increase) their market shares.
  
  
• Retirements of existing fossil-fueled power plants are reduced to address the uncertainty in the price of generation services in competitive markets. It is assumed that existing fossil-fueled power plants will be retired if their operating costs are greater than 6 cents per kilowatthour. In the other cases, plants with current operating costs greater than 4 cents per kilowatthour are assumed to be retired early because they would not be competitive given the costs and performance of new generating sources. The higher cost criterion used in this analysis allows more fossil plants to be available over the projection period. This assumption reflects the uncertainty about market prices for generation services in a competitive market as well as the value of having higher cost capacity available to provide ancillary services such as voltage stability and reactive power.
  
  
• Based on estimates of elasticities observed in regulated market, a higher level of electricity demand is assumed to capture the uncertainty of predicting the effects that lower electricity prices would have on consumption. In addition, the potential reduction in regulatory oversight could cause demand-side management programs to be deemphasized, resulting in an increase in electricity demand above what it would be if such programs were in effect. New pricing structures, such as time-of-day pricing, could also increase demand. The growth rate for electricity sales (1.9 percent) is assumed to be close to the growth rate for the gross domestic product (GDP), which averages 2.1 percent per year from 1996 through 2015. In the AEO98 reference case, electricity consumption is projected to grow by 1.5 percent per year. In the high fossil case, residential and commercial sector consumption of electricity was adjusted to mirror GDP growth.

In the low fossil case, the additional assumptions include the following:

• The low fossil case assumes that legislation mandating a renewable portfolio standard (RPS) will be enacted. The standard is based on H.R. 655, Electric Consumers' Power to Choose Act of 1997 (Title I Section 113) submitted by Congressman Dan Schaefer (R-CO). This bill requires that 2 percent of new generation be produced from renewable sources by 2000, increasing to 3 percent by 2005 and 4 percent by 2010. The RPS results in higher levels of generation from renewable sources than projected in the AEO98 reference case. Higher generation from renewable sources dampens the demand for fossil fuels for a given level of electricity demand. (In March 1998, the Department of Energy announced the Administration's Comprehensive Electricity Competition Plan, which recommends an RPS calling for 5.5 percent of generation from renewable sources by 2010. This is about 20 billion kilowatthours more than is assumed in the low fossil case.)
  
  
• This case also assumes no additional retirements of nuclear capacity before their operating licenses expire beyond those already announced. It is assumed that uncertainty about the price of generation services in a competitive market will encourage utilities to postpone the decision to retire plants early. In the AEO98 reference case, about 18 gigawatts are retired 2 to 10 years before the plant licenses expire, based on the expected need to invest additional capital to refurbish major systems. In this analysis it is assumed that only Big Rock (1997), Haddam Neck (1997), Maine Yankee (1997), Browns Ferry (1997), and Zion 1 & 2 (2004),²⁰⁴ for which retirements already have been announced, will be retired early.

There are likely to be many innovative approaches to providing electricity services that develop under competition. For example, power from environmentally benign sources (i.e., green power) is currently offered in California. Because the quantitative impacts of these programs and others that improve the efficiency of delivering electricity services are not well understood at this time, they were not considered in the low fossil case.

Results

Electricity Capacity and Generation

Figure 24. Differences in Capacity Additions from the No Competition Case (Click graph to view full size)

Currently there is more than sufficient baseload capacity to meet electricity demand, and new baseload capacity will not be needed in significant quantities for several years. From 1996 to 2015, additions of coal-fired capacity range from about 20 to 49 gigawatts for all the cases analyzed. In contrast, additions of natural-gas-fired turbine  and  combined-cycle capacity range from about 256 to 324 gigawatts; however, the impact of new natural-gas-fired turbines (132 to 158 gigawatts) is less than the level of capacity additions would indicate because, unlike coal-fired plants, these units operate at low capacity factors.

Even with the dominance of gas-fired capacity additions in mind, there are variations in capacity choice among the cases of this study (Figure 24).

Figure 25. Electricity Generation by Fuel Type, 1997, 2005, 2015 (Click graph to view full size)

Table 19. Electricity Generating Capability (Thousand Megawatts)

Table 20. Electricity Supply, Disposition, and Prices (Billion Kilowatthours, Unless Otherwise Noted)

In the high fossil case, where capital costs are assumed to be recovered over a shorter period, coal-fired capacity additions are about 6 gigawatts less in 2005 than in the no competition case. In this case, gas-fired additions are about 22 gigawatts higher and are shared between turbines (14 gigawatts) and combined-cycle plants (8 gigawatts). By 2015, coal-fired additions are almost 12 gigawatts less than in the no competition case, and gas-fired additions are about 69 gigawatts higher. These changes in capacity additions indicate that the assumptions about competitive markets used in this case have a significant impact on fossil-fired capacity additions in the later years of the projection period.

The low fossil competition case (where the RPS is imposed and nuclear capacity is assumed not to be retired before operating licenses expire) reduces the need for fossil-fueled plants even with a higher level of electricity sales than in the no competition case. By 2015, coal-fired capacity is about 30 gigawatts lower and gas-fired capacity is about 23 gigawatts higher than in the no competition case. As a result, coal-fired generation is about 9 percent lower than and gas-fired generation is about 5 percent above the no competition case (Table 20).

It is interesting to note that the need for turbines is higher by about 12 gigawatts in the low fossil case compared with the no competition case because the higher level of generation from nondispatchable renewable sources requires that additional backup capacity be made available to meet peak requirements. These cases indicate that natural gas is expected to have an increasing share of electricity generation as demand levels grow and that coal-fired generation will be lower than would occur in regulated electricity markets, absent the assumption about additional demand growth under competition.

Electricity trade levels across the NEMS regions change modestly across the cases analyzed. Incentives for regional trade are driven by differences in regional generation sources and region-specific characteristics. The assumptions about increased transfer capability of the transmission network in the low and high fossil cases do not cause trading patterns to change because the cost differences are not sufficient to make trading economical. This analysis does not address the potential changes in electricity trade within a region that could occur in competitive markets.

Renewable Sources

Unless required by policies, the restructured electricity market is not expected to stimulate central station renewable energy technologies. Overall, the scenarios suggest that renewable sources will remain more costly than fossil-fueled alternatives through 2015 and will penetrate electricity markets further than they do in the reference case only to the extent compelled, such as by an RPS that mandates generation from renewable sources. The cases suggest that, if policies require increased use of renewable sources, average electricity prices will increase slightly. Under the assumed RPS (HR 655), most of the growth in renewable generation will be from biomass, geothermal, and wind.

The results suggest that renewable sources will garner only a minor overall portion of electricity supply under a range of electricity market conditions. In the absence of an RPS, nonhydroelectric renewable sources (including municipal solid waste) hold only a 2.4-percent share of total U.S. electricity generation in 2015; the hydropower share falls as low as 6.6 percent (Figure 26). Although increased overall electricity demand also raises generation from renewable sources, significant growth occurs only under an RPS. Whereas generation by RPS-qualifying renewable sources (biomass, geothermal, solar, and wind) is 74 billion to 76 billion kilowatthours by 2005 and reaches as much as 85 billion kilowatthours by 2015 with no RPS, it increases to 130 billion kilowatthours in 2005 and to 190 billion kilowatthours in 2015 with an RPS (Table 21).

In the high fossil case, defined renewable sources remain barely changed from their no competition case market share. If renewable sources are to expand more rapidly, the results suggest a need for some significant market change, such as accelerated improvements in renewable energy technologies, an RPS, successful green pricing programs (where consumers choose electricity suppliers based on their impacts on the environment), subsidies, or higher costs for competing technologies.

Figure 26. U.S. Electricity Generation Shares by Energy Source, 2015 (Click graph to view full size)

Finally, the results suggest that renewable sources are highly vulnerable to improvements in competing fossil- fuel technologies, as shown by the high fossil case. Compared with the no competition case, renewable sources fare about the same under competition absent a policy mandating higher shares.

The results also show the likely technology choices for expanded use of renewable sources under more rapid growth or RPS conditions. Biomass, wind, and geothermal are the likely "winners" among renewable energy technologies. Biomass-powered generation increases most, more than doubling from 46 billion kilowatthours in 1996 to 97 billion kilowatthours in 2015 in the RPS case; its capacity also increases significantly, adding more than 7 gigawatts of new capacity by 2015. Geothermal generation increases from 16 billion kilowatthours in 1996 to 52 billion kilowatthours in 2015 in the RPS case; its capacity also increases significantly, far more than doubling by 2015. Wind-powered generation also increases from 3 billion kilowatthours in 1996 to 38 billion kilowatthours in 2015, a leap of nearly 14 gigawatts of capacity by 2015 in the RPS case. Because biomass capacity operates a much greater proportion of the time than wind power and can compete in more regions than geothermal, biomass-fueled generation appears the most likely source for increased electricity generation under policies encouraging use of renewable sources. However, significant issues of cost and land use could arise if the growth of biomass becomes a reality (see Chapter 5).

Because they remain more expensive than both fossil and other renewable alternatives, solar technologies are minor contributors in all the cases and do not increase significantly. Further, because neither solar thermal nor photovoltaic technologies operate as intensively as fossil technologies (they have lower capacity factors), their contribution to total generation remains small. The use of photovoltaic technologies could grow much more rapidly if their cost declined or if electricity prices were higher than those projected in this analysis.

Electricity Prices

Electricity prices are projected to decline from 1996 levels for all of the cases analyzed, including the no competition case. Prices will decline even in a "no competition" market because investments in new capacity will be relatively modest compared with historical levels and because of expected decreases in the price of coal. Prices in the competition cases are further reduced due to improvements in the efficiencies of both plant operations and the labor force. An additional factor contributing to lower electricity prices in the competition cases is less construction of capital-intensive coal plants (Table 20). In competitive markets, electricity prices are expected to be sensitive to the price of natural gas because it is projected to be used to meet demand during peak periods.

Electricity Fuel Consumption

In comparing the cases, EIA found that total energy consumption for electricity generation essentially follows the overall demand for electricity, although the composition of the fuel demands is important in explaining differences. The AEO98 reference case has slightly higher overall consumption in the electricity sector in 2005 than in the no competition case, but the two cases are virtually the same in 2015 (Table 22), despite the fact that electricity demand in the AEO98 reference case is higher by 51 billion kilowatthours in 2015, up from only 33 billion kilowatthours in 2005 (Table 20). In part this reflects the lower efficiencies for coal-fired generation. In the no competition case, the assumptions with respect to the cost of capital provide an incentive for more coal-fired and fewer gas-fired capacity additions than in the AEO98 reference case. Because of the lower efficiencies for coal-fired generation, this translates into roughly the same consumption in the two cases in 2015, despite the higher demand in the AEO98 reference case. The tradeoff between coal and natural gas in the two cases leads to a slightly higher efficiency in total electricity production in the AEO98 reference case.

Table 21. Renewable Energy Capacity and Generation

In the low fossil case, coal consumption is lower by almost 2 quadrillion Btu in 2015 compared with consumption in the no competition case. Consumption of renewable and nuclear fuels is higher based on the assumptions used in the low fossil case, and natural gas consumption is about the same as it is in the no competition case. In the high fossil case, both coal and gas consumption are higher in 2015 than they are in the no competition case in 2005, but by 2015 coal consumption is about the same as it is in the no competition case. Natural gas consumption is about 2 quadrillion Btu greater because of higher electricity demand levels.

The average price of fuel used for electricity production in 2015 is projected to be about the same as in 1996 in all but the high fossil case (Table 22). In the high fossil case, an increase of about 11 percent in the average price is projected because of higher natural gas prices resulting from assumed higher drilling costs for onshore production. Natural gas prices increase slightly in the other cases but are offset by an almost 30-percent decline in coal prices between 1996 and 2015.

Table 22. Energy Consumption and Prices for Electricity Generation