Any climate policy is almost certain to target the electric-power industry, which is responsible for about 38 percent of U.S. carbon dioxide (CO2) emissions. Said policy especially would affect coal-fired power plants, which contribute about 82 percent of the electric power CO2 total. How would various policy options change the economic value of current and proposed generation assets? The answer to this question could affect the rapidly evolving generation mix profoundly, particularly at a time when utilities are planning to add more than 60 GW of carbon-intensive coal plants.
Determining the impacts of climate policy on the electric-power industry, however, is a complex problem that requires considering both regional diversity and the linkages among markets for fuel, power, and emissions. In particular, imposing a cost penalty on CO2 emissions from electricity generation would affect the net revenues of various types of power plants very differently. As these differences become more apparent, the value of current generating assets could rise or fall dramatically, and significantly alter preferences among investment options for future generation.
To examine this complex issue, EPRI has developed an analytical framework for assessing the economic impacts of climate policy on the electric power industry, and has applied the framework to determine how specific CO2 emission penalties would affect existing and proposed power plants.1,2 The primary goal of this analysis is to assess potential risks of CO2 policy for both fossil-fired and non-CO2-emitting electric generation, and to put these risks into perspective with another key risk—uncertain natural gas prices. A second goal is to understand the cost-effectiveness of reducing CO2 emissions from the electric-power sector in the near term, while the generation fleet is relatively fixed and natural-gas prices are high.
Imposing a price on CO2 emissions can increase greatly the operating costs of fossil-fuel generation. The size of the increase for a particular power plant will depend on both the type of fuel used and the plant’s efficiency. The effect can be significant: Over the lifetime of a coal-fired power plant, a CO2 emission penalty of $10 per ton could have a present value approximately equal to the overnight-investment cost of building the plant. Higher operating costs, however, do not tell the whole story. From a cash-flow perspective, what matters most is how a plant’s dispatch costs compare to market prices. Specifically, for some types of plants, net revenues actually may increase if rising wholesale-power prices exceed higher production costs, while other plants may experience sharply diminished revenues. Net revenues are needed to cover fixed operations and maintenance, depreciation, and recovery of the cost of capital.
The process of how emission penalties affect electricity market prices and net revenues for different types of generation is illustrated in Figure 1, which presents a simplified three-plant example. When there is no emissions penalty, the natural-gas plant is assumed to operate at the margin and set a market price equal to its dispatch cost of $50/MWh, allowing coal and nuclear plants to earn net revenues of $25/MWh and $45/ MWh, respectively. The right side of Figure 1 shows what happens when an emission penalty of $20/ton of CO2 is imposed. The coal plant is assumed to have a CO2 emission intensity of about 1 ton/MWh, so its dispatch cost increases $20/MWh, while dispatch cost at the gas plant—which has about half the emissions intensity—rises $10/MWh. As a result, the market-price increases to $60/MWh, which still allows the coal plant to earn net revenues of $15/MWh and increases nuclear-plant revenues to $55/MWh.
Actual regional markets, of course, contain numerous generating units, which are dispatched in the order of their production costs, from lowest to highest, as shown in Figure 2. Nuclear and renewable energy plants have the lowest marginal costs and are dispatched first, followed by coal plants, which in this example represent the largest single part of the supply curve. The sharply rising right-hand portion of the curve represents natural-gas and oil plants, which have considerably higher dispatch costs. For each hour of the year, the market price is established by where the amount of load intersects the supply curve. Plants to the left of that point are dispatched, and those to the right do not operate for that hour. Net revenue for each plant is determined by the difference between its marginal cost and the market price.
When the net revenues for each plant are summed over all the hours of a year, those toward the left end of the supply curve earn the most, both because they have lower costs and because they are dispatched more often. Assuming no CO2 emission penalty, further analysis shows that nuclear units at the far left would earn annual net revenues above $300/kW-year, while income for the coal units in the middle would range from $100/kW-year to $250/kW-year. Oil and gas units at the far right would have much smaller annual net revenues, approaching zero in the extreme.
Now suppose a CO2-emission price of $25/ton is imposed on power plants in the region represented in Figure 2. The result, shown in Figure 3, is to boost the dispatch prices of the fossil-fueled plants in proportion to their emissions rates. The effect is most dramatic on the coal plants, which experience an approximate doubling of production costs, to roughly $50/MWh. Gas plants, which produce about half the amount of CO2 per megawatt-hour as coal plants, have smaller cost increases. Production costs for non-emitting nuclear and hydro plants at the far left of the curve remain unaffected.
Further analysis shows that the overall effect of adding the $25/ton CO2 emission penalty is to raise the average wholesale price for the region by $21/MWh. As a result, coal plants are able to recoup most of their cost increases, with more efficient plants experiencing the least decline in net revenues. The cash flow of gas and oil plants, already marginal, also changes very little, but net revenues for nuclear and hydro plants jump by about 50 percent.
Both Figs. 2 and 3 represent a region of the Upper Midwest (ECAR-MAIN) dominated by coal, which means that rising CO2 prices have very little effect on total emissions, since the non-emitting nuclear and hydro plants already are running flat out and economically competitive gas generation is limited. As a result, even a CO2 value of $50/ton would produce only a 4 percent reduction in regional emissions given the current generation mix. Including customer response to higher prices could lead to greater reductions than redispatch for this region, depending on the price elasticity.
The situation is somewhat different for a region like Texas (ERCOT), with its supply curve dominated by gas. Here, greater availability of lower-emitting gas generation means that rising CO2 penalties can result in a shift in the dispatch order in favor of gas plants. Even so, reductions in regional emissions in response to CO2 costs are highly dependent on natural-gas prices. When gas is selling for around $8/MMBtu, even a CO2 value of $40/ton produces little emissions reduction. On the other hand, with gas at $4/MMBtu, a $20/ton CO2 price could generate a 20-percent reduction in emission through redispatch.
Regional differences also are apparent in the combined effects of emission penalties and changing gas prices on the cash flow of a hypothetical, highly efficient (9 MMBtu/MWh) coal-fired power plant, as shown in Figure 4. With natural gas at the 2005 average price of $8.24/MMBtu, in the diagram on the left, net revenues from the plant would start off much higher in the ERCOT region than in ECAR-MAIN, but fall more rapidly with rising emission costs. If the price of gas were to fall to $4.24/MMBtu, however, net revenues for the plant only would be marginally higher initially in the gas-dominated region and would quickly fall below those of a similar plant in the coal-dominated region with rising CO2 penalties.
Two major conclusions can thus be drawn from the analysis so far, which has focused on the current generation mix. First, the impacts of climate policy on the electric-power industry must take into account the net revenues of individual plants in a region and not just the impact of rising CO2 value on production costs. Second, only under the combined circumstances of available gas-fired generation capacity and low gas prices does a rising price for CO2 significantly affect the cash flows of efficient coal plants and lead to substantial reduction of regional emissions through redispatch.
The largest emission reductions to result from imposing CO2 costs will come over time by providing investment incentives for new generation that produces lower or no emissions. To explore this potential impact, the EPRI framework was extended to provide a multi-year analysis of the effect that investor-driven generation additions could have on utility net revenues and regional CO2 emissions.
Consider, for example, the impact of adding more nuclear generating capacity to the ECAR-MAIN region. Assuming a net-revenue threshold of $300/kW-year for such additions to provide the equity returns required by investors, the region already could support about 6,900 MW of new nuclear plants. The effect of adding this non-emitting capacity would be to shift the supply curve to the right—pushing gas plants even further toward the margin and significantly reducing net revenues for existing coal and nuclear plants. The additions also would reduce regional CO2 emissions by about 7 percent.
Emission penalties would magnify this impact greatly. A CO2 price of $10/ton, for example, would enable ECAR-MAIN to support almost three times as much new nuclear generation at the $300/kW-year net-revenue level (18,600 MW versus the 6,900 MW of economic additions estimated with no CO2 cost). The result, however, would be to push the supply curve far enough to the right that more coal plants would become marginal and virtually the entire coal fleet would have net revenues below $100/kW-year. Meanwhile, regional CO2 emissions would fall 19 percent. Clearly adding this much nuclear generation will take time. Figure 5 shows how emissions continue to fall as the price of CO2 rises and stimulates further investments in non-emitting generation.
If investors require a larger return on investment because of higher construction costs or greater perceived risk, the effect would be to shift both curves in Figure 5 to the right. For example, an investment threshold of $400/kW-year would imply that achieving a 19 percent reduction in emissions might require a CO2 price of around $18/ton instead of $10/ton. Similar analyses could be conducted for investments in other low- and non-emitting technologies, such as coal with carbon capture and storage or renewables.
Climate policy that puts a price on CO2 significantly increases the dispatch cost of fossil generation, particularly coal, but these cost increases largely are passed on to the wholesale market in the short run. Net revenues for the more efficient coal-fired generators stay relatively flat except under circumstances of available gas generation and low gas prices. In the near term when the generation mix is fixed and gas prices are high, putting a price on CO2 emissions results only in small reductions in CO2 emissions through redispatch of the existing generators, and the increased use of gas is likely to strain the gas market. In the longer term, a price on CO2 creates a powerful incentive to add new non-emitting generation. These additions greatly can cut emissions, but the extent of reductions is highly sensitive to the cost of these non-emitting technologies.
1. EPRI Technical Update 1013296, Program on Technology Innovation: A Conceptual Framework for Modeling the Impact of CO2 Policy on Generator Cash Flows, 2006.
2. EPRI Report 1012577, Program on TechnologyInnovation: Managing the Risks of Climate Policies: The Effect of a Carbon Price on Existing Generation and Evaluation of Emission Reduction Investments, 2006.