Discussion Paper - Harvard Kennedy School

Coal or Gas: The Cost of Cleaner Power in the Midwest


Many of the loudest debates in environmental policy have focused on two issues: climate change and transported air pollution. In attempting to resolve these issues, considerable attention has focused on coal-fired electricity generating plants in the Midwest. For advocates of a more proactive policy aimed at carbon reductions, these plants offer an attractive target with large potential reductions. For opponents, these proposed emission reductions raise the spectra of higher electricity prices and the potential erosion of the region's economic competitiveness.

The debate, though, has not focused on what it would cost to bring about change in the Midwest. At what cost would Midwest coal-fired electricity generating plants convert from using coal to relying on natural gas? This paper identifies the factors influencing the answer to this question and provides estimates of both their magnitude and the uncertainties surrounding them.

Since the Clean Air Act of 1970, Midwest and Northeast states have argued over how to allocate the economic costs of meeting the nation's air pollution reduction goals. This regional battle has been fought over acid rain, nitrogen oxides (NOx) and submicron particles, and is likely to be fought again over carbon dioxide (CO2).

The paper begins with the premise that the owners of these generators are self-interested and if the cost of burning coal is expected to be greater than the cost of burning other fuels, they will either convert their facilities away from coal or retire them in favor of either a new gas-fired facility or non-fossil-fueled options.

At the moment, Midwest coal plants can be operated at slightly more than 50 percent of the capital and operating cost of a new gas-fired facility. What factors would shift this cost differential? The paper identifies three: a) the cost of meeting the requirements of the Clean Air Act Amendments of 1990 (CAAA) and the new EPA reduction initiatives, such as additional NOx reductions and the proposed particle and ozone standards, b) the relative prices of natural gas and coal delivered to the plant, and c) the cost of carbon emissions, which is presently zero, but could rise if the United States decides to meet future international greenhouse gas reduction targets.

Conventional Air Pollution
Often, studies of the cost of reducing carbon emissions forget that coal-fired power plants must meet major new emissions reduction targets, and these will significantly increase their cost of operation. In fact, a small percentage of the older coal plants are likely to be closed as companies implement emissions cutbacks to meet the new CAAA requirements.

The paper looks at sulfur dioxide (SO2), NOx, suspended particles, and mercury and concludes that requirements to reduce those four pollutants could add between 0.7 cents and 1.36 cents per kWh to the cost of coal-fired generation in the Midwest, with our best guess being a 1 cent per kWh increase. That is, marginal wholesale costs will rise from about 1.6 cents to around 2.6 cents per kWh. This cost is still more competitive than that of a new gas-fired plant, but it substantially narrows the differential between coal and gas. There are many uncertainties surrounding these numbers, but three stand out:

1) Will the US government endorse national emissions trading for NOx? If it does, it could reduce by up to 50 percent the cost of meeting proposed NOx emission standards, depending on the severity of the new reduction requirements. In the short term, a lower standard will result in fewer benefits from trading, while a less stringent standard will result in greater benefit, but this will change as new NOx abatement technologies come on line.
2) How will the EPA decide to meet its proposed new ambient standards for suspended and submicron particles, assuming the standards survive the numerous court challenges? Will the schedule for compliance be as generous as it now seems? Some coal-burning generators could wait until 2014 to make abatement investments, effectively separating particle investment decisions from those for SO2 and NOx.
3) Will scrubber technologies continue to improve, and will the price of scrubbers serve to cap prices in the sulfur allowance market? If it does, sulfur allowances are unlikely to exceed $250 per ton for any length of time.

Natural Gas
Many Midwest utilities recognize that the cost of conventional pollution abatement will drive up the price for wholesale electricity, but argue that continued price volatility in the natural gas market, compared with low and declining wholesale coal prices, creates strong economic incentives to remain on coal.

Forecasting future natural gas prices is an exercise fraught with peril. Studies have shown, however, that these prices are primarily a function of projected resource levels. If one believes the resource is more limited, one will forecast higher prices than if one believes it is more plentiful. Based on a review of existing studies and conversations with experts in the industry, this study is more biased towards the higher resource predictions. As a result, we forecast long-term prices in the $2.50 to $3.00 ($1998) per mcf range.

The authors further conclude that the transmission and distribution investments needed to meet even the high growth scenarios do not dramatically exceed historical levels and could be made without significant increases in wholesale prices. The authors also find that a vast majority of the larger electric generation plants in the Midwest are within 10 miles of an existing major gas pipeline.

If, by the end of this decade, the United States decides to significantly reduce its carbon emissions and switch Midwest capacity from coal to natural gas, what will it cost? The study assumes that if the cost of continuing to operate an existing coal plant is higher than the expected cost of building and operating a new gas-fired facility, electricity generators will opt to change significant percentages of their capacity mix from coal to gas. What cost increases will be needed to realize these conditions? Since investments in conventional pollutants will not erase all of the competitive cost advantages of coal-generated power, especially in a period of volatile and rising gas prices, what additional cost increases might tip the balance?

If reducing carbon emissions is the goal of changing the fuel mix, then the most logical option for governments would be to take actions that would cause the market to assign a price value for carbon emissions. A tradable carbon permit program, a mandatory emission reduction, or a carbon tax would achieve this result.

The study attempts to determine the percentage of existing coal capacity that will become less competitive than new gas-fired capacity at different carbon costs. It looks at several scenarios and projects that more than 60 percent of the region's coal capacity will be uneconomic compared to gas at a carbon penalty between $22 per ton and $142 per ton. This range is huge and represents an upper and lower limit. The study's best estimate is that a $60-$85 per ton carbon penalty will come close to realizing the 60 percent target, increasing Midwest retail electricity prices by about 1.5-2.1 cents to around 10-10.6 cents per kWh.

The study concludes by making several recommendations, including the establishment of a "multi-pollution approach" to emission reductions. Previous efforts to persuade either regulators or emitters to address all the criteria pollutants plus carbon simultaneously have met with political and administrative resistance, but the long-term benefits of such an approach could have measurable economic and environmental benefits.

See .pdf file for full-text.

For more information on this publication: Please contact Environment and Natural Resources
For Academic Citation: Lee, Henry, and Shashi Kant Verma. “Coal or Gas: The Cost of Cleaner Power in the Midwest.” Discussion Paper, 2000-08, Harvard Kennedy School, July 31, 2000.

The Authors

Henry Lee