Cost-Benefit Analysis of Coal to Gas Electrical Facility Conversion
(NOTE: This analysis was a final conducted for a cost-benefit analysis class that emphasized non-market valuation. The following is a hypothetical analysis, not intended to reflect any specific site in particular.)
Energy markets in the United States are transforming. Coal fired electricity generation facilities are facing increased regulatory scrutiny from local, state, and federal governments. Additionally, the fracking boom in the US has caused a drop in natural gas prices, making natural gas cost competitive than coal for electricity generation. As a result, the percentage of electrical generation in the US from coal has fallen from 50% in 2008 to 32% in 2016 . To underscore that transformation in the US electricity market, this analysis will examine whether a utility should replace an aging coal plant with a new natural gas plant.
This analysis will take place in two parts: A utility level analysis and a social analysis. This is being done for two reasons. Firstly, this analysis is an attempt to capture a microchasm of the changes sweeping America’s energy markets, and every electrical facility replacement is fundamentally a business decision by utilities. Only one state in the US has banned coal fired power plant construction (Oregon), and no state is mandating that currently existing coal fired power plants in their jurisdiction shut down. The mass conversion of coal to gas in the US has been the result of business decisions. Secondly, it is important to highlight the disconnect between externalized social costs and business costs, to emphasize the importance of policy in correcting market failures that result in negative costs on society.
The proposal is a county regulation to mandate a coal fired power plant to transition away from coal immediately. The size of the facility is 140 megawatts, the standard size of a coal fired power plant that has been retired over the past decade, and the age of the coal facility is 40 years old, the average age of an existing coal fired power plant. The analysis will cover a 50 year period, mirroring the long term planning of utilities and to mirror the lifespan of a brand new electrical generation facility.
To consider option value in this analysis, three courses of action will be analyzed.
1) Keeping the existing coal facility open for eight years (48 years is the average retirement age for a coal plant) before replacing it with a new coal power plant. The 140 MW reflects the average size of a coal fired power plant being retired in the past 10 years, according the the EIA .
2) Immediately replacing the existing coal fired power plant with a new combined cycle natural gas facility of the exact same size.
3) Replacing the existing coal fired power plant with a wind energy generation facility of the exact same size.
Nuclear was not considered as part of the option value analysis because new nuclear power facilities have not been constructed in the US since 1979, making construction of a new facility to offset a relatively small coal plant extremely impractical. Additionally, no commercial nuclear power plant is 140 MW. The smallest nuclear power plant is 505 MW, meaning that there’s no basis to estimate construction costs of a nuclear power plant of such a small size.
Such a targeted land use regulation only targeting one electric facility in the county would certainly result in a takings lawsuit against the county for stripping their ability to generate electricity coal without compensation. That would require the county issuing the regulation to pay the utility for the cost of replacing their facility, a cost a local government would likely be incapable of paying. That reality is not reflected in this analysis, as this analysis is a measurement of the social and business costs of a hypothetical conversion.
Job creation will not be factored into this analysis, as the locations of the jobs created should the coal facility be replaced by a gas or wind power facility may be outside of the county, as the best wind sites are often not where coal plants are located. While a socially conscious government may note externalized health and social costs outside of their jurisdiction, governments typically are not concerned with job creation occurring outside their jurisdiction.
The primary source of data collection was from the US Energy Information Administration, a government agency that provides data on energy markets in the United States, for usage by academics and the energy industry. It is comprehensive, covering historical coal, electricity, and natural gas prices as well as future projections. It is not as comprehensive at providing emissions data, which was found from the State of Oregon Department of Environmental Quality Website. Other data used in this analysis was compiled from the Department of Energy and Environmental Protection Agency websites, and academic articles or news sources.
There is no consensus on what the social cost of pollution is. An EPA analysis came up with four different prices for carbon. A Stanford analysis placed the price of carbon as high as $220 per ton of CO2 emitted. The DOE’s analysis of shadow pricing of the health impacts of NOx, SO2, and PM 10 was given as a range instead of as a concrete number. Carbon in particular is difficult to measure because the predicted impacts of climate warming and the amount of warming per unit off carbon are dependent on projections and models. Even pollutants with measurable health effects, such as PM 10 have varying costs because the pollutant’s impact is greater or smaller depending on the population density of the surrounding area. For the sake of the analysis, a midpoint in the pricing of carbon, $35 per ton was chosen.
To estimate the pollution output of a 140 MW coal power plant, Oregon’s Boardman coal plant’s annual emission reporting to the Oregon DEQ was examined, and the emissions were adjusted from Boardman’s 518 MW output to the hypothetical 140 MW coal plant in this CBA:
Through that data, it was estimated that the annual emission of a 140 MW coal plant would be:
SO2: 2,028.7 tons
NOX: 881.5 tons
PM10: 174.19 tons
CO2: 1,839,603 tons
Those quantities were multiplied by the respective costs per ton shown in Figure 2 to calculate the social cost of emissions for the plant.
Natural Gas facilities do not emit SO2, NOx, or PM10. Carbon emissions from natural gas electrical generation is roughly half of an equivalent coal facility, making annual natural gas emissions roughly:
CO2: 919,801.5 tons
Of course, wind generation does not result in any direct emissions. However, as will be discussed below, that does not mean that wind turbines do not result in social costs.
Life Cycle Analysis
The emissions from plant operation are not the only social costs of fuel. With all fuel sources, impacts are felt from extraction to disposal, and to measure the complete social cost of transitioning from one fuel to another, they must be considered.
Coal mining, particularly if it done through mountaintop removal, has a staggering local health impact. Roughly 45% of coal mined in Appalachian region of the US (which makes up 20% of US coal production) is derived from mountaintop removal. Other forms of mining, such as strip mining in the Powder River Basin in Wyoming, have a significantly lower impact. It is difficult to account completely for mining analysis, as the sourcing of coal at a facility can be mixed, different extraction techniques are used to extract coal depending on region, the different extraction techniques have different impacts, and the population exposed to the impacts of mining varies.
A rough handicapping was made to estimate the social cost of coal mining. A widely cited study by Harvard researchers in 2011 measuring the health and social costs of coal mining claimed that coal mining as a whole costs communities $140 billion a year. That number was adapted and scaled down from total US electricity generation from coal ($430 GW) to 140 MW, the size of the coal plant. That was calculated at $45,581.39 for the facility annually.
Measuring the life cycle impact of wind was considerably easier, as the primary life cycle impact is manufacturing and transportation or offsets by nonrenewable sources when it the wind is not blowing, all of which are very easily accounted for. The other social impact of wind turbines is the number of birds they kill.
A British analysis of the life cycle costs of wind power approximated 460 grams of CO2 generated per kilowatt of wind, which qualifies as a social impact of wind. Bird deaths from wind power were not accounted for because bird mortality per MWH for wind is lower than with coal and gas. There were no unique bird killing properties for wind turbines.
The primary life cycle impact of natural gas is the social cost of fracking. Research on the impacts of fracking, from groundwater pollution to air pollution near wells to possible increases in earthquakes, to the destruction in forest ecosystem services to possible increases in premature births near well is relatively sparse, with only qualitative analysis performed on the impacts of fracking, with no entity yet placing a value on the impacts. Some studies by environmental groups even noted that STD transmission rates rose in fracking boomtowns. Without any concrete numbers, the estimated cost of fracking was set at $20,000 a well. Although that number is rough, the studies showed that fracking has a significant health, property value and environmental service impact that is very clearly in the billions, as was the consensus from multiple studies. Given that there are roughly 1.7 million fracked wells in the US, fracking costs the US $34,000,000,000 annually. Dividing the total US gas production from fracking (64 quadrillion btu’s) by the amount of gas burned in the 140 MW plant (29 billion BTU’s), and dividing the 34 billion by that, fracking had a social cost of roughly $15,808 annually for the plant.
The costs of building and operating a new electrical generation facility, including the opportunity cost of the capital used is accounted for through data provided by the EIA in Figure 3. By extrapolating the data, it can be estimated that the capital costs of building a 140 MW facility are as follows:
Natural Gas: $136,920,000
The non-fuel operating costs were as follows:
Natural Gas: $1,540,000
The 2015 fuel costs (determined from the variable operation and management costs) were as follows:
Natural Gas: $490,000
Those numbers set a baseline of costs for the analysis.
Notably, both natural gas and wind had lower operating and capital costs than coal fired power plants. At current prices, coal is still more expensive than natural gas. The only advantage coal had above gas was a being a possibly cheaper fuel source depending on the fuel price projections in the future, as will be discussed below.
To account for natural gas uncertainty, the EIA’s long-term energy outlook was used to provide a range of possible natural gas price outcomes (Figure 4.). It included both high and low estimates for prices of coal and gas. The high price prediction for the next several decades was a tripling in gas prices. The low price prediction would predict gas prices remaining roughly the same. The reference was used for the primary analysis, which estimated gas prices roughly doubling in 30 years, a percentage increase of roughly 2.5%. The range of predictions for coal in the EIA long term analysis were considerably more stable, with projected falling demand causing falling supply. At best, the price of coal will rise at the rate of inflation: 2.5%.
The information from the DOE regarding capital costs of construction and cost of operation of the various electricity sources appeared to show that wind was competitive with coal and gas electricity. However, that information did not account for the intermittency in wind. To ensure that wind power was indeed competitive, the levelized cost of electricity was compared between the various sources. Levelized cost is an analysis of the break-even price of electricity necessary after factoring in all cost variables, from waste disposal to intermittency to fuel costs to parasitic loads to decommissioning. The levelized cost, shown in Figure 5, confirmed that wind is competitive with natural gas and coal on energy markets after accounting for all costs.
Although Figure 3 already provided maintenance costs for the different energy options, those numbers reflected the average for facilities in 2015 alone. Maintenance costs do not rise perfectly in step with inflation, and different energy generation sources have maintenance costs rise at different rates. The table in Figure 6 showed the rise in maintenance and operation costs in coal, wind, and gas (wind qualified as “small scale” in the analysis) facilities over time. allowing a proper maintenance cost inflation rate to be set for each energy source to calculate the rise in costs and facilities age. They were set at the following rates:
Natural Gas: 0.7%
The maintenance costs at coal fired power plants rose faster than inflation, while natural gas and wind power did not.
Assumptions needed to be made in this analysis. The inflation rate was assumed to be 2.5%, reflecting the Fed’s continued policy of maintaining inflation slightly above 2%, and reflective of the average inflation rate over the past twenty years.
The discount rate of the project was set at what the International Association for Energy Economics uses for projects, which is 5.75%, and tied to the rate of return on equities, as the risk of investing in regulated utilities is extremely low . European nations choose an extremely similar rate when discounting energy projects, choosing 5.7% as their discount rate. 
The business and social externalities were factored into a spreadsheet, a variation of which is shown below
Given all the uncertainty around future natural gas prices and social costs, four analyses were performed for each energy source. A “most likely” analysis, a “high social cost” analysis, a “low social cost” analysis, and a “no social cost” analysis, to properly account for the range of analysis depending on possible costs.
Most Likely Analysis
The most likely analysis used the data calculated above in the analysis, and served as the principal analysis as to whether the regulation would be beneficial or not. The other models are sensitivity analysis.
Under this model, the total cost (social + business) of the following sources over the 50 year period would be:
Natural Gas: $994,569,473
High Social Cost Analysis
To maximize costs, the social discount rate was set at 3.5%, the standard social discount rate recommended in the textbook, the a higher estimate as to the social cost of carbon ($90) was provided, the high estimates of the social cost of pollution provided by the DOE was provided, and the life cycle social costs were doubled.
Under this model, the total cost of the following sources over the 50 year period would be:
Natural Gas: $3,454,111,504
Low Social Cost Analysis
The discount rate was set at 8%, the price of carbon was set at $6 per ton (what Pruitt is currently planning to make EPA’s social cost of carbon be). Other social costs were halved.
Natural Gas: $265,239,572.94
No Social Cost Analysis
Only business costs were considered in this analysis, and this was primarily performed to highlight whether a transition would take place even without the consideration of social costs, to show what would happen in the free market without any mandates or regulation. All social costs were set to zero. The only costs considered were those in the business analysis.
Under this model, the total cost of the following sources over the 50 year period would be:
Natural Gas: $165,489,016.29
This highlighted that without valuation of social costs, the cost of wind and gas is approximately the same (gas having the advantage because of its lack of intermittency), and coal still remains uncompetitive in a 50 year analysis. This shows that without valuing social costs, natural gas is likely to be chosen over wind if coal is disallowed.
Regarding Natural Gas Prices
Under the worst case scenario for natural gas relative to coal, in which natural gas prices rise 4% per year, and social costs are not considered, and the discount rate is high, the gas and coal prices are this:
Coal’s high capital costs, rising maintenance costs, and high fuel costs prevent it from remaining competitive with gas, even under extreme circumstances.
Coal isn’t competitive. As a fuel, more expensive than gas, it has higher maintenance costs, and the capital cost of constructing a coal facility is higher than constructing an equivalent gas or wind facility. Even under a high gas price scenario, it still isn’t competitive. It makes business sense for the utility to transition immediately to natural gas or wind, and that has been reflected in the rapid change in America’s energy markets noted in the opening of this analysis. Sensitivity analysis showed that increased pricing of social costs made wind far cheaper than the alternatives, while also demonstrating that wind has nearly identical pricing to natural gas if social costs were not considered.
From a non-business sense, the regulation would be massively beneficial. Regardless of whether wind or gas is used by the utility as the alternative to coal, the social benefits are massive, and scaling depending on how steeply the social costs of pollution are priced. The regulation should be implemented, as the social costs of coal are so high that implementing a regulation to speed up the transition away from coal, even if to only gas, is a massive benefit. While not all the social benefits of the rule change would be felt by those within the county implementing the regulation, humanity would benefit. Additionally, this analysis showed that pricing carbon rather than simply mandating a transition away from coal would incentivize the direct conversion of coal to wind, instead of natural gas.
United States. United States Department of Energy. Energy Information Administration. Annual Energy Outlook 2017. Vol. Jan. Series 2017. https://www.eia.gov/outlooks/aeo/pdf/0383(2017).pdf
“Average Power Plant Operating Expenses for Major U.S. Investor-Owned Electric Utilities, 2006 through 2016.” Accessed June 6, 2018. https://www.eia.gov/electricity/annual/html/epa_08_04.html.
White, Maeve. “Effect of Wind Turbines on Bird Mortality.” Stanford.edu. November 24, 2016. Accessed June 11, 2018. http://large.stanford.edu/courses/2016/ph240/white1/
Rudolf, John Collins. “Tallying Coal’s Hidden Cost.” The New York Times. February 17, 2011. Accessed June 11, 2018. https://green.blogs.nytimes.com/2011/02/17/tallying-coals-hidden-cost/?partner=rss&emc=rss.
Jackson, Robert B., Avner Vengosh, and Richard Davies. “The Environmental Costs and Benefits of Fracking.” Annualreviews.org. 2014. Accessed June 11, 2018. https://www.annualreviews.org/doi/full/10.1146/annurev-environ-031113-144051.
Dutzik, Tony, and Elizabeth Ridlington. “The Costs of Fracking: The Price Tag of Dirty Drilling’s Environmental Damage.” EnvironmentAmerica.org. October 2012. Accessed June 5, 2018. https://environmentamerica.org/sites/environment/files/reports/The Costs of Fracking vUS.pdf.
“EPA Fact Sheet: SOCIAL COST OF CARBON.” Epa.gov. December 2016. Accessed June 5, 2018. https://www.epa.gov/sites/production/files/2016-12/documents/social_cost_of_carbon_fact_sheet.pdf.
 Khatib, Hisham. “The Discount Rate — A Tool for Managing Risk in Energy Investments.” IAEE.org. Accessed June 6, 2018. https://www.iaee.org/en/publications/newsletterdl.aspx?id=304.
 Hermelink, Andreas H., and David E. Jager. “Evaluating Our Future: The Crucial Role of Discount Rates in European Commission Energy System Modelling.” Eceee.org. Accessed July 5, 2018. https://www.eceee.org/static/media/uploads/site-2/policy-areas/discount-rates/evaluating-our-future-report.pdf.
“27 Gigawatts of Coal-fired Capacity to Retire over next Five Years.” Eia.gov. July 27, 2012. Accessed June 11, 2018. https://www.eia.gov/todayinenergy/detail.php?id=7290.
“Capital Cost Estimates for Utility Scale Electricity Generating Plants.” Eia.gov. November 2016. Accessed June 5, 2018. https://www.eia.gov/analysis/studies/powerplants/capitalcost/pdf/capcost_assumption.pdf.