Comments on the Current Dilemma of the

U.S. Nuclear Power Industry

Anees S. Azzouni
President, Azzouni & Associates, Inc.
Sugandha D. Tuladhar
Vice President, NERA Economic Consulting


The United States currently operates 99 nuclear units with 5 additional units to possibly come online in the next 5 years.  In 2015, the total installed US nuclear capacity was about 100 GW most of which was built during an era of robust electricity demand growth and absent of renewables.  As per the Energy Information Agency (EIA), the US produced about one-third of the global nuclear power in 2015.  The global supply share of the US nuclear generation is expected to decrease significantly by 2040 to 19% from 32% in 2012[1].  One of the reasons for such a drop in US global share is the growth of US nuclear power generation which is projected to stay flat compared to a robust 2.3% global growth over the next 25 years.

Within the US, nuclear power has continued to play an important role in meeting baseload generation and in support of grid stability; however the role of nuclear power that supplied about 20% of total generation is expected to diminish over the next 25 years.  US generation is expected to grow by an annual average rate of 0.7% due to low electricity demand fueled by smaller economic growth; while nuclear generation is expected to decline over time.  The commercialization of new technologies in exploration and production of shale gas paved the way for low natural gas prices outlook with significant impact to the economic viability of nuclear plants.  In addition to sustained lower natural gas prices, the cumulative effect of tax credit policies favoring renewables and emissions reduction regulations that act as a disincentive to nuclear generation are having negative unintended consequences on nuclear generation.  This brief discusses these economic pressures and recommendations to mitigate such unintended consequences to nuclear generation.

Over the last several weeks if not months, a number of nuclear power plants proved unable to clear organized electricity markets’ capacity auctions.  This was followed in turn by the announcement or threat of early retirement of a number of nuclear power plants in the near future.  UBS, Moody's and Fitch identified twelve nuclear power units they considered "at risk" of early retirement due to their inability to compete in the wholesale power markets[2].  On May 6, 2016, Exelon announced that “it will move forward with the early retirements of its Clinton and Quad Cities nuclear facilities if adequate legislation is not passed during the spring Illinois legislative session, scheduled to end on May 31 and if, for Quad Cities, adequate legislation is not passed and the plant does not clear the upcoming PJM capacity auction later this month”[3].

There have been various announcements from 7 different generating companies over the past couple of years to close 14 nuclear units amounting to about 12 GW, primarily due to market conditions, from 2013 through 2025.  Moreover, the aging nuclear units that were built in the 1960s and 1970s will come to retirement age in the next decade or two.  The figure below shows the units that have been announced to be retired (in blue) and projected retirement capacity from 2030 through 2050.  From 2030 till 2040, we could see about 42 GW of nuclear capacity being retired 60% of which is in the regulated market; while from 2040 till 2050, we could see about 50 GW being retired 66% of which is in the regulated market.  In sum, the total amount of retirement till 2050 could be around 100 GW due to either old age or short term announcements not taking into account future policies or regulations that might impact nuclear capacity one way or the other.


While these retirements can to an extent be attributed to US Environmental Protection Agency’s (EPA) regulation to limit power sector greenhouse gas emissions[4], this should be interpreted as indicative of the mounting challenges facing baseload power plants in general and nuclear power plants in particular.  At a hearing on March 25, 2014 of the House Appropriations Subcommittee on Energy and Water Development, DOE Assistant Secretary for Nuclear Energy Pete Lyons attributed the cause for the shutdown to low natural gas prices, flat demand for electricity and, renewable mandates[5]. Notwithstanding the role climate policy plays, the challenges faced by baseload power plants should be attributed in the main to: major developments in combustion gas technologies, an extended period of low natural gas prices, increasing share of renewables in the generation mix, continuing low growth in electricity demand, significant growth in distributed generation, significant progress in energy efficiency, and the entry of organized markets.

Advances in combustion gas technology (CGT) over the past twenty five years coupled with availability of additional natural gas resources at affordable rates is probably the most disruptive event to have impacted power generation in recent times.  Advances in CGT and heat recovery steam generation ushered in the combined cycle power plant with thermal efficiencies approaching 60% to achieve the lowest heat rate and lowest emission levels of all utility scale fossil fuel fired power generation.  The same technology also ushered in industrial cogeneration with thermal efficiencies approaching 90% in the provision of both power and process steam allowing captive power generation to be competitive with utility provided power.  CGT additional advantage has been in its ability to respond to change in power plant load requirements and the speed at which response can be made.  This distinguishing characteristic made gas fired power generation the efficient and competitive choice in meeting variations in power plant load profiles, peaking power requirements, and ancillary services critical in meeting the changes introduced with the entry of organized markets, renewables, distributed generation, and energy efficiency.  For further reading on CGT and its impact on the power industry and the electricity market please see the reference quoted below[6].

Since 2008, the US has seen remarkable increase in natural gas production.   Optimism about shale gas resources and accelerated technological advancement in recovery has created a shale gas boom that propelled the oil and gas industry leading to lower wellhead prices of natural gas and higher level of production.  Shale gas is projected to account for more than 50% of the total US natural gas production by 2040.  Production of natural gas is expected to increase from 28.0 quadrillion Btu (quads) in 2015 to 42.4 quads in 2040.  The Henry Hub natural gas price is projected to stay within $5 per million Btu (MMBtu) over the next 25 years[7].  The steady increase in natural gas production over time and slower growth in key global economies will pave a way for an extended period of sustained low natural gas prices, see figure below.


A centerpiece of the US Administration’s regulation to limit greenhouse gas emissions is EPA’s Clean Power Plan (CPP).  In its final CPP, issued in August 2015, EPA establishes carbon dioxide emission performance rates for fossil fired electric steam generating units (generally, coal fired power plants) and for existing natural gas combined cycle units.  The CPP sets interim (2022-2029) and final (2030) statewide goals in the following forms: (i) Mass-based state goals measured in total short tons of CO2; (ii) Mass-based state goals with a new source complement measured in total short tons of CO2; and (iii) Rate-based state goals measured in pounds of CO2 per megawatt hour (lb/MWh).  The CPP provided distinct preference for carbon free renewables while other carbon-free source, nuclear power, was not provided the same preference. The CPP recognizes that nuclear plants under construction[8] should not be part of the goal setting calculation but recognized for compliance once online.  In addition, power uprates will also be recognized toward compliance[9].  However, as per NEI, the “best system of emission reduction” in the final rule does not incorporate the carbon abatement value of existing nuclear power plants or credit for nuclear plant license extensions[10].

Concern with climate change and carbon emissions in the United States have translated in renewable mandates with significant tax credit incentives and other monetary and investment incentives in support of mainly solar and wind resources.  Renewable mandates have translated into increased demand for renewable power placing renewable power high on the dispatching merit order and operating at zero or low marginal cost.  This trend is expected to continue.  EIA expects that for the period 2015 to 2017 utility scale PV solar power capacity will increase by 13 GW with utility scale solar power averaging 1.1% of total US electricity generation in 2017[11].  The EIA also expects wind power installed capacity to grow by 12% in 2015, 10% in 2016, and 11% in 2017 with utility wind power averaging 6% of total US electricity generation in 2017[12].  In 2015, renewable energy sources overall accounted for 10% of total US energy consumption and 13% of electricity generation capacity[13].  Over the period 2015 – 2040, the EIA expects electricity generation from both solar and wind to grow at 6 to 7% annually[14].

Two issues particular to renewable energy are of direct impact to the economics and demand on baseload generation and nuclear power plants in particular: intermittency and the competitive price of renewable power.  Renewable power mandates provide wind and solar power a competitive edge in the electricity market.  They operate at low or zero marginal cost securing them a high position on the dispatching merit order in a day-ahead competitive market.  Intermittency of solar and wind resources in addition impacts baseload generation through the later obligation to meet the residual power not met by solar and wind resources.  Current intermittency impact on the grid is understood to be limited due to the smaller share of renewable power in total power generation at present.  This effect, however, is likely to change as renewables approach a greater share of peak demand.  This situation has been registered in Germany and to a lesser extent in California imposing a significant load following role to baseload generation[15].  The cost and terms under which renewable power is made available poses a further significant competitive challenge to baseload generation.  Renewable power PPAs are currently being signed in the $40 - 50/MWh range[16].

Energy efficiency efforts in the United States dating back to the late 1970’s have resulted in significant energy savings both at the end use residential and commercial appliance level as well as industrial applications.  US delivered[17] residential energy consumption grew at an annual rate of 0.3% annually over the period 1980 – 2009[18].  This extremely low growth rate reflects a reduction in residential end use energy intensity of 37% over the period 1980 - 2009.  While this major drop in energy intensity is due to a host of factors including energy pricing and population geographical shift, it is primarily due to the introduction of energy efficient technologies as well as energy programs such as standards, building codes, incentives, and labeling.

It is important to point out that the reduction in energy intensity have occurred despite significant increase over this period in US population and number of households, increase in home size and significant increase in the use of lighting and electronics.  Homes built during the period 2000 – 2009 in comparison to homes built before 2000 consume roughly 2% more energy on average despite the fact that homes built after 2000 are 30% larger on average[19].  This low increase in new homes energy consumption is explained by the significant end use energy efficiency improvement of the period and significant increase in end use energy consumption of both air conditioning as well as electronics and lighting as the following graph shows:


Source: EIA 

A form of distributed generation of interest in the context of this article is combined heat and power (CHP).  A typical CHP application involves the generation of power and the utilization of waste heat to produce heat, steam, or hot water.  The typical fuel source is predominantly natural gas.  A CHP application is typically sized to meet the heat and power requirements of the facility in question with a two way grid connection allowing the import of residual and backup power requirements and the export of excess power.  The facilities CHP serves are primarily industrial and manufacturing but are increasingly institutional, commercial, and residential.  In 2011, total US CHP capacity was 70 GW or 7% of total US generation with 25 GW in the industrial sector, 2 GW in the commercial sector, and 43 % in the electric power sector with an average capacity factor of 57%[20].  Twenty three states recognize CHP as part of Renewable Portfolio Standards or Energy Efficiency Resource Standards with several initiating specific CHP incentive programs[21].  CHP adoption impacts the structure of electricity demand and the power system in a number of ways.  CHP adoption deprives utilities of a major share of industrial power demand.  CHP applications in addition contribute to the rise in demand for backup power and other ancillary services.  All of which in turn impact baseload and nuclear power generators load profiles.

Changes in the demand for electricity in the United States since the early 1980’s reflect the structural change of the US economy during this period and the changes in demand for power brought about by advances in combustion gas technologies and natural gas pricing and availability, entry and impact of energy and climate policies, and the role of market design and organized markets. 


The combined impact of the above changes appears to have had a significant impact on the profile of electricity demand.  In the US and particularly New England, the ratio of annual peak demand to average hourly demand has been increasing since 1993 as the graph below indicates[22].  The higher ratio translates into decreased utilization levels of generators resulting in fewer running hours or plants operating at lower output resulting in lower revenues and operating costs and increasing the importance of capacity market payments to generators.  The higher ratio is of particular significance to the competitiveness of baseload generation and nuclear power and natural gas peaking units. 



Source: EIA 

Baseload generation power plants designed and built over the past thirty years to meet baseload power demand were sized, designed, and operated for high plant capacity factor, maximum availability, and lowest operating cost.  Baseload generation power plants built through the 1990’s and beyond were designed and optimized to capitalize on economies of scale and size.  Such plants are built to meet stable and high demand for their output and to maintain high revenues and efficiency. Exposure to significant fluctuation in power plant demand particularly in the context of organized markets and generation divestiture leads to frequent unit recycling, lower thermal efficiency, and significantly higher operating costs.  Structural changes in the US economy starting in the 1980’s are resulting today in significantly varying demand on baseload plants.  Baseload plants today are required to act as backup to renewable power and meet the residual power requirements of industrial, institutional, commercial, and residential facilities with captive generation, spinning reserves requirements, captive generation backup requirements, reactive power requirements and other ancillary services at low wholesale electricity market rates.

Three aspects of nuclear power plants differentiate them from other baseload power plants: cost, age, and load following.  Nuclear power plants are characterized by high fixed costs, low variable costs, and fuel costs that represent a small fraction of electricity generating cost particularly when compared with fossil fuel fired power plants[23].  Accordingly, significant variation in load has significant negative impact on nuclear power plant operating costs, equipment lifespan, revenues, and profitability.  Nuclear power plants are on average 35 years old[24].  The cost of maintaining aged reactors is on the increase with annual expenditures associated with running a nuclear reactor in the United States averaging $ 36.27/MWh with single unit plants averaging $ 44.14/MWh[25].  These figures are not competitive with the marginal costs of natural gas plants and renewables in a day-ahead competitive markets.  It is important to note though that nuclear power plants are designed with strong capabilities to respond to changes in load requirements[26].  It is quite uncertain though as to whether this load following capability can compensate for the drop in nuclear power plants load factors.  Lower nuclear capacity factor translates directly into significant generation losses and lower revenues impacting plant economic viability.

The combined effect of the afore mentioned variables resulted in turn in: significant change in the structure of demand for electricity, major improvement in the competitive position of gas fired power plants whether combined cycle or peaking units, and the ascendency of both wind and solar to a position of must run on the power plant dispatching merit order.  This in turn have led to lower wholesale electricity prices, deterioration in baseload electricity plants load profiles and utilization rates and increase in their operating costs.  The key public policy and regulatory question to be addressed is as to whether the above mentioned events have resulted in the creation of stranded assets, as to whether stranded assets are the result of market failure(s), and if so as to how the costs should be apportioned among electricity ratepayers, utility shareholders, or customers responsible for change in the load profile of baseload generation.  A further question to be addressed is as to whether the survival of the nuclear power industry is critical for maintaining diversified generating sources and the national interest and necessitates governmental intervention on its behalf.

Evidence provided here and elsewhere does indicate the presence of stranded assets in the case of US nuclear power industry as a result of technological, market design, and public policy and regulatory decisions.  The market failure can be characterized by the probable incomplete accounting of externalities cost and the assignment of incentives in combating climate change by public policy and regulatory decisions’.  Both nuclear plants in the regulated market as well as divested nuclear generation assets are subject to market failure.    A hypothesis that can be posed the examination of which may add credence to the public policy question to be addressed as to whether nuclear power plants can still provide a counterweight to intermittent renewable generation, a safeguard against spikes in fossil fuel prices, and a reliable low marginal cost and carbon free source of electric power.  On the other hand, if the nuclear industry is relegated the role of residual power and ancillary services provider, the nuclear industry has a lesser chance to survive given present nuclear plant technology, unit size, regulatory constraints, and mode of operation. 


[1] In 2012, the U.S. produced 769 billion kWh of nuclear power from 103 GW installed capacity; while global nuclear production was 2345 billion kWh from 373 GW installed capacity.  In 2040, the US and global nuclear generation is projected to be 833 billion kWh and 4501 billion kWh respectively, EIA IEO 2016.
[2] Utility Dive, October 19, 2015
[3] Exelon Statement on Early Retirement of Clinton and Quad Cities Nuclear Facilities, May 2016, http://www.exeloncorp.com/newsroom/exelon-statement-on-early-retirement-of-clinton-and-quad-cities-nuclear-facilities
[4] EPA finalized the Clean Power Plan (CPP) Rule to limit carbon pollution from existing power plants. The CPP does not give credit to exiting nuclear capacity.  The Rule was stayed pending further judicial review by the Supreme Court in February of 2016 (EPA).
[5] House Appropriations Subcommittee on Energy and Water Development hearing, March 25, 2014
[7] EIA AEO 2016
[8] Under construction nuclear units are Watts Bar 2 in Tennessee, Vogtle 3 and 4 in Georgia and Summer 2 and 3 in South Carolina.
[9] NEI 2015
[11] EIA, Electricity and Heat Generation from Renewables, July 12, 2016, http://www.eia.gov/forecasts/steo/report/renew_co2.cfm
[12] Same source as Footnote 5
[15] Lokhov, A., “Load-following with Nuclear Power Plants”, NEA updates, NEA News 2011 – No. 29.2
[16] Bloomberg NEF, “U.S. PPA Market Outlook.” 07/08/15. GTM/SEIA, “US SMI Q1 2015.”
[17] Delivered energy excludes energy consumed in the conversion of primary to secondary energy sources.
[18] U.S. Energy Information Administration, February 18, 2015, www.eia.gov/todayinenergy/detail.cfm?id=20031
[19] U.S. Energy Information Administration, February 12, 2013, www.eia.gov/todayinenergy/detail.cfm?id=9951
[20] U.S. Energy Information Administration, October 4, 2012, www.eia.gov/todayinenergy/detail.cfm?id=8250
[21] U.S. Energy Information Administration, October 4, 2012, www.eia.gov/todayinenergy/detail.cfm?id=8250
[22] U.S. Energy Information Administration, February 18, 2015, www.eia.gov/todayinenergy/detail.cfm?id=15051
[24] EIA, How old are U.S. nuclear power plants, and when was the last one built?, February 4, 2016, https://www.eia.gov/tools/faqs/faq.cfm?id=228&t=21
[26] NEA (2011), Technical and Economic Aspects of Load-following with Nuclear Power Plants, OECD/NEA online report available at: www.oecd-nea.org/ndd/reports/2011/load-following-npp.pdf
Click to view a printable version of this article.