The climate for change3 June 2003
EPRI has assessed the potential use, on a competitive economic basis, of next-generation advanced nuclear reactors under a range of possible regulatory and market conditions. By Ted Marston and Chris Wood
In 2000, the Electric Power Research Institute (EPRI) published the Energy-Environment Policy Integration and Coordination (E-EPIC) study, which investigated the potential effects of future regulation of emissions of sulfur dioxide (SO2), nitrogen oxides (NOx) and carbon dioxide (CO2) on the US electricity generating system through 2050. The E-EPIC study found that these effects could be very significant, and that their timing could lead to inefficient and disruptive swings in fuel use, and limit the productivity of some investments in generating assets.
The E-EPIC study used the National Energy Modeling System (NEMS) developed and employed by the US Department of Energy's Energy Information Administration (EIA). One aspect of the E-EPIC study that was particularly affected by the decision to maintain EIA assumptions was the treatment of nuclear power. EIA has employed conservative assumptions regarding the cost and performance of nuclear energy technologies. These include cost assumptions for new plants based on older designs and construction experience (prior to the advent of standardised and pre-certified advanced reactor designs and an improved licensing process), as well as those regarding the performance of current plants and the percentage of current plants that would obtain a renewal of their operating licence for an additional 20 years. As a consequence of using the EIA assumptions, the E-EPIC study did not project any role for new nuclear generating plants in the electric generating mix over the time period addressed by the study and underestimated the role of the current fleet of operating reactors.
EPRI has recently re-examined the E-EPIC findings and conclusions related to future nuclear generation in the report Nuclear Power's Role in Meeting Environmental Requirements Evaluation of E-EPIC Economic Analyses. This study also uses NEMS as its primary analytical tool, but it departs from EIA's assumptions regarding nuclear generation available for deployment in the near future, relying instead on EPRI assessments of the most likely cost and performance characteristics for current and future generating technologies.
The primary aim of this study is to examine the competitive prospects for the next generation of advanced nuclear power generating technology, to measure the potential market penetration of new nuclear plants in the next three to four decades, and to determine the effects on the US energy system of adding new nuclear generating capacity. Specific cost estimates and market conditions have been updated based on a recent internal evaluation of the projected cost and performance characteristics of the next generation of nuclear generator design. Estimated costs are based primarily on the Westinghouse AP1000, but these costs are expected to be consistent with the assumed costs for other competing new advanced nuclear generating plants.
As discussed below, this screening study indicates that new advanced nuclear generation technology could achieve significant market penetration under a variety of possible future conditions. Besides examining the prospects for nuclear plant market penetration against competing electricity generation technologies, which are also improving over time, these analyses also provide a preliminary view of the potential role and effects of advanced nuclear generation in our overall energy system and economy.
The right conditions
Considering several major kinds of baseload generating technologies competing for selection in 2010 under the nuclear base case, "snapshot" busbar $/MWh cost profiles show that advanced natural gas-fired combined cycle (NGCC) technology is favoured, followed by advanced nuclear technology and two types of coal-fired technology, with renewable technologies wind and biomass lagging.
More realistically, a comparison of generating technologies would take into account the expected trend in fuel costs over time, such as the rising trend for natural gas prices projected under the base case. Using levelised fuel costs, advanced nuclear technology and 'solid fuel' coal technology (pulverised coal in the 2010 time frame) are seen to be as attractive as advanced NGCC, due to the projected increase in natural gas prices and a very slight drop in projected coal prices (Figure 1).
While wind energy systems (WES) appear to be close to being competitive with NGCC, WES have disadvantages not reflected in the $/MWh cost profile. These disadvantages include the intermittent nature of wind generation, and the limited availability of high quality wind sites in certain regions of the country.
A look further ahead
Looking beyond the base case technology decision in 2010 to consider a technology decision in 2025, the comparison of baseload technologies is altered due to projected advances in the technologies and changes in fuel prices (Figure 2). All of the technologies are projected to experience improvements in capital and operating costs. These improvements benefit the natural gas technology less than the other technologies, due its already lower capital costs.
The $/MWh busbar costs projected for a 2025 technology decision reflect a slight decline in coal prices and flat prices for uranium and biomass fuels, but reflect a significant increase in natural gas prices that more than offsets the improved efficiency for the advanced gas technology (which under the base case is still assumed to be NGCC). By the 2025 time frame wind appears to be more competitive than in 2010, and in fact increased additions of WES are projected, although WES are still hampered by the intermittency of generation plus regionally distributed high quality wind sites. The net effect of these projected changes is that advanced nuclear plants have a slight cost advantage over other technologies in this timeframe.
Base case deployment
Advanced nuclear technology was assumed, for the purposes of this study, to first be available for selection as a baseload generation investment option in year 2005. This means that the earliest possible in-service year is 2009, with 2010 more likely. The attractiveness of this technology under the base case is expected to increase over time due to declining capital costs, growing electricity demand, rising natural gas prices, and projected retirements of existing generators. Under the base case, generating capacity additions are dominated by natural gas-fired technologies (Figure 3).
For displaying the NEMS projected generating technology choices, the study divides the 2005-2035 period into two 15-year intervals: 2005-2020 and 2021-2035. Under the base case, projected advanced nuclear technology deployment starts in 2009, reaching 23.2GWe by 2020 and 135GWe by 2035. In addition, many existing nuclear plants have their lives extended and are still operating during this period. (For this study, all existing nuclear plants are assumed to operate until age 60 years and then be retired.)
It is necessary to look further to better understand why NEMS modelling makes the technology choices it does, how real world choices might confront additional issues, and how advanced nuclear technology might fare under conditions and criteria that are more complex and varied. Some important issues that should be examined include:
• Regional conditions in general.
• Regional generation mix and generation needs in particular.
• Non-baseload generation may be more attractive.
• Alternative technology advances.
• Growth in electricity demand.
• Expectations for fuel prices.
• Expectations for environmental penalties and restrictions.
• Market-based profit and risk versus system cost minimisation.
• Adjusting technology choices based on risk.
• Disadvantages of intermittent generation.
Consideration of the different issue areas summarised above can enhance insights into the prospects and needs for nuclear and other generating technologies. The present study's screening analysis was extended in several ways to examine some of the above issues. These analytical extensions considered how the following conditions might affect mid-term advanced nuclear technology penetration to 2020.
• 10% lower capital costs for advanced nuclear technology, beginning in 2005 when investment in the technology first becomes an option.
• Both lower and higher natural gas prices.
• A moderate carbon tax affecting the overall energy system (also examined out to 2050).
Using the 2010 $/MWh busbar cost profile (Figure 1) as a benchmark, it is straightforward to visualise that 10% lower capital costs would make advanced nuclear technology very attractive as an option for new baseload generating capacity. Three alternative natural gas price cases were developed to reflect varying levels of optimism regarding the cost and efficiency of gas exploration and production, as well as varying levels of optimism regarding the extent of additional reserves to be discovered and brought to market in the future. We can expect that lower natural gas prices (or at least the expectation of lower prices) would increase the already dominant role of natural gas-fired technology additions out to 2020. We also expect that higher gas prices would cause gas-fired technology to lose market share to advanced nuclear technology and coal technology.
Compared to the expected impacts of lower capital costs or varied natural gas prices, it may not be as intuitively clear what impact a moderate carbon tax of $5/tonne in 2011 rising linearly to $50/tonne in 2020 would have on nuclear plant penetration. Insight is provided by revised $/MWh busbar cost profiles for a technology choice in 2010, incorporating the impact of the carbon tax on fuel prices (Figure 4). These profiles indicate that with even this moderate carbon tax, the advanced nuclear technology considered here would have a clear advantage over coal and would also have an advantage over natural gas-fired technology in a baseload role. The relative economics vary regionally.
Nuclear penetration projected through 2020 bears out these expectations. With 10% lower capital costs, advanced nuclear technology becomes more attractive, starting from its first possible in-service year of 2009. When compared to additions under the base case, this lower cost results in substantially greater projected nuclear plant additions in every year through 2020, leading to cumulative deployment of about 62GWe versus about 23GWe under the base case (see Table above).
As expected, the range of natural gas prices examined produced a range of projections for nuclear plant penetration. Even under the low gas price case, the advanced nuclear technology is projected to achieve some penetration, with the first additions occurring towards the end of the 2010-2020 decade, reaching 7GWe by 2020. Under the high gas price trajectory, advanced nuclear additions essentially parallel the base case, but at a slightly higher rate, reaching 36GWe (as opposed to 23GWe) by 2020. With the highest gas prices, nuclear technology becomes much more attractive and gains a substantial share of the market for baseload additions earlier, reaching 20GWe of deployment by the middle of the 2010-2020 decade, and 72GWe by 2020.
Ultimately, carbon taxes (even of the moderate magnitude considered in this study) have the greatest impact, favouring additions of nuclear technology. Starting with the first possible deployment year of 2009, additions under the moderate carbon tax case are higher than under the other cases, and accelerate at a faster rate as the rising carbon tax and growing energy demand increase the value of low- and no-carbon energy sources, so that cumulative deployment reaches 108GWe by 2020.
Beyond advanced nuclear technology's potential for market penetration, it is important to consider its possible role and impact within the overall electric supply system. The study considers what additions, retirements, and use of different generating technologies are projected to occur under the base case versus under identical conditions but with advanced nuclear technology excluded. These analyses show that for the 2005-2020 period, total additions of non-nuclear technology are only slightly affected.
However, during the 2021-2035 interval, nuclear technology is projected to be very competitive and to achieve substantial market penetration, with 112GWe more added. This would more than replace those
pre-2000 nuclear generators that are projected to retire in this time period. In this period, rising natural gas prices make advanced coal technology competitive with and even more attractive than natural gas for baseload generation (see Figure 2). Thus, the major impact of the substantial deployment of advanced nuclear technology in this period is to greatly reduce the market opportunities for the next generation of coal technology.
Nuclear technology additions are also projected to impact the US electric supply system by affecting retirements of the least efficient fossil generators. This impact occurs in the later 2021-2035 period, when nuclear deployment is projected to have become considerable. However, the effect is perhaps counter-intuitive. The impact of nuclear deployment is actually to reduce the rate of projected retirements, especially for older gas-fired steam generators (some having limited oil firing). This can result because nuclear technology additions reduce the additions of coal-fired generators and gas-fired combined cycle (or successor technology) generators through 2035.
Several potential economic benefits of expanding nuclear power's 20% share of US electricity generation are reflected in the projected prices of electricity and other fuels. These prices will affect the costs of energy to households, industry and commercial energy consumers. In turn, they affect gross domestic product (GDP) and other measures of economic and energy efficiency.
Gas and electricity prices
Building new nuclear power plants will change the demand for fossil fuels, including natural gas. Lower natural gas wellhead prices contribute directly to lower electricity prices.
Changes in electricity and fossil fuel costs will create macroeconomic costs or benefits that can be measured
by calculating differences in the
consumer and producer surpluses between scenarios.
The addition of new nuclear power plants could lead to a net present value saving of $11.6 billion (1999 dollars). The net present value of saving in the electricity sector would be about $12.5 billion. These savings would be partially offset by a loss in producer plus consumer surplus in the fossil fuel sector of about $0.9 billion, owing to lower consumption and prices for fossil fuels.
In its Clear Skies Initiative, the Bush administration proposed that emission reduction targets should be expressed in emissions per dollar of GDP. The US GDP has increased significantly during the last decade and is projected to continue to grow in the future. Hence, on a percentage basis, emissions per dollar of GDP will fall more rapidly than emissions by themselves.
Between 2020 and 2030 a $50/tonne carbon tax is projected on average to achieve about a 2.5% per year reduction in carbon emissions per dollar of GDP. This can be compared to projected average reductions of 1.6% and 1.9% per year per dollar of GDP under the no new nuclear plants and nuclear base case scenarios.
Energy price volatility
Both electricity and fossil fuels have exhibited unprecedented price volatility in the last several years. High natural gas prices during 2000 and 2001 have demonstrated that the US gas supply and demand balance can change dramatically from month to month and year to year. Future periods of disequilibria in North American natural gas markets and resulting high price volatility are likely to recur.
Even though the frequency of such events is very difficult to predict, and even though price volatility is not easily modelled in equilibrium models such as NEMS, the historical price volatility experienced in US natural gas markets has been examined. A relatively diverse mix of different generation resources with production costs and prices that are not highly correlated with one another will tend to reduce the combined volatility of the overall portfolio of generating resources. To that end, the availability of new nuclear power plants would reduce dependence on fossil fuels and, potentially, contribute to lower natural gas and electricity price volatilities in the future.
It all adds up
When EPRI data for generation option economics is used in this analysis, including a carbon tax, then substantial additions of nuclear capacity would appear in the generation mix (Figure 5).
This has a major impact on total CO2 emissions from the US electric generators, which rise to 142% of the 1990 level by 2010, but then return to the 1990 level by 2030 and drop to 75% by 2050. Figure 6 illustrates carbon emissions from all energy related sources and by sector and fuel from 1990 to 2050 for three of the cases analysed in the report.
Note that allowing realistic assumptions regarding nuclear energy to be used in the analysis mitigates the severe swings towards gas generation, the premature retirement of coal plants, and the resulting unacceptable damage to the US economy resulting from the 1998 Policy Direction, using EIA analyses method and assumptions that effectively exclude nuclear energy.
This follow-on study to the important E-EPIC report published in 2000 does not provide all the answers related to optimising US electricity supply mix and future deployment strategy to simultaneously support our nation's environmental quality, energy security, and economic competitiveness goals. It does bound the problem, by showing that accelerated CO2 emissions controls similar to those specified in Kyoto are probably not acceptable to energy, environmental, and economic policy makers. The study does not answer the question: "What type of controls over what timeline is the
optimum balance among these three critical national goals?" It does show, however, that for one very simple and relatively modest approach to CO2 management (a moderate carbon tax), nuclear energy is an essential element of a workable strategy. It even shows that with no explicit CO2 management requirements at all, nuclear energy is still an essential element of an energy policy that requires substantial progress on other environmental emission control goals in a manner that does not cause significant harm to the economy.
EPRI is initiating a study to better examine various emissions control proposals, including ones that address CO2. It will focus on the role of non-emitting generation (renewable and nuclear energy) and the role they can play under a variety of market-based incentive arrangements.