Select Committee on Environmental Audit Written Evidence

Memorandum submitted by Karen Cousins


  The United Kingdom's recent Energy White Paper establishes the goal of reducing carbon dioxide emissions by 60% of 1990 levels by 2050. A limited number of realistic options exist to reduce carbon emissions from the electricity sector, including efficiency improvements, increased use of renewable energy sources, carbon sequestration plants, and nuclear power. This paper investigates how UK's economic and energy growth projections can be reconciled with the government's plans for carbon abatement, given that renewable energy has practicable limitations and feasible electricity efficiency improvements are constrained below 2%/year. Under the relatively plausible assumptions, it is found that the UK will be unable to meet electricity demand in 2050 without a large scale program of nuclear energy or carbon sequestration.


  Karen Cousins is a graduate student at the University of Oxford. Her thesis involved an in-depth analysis of energy policy and the energy market in the United Kingdom, and focused on reconciling the Government's carbon abatement ambitions with the country's energy needs. Particular attention was given to the potential and costs of nuclear power.


  1.  Electricity generation from fossil fuel energy sources is a major contributor to global emissions of carbon dioxide, a greenhouse gas that contributes significantly to climate change (Ansolabehere et al, 2003). Fossil fuel energy sources have formed the basis of the world's economy since industrialisation (IPCC, 2001a) and reducing carbon emissions from electricity generation is a significant challenge. A limited number of realistic options exist to reduce carbon emissions from the electricity sector, including:

    —  improved efficiency in electricity generation and use;

    —  increased proliferation of renewable energy sources such as wind, solar, biomass, and geothermal;

    —  carbon capture and sequestration at fossil-fuelled (especially coal) electricity generating plants; and

    —  increased use of nuclear power (Ansolabehere et al, 2003).

  2.  The threat of climate change is recognised by the United Kingdom's recent Energy White Paper, "Our Energy Future—Creating a Low Carbon Economy", (DTI, 2003a), which establishes four key goals for energy policy, one of which is to "place the UK on a path to reduce carbon dioxide emissions by 60% of 1990 levels by about 2050". The aspiration of achieving a 60% reduction in carbon dioxide emissions is an ambitious one, particularly given that at present over 70% of the UK's electricity is generated from fossil fuels.

  3.  Moreover, electricity consumption in the UK is steadily increasing. Although the switch from coal to gas has prevented significant growth in carbon dioxide emissions, from 2010 onward, as nuclear power stations begin to close, dependence on fossil fuel sources of electricity is projected to increase and carbon dioxide emissions are forecast to climb (DETR, 2001).

  4.  Meeting the carbon dioxide emission reduction target will require significant cuts in carbon emissions from electricity generation. Electricity generation is responsible for 38.1% of emissions in the UK, as shown in figure 1.3. Emission cuts from electricity generation are likely to require both energy efficiency improvements and a significant shift to carbon-free sources of electricity.

Figure 1.3: UK 2003 emission generation, by sector (%) Figure 1.4: UK 2003 emission generation, by fuel type (%)

Source: DEFRA, historic data (Estimated emissions of carbon dioxide (CO2) by UNECE source category, type of fuel and end user: 1970-2003[37]).

  5.  In spite of this, the Energy White Paper does not set targets for different generating capacities. Rather it sets out a range of policy measures intended to create a market framework that will deliver the stated energy policy goals most effectively. However, under current plans, a gap exists between expected electricity demand and supply if the 2050 carbon abatement target is to be achieved.

  6.  Indeed, the 2020 renewable energy target is insufficient to put the UK on target to meet its 2050 carbon abatement goal (Oxera, 2005a). Furthermore, the Renewables Obligation is underperforming and it is unlikely that the 2020 renewable energy target will be met (Mitchell and Connor, 2004). Additionally, over the same time period, nuclear capacity is forecast to decline from providing 23% of the nation's electricity requirements, to providing just 7%. As a result, the benefits of investment in renewable sources of energy will be offset, in carbon abatement terms, by the decline in nuclear capacity. Hence, in order to meet the 2050 carbon abatement target, an additional program of renewable energy, carbon sequestration, energy efficiency or nuclear build will be necessary (Oxera, 2005a).

  7.  The Energy White Paper considers nuclear power an important source of carbon-free electricity and keeps the nuclear option open. However, it suggests that nuclear power is an unattractive option for new, carbon-free generating capacity due to the current economics and unresolved issues of waste disposal.

  8.  In liberalised electricity markets nuclear power is not presently cost competitive with gas or coal as private investors are deterred by the high capital requirements and the long lead times of nuclear generation. However carbon emission credits give nuclear power a cost advantage over coal and gas under reasonable assumptions of reductions in capital costs, operation and maintenance costs, and construction time (Ansolabehere et al, 2003).

  9.  This paper investigates how UK's economic and energy growth projections can be reconciled with the government's plans for carbon abatement. The 60% target was a key feature of the Energy White Paper of 2003, and although it applies to more than just the electricity sector, the electricity sector might be expected to contribute pro rata.[38] More specifically this paper investigates whether an adherence to both the aspiration of reducing carbon dioxide emissions by 60% by 2050, and to the promotion of competitive energy markets, will force an increasing reliance on nuclear power.

  10.  In addition to investigating the electricity mix to meet the UK's electricity requirements in 2050 in the most cost-effective manner, this paper investigates how inter-generational equity concerns would alter the electricity mix. Hence, in considering the costs associated with nuclear power, both a time-constant and a time-declining discounting treatment will be applied.

  11.  Before considering the research questions, Section 2 outlines the methodology and presents the scenarios used in the research. Results are presented in Section 4. Section 5 provides a discussion on these results. Section 6 concludes by analysing the implications of the results for energy policy in the UK and presents recommendations for achieving the 2050 carbon abatement target at least cost.


  12.  Results in this paper are derived using a relatively simple model which calculates the expected profitability or subsidy requirements of different electricity mix scenarios. To achieve this, the model first determines the electricity demand and the carbon emissions limit from electricity generation in 2050. A number of sensible electricity mix scenarios, that meet both the electricity demand and the carbon emission targets, are then constructed. The model then determines the present value, in 2005, of each scenario.

2.1  Key assumptions

  13.  The electricity price forecast model, which is at the core of this paper, required a number of assumptions regarding available technology and associated costs. The analysis assumes that in future, six different types of electricity generating technology will be constructed. These are Combined Cycle Gas Turbines (CCGTs), Integrated Gasification Combined Cycle (IGCC) coal power plants, nuclear power plants, offshore wind power plants and CCGT and IGCC coal power plants with carbon capture capabilities.

  14.  This selection reflects the expectation that new coal and gas power plants will use the latest available technology in order to maximise the electricity output/emissions ratio, as well as to achieve economic sustainability. Offshore wind is the only form of renewable energy included in the model as offshore wind it is expected to make the most significant contribution to electricity generation from renewable sources in the UK in the medium to long-term (Gross, 2004).

  15.  The cost structure of all power plants is assumed to consist of the following four cost elements: capital expenditure, operation and maintenance ("O&M"), fuel (except for wind) and the cost of carbon (except for wind and nuclear). For nuclear power plants, there is also a cost for decommissioning after the plant is closed down.

  16.  Capital expenditure, which includes all start-up costs, largely planning and construction, is expected to show the most variation. Variation in capital expenditure with time is shown in figure 2.1:

Figure 2.1


  17.  Nuclear capital expenditure is significantly above the capital costs of all other technologies modelled. Capital costs for nuclear energy are based on estimates by Oxera (2005b). Thus, capital expenditure for the construction of the first nuclear plant is estimated at £1,600/kW, decreasing to £1,200kW from the third plant onward. Additional costs are included as follows: £100 million each for FOAK[39] and public enquiry costs, and an allowance of 10% for over-run costs. Thus, capital expenditure for nuclear energy is estimated at £1,980/kW initially, decreasing to £1,430/kW in the long-term.

  18.  These cost estimates are conservative. The current nuclear reactor in Finland is being built on fixed-price contract of £1,140/kW. Further, the PIU Energy Review (2001) indicates that capital expenditure for current nuclear technology is in the region of £1,400-£1,700/kW and that capital expenditure may decrease to £690/kW in the long-run.

  19.  For offshore wind turbines, the 2004 capital expenditure is set at £1,140/kW. This number is based on Dale et al (2000) and includes £1,000/kW for the plant itself and £140/kW for necessary transmission and distribution ("T&D") investments. Dale forecast the costs of wind turbines to fall to £600/kW by 2020 and to £400/kW in the long-run, with the cost of T&D remaining constant.

  20.  CCGT gas plants are a relatively mature technology, and no major cost savings are expected. The PIU (2001) indicates current capital expenditure is £270/kW. Capital expenditure is expected to decrease to £250/kW in the long-run (PIU, 2001).

  21.  Capital costs for CCGT plants with carbon capture capabilities are assumed to be £690/kW (Tzimas and Peteves, 2005). The assumptions made for the model are that, due to limited technological learning, initial cost savings are of the order of 5% between 2004 and 2020. Post 2020, once the technology is established, it is assumed significant cost savings are possible, and that the costs of the capture plant (but not of the power generating plant) will decline by 50%. Thus the lower limit for CCGT with capture is £460/kW.

  22.  Marsh (2003) estimates current capital expenditure for IGCC coal power plants as £625/kW. The DTI (2001) has indicated that a 20% reduction in capital expenditure by 2020 is feasible. No cost reduction forecasts are available after 2020, but the model assumes a 50% reduction over the 2004 level in the long-term. This is conservative considering the higher expected reduction for other new technologies, ie offshore wind and nuclear.

  23.  For IGCC plants with carbon capture capabilities, Tzimas and Peteves (2005) estimate capital expenditure as £1,310/kW. No forecasts are available, but based on the same assumptions used for gas carbon capture plants, a reduction to £655/kW is considered feasible in the long-run.

  24.  As shown in figure 2.2, the O&M costs are less variable than capital expenditure.

Figure 2.2


  25.  The only O&M costs expected to decrease are those of wind farms (Dale et al, 2004). The anticipated decrease is based on the expectation of larger wind farms and thus higher operating efficiency. All other O&M costs are based on reports by Marsh (2003) for the DTI, and Oxera (2005, for nuclear), reflect the complexity of the plant in question and are not expected to change significantly.

  26.  Fuel costs are assumed to remain constant over time. Fuel prices are affected by many uncertainties and an attempt to quantify these uncertainties is beyond the scope of this paper. The current levels are summarised in figure 2.3:

Figure 2.3

FUEL COSTS, 2004-2050 (p/kWh)

Source: DTI, historic data (Average prices of fuels purchased by the major UK power producers and of gas at UK delivery points[40]); Oxera, 2005b.

  27.  The cost of carbon is a cost factor of increasing relevance for coal and gas powered plants. Forecasts from The Carbon Trust (2004) for the market price of carbon traded under the EU ETS are included when analysis uses market discount rates. This forecast sees the price increase from a current level of

5/tonne of carbon dioxide to

25/tonne of carbon dioxide in 2013. The price is forecast to remain flat post-2013. When analysis is based on the social rate of time preference, Clarkson and Deyes' (2002) estimate of the social cost of carbon is incorporated. Thus a value of £74/tonne of carbon is included to represent the social cost of carbon in 2005. This value is inflated by £1/tC per year.

  28.  As nuclear power plants need decommissioning at the end of their operating life, this cost item has been included in the model. Oxera (2005b) estimates the net present value of decommissioning costs in the final operating year of a nuclear plant as £500/kW. This value has been used for the model.

2.2  Other assumptions

  29.  The economics of the different technologies modelled are influenced by capacity utilisation and carbon emissions, as well as costs.

  30.  Capacity utilisation varies significantly by technology reflecting technical aspects and resource availability. See figure 2.4.

Figure 2.4


Source:Dale et al, 2004; DTI, 2002; Marsh, 2003; MIT, 2003.

  31.  Carbon emissions relative to electrical output vary considerably by technology with nuclear and wind power at an obvious advantage in this regard (see figure 2.5).

Figure 2.5


Source: Marsh, 2003.

2.3  Scenario Construction

  32.  The model considers eight different scenarios grouped into two sets of four. The first group considers an electricity demand base case in which the same efficiency improvement rates that have been achieved over the past decade are forecast to continue. The second group considers a lower electricity demand situation that is as a result of electricity efficiency improvements. The four scenarios in each demand group differ in the way that wind, nuclear and carbon sequestration are used to meet the 2050 supply target without violating the carbon emissions limit.

2.3.1  Electricity demand

  33.  Electricity demand in 2050 is forecast as a function of GDP and electricity intensity. Assumptions relating to GDP growth are in line with forecasts by HM Treasury (2003). Thus it is assumed that GDP will grow at 2.5% for the next five years, before levelling out at a long-term growth rate of 2.25%. These assumptions result in a doubling of GDP over the next 35 years. Economic output is linked to electricity consumption via the Electricity Intensity Ratio[41] (EIR), which has decreased steadily over the past decade, reflecting continuous improvements in electricity efficiency. The average increase in efficiency over the past decade is 1.16% and is used to forecast the base case electricity demand in 2050 of 563TWh in 2050. More optimistic electricity efficiency savings of 2% per year, which are considered achievable but optimistic (Blok, 2004), result in demand of 377TWh in 2050. See figure 2.6:

Figure 2.6


Source: Office of Statistics, historic data (Primary energy consumption, gross domestic product, the energy ratio[42]).

2.3.2  The emissions limit

  34.  It has been assumed that the 60% target will be applied to the economy and its sub-sectors, the electricity sector included, in equal proportion. As such, the 2050 emissions limit for the electricity sector is determined as 22.3MtC/yr as is shown in figure 2.7. All scenarios modelled further assume that coal and gas generate the maximum amount of electricity possible without violating the 2050 emissions limit for the electricity sector.

Figure 2.7


Source: DEFRA, historic data (Estimated emissions of carbon dioxide by UNECE source category, type of fuel and end user: 1970-2003).

  35.   Gas and coal generation. In all scenarios gas and coal are assumed to be employed in a 2:1 ratio. This assumption reflects the superior attractiveness of gas, but also the important role of coal generation in diversifying the electricity mix (DTI, 2003a). Given this ratio, and the relative emissions to output ratios shown in figure 2.5, the maximum output from gas and coal generating plants, in the absence of carbon capture, can be determined as follows:

  36.  Inserting the known values into (3) it follows that coal can supply up to 55TWh in 2050, and given ec, this will result in 11.1 Mt of carbon emissions. It follows from (2) that gas can supply up to 110TWh, which, given eg, also generates 11.1 Mt of carbon emissions.

  37.  Thus, in the absence of carbon capture, gas and coal can supply 166TWh of electricity in 2050, whilst jointly generating 22.3MtC, which is the 2050 carbon emissions limit. Under the base case scenario, this is equivalent to 29.4% of electricity demand in 2050, 563TWh, and 43.9% of electricity demand, 377TWh, in the lower demand case.

  38.  Under both the base case and the lower electricity demand scenarios, gas and coal alone are unable to meet the 60% target and simultaneously satisfy electricity demand. Substantial scope remains for other technologies.

2.3.3  Construction schedules

  39.  To facilitate the analysis, it is assumed that the capacity required in 2050 will be built in four 10-year intervals starting in 2010, with a quarter of the capacity needed in 2050 built in each construction phase. The construction mix in each period reflects the final electricity mix in 2050. An underlying assumption has been made that, for a technology to be reliable on a large scale in 2050, investment in the technology is required for a significant period of time.

  40.  The construction schedules ignore the phasing out of current generating capacity. Precise matching of new capacity requirements with the phasing out of existing power plants goes beyond the scope of this paper. However, it is assumed that an even distribution of new capacity over a 40 year period approximates the requirements of both new demand and phasing out of existing capacity. It is further assumed that all current power plants will be out of service by 2050. The model does not include the cost of replacing any capacity post-2050, but given the long discounting period it is unlikely that such costs would impact in a significant fashion on present investment decisions.

  41.  The model does however include the costs of replacing new capacity that phases out before 2050. This is the case for all plants built in 2010, with the exception of nuclear which does not require replacement before 2057, as well as for wind farms built in 2020. The phasing out of newly built capacity is summarised in table 2.1:

Table 2.1

    Construction year
Technology 2010 2020 2030 2040 Comments
CCGT 2042 2052 2062 2072 replace first generation in 2040
CCGT (CC) 2042 2052 2062 2072 replace first generation in 2040
IGCC 20462056 2066 2076 replace first generation in 2040
IGCC (CC) 2046 2056 2066 2076 replace first generation in 2040
Nuclear 2057 2067 2077 2087 no replacement needed before 2057
Wind 20312041 20512061 replace first generation in 2030, second in 2040


  42.  The scenarios used in the model illustrate how the gap between electricity demand, and supply by traditional fossil fuel generating technologies, can be filled by carbon-free generating capacity.

  43.  Scenarios 1 and 5. Scenarios 1 and 5 assume that the entire supply shortfall can be covered by wind. The two scenarios differ only in the overall demand, and therefore in the amount of electricity supplied by wind, as shown in figure 2.8:

Figure 2.8

  44.  Scenarios 2 and 6. Scenarios 2 and 6 consider significant investment in nuclear power. As wind energy is expected to contribute to the electricity mix regardless of investment in nuclear power, nuclear only covers 80% of the supply shortfall in these two scenarios. The scenarios vary only in the overall electricity demand in 2050, as can be seen in figure 2.9:

Figure 2.9

  45.  Scenarios 3 and 7. These two scenarios model more realistic wind scenarios than was represented by scenarios 1 and 5. The practicable potential for offshore wind power generation in the UK in 2025 has been estimated at 100TWh annually (Gross, 2004). No equivalent estimate exists for 2050, but a generous 25% improvement assumption[43] yields an annual limit of 125TWh/yr. Scenarios 3 and 7 assume wind power supplies this upper limit in 2050, and that nuclear power supplies any shortfall. As a consequence of these assumptions, scenario 3 does not differ substantially from scenario 2. However, the impact of the assumptions is more starkly illustrated by the differences between scenarios 6 and 7, as shown in figure 2.10:

Figure 2.10

  46.  Scenarios 4 and 8. In the final two scenarios, wind is again modelled to supply up to a realistic limit of 125TWh/yr, however scenarios 4 and 8 assume significant investment in carbon sequestration technologies rather than in nuclear power. The assumption of a gas to coal ratio of 2:1 is maintained, and the percentage share of electricity supplied by plants with carbon sequestration capabilities necessary in order to meet both the electricity demand and the carbon emissions target is computed determined. This procedure is illustrated in table 2.2:

Table 2.2

Scenario 4Scenario 8
2050 demand (TWh)563 377
Less: wind supply (TWh)125 125
Needed from coal and gas (TWh)438 252
Gas to coal ratio2:1 2:1
Gas supply (TWh)292 168
Coal supply (TWh)146 84
Share of carbon capture in coal and gas 73.6%40.7%
Gas supply—traditional (TWh)77 100
Gas supply—with carbon capture (TWh) 21569
Coal supply—traditional (TWh)39 50
Coal supply—with carbon capture (TWh) 10834
Gas emissions—traditional (MtC)7.77 10.07
Gas emissions—with carbon capture (MtC) 3.521.12
Coal emissions—traditional (MtC) 7.7710.07
Coal emissions—with carbon capture (MtC) 3.221.03
Total emissions (MtC)22.3 22.3

  47.  The critical element in the above analysis is the share of carbon capture in coal and gas output. This is calculated in the following manner:

Given values:

      EL = Emissions limit

Sc = total supplied by coal

Sg = total supplied by gas

ect = emissions to output ratio, traditional coal

eccc = emissions to output ratio, coal with carbon capture

egt = emissions to output ratio, traditional gas

egcc = emissions to output ratio, gas with carbon capture

Needed value:

scc = share of carbon capture in overall coal and gas supply

  The total emissions output can be set equal to EL and is the sum of all the emissions generated by the four different technologies:

  48.  Substituting all known values into (5) yields the share of carbon capture in overall coal and gas supply: 73.6% for scenario 4 and 40.7% for scenario 8. Consequently, the two scenarios differ in the way that carbon sequestration is used and also in the impact of wind supply, as shown in figure 2.11:

Figure 2.11

  49.  Scenario summary. To summarise, the eight scenarios are compared in figure 2.12, which illustrates the relative shares of the different generating technologies by scenario. Most interesting for the analysis in subsequent sections are scenarios 3 and 7, in which coal, gas and wind supply electricity up to their maximum. It is thus apparent that these three technologies alone are insufficient to meet future electricity demand—indeed, even in the lower demand case, in the absence of additional generating capacity, a supply gap of almost 30% exists which is filled by nuclear generation.

Figure 2.12

2.5  Discounting

  50.  While conventional approaches to discounting value a gain or loss in the future at less than the same gain or loss today, ethical controversies rage over the applicability of fixed-rate discounting approach to social issues (Heinzerling, 2000; Ackerman and Heinzerling, 2002). These controversies have particular applicability to nuclear build as nuclear energy has both positive and negative environmental attributes that are important in the long-term. While nuclear power can provide large amounts of carbon-free energy in the face of worsening climate change, negative environmental effects are associated with the decommissioning of nuclear power plants, the storage of nuclear waste and the potential for catastrophic nuclear accidents in the future.

  51.  As nuclear investment decisions will be made by the market, a fixed discount rate will be used to evaluate the investment potential, and social concerns will play no role in the decision making process. Consequentially, risks to future generations will be valued to a lesser extent than risks to the present generation. However, recent advances in discounting theory suggest that the correct social discount rate is not a single number, but a value that declines with time (Pearce et al, 2003). Official guidance to Ministries notes the power of time-varying discount rates when evaluating investments and policies with costs and benefits that accrue over more than thirty years hence (HM Treasury, 2003). Consequentially, this paper investigates the costs of energy investments using both a constant market discount rate, and the declining social discount rate scheme advocated by HM Treasury in the Green Book.

  52.   Fixed rate discounting. Private investors use a fixed discount rate, which reflects the risk of an investment, in evaluating investment potential. Oxera (2005b) quote the average expected return for a utility company as 8-12% annually. This is reflected in the discount rates used in the model. Gas, as the safest of the six technologies, is discounted at the lower end of the range, while nuclear is discounted at 15%, reflecting the extra-ordinary risks involved. This is in line with an Oxera (2005a) study that argues that nuclear investors expect a return of between 14% and 16%. Figure 2.13 shows the discount rates used for the different technologies, and reflects their expected risk:

Figure 2.13

  53.  These discount rates are used to discount all cash flows relating to a project from the day of its initiation, ie the first day of planning, to the last day of operation. Once the net present value of a project at initiation is determined, this value is discounted back to 2005 at a discount rate of 10%. This is the average return of a utility company and as such is deemed applicable to any money not yet committed to any specific project.

  54.   Green Book Method. The declining social discount rate scheme advocated by HM Treasury in the Green Book recommends a discount rate of 3.5% for the first 30 years of a project, and 3% for the next 45 years. Afterwards, it declines further, but is no longer relevant for this analysis. This discounting treatment requires risk to be dealt with before discounting takes place. As such, conservative assumptions of costs and benefits are appropriate for use with this discounting scheme.

  55.  The use of a declining discount rate scheme has two important consequences for the analysis of the future electricity mix.

  56.  First, a unified rate is used for all technologies. Thus nuclear energy is no longer penalised by a high discount rate. The investment risk of nuclear generation is accounted for by the use of conservative assumptions and an allowance for cost over-runs.

  57.  Second, declining interest rates increase the impact of future cash flows. As most negative cash flows involved in power generation occur during the planning and construction phases and profits are generated in later periods, an effect of declining discount rates is increased profitability. Hence, the longer the lead-time, the later a technology reaches its profitable phase, and the more likely its profitability will improve as a result of lower discount rates. Similarly, projects with long lifetimes are likely to benefit from low discount rates due to distant positive cash flows, which become more significant using a declining-rate scheme.

  58.  Figure 2.14 shows that, relative to other generating technologies, nuclear power plants have the longest lead- and life-times. Wind has the shortest. Consequently, nuclear benefits to a greater degree than wind from low and declining discount rates.

Figure 2.14

Source: Dale et al, 2004; MacKerron, 2004, Whittington and Bellhouse, 2000.

2.6  Profitability

  59.  In order to compare the profitability of the various technologies and scenarios modelled, a wholesale electricity price of 3p/kWh is assumed. This price is used in analysis by Oxera (2005b) as a long-term average wholesale electricity price and is consistent with a gas price of 28p/therm. The expected profits or losses from different generating technologies calculated are then discounted to 2005 for comparison.

  60.  In the absence of government subsidies it can be expected that investors will only pursue those projects that, based on traditional discounting methodology, offer positive net present values ("NPV"). To the extent that a scenario generates a negative NPV, it can be argued that this reflects the size of the government subsidy, either in form of direct subsidies or in form of risk guarantees, required to make it attractive.


  61.  This section presents the results of the discounting exercise carried out by the model described in the previous section. The costs of the different technologies are compared before the costs of the scenarios are examined.

3.1  Technology Comparison

Figure 3.1

Green Book method

Note: Cost defined as the price in pence per kWh that yields a NPV of zero.

  62.  The costs of the six different technologies evaluated are summarised in figure 3.1 in terms of pence per kilowatt hour. The figure shows the electricity price that is required to generate a net present value of zero for each technology. Thus, it reflects not only the costs of capital expenditure, O&M, fuel and carbon, but also the risk associated with each technology.

  63.  Costs for plants initiated in 2010, 2020, 2030 and 2040 are shown. As discussed in the previous section, capital expenditure costs in particular, are expected to decrease with technological learning. Consequentially, plants of all technologies built benefit from lower costs with time under the market method.

  64.  As becomes apparent from figure 3.1, under the market method, only CCGT power plants are able to generate electricity in 2010 at costs that are below the assumed price level. Despite significant cost savings, this does not change in 2020. Further analysis reveals that due to continuous cost savings, wind turbines built in 2030 will be able to generate electricity profitably, at costs of 2.88p/kWh. Similarly, in 2050 IGCC plants will generate electricity profitably, at costs of 2.90p/kWh. However, neither nuclear plants nor plants with carbon capture ability are expected to become profitable technologies if the wholesale electricity price remains at 3p/kWh.

  65.  Of the technologies that are likely to be needed to fill the supply gap left by traditional fossil fuel generating capacity and wind, gas plants with carbon capture technology are the cheapest option, followed by nuclear and coal plants with carbon capture technology.

  66.  The outcome changes significantly when the declining discount rate scheme recommended by HM Treasury, and the social cost of carbon recommended by DEFRA, are used. Under this treatment, all technologies except traditional gas and coal, which increase due to the application of the social cost of carbon, become considerably cheaper. Further, technologies discounted at higher discount rates under the market method benefit the most.

  67.  Consequentially, nuclear is the biggest beneficiary under the social cost method, in which low, declining discount rates are used and the social cost of carbon is included. Indeed, nuclear is not only profitable but becomes the cheapest technology. [44]All technologies except traditional coal and coal with carbon capture technology are profitable when Green Book discounting is used.

  68.  These results should be interpreted with caution. The rationale for using declining discount rates is to give more weight to the costs and benefits experienced by future generations. However, potential environmental costs are difficult to quantify. In particular, the costs of climate change and externalities associated with nuclear power pose a challenge. While the model includes the social cost of carbon, climate change has the potential to cause irreversible damage to ecosystems that is impossible to quantify in monetary terms. In respect of nuclear energy, the inestimable societal costs nuclear energy has the potential to impose[45] remain external to the model. Thus, the results of the analysis based on the Green Book discount rates have the potential to mislead as some, but not all, social costs have been included.

3.2  Scenario Comparison

  69.  The technology prices discussed above determine the profitability of the different scenarios analysed. The net present values of the various scenarios are summarised in table 3.1. For ease of comparison, key demand and supply parameters are given as well.

Table 3.1

2005 NPV, BY SCENARIO (£bn)

  70.  The first key insight is that none of the eight scenarios generate a positive net present value when using the fixed discount rate method. From this it can be deduced that either the assumed wholesale price of 3p/kWh is too low, or that government intervention will be necessary to achieve the emissions reduction target in 2050.

  71.  The need for government support is minimised in scenarios 1 and 5, which rely on wind energy to a degree that is improbable. [46]Further analysis reveals that those scenarios that rely heavily on nuclear power require less government support than the carbon capture scenarios.

  72.  This seems counter-intuitive as nuclear is not the cheapest option on a per kWh basis. The apparently contradictory result is caused by a peculiarity of discounting. Indeed, while the high discount rate of nuclear drives up the price at which a nuclear power plant can break even, it also discounts any losses generated by a nuclear plant at a higher rate.

  73.  Under the declining discount rate scheme, all eight scenarios are NPV positive, and therefore economically feasible. The two scenarios relying most heavily on nuclear are the most economically attractive. This is a consequence of the low price per kilowatt-hour for nuclear under the Green Book discount rate scheme.

  74.  It is however highly unlikely that the discount rates recommended by HM Treasury would be utilised in assessing privately financed electricity investment without government support to reduce the risks to investors. This could be achieved by a government offer to finance packages at the discount rates set out in the Treasury Green Book, and would result in the taxpayer carrying the risk in excess of discount rates recommended by the Treasury, should costs exceed expectations or benefits disappoint.


  75.  The results of the model indicate that the British government has some difficult choices to make if it is to meet the 2050 carbon emissions target. The limits imposed on coal and gas generation by the 2050 target, together with the limit imposed on renewable energy generation by its practicable potential, necessitate massive energy efficiency savings if the 2050 target is to be met without a large-scale nuclear energy or carbon sequestration program. Indeed, in order to achieve this, demand must be contracted to 1995 levels. [47]

  76.  The energy efficiency scenario used assumed efficiency savings of 2% per year. Blok (2005) suggests year-on-year efficiency savings of this level are possible but rarely achieved. Hence, the efficiency improvements assumed are considered optimistic and it is improbable that sustained efficiency improvements beyond this level will be achieved. [48]

  77.  The prospects for renewable energy sources are limited by what is practicable. DTI (2000) estimates for 2025 indicate that the practicable limit to generation by wind (both off-shore and on-shore) is 108TWh/yr, whereas all current renewable energy technologies have a practicable limit of 230TWh/yr (Gross, 2004).

  78.  If the UK could achieve electricity efficiency savings of 2% per year until 2050 and realise the practicable potential suggested by the DTI (2000) in wind (both offshore and onshore), biomass, BIPV, wave, tidal and small hydro, [49]notwithstanding uneconomic generating costs and ignoring any extra generation requirements due to intermittency, the electricity demand in 2050 could be met.

  79.  At present, this scenario seems impossibly optimistic.

    —  The Renewables Obligation is underperforming and is unlikely to achieve its objectives (Mitchell and Connor, 2004).

    —  The goals of the Renewables Obligation are insufficient to put the UK on track to reduce carbon emissions by 60% by 2050, especially in the face of declining nuclear contribution to the electricity mix and growth in electricity demand (Oxera, 2005b).

    —  The costs of wind generation, considered the most promising renewable energy technology, increases significantly as wind penetration approaches 30% (Dale et al, 2004)—this cost increase is difficult to model and not included in the analysis in this paper.

  80.  As a consequence of practicable limitations on electricity efficiency improvements and renewable energy penetration, either a large scale carbon sequestration or a nuclear energy program will be necessary to meet both future energy requirements and the 2050 carbon abatement target. [50]

  81.  The economic analysis completed indicates that scenarios that rely on nuclear power to meet the shortfall from traditional fossil fuel generation and wind energy require less government support than scenarios that rely on carbon capture.

  82.  The argument for nuclear is advanced as a carbon sequestration strategy increases the nation's dependence on imported fossil fuels, while a strategy of nuclear energy goes some way to meeting the UK government's stated requirement for energy diversity (DTI, 2003). The value of this cannot be quantified in an economic sense however it is undoubtedly a factor that weighs in favour of nuclear energy.

  83.  Carbon sequestration is similar to nuclear energy in that a large-scale project involving considerable investment is necessary for implementation. Unlike nuclear energy however, carbon sequestration is an unknown quantity. Only one carbon sequestration project operates worldwide (Tzimas and Peteves, 2005). Widespread deployment of carbon capture, transport and injection infrastructure carries unknown economic risks and significant scientific and technological challenges.

  84.  For the reasons discussed above, all else being equal, nuclear power is an attractive option in the transition to a low carbon economy. However two important constraints exist.

    —  The risk premium required by the private sector for nuclear investment results in unprofitable generation at a wholesale electricity price of 3p/kWh—a nuclear plant built in 2020 would generate electricity at a cost of 5.8p/kWh.

    —  The issue of nuclear waste is unresolved.

  85.  Risk has a dramatic effect on the cost of nuclear generation. This is well illustrated by the example of the new Finnish reactor, which is being built on a fixed-priced contract by the French company, Areva. Areva is partially owned by the French government, and as such, French tax-payers bear the risk of cost over-runs. This arrangement has enabled the private consortium, TVO,[51] which financed the Finnish reactor to use a low, 5% discount rate to do so.

  86.  The effect of the discount rate on profitability is shown in figure 4.1, which illustrates that nuclear power stations built in the UK in 2020 could be financed profitably at a discount rate of 8%.

Figure 4.1

  87.  The risk premium required for nuclear investment is a consequence of uncertainties surrounding new nuclear technologies, the possibility of cost over-runs and the dependence of the wholesale electricity price on the price of gas, ie price sensitivity (Oxera, 2005b). Indeed, if the nuclear waste issue can be resolved to the extent that privately financed nuclear build is desirable, government support for nuclear investment must address these uncertainties.

  88.  Uncertain new technologies. While new nuclear technologies are predicted, by the nuclear industry, to be able to supply electricity very cheaply, these designs have yet to be built anywhere in the world (PIU, 2001). The private sector, which considers the risk of investment in nuclear relative to investment in gas, requires a high return on investment to bear this technology-related risk.

  89.  The government on the other hand must consider the social cost of carbon and the imperative to reduce carbon dioxide emissions. As such, it is required to choose between two uncertain technologies if it is to meet its own emission reduction target while supplying sufficient electricity for economic growth. Hence, the risk profile faced by the government is markedly different from that faced by private investors and government intervention in favour of either nuclear energy or carbon sequestration is necessary for the 2050 target to be met.

  90.  Cost over-runs. A second generation of nuclear reactors could be expected to suffer less from cost over-runs. However, cost over-runs have been a feature of nuclear power in the UK and elsewhere to date. Consequently, this paper has been conservative in the assumptions made regarding the costs of nuclear energy—upper estimates of costs have consistently been used. Further, a 10% allowance for cost over-runs has been made.

  91.  Uncertainties over costs continue to affect potential nuclear investment regardless of cautious estimates. If the UK government decides nuclear energy is to be part of a solution to carbon abatement, this risk must be addressed by government. This can be done either through a cost guarantee or by a government-owned company building the plants on a fixed-priced bid.

  92.  Cost over-runs are not restricted to nuclear investment and could be expected to afflict a large-scale carbon sequestration project. As such, similar measures would be needed to reduce the risk of investment in a carbon sequestration project and encourage private sector involvement.

  93.  Price sensitivity. As a consequence of the low cost of gas generation in the liberalised market, the price of electricity is dependent on the price of gas. Part of the risk associated with nuclear investment is concerned with gas prices, and thus electricity prices, falling. The realised price of electricity is of particular risk to nuclear investors due to the long lead- and lifetimes of the nuclear power plants, although this risk is also present in a large-scale carbon sequestration project.

  94.  Private investors will require assurances that investment in a nuclear power plant will not result in an unprofitable source of generation several years hence due to a fall in gas prices. In this regard, government could prevent gas prices falling below a price floor by using either a carbon tax, or a "smart" import tariff that is dependent on the price of gas. [52]

  95.  However, a government decision to support nuclear new build will consider more than just economics. Ultimately, government may decide that supporting nuclear investment carries too much political risk, in which case commitment to carbon abatement will suffer, and the 2050 target will almost certainly fall by the wayside.


  96.  Uncertainty is inherent in any study which attempts to predict the future. This paper required assumptions regarding, among other factors, energy efficiency improvement rates and the practicable potential of renewable energy sources. In either case, significant deviation from the numbers used, would affect the results of the model in a relevant manner. However, the best case scenario (ie high energy efficiency savings and high penetration of wind energy) results in a 23% short-fall in supply in the absence of a nuclear energy or carbon sequestration program. Thus, significant changes to either or both of these numbers (in the same direction) would be required to alter the conclusion that either nuclear energy or carbon sequestration will be required to achieve the 2050 target.

  97.  Further study regarding the practicable potentials of renewable energy technologies past 2025 is necessary. Decisions regarding the generating mix are long-term decisions, and full information in this regard is essential.

  98.  It must be noted, that an economic analysis of the sort undertaken for this paper has limitations in accounting for social costs. Indeed, it is impossible for the social cost of carbon included in the analysis to account for the worst consequences of climate change, such as irreversible damage to the climate system and ecosystems. Similarly, certain societal costs nuclear energy has the potential to impose[53] remain external to the model.

  99.  These limitations notwithstanding, the analysis suggests that reconciling the UK's economic growth projections with the government's plans for carbon abatement will not be easy.

    —  The carbon abatement target imposes restrictions on fossil-fuelled electricity generation.

    —  Renewable energy has practicable limitations.

    —  Feasible electricity efficiency improvements are constrained below 2%/yr.

  100.  As such, the UK will be unable to meet electricity demand in 2050 without a large scale program of nuclear energy or carbon sequestration. There are compelling arguments for investment in new nuclear generating capacity:

    —  Nuclear energy is more cost effective than a carbon sequestration program (involving both gas and coal sequestration), although both would require government subsidy.

    —  A large-scale nuclear power program supports the government's desire for energy diversity, while a carbon sequestration program requires increased dependence on imported fossil fuels.

    —  Nuclear energy has benefited from much technological learning. Carbon sequestration, by contrast, is untried on the scale necessary to meet the predicted shortfall.

  101.  However, management of nuclear waste is a major weakness in the case for nuclear. Consequentially, whether the UK's national emission reduction targets will force an increasing reliance on nuclear power depends on satisfactory resolution of the nuclear waste issue.

  102.  The analysis presented in this paper suggests that an environmental toll will be paid for electricity usage in the UK over the next fifty years. Whether that toll will be in carbon emissions or nuclear waste is a decision for the country's politicians.

  2. Ackerman, F and Heinzerling, L 2002. Pricing the Priceless: Cost-Benefit Analysis of Environmental Protection. University of Pennsylvania Law Review, 150: 1553-1584.

Ansolabehere, S, Deutch, J, Driscoll, M, Gray, P, Holdren, J, Joskow, P, Lester, R, Moniz, E, Todreas, N. 2003. The Future of Nuclear Power, An Interdisciplinary MIT Study. Available online at: Accessed 6 August 2005.

Blok, K 2004. Enhanced policies for the improvement of electricity efficiencies. Energy Policy 33: 1635-1641.

Carbon Trust. 2004. The European Emissions Trading Scheme: Implications for Industrial Competitiveness. Available online at: Accessed 6 August 2005.

Clarkson, R and Deyes, K 2002. Estimating the Social Cost of Carbon Emissions. London: The Public Enquiry Office.

Dale, L, Milborrow, D, Slark, R, Strbac, G 2004. Total cost estimates for large-scale wind scenarios in UK. Energy Policy 32: 1949-1956.

Department for Environment, Food and Rural Affairs (DEFRA). 2001. Managing radioactive waste safely. London: The Stationery Office.

Department for Environment, Transport and the Regions (DETR). 2000. Climate Change: The UK Program. London: The Stationery Office.

Department of Trade and Industry (DTI). 2000. New and renewable energy in the UK: Prospects for the 21st Century. Available online at: Accessed 6 August 2005.

Department of Trade and Industry (DTI) 2001. In: House of Lords; Select Committee on the European Union. 14th Report—Energy Supply: How secure are we? London, The Stationery Office.

Department of Trade and Industry (DTI) 2002. Quantifying the system costs of additional renewables in 2020. Available online at: Accessed 6 August 2005.

Department of Trade and Industry (DTI). 2003a. Our Energy Future—Creating a Low Carbon Economy. London: The Stationery Office.

Department of Trade and Industry (DTI). 2004. Renewable sources of energy. Digest of UK Energy Statistics, chapter 7. London: The Stationery Office.

Gross, R 2004. Technologies and innovation for system change in the UK: status, prospects and system requirements of some leading renewable energy options. Energy Policy, 32: 1905-1919.

Heinzerling, L. 2000. The rights of statistical people. Harvard Environmental Law Review, 24: 189-208.

HM Treasury. 2003. The Green Book: Appraisal and Evaluation in Central Government. London: HM Treasury.

Intergovernmental Panel on Climate Change (IPCC). 2001. Climate Change 2001: The Scientific Basis. Geneva: United Nations Environmental Program, World Meteorological Organization.

MacKerron, G 2004. Nuclear power and the characteristics of "ordinariness"—the case of UK energy policy. Energy Policy 32: 1957-1965.

Marsh, G 2003. Carbon Dioxide Capture and Storage: A win-win option. London: The Stationery Office.

Mitchell, C and Connor, P 2004. Renewable Energy Policy in the UK 1990-2003. Energy Policy 32(17): 1935-1947.

Oxera. 2005a. Plugging the carbon productivity gap. Available online at: Accessed 6 August 2005.

Oxera. 2005b. Financing the nuclear option. Available online at: Accessed 6 August 2005.

Pearce, D, Groom, B, Hepburn, C and Koundouri, P 2003. Valuing the Future: Recent advances in social discounting. World Economics, 4(2): 121-141.

Performance and Innovation Unit (PIU). 2001. The energy review. London: The Stationery Office.

Tzimas, E and Peteves, S 2005. The impact of carbon sequestration on the production cost of electricity and hydrogen from coal and natural-gas technologies in Europe in the medium term. Energy, 30: 1672-2689.

4 September 2005

37 Accessed 6 August 2005. Back

38   This paper assumes a 60% reduction in carbon emissions will be required from the electricity sector. Back

39   Only applicable to the first nuclear power plant built. Back

40   Available online at: Accessed: 6 August 2005. Back

41   The ratio of electricity consumption and GDP. Back

42   Available online at: Accessed 6 August 2005. Back

43   Practicable potential is also influenced by space availability, efficiency and uptake; hence the 25% improvement assumes significant improvement in one of these areas. Back

44   This is the case even if the social cost of carbon is omitted. Back

45   That is, the effect of nuclear catastrophe on human health and society, and on the environment. Back

46   Both these scenarios require more generation from wind than is practicable, ie more than 125TWh of wind generation. Back

47   Assuming a 2:1 ratio of gas:coal, gas can provide a maximum of 110TWh/yr, coal can provide a maximum of 55TWh/yr and wind is limited by a practicable potential of 125TWh/yr. This amounts to 290TWh/yr, which is equivalent to the electricity demand in 1995. Back

48   To contract demand to 1995 levels, a yearly electricity efficiency improvement of 2.5% is needed-significantly above both the 1.7% achieved in the last decade and the 2.0% Blok (2005) considers possible but unlikely. Back

49   The costs of renewable technologies other than offshore wind have not been assessed. In the case of onshore wind, due to limited practicable potential of 8TWh/yr-the majority of the growth in wind to 2050 is expected to be offshore. In the case of all other renewable energy technologies this is due to prohibitive costs. Back

50   This paper has assumed that the UK government will either support a large scale program of nuclear energy or a large-scale program of carbon sequestration, but not a combination of the two, as cost reductions are dependent on economies of scale. Back

51   One of TVO's six shareholders is the government-owned utility company, Fortum. Back

52   That is, an import tariff that acts to ensure a minimum price but is absent at gas prices above this minimum level. Back

53   That is, the effect of nuclear catastrophe on human health and society, and on the environment. Back

previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2006
Prepared 16 April 2006