New whole system metrics can improve the targeting of policy support for emerging zero carbon technologies
Comment by Sarah Keay-Bright
Senior Advisor - Markets, Policy and Regulation
Policy makers need to ensure that they get value-for-money from the policy support they provide to promote innovation and early deployment of emerging (low carbon) technologies. So when policy makers make decisions about allocating policy support they need to take a view on which technologies are most promising or ‘deserving’ of support for innovation, early deployment and cost reduction. A crucial input to these decisions should be an assessment of the value that the technology can offer to the whole system as it transitions to a net zero future.
Unfortunately, the methodologies we use to design innovation support policy and to compare technologies were originally designed for traditional centralised generation technologies. Many commonly used metrics like ‘levelised costs’ fail to take account of the different characteristics of technologies and the need for a complementary mix of resources for the more variable and decentralised zero carbon electricity systems of the future.
This blog describes a framework methodology and metrics that can be used to estimate the full set of cost and benefits associated with a wide range of low/zero carbon technologies, including generation, storage, interconnection, demand reduction and demand response. This can enable much more clear-sighted and even-handed choices to be made about the allocation of policy support (or subsidies) for innovation and early deployment of immature technologies – taking account of their potential to bring ‘whole system’ value in a zero carbon future.
Around the world, it is increasingly common practice to use whole system models which take account of wider impacts but it is less common to include whole system impacts in all relevant decisions. In particular, demand-side technologies are typically given less consideration in both modelling and in the design of policy support measures or subsidies.
The levelised cost of energy (LCOE) metric is often used to “show” one technology (usually generation only) is cheaper than another and this implicitly colours allocation decisions about financial support (see Figure 1). LCOE is calculated as the discounted sum of all lifetime costs of a generator – including initial capital costs, fixed running costs, and variable running costs (such as fuel) – divided by the discounted sum of electrical energy generated over its lifetime. But, crucially, it does not take account of the wider impacts of the technology on the electricity system.
Figure 1: LCOE for different technologies
When designing innovation support policies, it is also important to take account of the ‘implicit support’ that technologies might be gaining through current market arrangements. This includes, for example, the value of risk transfers (for example, the benefit that investors enjoy from revenue stabilisation by winning a contract for difference) or the value of externalities (such as carbon emissions) that are not costed under current market and policy arrangements. Implicit support – usually due to imperfect market design and other policies – can be significant, favouring some technologies more than others.
Policymakers should aim to ensure that the wider electricity market and policy framework aligns the monetary incentives faced by investors with their wider impact on the system, ideally by designing a market framework that internalises system costs and externalities as much as possible.
In the meantime, while market reforms are developed and implemented, in designing innovation and early deployment support measures for promising technologies, it will remain important to take account of both:
the likely whole system cost and value impacts of technologies; and
the value of implicit support and risk transfers enjoyed under current policy and market arrangements.
This can be achieved by using new metrics that estimate these wider costs and impacts, according to a structured and objective methodology developed by ESC (in collaboration with Frontier Economics and the legacy of the Energy Technologies Institute). The methodology we have developed can also include the value of implicit support and risk transfers under current policy and market arrangements – within and outside the electricity sector too. BEIS uses this methodology in the “enhanced levelised costs” reported in its latest (2020) Eldectricity Generation Costs update.
The methodology and metrics for whole system costing
In 2018, the Energy Technologies Institute (ETI) (now partially incorporated in Energy Systems Catapult) commissioned Frontier Economics to develop a robust framework for comparing the whole system costs and benefits of electricity generation, storage and interconnection investments in Great Britain. The project produced transparent decision support tools and new metrics designed to assess the following:
The framework comprises 6 steps that calculate whole electricity system costs (WESC) based on measuring the change in costs of constructing and operating an electricity system that results from the addition of a given quantity of a particular technology to that system. The framework’s six steps include:
Decide on the scope of costs and benefits, which should include the impact on the whole electricity system costs. (Potentially the approach could also include costs and benefits that arise in other sectors (e.g. spillover benefits for decarbonising other parts of the energy system) or a valuation of innovation, air quality, fuel poverty or strategic security of energy supply benefits.
Define a baseline scenario for the electricity system, which should represent the most likely future development of the energy sector, including likely policy developments.
Decide on the size of the investment increment – small increments inform assessment of individual investment decisions while large increments are more relevant to broader changes in policy or strategy.
Set up detailed models of electricity system and wider energy system.
Assess the technologies on a level-playing field, by abstracting from current market arrangements (i.e. separating out the implicit support due to risk transfers and unpriced externalities).
Produce metrics that allow overall costs to society and to consumers to be assessed.
For step 1, research by Frontier Economics for DECC in 2016 established an exhaustive and non-overlapping framework for breaking down the electricity system impacts of technologies based on wider literature. This provides the basis of the breakdown of electricity system costs and benefits for WESC:
Technology direct cost
Capital and operational costs associated with the incremental technology.
Capacity adequacy impacts
To the extent, existing capacity can be retired, or new capacity forgone to ensure the same level of security of supply and carbon intensity as the counterfactual, there is a cost-saving to the system.
If the incremental capacity impacts the uncertainty of supply, it will affect how generators in the rest of the system are called on to help support system stability by altering their output. It will also affect the extent to which they need to be prepared to do so at short notice, potentially affecting their staffing, fuel, and/or maintenance costs.
The incremental technology may require investments to reinforce or extend the existing grid, and changes to power flow may increase or decrease power losses due to transmission and distribution. It is also possible that technologies can free up headroom on the grid, creating network benefits.
Displaced generation impacts
Outputs from the incremental technology can displace higher marginal cost generation, producing variable cost savings e.g. fuel, carbon. The scale of this is diminished if generators in the rest of the system operate less efficiently, or the incremental technology is curtailed. The category includes the impact on variable costs of ensuring that the same carbon intensity is maintained.
In step 5 of the framework, the value of indirect support provided through current market and policy arrangements is estimated. Indirect support includes externalities such as carbon dioxide emissions or network congestion which remain unpriced (or incompletely priced) under current market arrangements. Crucially it also includes an estimate of the value of implicit risk transfers provided through current support mechanisms (e.g. the value of revenue stabilisation provided to investors through Contacts for Difference (CfDs)). While CfDs provide direct support in the form of revenues, its long-term contract with Government as counterparty reduces risk for the investor and in turn, reduces the cost of finance.
Frontier Economics developed initial methodologies for estimating the value of these forms of implicit support (set out in Frontier Economics, 2018). Further refinement of these methodologies may be appropriate to take account of latest evidence (e.g. on cost of capital benefits).
Figure 2: Estimates of indirect support
Source: Frontier Economics, 2018
The modelling results reveal material impacts for different technologies
Models run for the GB power sector, using an integrated investment and dispatch model for investments made in 2025, found material impacts on the value-for-money ranking order when the new metrics were compared to the traditional LCOE and strike prices used by the Government at the time. When ‘strike price equivalents’ are compared under current arrangements (i.e. in 2018) and under a level playing field, which removes the impact of indirect support provided by current market arrangements, it’s easy to see that different technologies enjoy different levels of benefit from the implicit support due to market inefficiencies.
Figure 3: An illustrative example comparison of ‘strike price equivalents’
Source: Frontier Economics, 2018. Note: Model runs are highly sensitive to assumptions and specific to the assumed energy system context and care must be taken with interpretation – results are not generic alternatives for LCOE. Note that onshore wind was assumed to come online in Scotland – its higher cost is partially due to the resulting higher transmission costs and risk of curtailment.
A new metric to include the demand side
The Recosting Energy project – led by Laura Sandys, to which Energy Systems Catapult is a contributor – recently commissioned Frontier Economics to update and apply this whole system methodology to a wider range of demand-side technologies.
Using WESC, it’s possible to compare the cost-effectiveness of demand-side measures against supply-side investment as part of a whole system transition to net zero and identify the circumstances where demand-side measures may be more cost-effective. The £/MWh metric is particularly useful for comparing technologies with broadly similar roles, primarily providing energy (rather than capacity) to the system. Using £/MW may be a more relevant metric for comparing technologies that provide capacity.
As part of an exploratory modelling exercise, WESC was applied to several generation, storage and demand-side technologies hypothetically added to the system in 2025. The blue line in the Figure below is a net value, being a product of costs and system benefits, which indicates how much additional cost would be incurred on the electricity system if a sufficient amount of each technology would be built to produce 1MWh over its lifetime.
Negative values indicate a technology that, when added to the system, reduces costs and so, in the chart below, technologies with negative WESCs are more beneficial than those with positive WESCs.
The more negative (or less positive) the figure, the most cost-effective the technology is at providing an MWh of energy. A word of caution though: The generator that is most cost-effective at providing 1MWh of energy may not be the most cost-effective at meeting other needs or outcomes, such as providing capacity. For example, a solar plant may be able to provide cheap energy, but adds very little (if anything) to capacity adequacy. It cannot, therefore, be said that the generator with the lowest WESC is the “best” type of generation overall.
In general, no single metric can fully capture the complementary nature of different technologies – for example, how peaking and baseload technologies can work together as part of the system. WESC is useful for comparing and assessing technologies from a whole systems perspective as the whole-system modelling that underpins WESC can be used to inform on the optimal mixture of technologies. Qualitative analysis can also supplement modelling and quantitative analysis for a broader system view covering a wider range of outcomes e.g. the System Value framework developed by the World Economic Forum with Accenture.
It’s easy to see in the chart that the technologies with negative values, on the left, happen to be DSR technologies. That’s because flexibility was assumed to be a by-product of EVs and heat pumps, with no fixed costs or variable costs associated with providing flexibility services. Note, however, that different forms of DSR may require additional costs such as communications and control equipment.
When more significant direct investment is involved, low-cost energy efficiency measures like LED lighting compete with low-cost generation such as wind and solar. The far ends of the chart, left and right, involving larger investment moves us into the domain of providing capacity at valuable times.
Figure 4: Whole System Electricity Costs (WESC) (up to +-£300/MWh)
The Value of Lost Load (VOLL) is a commonly used metric in power sector regulation and often regulators make crude assumptions about consumers’ VOLL and their willingness to pay. We all know that many different types of consumers exist, ranging from big industry to one-person apartments. Consumers value energy very differently and for any consumer, this changes throughout the day and at different times of the year and can depend on many factors.
Some forms of DSR may therefore incur significant costs. The modelling illustrates a domestic DSR technology (“DSR-Other Domestic”) requiring significant investment comparable to OCGT. However, due to the system benefit, it provides to the distribution network, the WESC is comparatively much less. On the left-hand side of the chart, it’s striking to see how significant the distribution network benefits can potentially be for heat pumps that provide flexibility, in this case providing system benefit of near to £2,500/MWh. The value is dependent on how much congestion exists in the locality and if network reinforcement can be avoided – the results above assumed that the distribution network would require imminent reinforcement in the absence of DSR. The value expressed per MWh terms is also sensitive to how often DSR is activated. In this modelling, DSR was used infrequently, and so the benefits in terms per MWh of shifted energy are particularly high. To compare such technologies that provide valuable capacity to the system in times of high residual demand and high system stress, it can make more sense to apply WESC on a per MW basis to assess firm capacity.
Application of WESC
Applying WESC to the demand-side reveals that the benefits to the system (and for service providers) can be material. For example, the whole system benefits of carrying out DSR to shift the charging of an electric van might be worth up to £504 per van per year (see Table 1 below) if the van is in a location requiring distribution network reinforcement. By contrast, the benefit would be around £84 without such gains on the distribution network.
Table 1: WESC components for depot-based electric vehicles
Value per MWh
Value per van per year
Technology own variable costs
Technology own fixed costs
Capacity adequacy costs
Displaced generation costs
Distribution network costs
Source: Frontier Economics, 2020.
The modelling undertaken for the ReCosting Energy project directly compared some supply and demand-side options using WESC to illustrate the cost savings that could be made – and to show how those savings arise in terms of the Frontier Economics/DECC categories of system costs. Comparing large solar to low-cost non-domestic energy efficiency measures in providing energy to the system revealed a major cost difference (see Table 2). Of course, individual energy efficiency measures vary widely in cost, but the point is that the market and policy framework needs to drive optimisation of supply and demand and so such options should be compared.
This modelling exercise also highlighted the paucity of data currently available on the costs and benefits of demand-side measures, especially compared to data available for generation. This needs to change if the demand-side and supply-side are to be analysed, compared and treated more equally in future.
Table 2: WESC components compared for solar and non-domestic efficiency
Technology own variable costs
Technology own fixed costs
Capacity adequacy costs
Displaced generation costs
Source: Frontier Economics, 2020.
In future, what will be crucial for the demand-side, is the extent to which the true value of flexibility and optimisation can be revealed in markets through accurate prices AND the extent to which this value can be captured and flow through to consumers. Laura’s ReCosting Project and the Catapult’s RethinkingElectricity Markets set out several recommendations for reforms that could improve the ability of markets to reveal the true value of flexibility and optimisation across the electricity system in time and spatially.