LOW CARBON POWER
This paper builds on analysis of some fundamental issues, for the future of a low carbon power sector, carried out by the author in the course of work within the Oxford Martin Programme on Integrating Renewable Energy, and also in work for the Energy Technologies Institute (ETI), due to be published by the ETI in the near future. My current intention is to develop some of these ideas further, with recommendations, in the context of the Oxford Martin School programme.
This paper tries to provide a high level perspective on challenges for the integration of renewable energy sources within a low carbon electricity sector. The conceptual approach is to start from a perspective of the system economics for the power sector as a whole, and several clear sources of market failure in three main domains of investment, operations and allocative efficiency. This last is the consumer and behavioural dimension, and included issues of tariffs and demand management. This approach leads to numerous implications for markets, governance and regulation. These reflect practical commercial, market and organisational questions, and will suggest corresponding structural, institutional and regulatory changes of a wide ranging nature.
Analysis of the systems economics of the renewable energy challenges is essentially about interactions and complementarities within the power system. This covers other sources of generation, including residual fossil fuel generation (at least in a period of transition), nuclear, and carbon capture and storage (CCS), storage and demand side measures. It also needs to take account of how future use of power is expected to develop in energy economies that are intended to become wholly low or zero carbon. This implies looking at future electricity penetration in transport and heat, and the particular challenges posed in additionally meeting these kinds of demand.
The Grubb Paradigm. Investment, Operations and Consumption.
Michael Grubb’s analysis of climate issues uses a categorisation of climate mitigation and energy policy options within three domains, loosely summarised as:
· transformative, strongly linked to all the major issues of technology policy and strategic investment, and generally linked to timescales measured in decades. This domain is fundamentally strategic and policy driven.
· optimising, strongly associated with the operation of the major energy industries, with markets and prices, and the assumptions and methods of neoclassical economics as a major pillar of policy; and both short and medium term timescales.
· consuming, or part of the “satisficing” domain within the Grubb paradigm, with a focus on day to day consumer behaviour and choices rather than energy industry operations. Relevant policy in this domain has often focused much more on regulatory solutions, through building standards and appliance labelling for example. The targeted time dimension is the more immediate, including behaviours relating to the ways consumers buy and use energy.
In this context our analysis also distinguish three domains. These correspond to the Grubb paradigm, at least superficially, but will be described more simply as the investment domain governing the way investments are chosen and made, the operations domain covering how the energy industry capital stock is utilised, and the consumption domain dealing with consumers and their choices. In each of these domains there is likely to be some combination of market driven choices and decisions, policy intervention or formal coordination, and regulation. Investment tends to be a dominant concern but cannot be isolated from the matter of confidence in the future arrangements governing both operation and consumption.
The basic economic and practical objective, in responding to several substantial market challenges (and potential market failures) is efficiency and effectiveness, in investment, operation and consumption. This means considering how to:
· finance and deliver an efficient mix of investment, in the quantity needed, and without incurring excessive capital costs.
· incentivise and deliver efficient and secure operation of those investments once made, together with operation of the existing capital stock.
· achieve allocative efficiency by ensuring consumers can respond to cost reflective pricing, with appropriate technical and commercial options, to allow a potentially large contribution to getting an efficient low carbon economy.
2. MARKET FAILURES/ MARKET CHALLENGES.
We can define six challenges to a low carbon objective that stem from the system economics of the power sector.
2.1 Failure to internalise the CO2 externality.
An annoying and inconvenient truth: “If damaging externalities are not internalised in prices, there is no basis to assume that free markets and free trade will improve human welfare”. In the absence of adequate carbon pricing a purely market-based approach is unable to deliver low carbon investment or even to ensure the most carbon-efficient operation of an existing stock of plant. Nor, in systems with significant fossil fuel content, can the market be relied on to provide the right signals to consumers for efficient use of energy or the optimal choices between alternative energy vectors. The carbon pricing issue, whether it is treated as explicit or implicit, impacts on all three domains - investment, operations and allocative efficiency.
SOCIAL COST OF EMISSIONS AND THE PRICE OF CARBON
The notion of a generally agreed measure of these externalities as a social cost, and its application to the pricing of carbon, may be illusory. Policy makers are increasingly influenced by an appreciation of the potentially catastrophic scale of failure to address climate issues, and it is global risk management concerns, rather than the arguably inadequate and incomplete tools of cost benefit analysis, that drive policy. Moreover cap and trade schemes have hitherto been of very little significance. Nevertheless measures do exist, have been embodied for example in UK Treasury guidelines and do serve a useful purpose as a very loose benchmark against which to evaluate, at least superficially, costs of failure to deal with the externalities.
The largest adverse consequence of this failure may be the effect on low carbon investment incentives, the most notable being failures to incentivise the carbon capture investments widely recognised (not least by ETI) as key to a least cost decarbonisation of the power sector . But it can also lead to perverse operational choices, an obvious recent anomaly (in Northern Europe) being the closure of highly efficient gas power stations, while coal stations continue to operate baseload due to cheap coal. In climate terms this is an expensive anomaly, even though the immediate financial implications for the utilities were relatively trivial.
It is easier to attach a value to current operational anomalies than to investments foregone. Recent UK experience is an illustration of the scale of the issue at the operational level. Between 2009 and 2012 the substitution of coal for gas, induced by changes in gas/coal price relativities, increased UK coal consumption by about 15 million tonnes. This increased CO2 emissions by around 20 million tonnes, to which past Treasury guidelines might have attached a notional “social cost of carbon” value of around £ 1.2 billion. Actual fuel savings to generators are likely to have been at most 10% of this amount. In other words this single failure to price carbon will have generated, notionally, a long term “social cost” of around £ 1 billion.
If investment choices were also to be based on a projection of very low carbon costs, corresponding measures of welfare loss would be of a much larger magnitude, even on the basis of these very conservative measures of social cost. The real cost however would be comprehensive failure to meet emissions targets, and consequences of that. The inability of existing market structures to incentivise and bring forward low carbon investment provides a primary justification for intervention.
Consistency between vectors is particularly important in contexts of consumer choice. A UK inconsistency has been Treasury guidance, for public sector projects or public policy purposes, which differentiates between “market” and “non-market” sectors. Inclusion of aviation in the carbon traded sector attaches a lower value to CO2 in aviation than in road transport. So hypothetical schemes to promote biofuels would show a higher return if they displaced diesel as a road fuel (excluded) rather than aviation fuel (included), even though, ceteris paribus, the opposite might prima facie be better policy.
A number of points can be made in the context of low carbon policies.
i. Even modest carbon prices may assist attainment of low carbon objectives, even if they remain insufficient to incentivise investment.
ii. An important role for carbon prices is to counteract “rebound effects”, where energy efficiency induces lower costs to the consumer and additional use.
iii. Consistency matters. Inconsistent or incomplete coverage can lead to “leakage” between sectors or geographies, with perverse effects.
Implications for a low carbon power sector. This is a major issue for the system economics calculations around finding the best or even the (most) feasible choices for combinations of renewables, other low carbon sources, and residual fossil plant. There may be fundamental differences between systems and policies that work well for (say) an 80% emissions target, with a substantial but still modest implicit cost of carbon, and those for a 100% target with a very high implicit cost. In the first instance analysis can start just by exploring how sensitive results are to the target or to the carbon price, but conclusions may well remain very sensitive to this parameter.
2.2 The investment problem for infrastructure assets.
This relates primarily to the investment domain, as a problem of infrastructure, policy, regulation and risk. Economists describe it as a “time inconsistency” problem. Ex ante, policy makers want to make sure an investment is made. Ex post, with investments made and costs sunk, political, regulatory and consumer priorities turn rapidly towards demand for lower prices. This applies to energy and other infrastructure investments where costs of fixed assets are recovered directly through charges to consumers.
Conceptually at least, low carbon technologies do not alter the problem. However the possible scale and extent of policy interventions, observed policy changes or reversals, higher costs of low carbon alternatives, and controversy over climate change or energy policy generally, all tend to increase policy, regulatory and market uncertainty. Renewables have been particularly dependent on policy interventions and also vulnerable to market rules. It will also be a near universal issue, except perhaps for the smallest systems. In less developed economies it will be closely bound up with more general social conditions and the strength of institutions.
The solution for developed economies, history suggests, requires a selection from, or some combination of, vertically integrated monopoly, long term contracts for generators with secure counterparties, and Government guarantees. All of these possibilities will have strong implications of a legal and regulatory character, particularly in systems which have moved towards unfettered free market approaches. They pose questions for competition law, “state aids”, regulation, and the construction of workable contracts with reliable counterparties.
COST OF CAPITAL.
It is worth reiterating some of the principles of capital theory as commonly deployed. The conventional CAPM model equates the cost of capital (CoC) to a risk-free rate plus a risk premium. In this context CAPM ignores project specific risk and defines a “beta” risk premium in terms of correlation with the movement of the market as a whole. Since investments in mitigating or adapting to climate change can be deemed to be essential there is no reason to suppose any market correlation. As with public utilities, which also have a very low correlation with the market, and hence a low beta and cost of capital, the CoC should be close to the risk-free rate.
The discrepancy arises from confusions between how to treat risks attaching to construction and commissioning, often of untested technologies with difficult site conditions, which are likely to be high but project specific, on the one hand, and how to treat the risk to the value of the investment once in place. In a regulated context the latter should be very low. Project specific risks, in principle, are diversifiable and so do not add to the cost of capital. They may affect the cost of the project through the incorporation of a contingency allowance in cost estimates.The answer is financing structures that separate the two dimensions of risk. Risk in the pre-commissioning phase may well need to allow for contingencies and appraisal optimism, not relevant to use of a discount rate per se, and this is reflected in the capital value at commissioning. But return on that capital needs to be much closer to the risk-free rate and, on the assumption that regulatory and policy risks can be eliminated, will be much closer to any social cost of capital used in policy appraisals.
But the biggest source of risk for potential investors remains policy and regulatory uncertainty. The cost of this risk is borne by consumers, under any market or regulatory structure; if imposed on investors it will be reflected, if investment is forthcoming at all, in a much higher cost of capital. As this risk sits almost entirely within the control of governments and regulators, its imposition on infrastructure investors is entirely self-defeating in terms of efficiency or value for money.
Analogies. Pension funds, who diversify their “person specific” risk across millions of individuals, are often required to value future liabilities at a risk-free rate, ie government bonds. Similar practice is often applied to nuclear liabilities, which are emphatically not discounted at 10% over 100 years. The reason in each case is that the liabilities are “certain” rather than market correlated risks.
Implications for a low carbon power sector. Responses to infrastructure investment questions define the institutional and governance context for the power sector as a whole. Low carbon policies generally will accentuate moves away from liberalised market regimes, towards more public involvement, government guarantees, long term contracts and regulated monopoly approaches. Law and governance issues may be particularly important in developing country contexts.
This is also critical for the cost of capital (CoC), another key parameter in any technical economic appraisal of the system economics. Appraisals should be based primarily on low discount rates, although we can show the effect of varying CoC assumptions on policy and investment choices.
2.3 Paying for capacity. Zero or negative system SRMC.
This is a very specific problem for generation or capacity investment. Not covering capital costs is a major, intrinsic and largely unresolved problem for SRMC-based wholesale markets, sometimes known as “energy only” markets. The typical question is how to recover the capital cost of fossil peaking plant that operates for a few hours a year, during which it earns an SRMC based wholesale price that is set by its own fuel cost. There is clearly no profit even in the few hours for which it operates. It is a major issue even in fossil-based systems, but it will be accentuated dramatically in low carbon systems; many renewables have zero marginal cost and are often “at the margin”, setting SRMC and hence the wholesale price at zero. Resolution of this issue by external intervention in the market, eg by subsidising additional capacity, further distorts the market, possibly in unpredictable ways, and further undermines the confidence of investors in market based price signals.
“Market” solutions are often assumed. The simplest is assumption of reliance on “scarcity pricing”, sometimes combined with the assumption of a degree of market power by generators. Scarcities produce “price spikes”, attracting instant political and regulatory attention. This reinforces the “time inconsistency” problem, and undermines confidence that investors might otherwise place in a “market price”.
Market interventions to ensure adequate capacity most commonly involve the introduction of a capacity payment. However capacity payments and capacity auctions are emphatically not a pure “market” solution. They are themselves the product of a central intervention, and pose questions of what entity is responsible for making the capacity payments, how much capacity is required, who is to conduct the auction, how is it to be conducted, what should be the contract lengths, how to define and monitor what is being supplied, and a host of questions on how to measure and compare capacity from very different sources (eg wind, nuclear or fossil). In other words this implies that some agent is necessarily required to make a large number of important decisions on generation investment, inter alia bearing on technology choice, and to make substantial financial commitments.
Introduction of capacity markets can therefore be seen, correctly, as a substantial first step towards central direction of investment in generation, and towards the installation of a “single buyer” or central purchasing agency.
Capacity choices determine standards for security of supply
Decisions on how much capacity are closely tied to any setting of standards for generation security (capacity adequacy), by governments and/or regulators. A particular form of intervention, adopted in the UK in 1990, was to set a penalty charge for failure to supply, constructed around a value of lost load (VOLL). This was intended as a minimal intervention, and to mimic how a market might operate under conditions of capacity shortage, with VOLL as the critical parameter in setting the security standard expected by consumers. It was a clever administrative device. Relating capacity to VOLL makes it explicit that decisions on capacity represent a balance between cost and capacity adequacy, with a higher VOLL leading to a higher standard and conversely. In other words the whole issue can also be described in terms of setting standards for generation security.
Implications for a low carbon power sector. “Energy only” or SRMC based markets, cannot properly reward future investment. This is a problem not just for generation per se but for any facility that provides additional capacity or its equivalent. This includes storage, transmission, interconnection and load management, all of which are relevant to the question of integrating renewables.
This problem of wholesale markets also reinforces the general infrastructure investment problem, and provides another more technical argument on the need for market reform, and for institutional structures that do not rely only on market solutions.
However the discussion also shows decisions on capacity adequacy are inextricably linked to the choice of what constitutes the right level of security across the system. This will become a much more complex question with sophisticated technologies for load management, and may reflect much more differentiated levels of capacity adequacy for different applications and different consumers.
But chosen levels of security, or capacity adequacy, will also have a large effect on total electricity costs and hence consumer prices, and on the optimum mix of plant. A higher proportion of capital cost makes this an issue for low carbon systems, and intermittent renewables, that is much bigger than for conventional systems.
2.4 Optimising or coordinating generation under complex constraints.
This is primarily an operational issue, although operational issues can feed back into strategic investment choices, and it has implications for prices and tariffs, and hence for allocative efficiency. Conventional wholesale markets applied to low carbon systems will not deliver either efficient system operations or meaningful price signals. The reason is that most low carbon technologies have cost and operating characteristics that are very different from fossil generation and are incompatible with the assumptions behind current wholesale markets. It may be possible to remedy this problem in certain well-defined conditions, but theoretical considerations suggest it will often not be possible.
This is a novel and largely unexplored problem for the power sector, and deserves analysis and explanation in more depth. The issue is also important because it has the potential to invalidate much of the conventional wisdom on how to organise market structures across the board. It is an issue separate from but additional to that of zero marginal cost and “missing money”, and a deeper and possibly more intractable problem. A short further exposition is provided in the Annex.
Merit order operation is unlikely to be an adequate approximation for efficient operation of future low carbon systems, which incorporate substantial inflexibilities (eg nuclear plant and possibly CCS), substantial storage or time shifting options (including the demand side), and stochastic supply (eg wind power). These are new issues in that they are accentuated in systems that do not have the flexibility of fossil plant, where both storage and demand shifting play a large role, and where there is an increasingly important stochastic element. It is often claimed for example that the UK’s BETTA market has penalised less flexible plant, not necessarily fairly.
These complications predispose to a contract based command and control system in which a system operator (SO) seeks to optimise and balance choices, including those related to system security. A later section argues that there is a counterpoint to the strengthening of existing real time “command and control” by the SO, in the form of much more consumer involvement through innovative approaches to competitive supply, and a larger role for retail competition.
Merit order type wholesale markets have been the intellectual and practical lynchpin of liberalised electricity markets as a whole. Loss of easy to understand wholesale markets will make it much harder for generators to anticipate the likely operating regimes for their plant or to have confidence in its fairness. Combined with the zero wholesale price issue, this will make generators much more likely to rely on contract terms both to cover capital costs and to govern their operating requirements.
Implications for a low carbon power sector. This issue is sufficiently important to the operations of a low carbon system with renewables that it deserves more analysis both at a theoretical level and empirically. One large unknown is exactly how operational choices will be determined in complex low carbon systems, and what role might be played by different markets. Loss of confidence in conventional wholesale markets as the lynchpin of commercial structures will also have significant regulatory and wider implications.
2.5 Coordination of the investment mix.
This relates primarily to the investment domain but in a context that has to take into account future balancing and compatibility problems in system operations.
In a fossil fuel dominated generation sector, all plant is, at least in the essential respects of how it responds to being part of a complex power system, very similar. Even so, and even within market driven systems, some limited coordination, often in relation to transmission, has been implicit, eg in regular forward looking statements from the UK’s National Grid.
However whether markets can in future sort out coordination issues to provide workable combinations of technically very different forms of generation is a more open question. There is for example substantial debate over practical problems in managing systems with substantial volumes of intermittent wind and/or inflexible nuclear. Moreover potential new or greatly increasing loads, like battery charging or storage heating, also transform the possible choices, but their impact depends heavily on exactly how these demand side options themselves are developed. These issues are widely recognised, for example in a recent call by the Institution of Engineering and Technology (IET) for a “system architect”. Choice is not just about comparing costs, but about technically feasible and economic combinations.
Implications for a low carbon power sector. From a system economics perspective the general task of finding the right balance of investments, including those necessary for the future penetration of electricity in new applications, is the central problem. The box below describes some typical questions.
Low carbon power systems raise new coordination issues. Items on the agenda are:
Siting of wind turbines. Geography has a major effect on the economics of a wind facility, most obviously in the quantity and reliability of the wind and in the cost of construction, particularly offshore with different depths and sea conditions. However “firm” wind capacity in aggregate, and hence reliability, is improved through geographical diversity, so that concentration in a few favoured locations is not necessarily the best choice.
Question. What can we say about the parameters on diversity, in relation to costs, feasibility, etc, and how much does this depend on the load characteristics?
Question. What percentages of intermittent generation are consistent with workable power systems, and how does this relate to differing combinations of other plant, load management, storage, etc.
Solutions, in terms of demand management, storage and generation, need to bridge both diurnal and seasonal variations.
Question. Are different forms of storage better located locally or even with individual consumers, or centrally to capture economies of scale with a grid location and under direct system operator (SO) control?
Question. What are the desirable characteristics, in technical and economic terms, of storage options for different applications?
Electric (or hydrogen) vehicle development will have a profound influence on storage needs and the potential for load management through possible battery charging regimes. The extent to which re-charging is on domestic premises, or “rapid re-charging” at service stations, will have major implications for load balancing, the mix of generation, and reinforcement of local distribution networks. In either instance there is a potential requirement to “manage” load by some combination of price or administrative means (advance booking).
Question. What are the potential profiles of this transport load, and how should they be managed or influenced?
Very similar issues will arise for electricity used to heat buildings, with heat pumps and storage heaters posing new questions.
Question. What will the profiles look like in cold countries? And what does this tell us about the economics of seasonal storage?
Question. Is there a rationing or compulsion element in the heat sector, eg how many households can be allowed access to “off-peak” night storage tariffs, and will there be local distribution network limitations on heat pumps?
2.6 Communicating cost structures as tariffs and incentives for consumers.
This starts as an issue of allocative and operational efficiency. Market structures in general do not provide for complex interactions between choices on supply and storage, and demand side options open to smaller consumers. In the UK this arose from adoption of load profiling. With load profiling, all consumers of a particular type are assumed to have the same time profile in their consumption pattern, implying a homogenous mix of peak/ non-peak, day/night and winter/summer loads. The supply business is then essentially commoditised. All suppliers provide the same product, with differentiation only on price. This undermines, or rather excludes from the market, any competitive benefit from offering consumers a truly differentiated service. Profiling inhibited UK development of sophisticated metering and control systems and tariffs, arguably for a generation.
Nor are these features of future low carbon systems currently accommodated in conventional vertically integrated utilities, still constructed mainly around the technical features of fossil generation and a “predict and provide” approach to determining capacity needs. Future low carbon scenarios will require cost and price signals to be consistent, cost reflective and supportive of demand side management. This will be true both in terms of aggregate supply and demand in national systems and, increasingly also at smaller scales, eg within local distribution networks (LV for the power sector).
If markets are to produce efficient outcomes then they must contain the means for complex cost structures and network/ system characteristics, in production and distribution, to be combined and integrated with the right incentives for the many forms of decentralised contribution to balancing and stabilising the system. This includes appropriate tariffs for consumers engaged in demand management, for “prosumers” (who both produce and consume), and “prostumers” (who produce, consume and store). However even simple versions of “real time pricing” face major practical obstacles, mainly because consumers generally show little appetite for time or effort intensive economic calculations in relation to what they will continue to regard as an “on demand” utility service. There is also potential dynamic instability if simple peak load pricing just shifts peak to another period rather than managing and shaping load. A more sophisticated and radical approach may be required.
Developments in communication and control technology have created an explosion of possibilities in metering and service provision. They allow for the application of sophisticated TOD metering – even real time pricing – previously seen as impractical or impossibly expensive, as well as sophisticated approaches to remote control of individual appliances. Given the interactive nature of these possibilities, utilities need to consider how end use should be incorporated into processes for the secure and efficient operation of the system.
Distribution engineers have identified further questions for the medium and low voltage distribution networks, if they have to handle substantially greater volumes of electrical throughput, with new types of load, and possibly with very different dynamic characteristics. So there are likely to be future problems of local load balancing, as well as aggregate demand/ supply management at higher or national system levels. These too may need to be reflected in the design of network use of system charges and in control measures to manage local stability. Network tariffs are another major area where cost recovery may have to be done differently in future, and not simply by averaging total costs over total units sold.
There are happily many technical and other options for managing these interactions more effectively while at the same time moving towards a “service” rather than a commodity approach. These include the automation of responses within individual appliances on the consumer premises, to respond either to price signals or to remote control by a supplier/provider. These are discussed in more depth below.
New Models for Retail Supply to Consumers
The conventional utility model has consumers able to treat electrical energy supply as “on tap”, with limited or no differentiation between applications (e.g. as between lighting, heating or mechanical power). Tariffs and prices for the most part approximate to an averaging of the costs of supplying electricity, with limited ability to differentiate on grounds of differing incremental costs.
One means to meet the challenges above is to redefine the “consumer offering”, defining electricity as a set of services, rather than a homogeneous commodity. This requires starting with a clean sheet in defining the nature of the services that consumers will want, and the basis on which they pay. So, for example, a consumer wanting to charge electric vehicle batteries might request 75 kWh to be delivered in a specified period, over several hours or even several days (eg a weekend), and the consumer’s contract might specify that this requirement will be met in full but with timing that is “at the supplier’s discretion”. Corresponding but different arrangements could apply to the purchase of power for heat, for refrigeration, and some other uses, reflecting in each case the nature of the load. Commitments to individual consumers would be made by energy service companies who would be able to aggregate consumer requests and feed them in as part of the SO’s system optimization routines. Such services might even be packaged with the provision of appropriate equipment (eg storage heaters).
The role of suppliers is then to act as aggregators, and their essential function will be to manage the complex interaction between consumer loads and system balancing requirements, including shaping and managing the pattern of consumption. This provides a major opportunity for a much more innovative approach to all aspects of metering and for the terms on which consumers purchase power. Suppliers could at the same time enter into individual contracts with generators, or a system operator or other central agency, which would reflect the economic benefits of their ability to shape consumer loads. They would also take responsibility for managing loads within network constraints at lower voltages, ie within local distribution networks.
This would have some powerful advantages. First it would allow consumers to purchase power for particular usages in ways more akin to their purchase of other goods and services, as opposed to perpetuating the “instantaneous commodity” characteristics that have hitherto been a unique and constraining feature of the power sector. This would also correspond to what consumers actually want and need from the utility. At the same time it would help make the services more affordable. Consumers could still choose to take some power “on tap” and would normally pay a higher price for this. But many of the issues associated with administrative setting of security standards would become much less significant. To a much greater extent security levels would be chosen in a market.
This change is enabled by one set of technologies – those that surround metering, separate identification of different loads within each consumer household or business, and remote control. But it also helps to resolve the problems posed by another set of technologies, those linked to intermittent or inflexible sources of non-fossil and distributed generation.
The major new loads of particular interest in this context are transport and heating, since each may pose particular, but different, sets of problems. In the UK, the ETI has suggested that there may be a need to add a hydrogen economy to this list.
Penetration of electricity into the transport sector has been widely assumed to be a means of flattening load shapes and managing intermittency, but a Norwegian experience illustrates that it may create some new questions of its own, both locally and nationally.
Heating loads may be generally assumed to be manageable in a relatively straightforward way in terms of a diurnal load shape, if one assumes significant on premises potential for short term heat storage, but have additional characteristics that deserve attention. These include the sheer scale of potential heat load, its strongly seasonal character, and the potential for a big accentuation of the peak in response to cold weather, a problem that is particularly acute for heat pumps.
This raises questions about the appropriate level of supply security for heat load (with possibly a very high cost attaching to sufficient capacity margins of low carbon plant), and also of possible needs to determine quotas of different types of heating at local level, in order to manage load on local networks.
Implications for a low carbon power sector. The most significant relate to the nature of the “deal” for consumers, and how this can be made consistent or compatible with the need to balance the system as a whole, whether this is at an aggregate grid level or at much lower voltages.
New approaches are likely to reflect the many developing options for real time metering, automated controls, and remote control over loads, pointing towards a “service provision” approach rather than current models for delivery of a commodity. But this leaves many questions and options.
1. The basis for sale to consumers may reflect some combination of:
· more sophisticated time of day tariffs
· different tariffs/ contract terms for different loads/ household circuits
· automated responses for particular appliances
· remote control by an aggregator or system operator
· “administrative rationing” for certain loads, eg booking a time slot for battery charging
· different guarantees of reliability of supply for different loads
2. Suppliers may have “aggregation” of loads as an important function, or there may be a reversal of 21st century trends, with the supply function returning to local distribution companies.
3. The basis of network (use of system) tariffs need to change. Traditional averaging of total costs over all kWh will need to give way to a much more cost reflective approach, in which retail consumers pay for the actual service that they value from the network (local or grid), ie as back-up or for peak loading.
 Planetary Economics, Michael Grubb, 2014
 The quote is attributed to Michael Grubb, but is no more than a simple re-statement of a fundamental idea in economics.
 €15/ tonne may be sufficient to induce early gas for coal substitution in existing plant, for example. One claim made for the EU ETS is that companies are now basing their planning on an expectation of much higher carbon prices, eg. €40/ tonne, even if these are not sufficient to induce the major generation investments. Both these numbers were quoted in a recent address by the Director of the Commission’s Directorate for Climate Change Action
 In relation to industrial energy prices, this is often expressed as concern over international competitiveness
 A recent example has been withdrawal of CCS funding in the UK.
 For various reasons, such as perverse incentives for renewables or actual costs of reducing output for nuclear, prices in these markets can even be negative.
 Indeed this has often been expressed as a serious criticism of such proposals by proponents of fully liberalised markets. It is an issue at EU level where the Commission has generally supported “energy only” markets, which of course avoid the issue of setting a Community wide security standard.
 It has been suggested that this is also a factor that should influence choice of CCS technology. Post consumption capture has been claimed to allow for more flexible operation of plant. Operating inflexibilities will tend to be associated with the chemical process aspect of CCS rather than combustion/ generation per se.
 The CALMU credit and load management unit was pioneered by Fielden and Peddie (then an Area Board Chairman) in the 1980s, and has enjoyed worldwide success. It died in the UK with privatisation and the adoption of profiling.