Governments around the world (from the UK1 and the EU2 to the USA3 and China4) agree that smart energy infrastructure will play a major role in meeting their energy objectives. Many of these energy objectives relate to common or public goods such as (technical) reliability, environmental sustainability and (political) security of power supply.
In the EU only Italy, Sweden, Finland and Denmark have
reached high levels of smart meter penetration yet.
Overall it seems that everyone – including all market players – would have something to gain from the use of smart technologies. Nevertheless, the take-up of smart technology has been slow. For example in Europe, a Union of Electricity Industry (EURELECTRIC) survey at the end of 2010 found that the majority of EU Member States have yet to reach wide spread roll-out of smart meters.5
If all market players stand to gain from such a technological improvement, why aren’t we seeing greater, more rapid take-up of smart technologies in Europe?
The benefits of being smart
A smart grid is generally understood as deep integration of Information and Communication Technology (ICT) services into the electricity value chain.
In a fully unbundled electricity market, consumers buy their power from utility providers, who in turn buy power on the wholesale market from generators. The transmission of power between market participants is managed by the transmission system operator (TSO). The TSO (or some other independent entity, such as the electricity exchange) also organises the dispatch of generators and oversees the stability of the grid. In reality the distinctions between market players are often fuzzy because of vertical integration of varying depth from generator over TSO to utility.
Wholesale prices in the electricity market are essentially scarcity prices set in a uniform price auction by the marginal plant that needs to be dispatched to satisfy total demand. Cost-efficient and reliable power plants (driven by water, coal or nuclear fuel) provide base load power, while more expensive but flexible plants (mainly gas powered CCGT6 ) are used to provide additional peak load power and to step in for fluctuation from less predicable energy sources such as wind and solar. The less generation capacity is idle, the more expensive electricity becomes.
However, without smart grid technology, the price signals of the wholesale market – e.g. price spikes during peak hours when electricity is scarce – cannot be passed on to consumers in the retail market. By contrast, in the smart grid, fine-grained information about demand and supply, about equipment availability and about pricing will be gathered, communicated and processed by all market participants. Thus, with deployment of smart grids and smart meters, consumers would receive real-time price information. Consumers are then able to employ so called (automated) demand side responses (DSR), which take advantage of real-time price information to time energy consumption and thus lower overall expenditure on electricity.
From an industry perspective the smart grid will improve the efficiency and reliability of energy supply, as the TSO and utilities will be able to detect and address problems and failures much earlier. Remote control and maintenance will also generate cost savings. The otherwise challenging integration of additional, heterogeneous generators into the grid will become easier. This will be important when generation becomes more distributed with the rise of small alternative energy generators (such as solar and biofuel) and facilities which will be consumers of power at some times and suppliers (generators) at others.
Detailed information about electricity demand provided by the smart grids may better inform generators’ investment decisions for new power capacity. Incumbents and new entrants will be able to precisely target demand with local generation, non-conventional energy sources and energy storage solutions.
If you’re so smart …
Obviously, investment in smart technology is costly, and will only be undertaken if benefits exceed costs. Because many of the benefits are external to decision makers – in particular environmental benefits and potential future innovation enabled by the smart grid are not internalised in investment decisions – one should expect there to be less investment in smart technologies than is socially optimal. It should not therefore be surprising to find that “there is currently a considerable gap between present investment and optimal investment in smart grids in Europe”, as has been noted in the EC’s Communication on Smart Grids.7
However, even ignoring the wider societal benefits of improving energy efficiency, which do not necessarily affect individual consumers’ and producers’ decisions, the cost saving potential of smart grids in combination with DSR should be sufficient to encourage investment in the new technology. Yet, there seems to be a reluctance of all market participants to become smarter.
One important factor that might help explain this is that the full benefits from the use of smart metres will only be enjoyed when utilities provide real time tariff information that allows consumers to apply effective DSR. However, utilities may only invest in smart grids and offer real time tariffs if they observe that take-up is sufficiently high. In addition, effective DSR will also require some amount of automation, e.g. via household appliances with built-in DSR capability, but the development of such technology requires sufficient demand and certainty about technical standards.
All of the parties involved – utilities, consumers, manufacturer of DSR-capable appliances – would benefit from economies of scale, yet none of these parties may have sufficient incentives to make the initial investment to create the network of compatible technologies and services in the first place. The need for co-ordination in order to reap the full benefits in the presence of network effects is thus yet another reason why parties may be reluctant to make investments.8 This means that regulatory intervention and incentives such as public funding of roll-out of smart metres or a clear mandate on roll-out of smart metres may be necessary in order to start the bandwagon.
Of course, such intervention needs to be alert to the risk of picking or pushing the market towards the wrong technology, which in the best case would be a waste of resources and in the worst case may frustrate further investment for years to come. In Sweden for instance, where the government’s focus was on smart meter roll out9 rather than smart grids, meter readings were transmitted via power lines and out-dated technologies hence measurements were often delayed and real-time tariffs not made available to consumers. Gabriele Riedmann de Trinidad, head of Deutsche Telekom’s business area for energy notes that “most investment made in smart metering (in Sweden) will need to be removed and new equipment installed.”10
Choosing how to intervene is likely to be tricky, however, it is clear that – at the very least – the regulatory environment should be (a) conducive to supporting risky investments and (b) respectful of the need for co-ordination between market participants.
Investment in smart grids in particular is largely sunk thus irrevocable and subjected to high levels of risk driven by uncertainty over technology standards, eventual take up and success. Regulation is only applied in the case that the product is successful, thereby curtailing up-side returns and creating an asymmetric risk as illustrated below.
In order to avoid creating such asymmetric risk, the instances of failures have to be considered as counterfactuals in order to determine the appropriate rate of return for such a risky investment. For instance, imagine a firm with a Weighted Average Cost of Capital (WACC) of 11%. In the case that the investment is a success, the firm would make a 11% return. However if it is unsuccessful, the firm loses the full sum of its investment. If the firm was only allowed a regulated rate of return of 11%, this firm would not invest as the expected rate or return on investment given the counterfactual of failure cases would fall below its WACC of 11%, which would be loss making.
Therefore, if returns are regulated down to the cost of capital, TSOs would have a strict incentive to wait and see, as this serves to curtail downside risk to the TSO. The TSO will only invest when it observes a positive state of the world with substantial upsides. As the TSO enjoys an option value from a wait and see approach, investment in smart grids would be delayed.
From the suppliers’ perspective, a key issue is the question of whether they will be able to recoup their investments in smart grid infrastructure. TSOs and utilities would have to invest in the roll out of smart grids. They will need to develop and install smart meters and communication networks (via fibre, wireless or power cable) as well as IT systems to process and interpret the raw data. Owing to the infancy of such technology and systems, the lack of industry standards and consumer inertia to take-up smart energy products, such investments are very risky. TSOs and utilities potentially face substantial downsides, and investment will take place only if there is a prospect of a correspondingly large upside to compensate investors for the risk they take.
Yet, an allowance for risk premiums is not widespread under current European electricity regulatory frameworks. In fact, existing regulatory frameworks in the energy sector are typically focused on managing competition rather than encouraging investment and innovation. Traditional price cap regulation for instance may constrain investment incentives as there is a time delay before new capital expenditure (CAPEX) on investment is, if at all, incorporated into a firm’s regulatory asset base in the new charge control period. In addition, it is also uncertain if research and development expenditure would be accounted for in an efficiency analysis. A rate of return regulation on the other hand may narrowly focus on reducing operating cost rather than encouraging increase expenditure on investment in order to improve long-term efficiency.
Indeed, the EURELECTRIC survey found that the lack of investment in smart grids is a result of:
- “Sub-optimal rates of return and regulatory instability are hampering investment in smart distribution grids;
- The roll-out of smart meters is being delayed by a lack of clarity regarding the roles and responsibilities of individual market players;
- Regulators are taking a narrow view when evaluating cost efficiency, penalising extra expenditure on R&D or smart grid pilot projects and encouraging business-as-usual expenditure instead.”11
Moreover, in addition to being extremely risky, investments in smart technology are also largely sunk. This means that there is potentially a substantial benefit to delaying investment, as every investment decision implies giving up the option to change or modify the investment plan in the future. Even for investments that are NPV-positive, investors may prefer to adopt a wait-and-see strategy and delay their investment because of the option value associated with doing so.
Even without looking at the curtailing of returns, the upside of investing in smart grid infrastructure is sometimes smaller for incumbents than one might think – in particular if they can simply pass on higher cost associated with inefficiencies in generation and distribution to consumers.
For example, incumbent generators generally have long planning cycles for generating capacity and it is likely that they have already invested in future capacity based on current predicted peak demand. Therefore, if smart technologies were to take off rapidly and cause a significant shift in load distribution, generators may end up with stranded assets.12 There may need to be a regulatory mechanism for dealing effectively with such asset stranding. Where TSOs are vertically integrated, the desire to protect the generating business from more competition might slow down investment. For overcoming these impediments to investment it is necessary to deal effectively with such distorted incentives and to make sure through appropriate pricing mechanisms in the wholesale market that generators do not enjoy a position of market power in terms of their impact on prices.
Even though the potential consumer benefits from the use of a smart meter may often exceed the cost of the meter, smart metering seems to suffer from the same problem as the energy saving light bulb. Despite potential benefits there is a lack of consumer interest and take-up. The European Commission argued that its ban of incandescent light bulbs would directly reduce consumers’ energy bill, but also create sufficient demand for the prices of fluorescent bulbs to decrease and new lighting technologies to be developed . As in the case of the energy saving light bulb, this may be driven by a combination of low awareness, myopic behaviour, strong concerns over the novelty of the technology and relatively high upfront costs. There is the potential risk of investing in equipment that is incompatible with the technology that will eventually prevail, and will then have to be written off.
Overcoming these problems may require that utilities rather than consumers make the investment, and recover the cost through higher consumer charges over time. They will only do this, however, if they can be sufficiently certain about their pay-back. This in turn may mean that they might need to be permitted to sign up consumers for longer-term contracts, with some arrangements in place for the case where customers want to switch to another supplier earlier in their contract (e.g. termination payments from the customer, or transfer payments from the new supplier). Of course, such arrangements may somewhat go against the grain of trying to lower consumer switching costs, but they may be necessary in order to support risk sharing which in turn is needed to overcome co-ordination problems.
Specific concerns about the privacy of information collected from smart meters adds further to the barriers of take-up. These concerns will have to be addressed to establish consumer trust in the technology.
Annex I.2 of the Electricity Directive requires that Member States define an implementation plan and timetable for the roll-out of smart metering systems by September 2012.13 TSOs in Sweden, Finland, Italy, France and Spain have a legal mandate to roll out smart meters. While this is missing in other Member States and in Norway, other means of achieving smart meter roll-outs are also being considered. In September 2011, the UK Department of Energy and Climate Change announced a tender worth up to £4.5bn for the roll out of 30 million smart meters across the UK within 4 years. An end of 2012 deadline to develop standards for implementation of smart grid services has been set by the March 2011 Commission Mandate to European standardisation organisations.14 With the current Directives and motions in place, one should expect greater roll-out and take-up of smart meters in the near future.
However, creating legal mandates and obligations – both in terms of rolling out smart grids, and developing standards – are only one side of the coin. At the same time, the regulatory environment needs to provide the right incentives for those who have to make the investments. In tackling the associated regulatory issues, the Electricity Directive and the Energy Services Directive set out to move Member States away from regulatory regimes that encourage ‘volume-based’ business models to one that yields quality and efficiency improvements.15 Specifically Article 10(1) of the Energy Service Directive requires Member States “to remove any tariff incentives which unnecessarily increase the volume of transmitted or distributed energy”.16 Regulators across Europe are beginning to adapt their regulatory frameworks in order to encourage investments in new technologies – Portugal, Great Britain and France combine a revenue cap regime with a planned cost approach (CAPEX agreed at the start of the regulatory period thereby removing the risk associated with whether the CAPEX on investment would be incorporated in the charge control) while Austria allows an additional mark up over Weighted Average Cost of Capital (WACC) if book value (capitalisations post 2009) increases (essentially allowing a higher return if new investments were made).17
Supporting risk-sharing arrangements amongst the parties – potentially across the value chain – might be helpful in order to provide investment incentives. Such arrangement might take a variety of forms, such as co-investment schemes that allow parties to jointly take-up the risk, or flexible contract terms and long-term contracts that provide greater certainty over the upside of investment.
Such arrangements might of course involve some preferential treatment of certain counterparties, which means that general non-discrimination obligations need to be applied carefully. On one hand, one needs to acknowledge that volume commitment or co-investment may justify different terms and conditions. At the same time, however, one needs to be alert to the risk that risk-sharing arrangements might be used in order to foreclose the market and could distort competition to the detriment of consumers. Using standards for exclusionary purposes might, for example, fall into the latter category.
The regulatory regime should also be conducive for the transition from old to new technology standards. Where price signals are relied upon to encourage the transition to smart technology, this could require a temporary deviation from cost based price regulation for the legacy product.
Given a conducive regulatory framework, private incentives of market players should drive future industry development once teething issues such as industry standards and uncertainty over take-up have been overcome.
- Department of Energy and Climate Change, 27 July 2010, ‘Annual Energy Statement’. [↩]
- Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, 12 April 2011, ‘Smart Grids: From innovation to deployment’, COM(2011) 202 final. [↩]
- Executive Office of the President, National Science and Technology Council, June 2011, ‘A Policy Framework for the 21st Century Grid: Enabling our secure energy future’. [↩]
- Premier Wen Jiabao, 5 March 2011, ‘Report on the Work of the Government’, Fourth Session of the Eleventh National People’s Congress; see also press release of 14 March 2011, ‘China sets to build smart grid to tap renewable energy’. [↩]
- EURELECTRIC, 2011, Regulation for Smart Grids, D.2011/12.205/3. [↩]
- Combined cycle gas turbine. [↩]
- EC Communication, ‘Smart Grids: From innovation to deployment’, COM(2011) 202, page 4. [↩]
- The solar industry was facing a similar ‘chicken and egg’ problem where, according to Deloitte, lower costs were necessary to stimulate demand but were only achievable through economies of scale or market subsidies (Deloitte, 2009, ‘Solar’s Push to Reach the Mainstream’. [↩]
- Sweden was the first country to reach 100% penetration of smart meters in July 2009. [↩]
- Guardian Article, 13th April 2011, ‘Five destinations embracing smart meter technology’. [↩]
- EURELECTRIC, 2011, Regulation for Smart Grids, D.2011/12.205/3, Executive Summary. [↩]
- Alternatively, generators might refrain from investing now because of such uncertainties, which might lead to underinvestment and shortage of capacity (thus higher electricity prices). [↩]
- EC Communication, COM(2011) 202, page 9. [↩]
- The European Committee for Standardization (CEN), European Committee for Electrotechnical Standardization (CENELEC) and European Telecommunications Standards Institute (ETSI). [↩]
- Directive 2009/72/EC of the European Parliament and of the Council concerning common rules for the internal market in electricity and Directive 2006/32/EC of the European Parliament and of the Council on energy end-use efficiency and energy services. [↩]
- European Commission, Energy Services Directive, 2006/32/EC. [↩]
- EURELECTRIC, 2011, Regulation for Smart Grids, D.2011/12.205/3, Section 4.1. [↩]