Development of Best Practice Recommendations for Smart Meters Rollout in the Energy Community

Transcription

1 Development of Best Practice Recommendations for Smart Meters Rollout in the Energy Community Customer KEMA International B.V February This report was financed by the Energy Community.

2 Development of Best Practice Recommendations for Smart Meters Rollout in the Energy Community By order of: Am Hof 4, 5th floor A-1010 Vienna, Austria Submitted by: KEMA International B.V., Utrechtseweg 310, 6812 AR Arnhem, The Netherlands Authors: David Balmert, Daniel Grote, Konstantin Petrov Bonn, February 24, 2012

3 Table of Contents 1. INTRODUCTION SMART METERING DEFINITION AND FUNCTIONALITIES Definition of Smart Metering Description of a Smart Metering Infrastructure The Meter and Associated Devices In-House Display Communication and Data Processing Infrastructure Consumption Feedback Mechanisms Minimum and Optional Functionalities RELEVANT EU FRAMEWORK AND EXPERIENCE EU Legal Framework and Requirements Overview of State of Smart Metering in the EU Experience in selected EU Member States Italy Sweden Germany Austria United Kingdom POTENTIAL BARRIERS FOR SMART METERING DEPLOYMENT Consumer Resistance Legal/Regulatory Barriers Revenue / Tariff Setting and Incorporation of Costs of Smart Metering Implementation of Time-of-Use Pricing Use of Standard Load Profiles Other Technical Regulation Smart Meters Rollout in the Energy Community Page 2/107

4 4.3 Economic Barriers Technical Barriers STRUCTURE AND SET-UP OF A COSTS-BENEFIT ANALYSIS Definition of Costs and Benefits Potential Costs of Smart Metering Benefits to Network Operators Benefits to Suppliers Benefits to Consumers Benefits to Society Set-up of Cost-Benefit-Analysis Definition of Input Parameters and Assumptions Definition of Smart Metering Roll-Out Scenarios Capturing of Costs and Benefits Calculation of Net Benefits Sensitivity Analysis Important Determinants of a Cost-Benefit Analysis International Experiences with Cost-Benefit-Analysis on a Smart Metering Roll-Out MARKET MODELS AND REGULATION FOR METERING Metering Market Models Basic Metering Market Models Pros and Cons of Metering Market Models Deployment Strategies Deployment Speed and Penetration Ratios Voluntary Smart Metering Roll-Out Mandatory Smart Metering Roll-Out Role of Regulation Ensuring Overall Efficiency and Security Smart Metering in the Price Control Process Smart Meters Rollout in the Energy Community Page 3/107

5 6.3.3 Tariff Setting Enhanced Regulatory Performance NEW SERVICES Innovative Tariff Schemes Home Automation Services Consumers as Active Market Participants Other New Services ROLL-OUT PLAN FOR SMART METERING DEPLOYMENT SUMMARY Appendix 1: Major Legal References Italy 106 Sweden 106 Germany 106 United Kingdom Smart Meters Rollout in the Energy Community Page 4/107

6 1. INTRODUCTION The aim of this project was to develop a set of best practice recommendations for rolling-out smart metering in the Energy Community (EnC), taking into account recent experience from the European Union as well as the specifics of Energy Community Contracting Parties. The Energy Community Contracting Parties are obliged to implement the European Union's acquis communautaire on electricity, gas, environment, competition, energy efficiency and renewables. Provisions on the roll-out of intelligent metering systems are in particular set out by Annex I of Directives 2009/72/EC and 2009/73/EC requiring Member States of the European Union to install by 2020 intelligent metering systems for electricity consumption for at least 80% of customers where such a roll-out is assessed positively, and to prepare a timetable to implement intelligent metering systems within 10 years. For gas consumers no fixed targets are stipulated. Implementation of such metering systems is required and a timetable should be set up, however a firm time horizon is not provided. With the Energy Community's decision from October 6, 2011, the Third Package (Directives 2009/72/EC, 2009/73/EC and associated regulations) is to be implemented in the Energy Community's legal framework. 1 According to the Ministerial Decision, the cost-benefit assessments regarding smart metering deployment as specified in Annex I of the Directives are due by January 1, 2014 (Article 6). The same timeline for a roll-out after a positive assessment of costs and benefits deployment of intelligent metering systems to 80% of consumers by 2020 however applies to the Contracting Parties of the Energy Community as well as for EU Member States. As mentioned in the terms of reference (ToR) of this project, the recently created review of smart metering roll-out for electricity in the Energy Community 2 shows that large-scale implementation of smart metering has not yet taken place in the Energy Community and that all Contracting Parties still need to carry out the economic assessment of the long-term benefits and costs of smart metering implementation. Smart metering is mainly perceived to be an electricity topic. Due to the properties and consumption patterns of electricity, this is certainly the most promising area of application for smart metering. However, it can also be relevant for other network industries such as gas, water and district heating. EU Directive 2009/73/EC explicitly stipulates the requirements for Member States to draft a roll-out plan also for gas smart metering. Although not providing a deadline, the 1 Decision of the Ministerial Council of the Energy Community, D/2011/02/MC-EnC: Decision on the implementation of Directive 2009/72/EC, Directive 2009/73/EC, Regulation (EC) No 714/2009, Regulation (EC) No 715/2009 and amending Articles 11 and 59 of the Energy Community Treaty. 2 Energy Community Regulatory Board (2010): A Review of Smart Meters Rollout for Electricity in the Energy Community; available at: Smart Meters Rollout in the Energy Community Page 5/107

7 Commission states that this should be done within a reasonable timeframe. In only a couple of EnC Contracting Parties, gas plays a role as energy carrier in the household sector. The relevance of smart metering for gas thus needs to be assessed in each case separately. Additionally, successful smart metering deployment for other energy carriers is dependent on the market structure (if, for instance, both commodities are supplied by the same or affiliated network operators, or if metering for both belongs to the same metering service provider). The report at hand is mostly focused on electricity, in many cases though, results are also valid for gas. In line with the objectives of this project as specified in the ToR, the report is structured in the following way: Chapter 2 includes an overview of different terms subsumed under intelligent metering, provides an unambiguous definition of smart metering and describes the typical set-up of a smart metering infrastructure. Chapter 3 describes in brief the relevant EU legal framework and gives an overview of the state of smart metering in the EU in general and in five example countries in particular. In chapter 0, potential barriers for successful smart metering deployment are highlighted and discussed. Chapter 0 discusses the costs and benefits of smart metering and the approach in setting-up a cost benefit analysis. The chapter provides some purely indicative cost and benefit estimations as seen for example in actual cost-benefit studies. Chapter 0 describes the different models for the metering sector and discusses the role of regulation in smart metering as well as the role smart metering could play for regulation. Closely linked to the benefits of smart metering are the new services which may emerge from deployment, as is discussed in chapter 7. Chapter 0 describes the general structure and characteristics of the smart metering deployment schedule. Chapter 0 concludes the paper by summarizing the recommendations for smart metering roll-out in the Energy Community. The report is accompanied by a template for the scope of a country-wide cost-benefit analysis, showing the general set-up of a cost-benefit analysis, the steps to perform and the resulting deliverables; the template is supplied separately. Smart Meters Rollout in the Energy Community Page 6/107

8 2. SMART METERING DEFINITION AND FUNCTIONALITIES 2.1 Definition of Smart Metering The term smart metering is used in a variety of ways. The relevant EU Directives 3 promoting the deployment of smart metering use the term intelligent metering, further technical terms often used in parallel are Automated Meter Reading (AMR), Advanced Metering Infrastructure (AMI) or Automated Meter Management. AMR AMI AMM Smart Metering Enhanced Functions and Services Figure 1: Smart Metering & Co Source: KEMA In order to provide clarity for any discussion or public consultation about the design of a smart metering system and the subsequent roll-out decision it is necessary to provide an unambiguous definition of smart metering. As the term itself can be and is used in a broad context, it is recommended to define smart metering through its components and through the functionalities provided by such a system. This shows that there is a clear distinction to be made between a smart meter and smart metering. The smart meter is the individual appliance installed at an energy consumer s house or facility, primarily metering the consumer s energy consumption. Smart metering is an actual application of smart meters on a larger scale, i.e. the application of a general principle rather than an individual appliance. 3 The most relevant Directives would be Directive 2009/72/EC of the European Parliament and of the Council of13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC, and Directive 2009/73/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in natural gas and repealing Directive 2003/55/EC, cf. chapter 3.1. Smart Meters Rollout in the Energy Community Page 7/107

9 As stated above, the EU legal framework refers rather to intelligent metering systems. The European Commission's Interpretative Note on Directive 2009/72/EC provides a description of the Commission's understanding of such a metering system as defined by the ability to provide bi-directional communication between the consumer and the supplier/operator and to promote services that facilitate energy efficiency within the home. 4 Current discussions at European level focus among others on defining a set of standard functionalities. The European Regulators' Group for Electricity and Gas (ERGEG) published Guidelines of Good Practice (GGP) where a set of typical functionalities of a smart metering system is included. 5 Another approach to defining standard functionalities for smart metering is taken through a standardization mandate of the European Commission to standardization bodies 6. The ERGEG GGP explicitly aim to be in line with the official standardization process through this mandate. In line with the general direction taken by the debate in the European Union, we would like to define smart metering in general as follows: Automatic reading, processing and transmission of metering data, Possibility of bidirectional data communication in real-time (or with only a small time lag), Support of additional services and applications, e.g. home automation, remote (dis-) connection of supply or load limitation, and Remote update of meter firmware to enable new services, communication protocols, etc. The definition should be taken as a starting or orientation point for the debate. It may evolve over time (it most certainly will) and it may differ between countries, (pilot) roll-out projects, etc. The next sections describe the typical set-up of a smart metering infrastructure and give an overview of the typical minimum and optional smart metering functions. 4 Interpretative Note on Directive 2009/72/EC Concerning Common Rules for the Internal Market in Electricity and Directive 2009/73/EC Concerning Common Rules for the Internal Market in Natural Gas Retail Markets, Commission Staff Working Paper, Brussels, 22 January 2010, p. 7 5 EREGEG, Final Guideline of Good Practice on Regulatory Aspects of Smart Metering for Electricity and Gas, Ref: E10-RMF-29-05, Brussels, 8 February European Commission, Standardization mandate to CEN, CENELEC and ETSI in the field of measuring instruments for the development of an open architecture for utility meters involving communication protocols enabling interoperability, M/441 EN, Brussels, 12 March Smart Meters Rollout in the Energy Community Page 8/107

10 National regulatory authorities 7 should provide an unambiguous definition of smart metering, based on the (national) policy objectives of a smart metering deployment and in line with European standardization efforts. 2.2 Description of a Smart Metering Infrastructure In line with the distinction between an individual smart meter and the general concept of smart metering, it is important to point out that the implementation of smart metering requires a complete infrastructure. A typical smart metering infrastructure basically consists of the following elements: Metering device and associated devices on the consumer's premises (optionally connected to a smart home unit controlling household appliances, for instance based on tariff information, in case demand side management is applied) Optionally, a graphical display within the consumer's living space providing actual real-time meter data and eventually information on tariffs or other relevant data (Inhouse display, IHD) Communication and data processing infrastructure between the devices on the consumer premises and the back-end systems, Including eventually different communication solutions, depending on situation and approach Energy data management (EDM) systems at the metering operators 8 back-end that provide necessary data to billing and invoicing systems of the supplier and optionally to the consumer, e.g. on a web page 7 The report uses the terms 'regulator' and 'regulatory authorities' mostly in an interchangeable way, although typically either the national regulatory authority or for instance a ministry could be implied. Apart from for instance setting network tariffs, there is in most cases no default responsibility. The authority or governmental office which is responsible is thus dependent on the regulatory and legislative framework in a particular country. What can be typically observed is that the overall responsibility for a roll-out decision falls into the responsibility of the government or if the roll-out is stipulated by law legislative bodies. The role of the national regulatory authority is typically much more focused on execution and support of governmental or parliamentary decisions, i.e. for instance providing a cost-benefit analysis, deciding about functional requirements or preparing a smart metering deployment plan. 8 As will be assessed in detail in chapter 0, several entities may in principle be suited to take over the responsibility for metering, although in general it will be the DSO. If in the present report the role of the metering operator is referred to, this may mean a stand alone metering service or any other market player (most likely the DSO or eventually the supplier) taking over the responsibility for metering. Smart Meters Rollout in the Energy Community Page 9/107

11 Household appliances Communication module Consumer Home Area Network Inhouse Display Smart Home Unit Wide Area Network Network Operator/ Metering Operator Electricity Meter GSM GPRS EDM Gas Meter District Heat DSL PLC Meter data storage and management Data transmission Figure 2: Smart Metering Infrastructure Source: KEMA The above figure shows an exemplary fully fledged smart metering infrastructure. As indicated above, in practice some of the depicted elements are not a mandatory part of smart metering The Meter and Associated Devices Traditional meters in the residential sector for electricity are electromechanical induction meters; for gas bellows-type flow meters are used. The electricity meter uses a very small amount of the electricity flowing through to drive the meter, the gas meter uses the energy of the gas flow, i.e. in principle it is purely mechanical and does not require an electricity supply. Traditional meters measure only the amount of electricity or gas flowing through them. The meter data, when read, does not allow any conclusion on variations of consumption over time since the last meter reading, only the accumulated amount of energy consumed since the last meter reading is shown. In order to deploy smart metering these traditional meters are replaced by modern electronic solid state meters; for gas electronically refitted bellows-type meters are also regularly used. These meters are able to measure the load over time. Smart Meters Rollout in the Energy Community Page 10/107

12 Electricity Gas Traditional Meters Smart Meters Figure 3: Examples of Traditional and Smart Meters for the Residential Sector Sources: Görlitz (smart electricity meter), Diehl (smart gas meter) Metering data (e.g. a load curve) is crucially required in order to enable the required smart metering functionalities, and needs to be available not only in digital format. In addition, the meter needs to be able to process the data, submit it and to receive and process data as well, e.g. requests for actual meter data, tariff information, commands for a remote (dis- )connection and/or load limitation, firmware updates, etc. This data is transmitted between the metering device (or the associated communication module) and the back-end EDM system through the wide area network (WAN). Moreover, meter data as well as data received by the meter can be transmitted directly to the other appliances on the consumer's premises, using the home area network (HAN). A consumer may use an in-house display to have access to this information somewhere in his living area (whereas the meter is typically hidden away in the basement or in some meter closet). Additionally, the meter may be connected to a smart home unit, which controls household appliances based on the information received, for example switching on the dishwasher when electricity prices are low. Electricity smart meters are often able to measure the electricity withdrawn from the grid as well as electricity injected into the grid. Thus, in cases where a distributed generation unit, e.g. PV panel or micro CHP, is used, a separate meter would not be required. The smart meter set-up on the consumer's side may follow two different approaches. The communication technology may be included in the smart meter or the set-up may be modular, with a rather simple electronic meter and a separate communication module containing the bigger part of the set-up's "intelligence". Such a set-up may be beneficial in cases where Smart Meters Rollout in the Energy Community Page 11/107

13 only parts of the infrastructure need to be replaced, e.g. when components have different lifetimes, or additional metering data need to be integrated. In cases where a roll-out for smart metering for gas is also considered, an assessment should be made as to whether an integrated approach would be favorable. In a multi utility set-up, the communication infrastructure is shared, avoiding costly redundancy. There are two possible approaches: firstly, one of the meters (typically the electricity meter) has an interface to connect additional meters, or the modular approach is taken (multi utility communication, MUC) and the communication module is able to connect several meters (multi utility communication controller, MUC-C). Electricity Meter Network Operator/ Metering Operator MUC-C WAN Gas Meter EDM Figure 4: Multi Utility Setup with MUC-C Sources: KEMA, Hager (electricity meter), Dr. Neuhaus (MUC-C) The MUC-C is the key device in a multi utility set-up. It functions as a data hub between the metering devices, other appliances like the in-house display and the communication towards the back-end systems In-House Display If the smart metering deployment targets a change in consumers' behavior, feedback to the consumer is essential. The most direct feedback available is the so-called in-house display. A graphical display located in the consumer's living space connected to the metering or communication device showing consumption data in real-time (ideally) and also showing information externally pro- Smart Meters Rollout in the Energy Community Page 12/107

14 vided e.g. actual tariff information, etc., enables the consumer to see the associated costs of consumption as well as the effects of any behavioral changes. Alternatively to a dedicated in-house display, a web page showing the data may have the similar effect. However, if a webpage is used, the data needs to leave the house and is processed by the back-end systems. An in-house display directly connected to the meter would allow for real-time information to the client, whilst at the same time data transmission to the back-end would not need to be in real-time. Depending on the objectives of smart metering deployment or the approach chosen, the inhome display is not necessarily seen as a mandatory part of a smart metering infrastructure in all cases Communication and Data Processing Infrastructure A communication infrastructure connects the metering devices to the customer s other devices, e.g. the in-house display, and also to back-end systems of the metering provider. The communication to the consumer's other appliances uses the home area network (HAN) based communication technologies such as wireless/wired M-Bus, ZigBee, power line communication (PLC) or the Internet Protocol (IP). Optionally, data can be transmitted to the consumer only indirectly through the EDM system at the back-end. In the latter case the information can be provided to the consumer using web portals, short message services or the invoice. If real-time provision of meter data to the consumer is intended, the direct transmission to consumers is preferable. Thus the transmission interval to the back-end can be chosen depending on the needs, e.g. only daily, and possibly sensitive real-time data may remain within the consumer s sphere. For communication between metering devices and the back-end system, typically the following technologies are used: Power line communication (PLC) GSM/GPRS/UMTS based mobile phone technology Broadband internet connections (DSL) In many set-ups a combination of more than one technology is applied. When rolling out a smart metering infrastructure, the basic decision would be whether data is transmitted directly from the consumer s site to the EDM or bundled locally with a so-called data concentrator and forwarded to the EDM from there. Data concentrators are recommended in more densely populated areas when a massive rollout of smart metering takes place or is planned. In that case the individual metering installations are connected to data concentrators with PLC, with up to several hundred meters per data concentrator. The data concentrator is for instance installed in a distribution substation Smart Meters Rollout in the Energy Community Page 13/107

15 and bundles, checks, 9 processes and stores meter data, and transmits and receives data to and from the back-end systems E-Meter PLC GSM / GPRS M-Bus G-Meter Data Concentrator GSM / GPRS / ISDN 2 EDM IHD Figure 5: Exemplary Set-up Combining PLC and GSM/GPRS Technology Source: KEMA PLC is the collective name for techniques which enable telecommunication using the electricity distribution network as a communication channel. A common application of PLC is the reading of metering data from energy consumers. For this purpose, equipment is installed in the consumer's electricity meter which can transfer the recorded metering data to the data concentrator. The range of PLC is limited. It can only be used for local connections on the consumer's premises or to the data concentrator. The range of PLC is limited by the fact that the data cannot be transferred through power transformation. To connect a data concentrator to the back-end alternative communication channels are required, e.g. mobile phone technology, as depicted in the above figure. Alternatively ISDN or DSL technology may be used. GSM (Global System for Mobile Communications) is a digital mobile telephone standard. GPRS stands for General Packet Radio Service, which is an extension technology to the existing GSM network. Universal Mobile Telecommunications Systems (UMTS), the next generation technology also based on GSM standards achieves higher data transmission rates. Data can be sent and received efficiently and rapidly using mobile phone networks. 9 For example for fraud detection a check could be made to determine if the sum of supplied energy (sum of all individual meter readings) equals the sum of consumed energy (in total, measured at a substation). Smart Meters Rollout in the Energy Community Page 14/107

16 Using mobile phone communication exhibits relatively high investment and operation costs per single communication unit, but can be operated without requiring substantial additional infrastructure. 10 Mobile phone technology is thus deployed either for direct connections between meter devices and EDM, if the population or smart metering density does not allow for usage of data concentrators, or to connect data concentrators to the EDM. Alternatively, the consumer s broadband internet connection (DSL) can be used to transmit meter data directly to the EDM. Modern electricity meters have TCP/IP output, in principle allowing direct access to the worldwide internet. The metering data is then sent via the Internet Protocol (IP). Meter TCP/IP DSL Router Internet EDM Web-Portal/IHD Figure 6: Exemplary Set-up Using DSL Technology Source: KEMA Such a solution does however require software and more processing power inside the meter to allow communication to take place properly. If this solution is selected, no further hardware is required beyond the consumer s modem. Although this solution is dependent on the existence of such an infrastructure at the consumer's premises and means that this infrastructure is under control of the consumer itself, it is comparably cheap. A web portal or an in-house display also using the IP communication may be connected to the back-end through the internet or it may also display data directly received from the meter. 10 I.e. the metering provider does not need substantial infrastructure; nevertheless a mobile communication network is required. Smart Meters Rollout in the Energy Community Page 15/107

17 2.3 Consumption Feedback Mechanisms In many cases the roll-out of smart metering is driven by the aim to achieve energy savings and to reduce carbon dioxide emissions. In the European Union this political aim is expressed in the target, finally approved by the European Parliament and Council in December Smart metering will not necessarily save energy as such, but by using smart metering, effective consumption feedback (including costs of used energy) can be provided to the consumer. Additionally, new tariff schemes can easily be adopted. Thus, changes in the consumer's behavior can be triggered. As described above, the feedback can be provided in real-time using an in-house display or less frequently, for instance with a monthly invoice based on real meter data. As the survey on smart metering conducted by the Energy Community Regulatory Board (EnCRB) shows, monthly meter reading (of conventional electricity meters) is already the rule in almost all EnC contracting parties. 12 In general two forms of feedback are distinguished: 13 Direct feedback, e.g. through in-house energy consumption displays or web-based information portals with real-time information Indirect feedback, e.g. through frequent invoicing based on actual meter readings, e.g. enhanced with comparison values from peer groups, energy savings advice or information on energy savings achieved so far 11 Cf. Commission welcomes adoption of climate and energy package, press release, IP/09/628, Brussels, 23 April Energy Community Regulatory Board, A Review of Smart Meters Rollout for Electricity in the Energy Community, 2010, Table A comprehensive overview of different types of feedback is for instance provided in Darby, Sarah, The Effectiveness of Feedback on Energy Consumption, Oxford, Smart Meters Rollout in the Energy Community Page 16/107

18 In-house display Monthly information for residential costumers Figure 7: Examples for Improved Feedback Source: CER Direct feedback is characterized by simple access to the data, e.g. through an in-house display, as shown on the left side of the above figure, or a web-based portal which provides real-time information. Direct feedback enables the consumer to monitor energy consumption and associated costs constantly and to see immediately results of any changed behavior. For direct real-time feedback to the consumer, it is not necessary to submit real-time information to the back-end systems. Very simple systems measuring only the amount of electricity consumed and sending this information to an in-house display can be easily fixed to the traditional electricity meter without deploying a full-scale smart metering infrastructure. 14 Indirect feedback is typically based on data which is already processed, i.e. a first analysis is already carried out by the metering operator or supplier. The usual invoice is a typical example of indirect feedback. Smart metering can improve this indirect feedback by providing accurate meter data, providing this data more frequently and by providing additional data which can be used to enhance the invoice, enabling and motivating the consumer to take action to decrease energy consumption, as shown on the right side of the above figure. 15 Such an invoice may include an analysis regarding the time of day energy is typically used, how much money could be saved by shifting parts of the demand to off-peak times, consumption benchmarks or historical comparison data. Long-time surveys show that direct feedback seems to lead to faster and higher energy savings but that the change in consumption behavior lessens over time once the novelty wears 14 Cf. for instance where a simple technical solution is offered. 15 The isolated impact of improved billing was for example tested in a Norwegian project, c.f. Wilhite, Harold, Hoivik, Asbjorn, Olsen, Johan-Gemre, Advances in the use of consumption feedback information in energy billing: the experiences of a Norwegian energy utility, Smart Meters Rollout in the Energy Community Page 17/107

19 off. Indirect feedback seems to result in smaller but more sustained changes in consumption behavior. The requirement for direct feedback using real-time information is not included in the EU legal framework. However, the need for the smart metering infrastructure to include an interface to connect for instance an in-house display is referred to in the set of standard smart metering functions as defined under Mandate M/441 and the ERGEG GGP (cf. section 2.1), and is also mentioned in a European Commission survey on common functional requirements of smart metering among the most advanced Member States. 16 These sources allow the conclusion that at least the possibility to connect an in-house display is advised, whereas the provision of the display itself is sometimes left to the consumer. The latter survey shows that for example in Austria and Belgium an interface to connect an in-house display is required, whereas the consumer is responsible for providing the display itself. The UK goes one step further and has included such a display in the general design of the smart metering set-up. Feedback, in direct or indirect form and supported by new tariff schemes, is crucial if smart metering deployment should lead to energy savings by changes in consumption behavior. The different feedback options of smart metering should be incorporated in the cost-benefit analysis. 2.4 Minimum and Optional Functionalities As already indicated in section 2.1, the best way to define smart metering is to identify the set of functionalities and services provided by the smart metering system. However, apart from several basic features (e.g. bidirectional communication) there is no exhaustive list of what smart metering must provide. In the European Union however, a common set of typical standard functionalities is evolving. The processes and publications mentioned above are relevant here, i.e. the standardization mandate M/441, ERGEG's GGP and the European Commission's survey on common functional requirements. In this process a European set of standard or required functionalities for smart metering seems to be emerging. Basic requirements for any metering installation at EU level result from the Directive on measuring instruments from In addition, six smart metering functionalities have been identified under the standardization mandate M/441: European Commission, A joint contribution of DG ENER and DG INFSO towards the Digital Agenda, Action 73: Set of common functional requirements of the SMART METER, Full Report, October Directive 2004/22/EC of the European Parliament and of the Council of 31 March 2004 on measuring instruments. 18 Standardization mandate to CEN, CENELEC and ETSI in the field of measuring instruments for the development of an open architecture for utility meters involving communication protocols enabling interoperability M/441, SMART METERS CO-ORDINATION GROUP, FINAL REPORT. Smart Meters Rollout in the Energy Community Page 18/107

20 Remote reading of metrological register(s) Two-way communication Support of advanced tariffs and payment systems Remote (dis-)connection and load limitation Communication with other devices on consumer's premises Information through web portal/gateway to in-home display or auxiliary equipment The ERGEG GGP aim to be in line with the results of mandate M/441. They provide a list of services which should be facilitated by smart metering (twelve for electricity and nine for gas). Customer services Recommendations for Electricity and Gas Recommendations for Electricity only Information on actual consumption and cost, on a monthly basis, free of charge Access to information on consumption and cost data on customer demand Easier to switch supplier, move or change contract Bills based on actual consumption Offers reflecting actual consumption patterns Alert in case of exceptional energy consumption Interface with the home Software to be upgraded remotely Remote power capacity reduction/increase Remote activation and de-activation of supply All customers should be equipped with a metering device capable of measuring consumption and injection Alert in case of non-notified interruption Recommendations for Gas only Remote enabling of activation and remote de-activation of supply Table 1: ERGEG Guidelines of Good Practice on Customer Services Offered Through Smart Metering Source: ERGEG EREGEG, Final Guideline of Good Practice on Regulatory Aspects of Smart Metering for Electricity and Gas, Ref: E10-RMF-29-05, Brussels, 8 February Smart Meters Rollout in the Energy Community Page 19/107

21 The European Commission survey identifies altogether ten functionalities, split into those for customers, meter operators, supporting commercial aspects, privacy and security and distributed generation. The survey focused on eleven advanced Member States and was based on a questionnaire. Only those functionalities with at least eight out of eleven points were included in the final list, which can be seen below. Functionalities with high consensus For the customer Provides readings from the meter to the customer and to the equipment he may have installed Updates these readings frequently enough to allow the information to be used to achieve energy savings Allows remote reading of meter registers by the Meter Operator For the meter operator For commercial aspects of energy supply For security and privacy To allow distributed generation Provides two-way communication between the meter and external networks for maintenance and control of the meter Allows readings to be taken frequently enough to allow the information to be used for network planning Supports advanced tariff systems Allows remote ON/OFF control of the supply and/or flow or power limitation Provides Secure Data Communications Fraud prevention and detection Provides Import / Export & Reactive Metering Table 2: Common Smart Metering Functionalities According to a Survey Amongst Eleven Advanced Member States Source: DG ENER, DG INFSO 20 As can be observed when comparing the different lists, there is a very high redundancy in ERGEG's list and the European Commission's survey. It is important to point out that in practice not every metering device will or should necessarily support all these functions, it is rather a non-exhaustive list of the direction in which developments are presently heading. Additional functionalities which are often mentioned are: 20 European Commission, A joint contribution of DG ENER and DG INFSO towards the Digital Agenda, Action 73: Set of common functional requirements of the SMART METER, Full Report, October Smart Meters Rollout in the Energy Community Page 20/107

22 Transmission on demand of data on power quality/condition and voltage level Prepayment function Their application was also tested in the Commission's survey. From the results it becomes clear that although not implemented by a large majority, these functionalities are included or considered in many cases. Depending on the exact set-up no additional hardware components would be required. During a joint workshop held with the ECRB CWG and other representatives of the regulatory authorities of the EnC Contracting Parties during the project, the possibility to install prepayment systems together with smart metering was highly welcomed. 21 When national minimum functionalities are defined, it should be noted that whereas some of the functionalities may be defined as mandatory, others could provide extra benefits but may not be required in a basic set-up. In order to achieve a successful smart metering roll-out, the necessary minimal functional requirements need to be set in line with the country's objectives to ensure an efficient and functional approach and to enable new services to maximize potential benefits. However, if requirements are set too high, the resulting benefits may be achieved at high costs, because in several cases, more expensive hardware would be required for more functionality. The decision as to which functionalities are included in a definition of minimum requirements should be dependent on local circumstances, on the objectives behind smart metering deployment and in line with the general development in the European Union. European legislation on smart metering is driven by the objective of improving energy efficiency, achieving energy savings and facilitating distributed generation. Additional benefits from the reduction of fraud and defaulting payment may also be expected in parts of the Contracting Parties. So far, minimum requirements have for instance been defined in Denmark, Estonia, Finland, France, Greece, Italy, Spain, the UK and Sweden. 21 In the EU prepayment meters are only used in the UK and in Belgium, only in the UK a considerable number of customers are equipped with a prepayment meter. Currently around 11% of gas customers and 14% of electricity customers are equipped with prepayment meters. In the UK, coin-in-the-slot meters were used until the 1980s, since then they have been replaced by electronic forms of prepayment. Prepayment meters are mainly used for customer groups with a high risk of defaulting payment. Prepayment meters prevent customers from being in debt, however, the biggest concern with this form of prepayment is self-disconnection of customers when they run out of credit. Subsequently, a discussion on consumer protection revolves around prepayment meters. Additionally, as prepayment meters are more expensive than normal meters, electricity tends to be more expensive for those customers who use a prepayment meter, and who typically belong to poorer consumer segments anyway. So far, Ofgem estimated additional costs for a prepayment meter of 65 to 85 GBP. With smart metering, it is expected that extra-costs for prepayment meters can be driven down. With payments being made for instance through the internet or via phone calls, no card reading systems within the meter are required. If the smart meter provides for remote disconnection or load limitation anyway, no additional meter hardware is required for prepayment metering. Prepayment metering would simply be included as a software functionality of the meter. (c.f. Owen, Gill, Ward, Judith, Smart pre-payment in Great Britain, March 2010). Smart Meters Rollout in the Energy Community Page 21/107

23 A national decision on minimum functional requirements is recommended in order to ensure a successful and efficient smart metering deployment. The required functionalities should be chosen in line with the national policy objectives driving the roll-out and the EU development to identify a set of standard functions. 3. RELEVANT EU FRAMEWORK AND EXPERIENCE 3.1 EU Legal Framework and Requirements Two basic documents set the current EU legal framework for smart metering: the Directives on the internal markets for electricity and gas (2009/72/EC & 2009/73/EC, the so-called Third Package) and the Directive 2006/32/EC on energy end-use efficiency and energy services. Although not mentioning smart metering directly, the requirements stemming from Directive 2006/32/EC pave the road to metering systems which are at least more sophisticated than traditional meters. In Article 13 the Directive requires that final customers are provided with competitively priced individual meters that accurately reflect the final customer s actual energy consumption and that provide information on actual time of use. However, the requirement is subject to technical feasibility, financial viability and ability to potential energy savings. At the same time, energy saving targets are set in Article 4. Moreover, the energy services directive sets requirements for the level of detail and information to be provided to the electricity customer together with or in addition to the invoices. Invoicing is supposed to be based on actual consumption and to be performed frequently enough to enable customers to regulate their own energy consumption. Information on consumption should also include information on energy consumption in the previous year and a reference value of the consumer segment. 22 Directive 2006/32/EC was transferred into the Member States' national legislation quite differently and only in a few Member States has it led to a requirement to install smart meters. In 2009 the so-called Third Package for the further liberalization of energy markets was legislated in Directives 2009/72/EC for electricity and 2009/73/EC for gas. The requirement for smart metering is stipulated in Annex I of both Directives. Annex I contains provisions to promote smart metering requiring Member States to ensure the implementation of intelligent metering systems [ ]. The implementation of those metering systems may be subject to an 22 Directive 2006/32/EC of the European Parliament and of the Council of 5 April 2006 on energy end-use efficiency and energy services and repealing Council Directive 93/76/EEC. Smart Meters Rollout in the Energy Community Page 22/107

24 economic assessment of all the long-term costs and benefits to the market and the individual customer or which form of intelligent metering is economically reasonable and cost-effective and which timeframe is feasible for their distribution. Such assessment shall take place by 3 September For gas, only the preparation of a timetable for smart metering implementation is required, subject to economic assessment. For electricity, a time horizon of ten years is set for the timetable. Furthermore, if roll-out is assessed positively, Member States are required to ensure that 80% of consumers are equipped with intelligent metering sys tems by Shortly before passing the Third Package, the European Commission had already mandated European standardization organizations to develop an open architecture hardware and software standard for smart metering systems enabling interoperability of meters (Mandate M/441 EN). 25 This mandate already led to a list of functional requirements for smart metering, as described in section 2.4. The European Regulators' Group for Electricity and Gas (ERGEG) launched a consultation process of regulatory aspects of smart metering, covering questions on the cost-benefit assessment, roll-out decision and functional requirements. In June 2010 ERGEG published Draft Guidelines of Good Practice on Regulatory Aspects of Smart Metering for Electricity and Gas 26, the final guidelines followed in February The Guidelines of Good Practice (GGP) define the minimum and optional services which smart metering should provide to electricity and gas customers and make suggestions on the conduction of economic assessment, the roll-out and on data security. Whereas the Directives described above set the requirements for smart metering installation, Directive 2004/22/EC on measuring instruments sets the general technical requirements for metering appliances. As such, the Directive is also relevant for smart metering appliances. It contains technical provisions for metering devices for electricity, gas, water and for other 23 Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC, and Directive 2009/73/EC of the European Parliament and of the Council of13 July 2009 concerning common rules for the internal market in natural gas and repealing Directive 2003/55/EC. 24 The question has been raised as to what exactly is meant by consumers - the number of consumers or the consumed volume. The official documents refer only to the threshold of "80% of consumers", thus we would expect that the number of consumers is meant. (c.f. Commission Staff Working Paper Interpretative Note on Directive 2009/72/EC Retail Markets, Brussels, 22 January 2010, and Council of the European Union, Addendum to "I/A" Item Note, Council document 10814/09). 25 European Commission, Standardization mandate to CEN, CENELEC and ETSI in the field of measuring instruments for the development of an open architecture for utility meters involving communication protocols enabling interoperability, M/441 EN, Brussels, 12 March ERGEG, Public Consultation Paper on Draft Guidelines of Good Practice on Regulatory Aspects of Smart Metering for Electricity and Gas, Ref: E10-RMF-23-03, Brussels, 10 June ERGEG, Final Guideline of Good Practice on Regulatory Aspects of Smart Metering for Electricity and Gas, Ref: E10-RMF-29-05, Brussels, 8 February Smart Meters Rollout in the Energy Community Page 23/107

25 liquids, heat meters, scales, etc. Based on the subsidiary principle, Directive 2004/22/EC only includes regulations covering the process until the metering device is offered on the market or brought into operation. Further requirements during the lifetime of the meter, regarding calibration, tolerances, etc. are subject to national legislation. Although the pace for smart metering deployment is mainly set by the energy services Directive and the Third Package, other Directives also make reference to intelligent metering systems, thus encouraging (but not requiring) smart metering systems in order to promote energy savings and demand response (Directives 2005/89/EC and 2010/31/EC). Especially the recent Directive on energy performance of buildings (2010/31/EC) is very specific: Member States shall encourage the introduction of intelligent metering systems whenever a building is constructed or undergoes major renovation, [ ]. Whilst the energy services Directive and the Third Package aim to promote intelligent metering throughout the buildings / metering inventory, Directive 2010/31/EC explicitly refers to new buildings and those undergoing a major renovation. 3.2 Overview of State of Smart Metering in the EU Although the Third Package sets a common target for the deployment of intelligent metering systems in all EU Member States, the development is quite diverse, with different countries applying different approaches in terms of market model, technologies and objectives. So far, only two countries have completed a full roll-out of smart metering systems, i.e. Italy and Sweden, and in both cases the degree of smartness of the metering systems tends mostly to remain on the level of remote meter reading. In other Member States, the decision for a mandated roll-out has been already taken, i.e. France, Spain and the UK, and massive deployment is expected to start soon. In the majority of Member States, so far no mandated full roll-out has been decided upon. In some, smart metering deployment has been left at least until now to individual (voluntary) decisions of market parties (e.g. Slovenia, Denmark) or is required only for specific consumers as mandated by the Directive on energy performance of buildings (e.g. Germany), in other countries the final roll-out decision is still pending (e.g. Austria). Some countries have not considered smart metering at all, until now, whereas in others the whole idea is subject to a very broad and public discussion, as for instance in the Netherlands where data privacy issues toppled the attempt to have a mandated roll-out. Motivations behind a smart metering roll-out are also diverse. Although by now European legal requirements are setting the pace of the process, Italy and Sweden have already accomplished a massive roll-out. The fundamental motivation from the European perspective is Smart Meters Rollout in the Energy Community Page 24/107

26 to use smart metering as a tool for better informing and educating final consumers about (their) energy consumption, and to achieve a higher level of awareness and thus energy savings and reduced greenhouse gas emissions. Thereby, smart metering is embedded into the EU's 20/20/20 targets. In Italy, starting the roll-out already in 2000, the basic goal was to reduce non-technical losses, which back then were a huge problem for Enel. In essence, many meters were removed before they could be read. Subsequently, the focus of Enel was on remote meter readings with sufficiently short intervals and anti-fraud mechanisms and less on consumption feedback to consumers. In Sweden, the legal requirement to provide monthly invoices based on actual meter readings from 2009 onwards provoked DSOs to roll-out smart meters, however, the main function to facilitate the goal was remote meter reading and many meters installed at the beginning are probably not really smart meters. The roll-out decision and the technical and regulatory framework is based on individual (policy) and local circumstances. Sweden is not very densely populated, manual meter reading traditionally took place once a year. Manual monthly meter reading would have been extremely cumbersome in the rural areas of Sweden. In other countries, with high population density and/or a cheap labor force available, manual monthly meter reading is quite common. This is also the common approach in many EnC Contractual Parties. The motivation for the roll-out and the local circumstances thus clearly influence the roll-out decision and thereby also the chosen technologies and required functionalities. The following figure taken from the SmartRegions (a EU funded project to promote best practices of innovative smart metering services) Landscape Report on the state of smart metering in the European Union distinguishes five groups of countries, according to their progress in smart metering implementation and the existence of a legal and regulatory framework. The legal and regulatory framework in this context refers to whether the framework not only provides clear guidelines for smart metering deployment but also whether smart metering deployment aims at energy savings or peak load shifting. Smart Meters Rollout in the Energy Community Page 25/107

27 Figure 8: State of Smart Metering in EU Source: Renner et al. 28 The SmartRegions report gives the following classification of the five groups depicted above. "Dynamic movers" are those countries in which the decision for a mandated full smart metering roll-out has already been taken or where major pilot projects are already in place and should result in a mandated roll-out. The countries classified as "market drivers" leave the deployment of smart metering to individual decisions of distribution system operators or suppliers, either because of technical considerations or following customer demands. "Ambiguous movers" have the required framework largely in place but the roll-out has nevertheless not taken place in general. "Waverers" are those countries where the topic of smart metering is already on the agenda but has not yet come to tangible results. The "laggards" are the countries where smart metering is not yet under discussion, however due to the requirements of the Third Package, these countries will have to make progress. 28 Renner et al., European Smart Metering Landscape Report, SmartRegions Deliverable 2.1, Vienna, February Smart Meters Rollout in the Energy Community Page 26/107

28 3.3 Experience in selected EU Member States Italy Italy was the first EU Member State to opt for a large scale smart metering roll-out. At that time no legal or regulatory framework for smart metering was in place. The decision for the first major and (still) one of the largest roll-outs was taken by the incumbent energy supplier Enel. Although Enel was formally privatized in the early 1990s, it has since been under government control. As mentioned above, the basic motivation for smart metering deployment for Enel was a reduction in non-technical losses, i.e. fraud. Additionally the roll-out was driven by expected savings or revenues in the areas of purchasing and logistics, field operations and customer services. In 2001 Enel started to deploy smart meters throughout its low-voltage customer base, i.e. around 30 million meters, covering roughly 85% of the Italian household market. For the Italian smart metering roll-out a smart meter was installed, already providing functionalities like bidirectional communication, remote (dis-)connection and load limiting. The smart metering set-up was based on power line communication between the smart meters and data concentrators. For further communication towards the back-end, IP communication was employed. The technology used, as well as interfaces and communication formats were mainly proprietary, which later created issues with interoperability. The metering interval is one hour. In 2006, the regulatory authority monitored the developments and started to set up a legal framework for smart metering along with a mandated roll-out, also setting minimum functional requirements. The justification for the full roll-out was different from the original Enel s one (Enel as well had to adapt its system to the new requirements). It was to have a tool to transfer the price signal to end customers and to influence their consumption behavior by means of a time dependent pricing. In order to achieve this, amendments to the existing load profiling methodology had to be undertaken and could be done only by using smart metering on national base. The mandatory roll-out requirement was included in the Energy Law and the responsibility for the metering roll-out was assigned to the DSOs. The mandatory roll-out was supposed to start in 2008 and was targeted to achieve 95% by mid To cover the costs for the roll-out, a separate metering charge has been levied since 2004, although only granted to those DSOs who actually deploy smart meters. An additional incen- 29 Major legal references for each country mandating/regulating smart metering roll-out are provided in Appendix 1. Smart Meters Rollout in the Energy Community Page 27/107

29 tive for DSOs to comply with the national roll-out target is that the allowed revenue from metering charges is only granted to DSOs who fulfill the target penetration rates for each year. In order to promote smart metering even further, an additional monetary incentive was set for DSOs rolling out smart metering faster than targeted. The roll-out is almost 100% complete. However, given the original aim to reduce fraud and save on process costs, the whole set-up of metering and metering services is somewhat lacking in customer feedback and thus will probably only limitedly incentivize consumers to reduce their electricity consumption Sweden The Swedish system is characterized by very high electricity consumption per capita (>10,000 kwh) as electricity is also widely used for heating purposes. Around 2000, metering became an important topic on the political agenda in Sweden, caused by increasing energy prices, power conservation, incomprehensible energy bills (which were inaccurate and did not correspond to actual consumption), and the strong desire to have a correlation between energy costs and energy consumption. In May 2002, the Swedish regulatory authority performed a cost-benefit-analysis on the introduction of monthly reading of electricity meters, which provided benefit estimations of about 60 million per year, resulting from reduced electricity consumption due to improved feedback to consumers. Total costs were estimated to be around 1 billion. 30 Thereafter it was decided that every user with an average annual energy consumption of more than 8,000 kwh should have their electricity meters read at least once a month by By July 1, 2009, all meters were legally required to be read monthly. With the previous practice of annual meter reading and the large sparsely populated areas of Sweden, this resulted in the deployment of smart metering. Smart metering (or in many cases more simple remote reading of hourly values) is now deployed all over the country. When deployment of remote meter reading started, some of the technologies currently available were not on the market at that time. Nowadays, DSOs typically deploy smart meters and also assess additional benefits from full-fledged smart metering. Around 10%-15% of the installed meters are in metering systems which are only capable of monthly remote meter reading. It is expected that these meters will be replaced significantly before the end of their economic lifetime. Pilot projects are evaluating the potential of real-time tariffs and new services. Interestingly, a comprehensive cost-benefit analysis for a national smart metering deployment never took place before the roll-out, only the benefits of monthly meter reading were 30 Statens energimyndighet, Månadsvis avläsning av elmätare, ER 12:2002, 2002, %20elm%C3%A4tare.pdf Smart Meters Rollout in the Energy Community Page 28/107

30 assessed. The decision to go for an advanced metering system was taken by the DSOs individually; however it resulted in a more or less uniform approach all over the country. The metering sector is not liberalized, responsibility for metering lies with the DSOs. DSOs have also had to bear the costs of smart metering deployment so far, although recently the regulator had to grant increased network charges. There are no rules for third-party access to consumption data (e.g. by independent service providers) and there is no legal obligation for the interoperability of smart metering systems, or for exchangeability of meters Germany In Germany, the metering market was liberalized in 2008, opening up the metering market to competition. The consumer is entitled to choose its own metering service provider. However, apart from cases where metering was now offered by new entrants on the supply market, thus taking over the role of the metering service provider, the default responsibility for metering remained with the DSOs. In theory, the German legislative framework even distinguishes two roles, the metering infrastructure operator and the service provider, but with regards to smart metering, this distinction is not made. Basic requirements on where or when smart metering needs to be deployed are set directly by the Energy Act. With the metering sector liberalized, the introduction of smart metering in Germany was relying on the voluntary roll-out of smart metering by DSOs, suppliers or independent metering providers. This approach has so far not been very fruitful. Smart meters were only deployed in the beginning by DSOs in smaller or larger pilot projects (up to meters). In 2009 only one offer for smart metering was commercially available throughout the country, offered by one of the new entrants on the supply market (although affiliated to one of the large incumbent suppliers/generators). So far, although some cost-benefit analyses have been conducted by authorities, among others by KEMA for the Ministry of Economics 31, no assessment was meant to be the one demanded by Annex I of Directive 2009/72/EC. Subsequently, a mandated roll-out has not yet been decided upon. Taking into account the requirements of the Third Package though, the Energy Act has since been changed and has increased the requirements for smart metering. The cost-benefit analysis as defined by the Directive is expected to be awarded by the regulatory authority during Q1/2012. By now metering providers (i.e. in most cases the DSOs) have to install smart metering systems if technically possible in all buildings undergoing major renovations, at all electricity consumers with an annual consumption of over 6,000 kwh (average household consumption 31 KEMA, Endenergieeinsparung durch den Einsatz intelligenter Messverfahren (Smart Metering), Endbericht, Study for Federal Ministry of Economics and Technology, November 2009, Bonn. Smart Meters Rollout in the Energy Community Page 29/107

31 is around 4,000 kwh) and where renewable or CHP generation units with a capacity of more than 7 kw are installed. In all other cases, smart meters have to be installed when technically possible and economically justifiable. In addition, smart meters have to be offered to all customers. The Energy Act also sets minimum requirements for invoicing. Among others the provision of historic data is required, and on demand information on where smart meters are used and monthly consumption should be provided free of charge. Smart meters are offered in many cases by DSOs. However, as DSOs are free to charge substantially higher fees for smart meters (installation fee plus metering fee), consumers seem to be hesitant in demanding their installation. A large assessment of running pilot projects from 2009 resulted in average electricity savings of around 6%. However, during this study it was entirely clear whether these were actual savings or rather savings expected, as parts of the pilot projects were still at very early stages Austria Smart metering has been under discussion in Austria for a couple of years now, however, no decision for a mandatory roll-out has yet been made. The starting point for the debate on smart metering was the aim of reducing electricity consumption. Already in 2008, the regulator published a report on potential measures to increase energy efficiency. 33 Among others, the report includes the recommendation of a full smart metering roll-out by Smart metering was explicitly meant to result in a higher quality of invoices, individualized tariff models and prompt information on consumption and thus together with improved consumer education lead to a decrease in consumption. Subsequently, the regulatory authority commissioned a cost-benefit analysis, which resulted in a positive net benefit for a smart metering roll-out. 34 The Austrian power industry also launched its own cost-benefit analysis, which resulted in a negative net benefit. 35 Controversial discussions on the smart metering roll-out took place between the regulator and the industry, mainly focused on the acknowledgement of costs for smart metering in revenue regulation. Although the ordinance on system charges already includes an explicit 32 KEMA, Endenergieeinsparung durch den Einsatz intelligenter Messverfahren (Smart Metering), Endbericht, Study for Federal Ministry of Economics and Technology, November 2009, Bonn. 33 E-Control, Grünbuch Energieeffizienz: Maßnahmenvorschläge zur Steigerung der Energieeffizienz, October 2008, Vienna. 34 PwC, Studie zur Analyse der Kosten-Nutzen einer österreichweiten Einführung von Smart Metering, June Capgemini, Analyse der Kosten Nutzen einer österreichweiten Smart Meter Einführung, January Smart Meters Rollout in the Energy Community Page 30/107

32 reference to smart metering, higher charges than for 'ordinary' metering devices are not granted. With the renewed Electricity Act from November 2010, the Ministry of Economics was authorized to decree the deployment of smart metering after a cost-benefit assessment has been carried out and the regulator and consumer advocates are consulted. So far, no roll-out decision has been taken. Only recently, the regulatory authority published an ordinance setting minimum requirements for smart metering, which are broadly in line with the requirements set by ERGEG. 36 With the technical ordinance in place, the ministerial ordinance mandating the roll-out itself is expected to be published during the first months of 2012, with the full deployment to start around 2014/ United Kingdom Discussion about the introduction of smart metering in the UK has been taking place for several years. From around 2005 there was no longer any real dissent on the question of whether smart metering for the residential sector should be deployed. In October 2008 the government announced the mandate of smart metering for the residential sector. This was followed by an announcement in December 2009 that 50 million smart gas and electricity meters are to be installed by In July 2010, the government's final plans for smart metering deployment were published for consultation together with the most recent cost-benefit analysis. 37 The proposed roll-out requires a full set of functionalities including the option to switch the meter to a prepayment scheme (which is already common in the UK). Other noticeable functionalities are: (1) HAN communication via open standards and protocols, (2) real-time information provided on a connected in-home display, (3) load management capability, (4) remote disconnection and a communication capability with on-site microgeneration. The minimum information to be shown for the in-home display is also defined. The proposed rollout scheme stipulates further that communication for smart metering is provided by a single company (DCC) and is to be used by all parties (central communications model) and that suppliers are responsible for the deployment of the metering devices themselves, as is depicted in the figure below. 36 Verordnung der E-Control, mit der die Anforderungen an intelligente Messgeräte bestimmt werden (Intelligente Messgeräte-AnforderungsVO 2011 IMA-VO 2011, October DECC/OFGEM, Smart Metering Implementation Programme Prospectus, Consulting and Supporting Documents, July 2010, Smart Meters Rollout in the Energy Community Page 31/107

33 Figure 9: Responsibilities for Smart Metering in the UK Source: Department of Energy and Climate Change (DECC) The overall responsibility for smart metering deployment is given to the Department of Energy and Climate Change (DECC), primarily working together with British regulator OFGEM but also coordinating with other involved parties. The technical specifications for smart metering in the UK were developed jointly with the industry in the first half of 2011 within the Smart Metering Design Group, resulting in a draft technical design able to provide the required functionalities. An earlier cost-benefit analysis conducted by DECC results in a net benefit (NPV) of 7.2 billion GBP (best estimate, in central communications model) for domestic and smaller nondomestic customers over the next 21 years, mostly stemming from energy savings and cost savings in industry processes. A best estimate for total costs is 9.7 billion GBP; total benefits are estimated at 16.9 billion GBP. Broken down into the market parties, the total consumer benefit for domestic and smaller non-domestic customers is given as 8.77 billion GBP, supplier benefits as 6.71 billion GBP and other benefits as 1.4 billion GBP. The most recent impact assessment conducted by DECC, published in August 2011, shows similar values with a best estimate for the net benefit of around 5 billion GBP for the domestic sector and 2 billion GBP for the small and medium non-domestic sector, varying slightly for the different policy options with regards to the integration of the communication interface into the meter. 38 Regarding data privacy, it has been stipulated that customers should be able to decide which information is used and by whom, except where data is required to fulfill regulated duties. The government acknowledges positive consumer engagement as crucial for suc Smart Meters Rollout in the Energy Community Page 32/107

34 cess, and subsequently plans to implement programs to promote consumer knowledge and awareness. Smart Meters Rollout in the Energy Community Page 33/107

35 4. POTENTIAL BARRIERS FOR SMART METERING DEPLOYMENT As experience gathered so far shows, the deployment of smart metering will face many barriers. Despite the alleged benefits of smart metering, market parties will not in all cases adopt smart metering voluntarily or willingly. Moreover, if they do aim to, their efforts may be hampered by existing barriers. In order to ensure a successful smart metering deployment, potential barriers need to be analyzed in advance and necessary steps to mitigate barriers need to be taken. In addition, developments which may endanger successful smart metering deployment need to be taken seriously throughout the process. 4.1 Consumer Resistance Apart from legal, regulatory and technical barriers which will be discussed in the next sections, consumer resistance may present a serious barrier to smart metering deployment. In contrast to other barriers, consumer resistance is probably the most difficult barrier to mitigate. Consumers may not perceive smart metering as positive, with consumer advocacy groups opposing smart metering roll-out. Examples of consumers heavily opposing smart metering deployment can for instance be found in the United States or in the Netherlands. In most cases consumer resistance can be observed to be driven primarily by two reasons: Consumers might fear that security and privacy of data gathered by smart metering cannot be guaranteed and hence unauthorized parties might have access to private data; they may also be against the authorized usage of the data. Consumers might also fear that they would have to bear the costs for deploying a smart metering infrastructure or that new (time-of-use) tariffs would lead to higher energy costs, whereas consumers' benefits might prove to be overestimated. The case of consumers opposing smart metering deployment due to the amount and level of detail of personal data gathered is highly relevant. The amount of data collected possibly allows very detailed conclusions on the lifestyle and daily routines of households, for instance when someone is at home, or in extreme cases where typical demand profiles of single appliances can identify what someone is doing. Many people may have concerns regarding the availability of such detailed data for the energy supplier or network operator. Additionally, the real-time transmission of this data from the consumer s site to the supplier's or network operator's back-end systems through the WAN (Wide Area Network) creates Smart Meters Rollout in the Energy Community Page 34/107

36 some vulnerability to unauthorized access, which did not earlier exist when this kind of data was simply not generated. As concerns regarding data security and consumer privacy can be easily understood, given the nature and amount of data gathered, they should be taken seriously. Moreover, the timely acknowledgement of concerns may be crucial in preventing issues endangering the success of smart metering deployment and in creating the necessary public acceptance. In the Netherlands for instance privacy concerns led to a serious delay in the roll-out scheme, when in April 2009 the Dutch Senate rejected a proposal for mandatory smart metering deployment. In its renewed proposal for smart metering deployment the government was forced to lessen the requirement for mandatory smart metering installation and to allow consumers to decide against smart metering. The revised legal framework from 2010 stipulated only a voluntary roll-out with various options for consumers to protect their data. Besides having a smart meter which is fully integrated into smart metering systems, consumers are now allowed the keep the traditional meter, to have a smart meter where no data is transmitted automatically or to limit the automatic transmission to supplier changes, relocation, annual billing and bi-monthly reading. The result of consumer resistance is a delay in the roll-out process and a less efficient roll-out as potentially a significant number of consumers may opt-out of smart metering and economies of scale and density may be lost. It is hence very important that provisions are implemented to ensure that data is not accessed by unauthorized parties and that there are clear regulatory provisions on how data is gathered, processed, stored and evaluated, and who has access to which data. In order to protect the data against unauthorized access, adequate measures (encryption, digital signatures) need to be taken. Data encryption is of particular importance when PLC technology is used to transmit data from the consumer s site to a data concentrator, as potentially every user connected to the same power line is able to intercept the communication between meter and data concentrator. Personal data should in general be protected by privacy law. Within the European Union certain requirements are set by Directives 95/46/EC and 2002/58/EC. 39 However, special attention should be given to smart metering as the amount of personal data collected (and the potential harm which could be caused with it) is much greater than ever before. Privacy standards and access rights should be in place before a smart metering roll-out is started. As not only the example of the Netherlands but also recent publications and discussions show, data protection and security are high on the agenda in discussions circling around smart metering roll-out, first of all from the point of view of consumes' privacy, but also from 39 Directive 95/46/EC of the European Parliament and of the Council of 24 October 1995 on the protection of individuals with regard to the processing of personal data and on the free movement of such data; Directive 2002/58/EC of the European Parliament and of the Council of 12 July 2002 concerning the processing of personal data and the protection of privacy in the electronic communications sector. Smart Meters Rollout in the Energy Community Page 35/107

37 the point of view of general security. 40 At present, the European Commission prepares a recommendation for the roll-out of smart metering in accordance with the requirements of Annex 1 of Directive 2009/72/EC and 2009/73/EC, which draws special attention also to the data protection issues. The list of functionalities given in Error! Reference source not found. includes also requirements for secure data communications with high consensus among EU Member States. Data protection and security can be achieved through two major instruments. Firstly, secure data communication (i.e. encryption of data transmission) should be in place, ensuring that data is not accessible for un-authorized parties. Secondly, a clear and functional legal framework should be enforced, setting out explicit rules on data access and handling as well as responsibilities to safeguard data protection and security. As outlined out above, the second major concern consumers may have is the fear that smart metering may lead to higher energy costs. There are two reasons for this concern. The first reason is the cost attributed to smart metering deployment. A full smart metering roll-out is a major investment in new metering and communication infrastructure. Although the extent of costs passed through to the final consumer is very much dependent on regulation, the end consumers may face an additional financial burden. It is the regulator's task to ensure that only the justified efficient costs are passed through to consumers and that these costs are shared with other parties gaining from smart metering deployment. Practice shows that one of the major expected benefits, i.e. the effect of smart metering on energy savings, will be felt by consumers. However, the potential energy savings may be rather uncertain and in any case will be unequally distributed. Nevertheless, consumers may benefit most from smart metering when they are given the means to reduce their energy consumption and to increase energy efficiency by identifying areas of high energy consumption and energy savings potentials. Additionally, consumers may be advantaged by new time-of-use tariffs imposing lower charges in off-peak times. Regulatory authorities may be reluctant to allow higher revenues to network operators to cover smart metering costs for different reasons. They may perceive the planned costs to be high, or for political or social reasons they may try to prevent price increases. Such a restrictive regulatory policy may undermine the success of the smart metering roll-out. Even if it is mandated by legislation, the scope of functionalities enabling energy savings will probably suffer from tight budgets. If consumer benefits resulting from smart metering deployment are higher than the associated costs, then passing efficient costs through to consumers is justified. To prevent consumer resistance and to mitigate consumer concerns requires an effort to increase consumer awareness of energy savings potentials and to strengthen their confidence in the proposed reforms in metering infrastructure. At the same time, political acceptance and so- 40 See for example Saurugg, Herbert, Cyper Security Austria, Smart Metering und mögliche Auswirkungen auf die nationale Sicherheit (Smart Metering and possible impacts on national security), Vienna, July Smart Meters Rollout in the Energy Community Page 36/107

38 cial affordability of any price increase cannot be disregarded and remain an important factor in the majority of the EnC Contracting Parties. There have been multiple examples in the past where consumers and politicians in these countries have been opposed to price increases regardless of whether these increases had been driven by objective economic reasons. The deployment of smart metering will naturally be accompanied by time-of-use tariff systems. Depending on tariff design and assuming low elasticity of consumption, higher energy prices in peak periods may lead to an overall higher bill despite lower prices in off-peak times. In the United States, it can be observed that much resistance stems from unexpected increases in energy bills. This was mainly due to the fact that time-of-use tariffs were implemented simultaneously with smart metering installations without properly preparing consumers for the new system. The obvious lesson from the US experience is that the introduction of new tariff schemes should allow sufficient time for consumers to prepare, and should be accompanied by considerable consumer information campaigns. The campaigns should provide information to consumers on the potential impact of tariff changes and explain how to adjust consumption behavior in order to reduce energy bills. We can conclude that strengthening consumer awareness, trust and knowledge is essential to mitigate consumer resistance. Smart metering deployment should be accompanied by an information campaign. This is particular relevant for the EnC Contractual Parties where in many cases customers are not well informed due to simple disinterest and/or lack of organized information channels using customer associations or industry information centers. The establishment of the smart metering deployment strategy should involve all stakeholders and take their views and concerns into consideration at an early stage. In addition, clear rules to safeguard data privacy should be set. If consumer benefits resulting from smart metering deployment are higher than the associated costs, then passing efficient costs through to consumers is justified. Regulators should ensure the integration of efficient cost into price control while at the same time ensuring that all parties benefiting from smart metering participate adequately in costs. 4.2 Legal/Regulatory Barriers Full-scale smart metering deployment is mainly driven by political decisions, starting with the high-level requirements of the Directives 2009/72/EC and 2009/73/EC and passing down to national laws, ordinances and regulatory decrees. Successful smart metering deployment is thus dependent on regulatory authorities, governmental and legislative bodies. These institutions have to play a significant part in assessing costs and benefits of smart metering deployment, setting up the roll-out scheme and monitoring the actual implementation. Without a Smart Meters Rollout in the Energy Community Page 37/107

39 clear legal and regulatory framework, market parties will be reluctant to commit themselves to the investments needed to set up a whole new communication and metering infrastructure. This will deter a smart metering roll-out and lead to inefficient results. Regulatory and legal barriers may stem essentially from two reasons: The necessary legal and regulatory framework is not in place at all; and The existing framework is insufficiently amended to incorporate smart metering and thereby contains provisions which hinder or delay smart metering deployment, and in this way increases costs or decreases benefits. The legal and regulatory framework will certainly have a decisive influence on the overall costs and benefits. This is particularly true in the case of a mandatory roll-out where the framework should set the responsibilities of market parties, time schedules, new tariff schemes, and minimum functionalities. Several of these aspects are discussed in the sections below Revenue / Tariff Setting and Incorporation of Costs of Smart Metering The lack of a consistent legal and regulatory framework sufficiently adjusted to foster a smart metering roll-out and to promote energy savings will pose a major barrier to successful smart metering deployment. The legal and regulatory framework should show commitment to the smart metering roll-out by governmental and regulatory authorities. Moreover it should explain clearly how the investment and operating costs will be accommodated in tariff regulation. Several EnC Contracting Parties apply different forms of rate-of-return and incentive-based regulation (usually revenue-cap regulation). Under these regimes, the regulator sets the allowed revenue of the regulated entities for each year individually, or for the whole regulatory period in advance, by assessing the capital and operating costs. In the course of price reviews and revenue setting, the regulator may also incorporate efficiency increase requirements in the allowed revenues. The use of such regulatory models and the fact that metering business remains regulated (and typically is and will remain part of the network operator) provide an appropriate platform for the integration of smart metering costs in the network price control. Setting up a smart metering infrastructure is a major investment. The responsible parties, i.e. in most cases the network operators, will not be willing to invest in smart metering if they are not certain that efficient net costs of the roll-out will be accommodated in their allowed revenues. Thus, the willingness of network operators is crucially required for a successful smart Smart Meters Rollout in the Energy Community Page 38/107

40 metering roll-out. The decisive factor will be the extent to which the allowed revenues and tariffs reflect the underlying costs for the provision of regulated services. In some cases we observed that regulatory authorities have been reluctant to allow higher tariffs in order to cover smart metering roll-out costs. However, this occurred in those cases where the roll-out decision was not (yet) politically mandated but rather based on the voluntary efforts of network operators. Normative pricing principles require primarily economic efficiency and cost recovery. However, introducing cost reflective tariffs often results in high price increases for small consumers. While the computation of cost reflective tariffs is a quantitative effort and depends mainly on the quality of available data and professional knowledge, their implementation for all customer categories has usually been a gradual process to achieve political acceptability and address social affordability. While the social and political constraints are understandable issues, in particular for the EnC Contracting Parties, it is a fact that non-cost reflective prices cause distortions in price signals and consumer behavior. In the context of smart metering, non-cost reflective prices may encourage energy consumption and undermine energy savings and the potential benefits. Therefore, it remains essential that regulators should strive to adopt end-user prices as well as network tariffs that reflect the costs of providing regulated services to specific groups of customers. The legal and regulatory framework should be aligned with the national smart metering roll-out plan. Moreover, the framework should provide market parties with certainty that efficient costs for building up and operating the required smart metering infrastructure are incorporated in the tariff regulation in the areas where such regulation applies Implementation of Time-of-Use Pricing Smart metering provides the technology to apply more sophisticated tariff schemes than simple one-zone or two-zone tariffs (c.f. section 7.1). Typically, three-zone tariffs differentiating between working days and weekends are used, as is shown in the following figure with an example from Canada where such tariff schemes are already widely applied. In theory, tariffs with a higher granularity or even dynamic pricing schemes are technically possible. Smart Meters Rollout in the Energy Community Page 39/107

41 Figure 10: Multi-Zone Tariffs Source: HydroOne Time-of-use tariffs should result in a better match of demand and generation, thus reducing the amount of peak capacity required and allowing for a more efficient use of available renewable energy sources. Dynamical pricing schemes may lead to even better results in systems with a very high share of intermittent generation. In addition, it seems that energy saved in peak times is not in all cases shifted to off-peak periods, but that as a result of load shifting the overall energy consumption decreases. If the implementation of time-of-use or dynamic tariffs is not sufficiently addressed in the legal and regulatory framework or even explicitly not allowed, benefits from smart metering deployment cannot fully unfold. Moreover, if the impact of new tariff schemes is not assessed in the social cost-benefit analysis, the case for smart metering might be hindered from the beginning. This is of particular relevance for many EnC Contractual Parties where retail price control continues to exist and regulated tariffs are frequently applied (for captive customers). In a liberalized electricity industry with functional competitive wholesale and retail markets where suppliers are free to agree with customers on the level and structure of their prices, regulation should focus on the monopoly network business only. The end-user prices would be subject to monitoring and ex-post control by the national competition authorities. Regulatory authorities are in charge of setting the allowed revenue and the rules for cost allocation and tariff setting for distribution networks, and retail market rules where markets are liberalized. Therefore, it is essential that regulatory authorities encourage and foster the development of innovative end-user tariff schemes by market players and regulated entities alike. Smart Meters Rollout in the Energy Community Page 40/107

42 4.2.3 Use of Standard Load Profiles Another regulatory obstacle may be found in the widely applied usage of standard load profiles. In a liberalized market environment, suppliers operating on the competitive retail markets typically use standard load profiles to schedule their energy purchases and network usage. With the implementation of smart metering, real-time data with high granularity will become available. If regulatory and legal frameworks continue imposing obligations on suppliers to procure and deliver energy into the network based on standard load profiles, the advantages offered by smart metering will not be realized. This is particularly relevant for the application of advanced pricing schemes (as discussed above) which can be supported by real-time metering and flexible demand response. If instead based on the metered data, procurement still has to be based on estimated standard load profiles, suppliers will not be advantaged by procuring energy for smart metering clients and thus no benefits can be passed on to consumers. Benefits could result where peak loads can be reduced or demand predictability is improved. The implementation of smart metering and time-of-use or dynamic pricing schemes should be accompanied by an amendment of the retail market rules, allowing suppliers to schedule real flows in line with actual demand Other Technical Regulation Technical regulations which have evolved historically may not correspond to the new smart metering technology or may no longer be in accordance with one other. If, for example, gas and electricity meters have different calibration periods and hence replacement cycles, additional costs may occur in a multi-utility approach. Additional costs may also occur by not aligning calibration periods of smart meters to the expected lifetime of communication modules. Due to potential consumer concerns regarding data privacy and in order to ensure interoperability of hard- and software and to foster competition and avoid stranded investments in case of supplier changes, clear rules on communication, data formats, security, handling and access are required. Such technical regulations should encompass for instance encryption standards, role-based data access, purge dates for gathered data but also regulations on data formats, communication protocols, and hard- and software interfaces. Especially the latter must not be defined in a legal or regulatory document but could also be achieved by voluntary industry agreement (c.f. Open Meter Project). In addition, the current development towards common standards at European level must be taken into account. Smart Meters Rollout in the Energy Community Page 41/107

43 National technical norms and regulations, for instance the general legal framework for metering and measurement should be reviewed and if necessary amended to take into account the requirements of the new technology. If, for example, legal provisions require the regular visual inspection of a metering device, this may decrease the potential benefits of smart metering. 4.3 Economic Barriers Manifold economic barriers have additional potential to delay smart metering deployment or to reduce the net benefit. These barriers need to be dealt with in order to ensure a successful and efficient smart metering deployment. One of the primary economic barriers is that whereas costs for installing and operating a smart metering infrastructure can be very clearly assessed, there is often high uncertainty regarding the benefits. Many benefits rely on assumptions and forecasts, such as the amount of energy which might be saved by smart metering. Other benefits are difficult to reliably quantify, such as a gain in leisure time for a consumer due to fewer errors in invoices and hence less time spent on the customer service hotline. Practical experience and historical data is not yet sufficient to enable a full evaluation of the economic benefits of smart metering, although this information is becoming more available compared to a few years ago. Apart from being in general ridden with uncertainties, benefits from energy savings and increased energy efficiency will also be distributed very unequally, depending on individual consumers' behavior, awareness, education, willingness and the consumers' leeway to reduce energy consumption. Given the relatively high certainty with regards to costs of smart metering deployment and the significantly lower certainty with regards to the benefits, a cost-benefit assessment carried out with the appropriate due diligence may lead to a result where costs may be overestimated whereas benefits may be underestimated. Altogether, this would result in a costbenefit analysis which is biased against a smart metering roll-out. In addition, a cost-benefit assessment from an individual perspective may lead to negative results as the scope of the assessment is for instance limited to benefits within the scope of the individual. Another serious economic barrier is a possible split between the cost bearing party and the beneficiaries. This would for instance be the case if costs were fully borne by DSOs within the existing network or metering charges, whereas energy savings as a major benefit occur on the consumer s side. Full-scale smart metering deployment is highly capital intensive and may expose the responsible party to significant financial risk. In many countries, governments do not plan to provide direct subsidies for smart metering and network operators rely solely on recovering their costs from regulated charges. Therefore regulators and governments should provide clear commitment and transparent arrangements. If the perceptions of Smart Meters Rollout in the Energy Community Page 42/107

44 regulators and network operators with respect to the cost recovery largely diverge, this will most certainly become a serious barrier to smart metering deployment. Finally, economic barriers may occur due to limited access to capital markets, if the network operator cannot provide the necessary credentials (e.g. government guarantee, rating). Economic barriers that may cause a delay in smart metering deployment should be properly addressed when assessing a roll-out and in particular after the roll-out decision has been taken. 4.4 Technical Barriers Besides the barriers discussed in the previous sections, technical barriers may also hamper successful smart metering deployment and need to be addressed. The main technical barrier with regards to smart metering deployment is the present lack of standardization. Commercially available smart metering components often lack interoperability due to highly proprietary solutions. Open hardware and communication standards are still under discussion and development. However, progress was already made under Mandate M/441, which was given to CEN, CENELEC and ETSI by the European Commission 41, or the Open Meter project, funded by the European Commission within the Seventh Framework Programme. 42 Established and open standards will enable a modular set-up of compatible devices. This is of particular relevance if consumers want to change their suppliers in a liberalized market, where incompatible metering components may result in barriers to supply changes and thus hamper competition. In the worst case, the lack of standardization (also with regards to new components installed at a later time) may even result in stranded investments. Only a limited number of manufacturers offer smart metering hardware. So far, the market is still small, but it has been developing rapidly in recent years. Hard- and software as well as communication infrastructure providers are organized in the European Smart Metering Industry Group (ESMIG). Many of the member companies are small, but also big players such as Siemens, IBM and ABB are present. Due to an absence of standards, manufacturers have developed a variety of proprietary solutions, while interoperability of devices produced by different manufacturers may be lacking. The existing industry structure and lack of standards lead to the following problems: 41 Standardization mandate to CEN, CENELEC and ETSI in the field of measuring instruments for the development of an open architecture for utility meters involving communication protocols enabling interoperability M/441, SMART METERS CO-ORDINATION GROUP, FINAL REPORT Smart Meters Rollout in the Energy Community Page 43/107

45 Due to the number of players and the limited size of the market, economies of scale are not achieved, resulting in fairly high costs for smart metering components. This is however expected to change in the future if a real mass market is established. Devices of different manufacturers are often incompatible. This carries the inherent danger of stranded investment, e.g. if at a later stage a gas meter shall be integrated, or if in the case of a supplier change, the meter is not compatible with the supplier s infrastructure. Devices are often designed as stand-alone, thus complicating a modular approach where all components could be replaced separately. The modular approach is further impeded by the lack of interoperability. The lack of modularity is hindering the easy adaption of a smart metering infrastructure to the specific needs of an individual roll-out scheme. Specially designed solutions are often needed to match individual requirements, resulting in higher specific costs. Cooperation between DSOs or Metering Service Providers and the smart metering supply industry is weak; development of smart metering could clearly be improved by better cooperation and a more goal-oriented component design. Further development of smart metering technology is expected in the future. Metering devices installed in recent years may need to be replaced before the end of their economic lifetime to enable new, innovative services. Installation of a smart metering infrastructure is a highly demanding technical task requiring a qualified labor force. A short-term large-scale roll-out might be hindered by a lack of resources. Furthermore, production capacities of smart metering suppliers are limited, which might lead to supply problems if a short-term roll-out is mandated in several European countries at the same time. In order to ensure successful full-scale smart metering deployment, it is recommended to achieve sufficient standardization of hard- and software, of communication protocols and of data formats. Standardization may be achieved by industry agreement, as for instance by the Open Meter project, by administrative arrangements initiated by competent authorities or by a combination of both. Smart Meters Rollout in the Energy Community Page 44/107

46 5. STRUCTURE AND SET-UP OF A COSTS-BENEFIT ANALYSIS As a roll-out of smart metering is associated with significant investment costs, a thorough assessment of all possible costs and benefits is required before a decision on the smart meter roll-out and infrastructure can be made. This applies in particular when a roll-out of smart metering is made mandatory by legislation (see chapter 6.2). As described in chapter 3, such an assessment of all costs and benefits in the form of a cost-benefit analysis is also suggested by the European legislation (Annex I of Directives 2009/72/EC and 2009/73/EC). In order to carry out a social cost-benefit analysis, detailed information on all possible costs and benefits is needed as input data. As the result of a cost-benefit analysis can be strongly influenced by the selection, definition and specification of the input data, this step is crucial in avoiding bias in favor or against a smart metering roll-out. The following section describes how costs and benefits can be defined and which factors influence the costs and benefits of smart metering. Section 5.2 explains potential costs of smart metering and sections 5.3 to 5.6 analyze the potential benefits for different stakeholders. The set-up of a cost-benefit analysis and its role is discussed in section 5.7. This report is accompanied by a template (submitted separately) that provides on overview of the set-up of a national cost-benefit analysis for a smart metering roll-out, highlights the major steps to carry out the CBA and incorporates an worksheet with the major input data Definition of Costs and Benefits Major costs associated with smart metering are the purchasing, installment and operating costs of the smart meters as well as the investment costs for advanced data collection and data communication tools. Major benefits typically associated with smart metering are energy savings due to increased efficiency or sufficiency and due to load shifting, reduced metering costs, improved security of supply and reduced non-technical losses. Figure 11 provides an overview of possible benefits of smart metering to major stakeholders, namely distribution network (or system) operators (DSOs), consumers, suppliers and society as a whole. 43 At present, a recommendation of the European Commission with regards to the preparations of a smart metering roll-out in accordance with Annex I of Directives 2009/72/EC and 2009/73/EC is in preparation. This recommendation will also include guidance on the necessary steps in order to perform a cost-benefit analysis for the roll-out of smart metering and potential parameters to be included in such analysis. Smart Meters Rollout in the Energy Community Page 45/107

47 Improved invoicing processes Simplified change of supplier Better information on consumption patterns Cost savings on the procurement side Reduced bad debt Suppliers DSOs Reduced fraud Lower costs for meter reading Faster fault detection and restoration Voltage quality monitoring Compliance with energy savings and emission reduction targets Reduced capacity demand Improved quality of supply Labour market effect Society Consumers Improved information on consumption Lower energy costs Innovate tariff schemes Accurate bills Improved customer service Integration of microgeneration Figure 11: Possible Benefits for Separate Stakeholders Note: The picture provides a generalized overview and assumes that metering is the responsibility of DSOs. Source: KEMA Costs and benefits of smart metering however very much depend on the technical specifications of the smart meters and the smart metering infrastructure rolled-out. More advanced smart metering systems with a larger range of functionalities (as described in chapter 2) could provide greater benefits and a larger range of benefits, but are also likely to be more expensive than basic smart metering systems. The technical specifications of a smart metering infrastructure on the other hand are strongly determined by the policy objectives pursued with the roll-out of smart metering. While it is possible to give a rough indication of the costs of different smart meters, it is not meaningful to provide general numbers of the benefits of smart metering per meter or customer, as these strongly depend on the individual specifications of the smart meters, the group of stakeholders concerned, country or regional specifics and a range of other assumptions assessed in a cost-benefit analysis. Drivers for implementing a smart metering infrastructure can be quite diverse. As described in chapter 3.3, in Italy a reduction in energy theft and fraud (i.e. a reduction of commercial losses) has been the main driver for a roll-out. A legal requirement to change from annual to monthly invoicing based on actual meter readings was the trigger for the Swedish distribution network operators to (voluntarily) roll-out smart metering (reduction in metering costs). In most European countries, the increase of energy efficiency (reduction of final energy consumption) driven by climate policy is at the top of the agenda, whereas in the US, due to a capacity shortage, demand side management and improved security of supply is the key driver for smart metering deployment. Smart Meters Rollout in the Energy Community Page 46/107

48 The costs and benefits arising from a smart metering roll-out also strongly depend on the local circumstances and the status quo of the electricity system. The potential of smart metering to contribute to load shifting and energy savings is strongly related to the types of energy consumption and the consumption patterns. Electricity used for heating and cooling for example can be shifted easily. However, this will require automation, which is less likely to be available for household consumers compared to industrial and larger commercial consumers. With cooling and heating as loads which are relatively easy to shift, the percentage of load which could be shifted is much higher in a country where for instance air-conditioning is widely applied. The ability to make energy savings also depends on the overall level of per capita energy consumption. In countries with very high (careless) energy consumption, the potential to significantly reduce energy consumption might be comparably high. Likewise, in countries where household budgets are typically very limited and energy costs are consuming larger shares of the budget, the incentive to realize cost cuttings, e.g. by energy savings or demand response measures is much stronger. The latter might be relevant for many EnC Contracting Parties. As a result, potential costs and benefits of smart metering can be quite varied in different contexts and countries. Most of the costs and some of the benefits related to smart metering can be estimated before making a roll-out decision. New services and functionalities likely to arise in the future and provide additional benefits cannot however be properly estimated before the roll-out has taken place. Manufacturers of products such as household appliances and the service industry for example will adapt to smart metering technology and will develop and offer a wide range of specially designed products and services, e.g. further increasing energy efficiency by intelligent household control or enhancing consumer welfare with increased comfort. The costs and benefits of smart metering may be unevenly distributed between the different stakeholders. Clearly costs and benefits directly affect the network operator or the supplier replacing the old meter with a smart meter and the customer whose old meter is replaced with a smart meter. But costs and benefits also affect (indirectly) other market participants, such as other network operators, generators, suppliers or customers and the society as a whole. Different stakeholders are likely to benefit to different extents from a deployment of smart metering. Costs might for example only be borne by one market participant (e.g. the customer), whereas benefits might be split across a larger number of market participants (network operators, suppliers, customers etc). Costs might also mostly arise in the shortterm, whereas some benefits of smart metering might only occur in the long-term. Smart metering is primarily an electricity topic, in particular of course in those countries where gas plays no or a negligible role in residential energy consumption. In many EnC Contracting Parties household gas consumption plays only a minor role compared to Western European countries like the Netherlands or Germany. The benefits of an application of smart Smart Meters Rollout in the Energy Community Page 47/107

49 metering are also generally greater for electricity than for gas. Benefits from load shifting for example are only applicable to electricity since fluctuations in electricity production and demand have to be balanced in much shorter time intervals than for gas, which generally varies at a much slower pace. Given the nature of gas usage for heating purposes, a load shift from peak to off-peak times would also make little sense. With regards to savings, the impact on gas consumption is more limited, as the purposes of electricity usage are manifold with plenty of individual and independent consumer decisions on whether or not use electricity on a daily basis, where constant or regular feedback will have the strongest effect. In our description of potential costs and benefits of smart metering we therefore focus primarily on smart metering for electricity. A roll-out decision for smart metering requires a thorough assessment of costs and benefits. Costs and benefits of smart metering however depend on a number of country specifics as well as the deployment strategy and objectives. A simple transfer of cost benefit assessments from one Contractual Party to another is therefore not possible. Furthermore, the distributional effects, i.e. different stakeholders facing quite different costs and benefits of a smart metering roll-out, also have to be taken into account. The categories of costs and benefits are identical to a significant extent for electricity and gas. However, some of the most significant benefits of smart metering, such as benefits from load shifting and energy savings, are much greater for electricity than for gas. 5.2 Potential Costs of Smart Metering Costs of smart metering include the costs for the smart meters and the costs of the communication and data processing infrastructure required to establish a true smart metering system. The costs for additional applications which provide further benefits of smart meters such as in-house displays, web portals or SMS notifications could also be considered as costs of smart metering. 44 Smart meters have to be procured, installed, read, serviced and maintained resulting in substantial capital and operation costs. The costs of smart metering strongly depend on the technical specifications of the smart meter and the communication technology used. The costs of the communication infrastructure required for remote meter reading such as Power Line Communication (PLC), GPRS/GSM or DSL depend on the existing communication infrastructure, the size of the network area and the customer density. 44 Costs of the cost-benefit analysis itself are not included in the cost-benefit analysis, as they arise in any of the analyzed scenarios, including the scenario to keep the status-quo and not roll-out smart meters. Smart Meters Rollout in the Energy Community Page 48/107

50 Stranded costs of the existing meters and metering infrastructure that become obsolete are methodologically not regarded as part of an economic cost-benefit analysis. However such costs, in particular when the conventional meters have been recently installed, have to be taken into account when the maximum allowed revenues are set by the regulator. As indicated in the previous section, costs of smart metering not only depend on the technical specifications of the smart metering infrastructure, but also on a number of country specific factors, such as the status quo of the existing local electricity system. Specifying the costs of specific components of smart meters can therefore only be indicative. The cost data presented below are obtained from pilot projects, manufacturers and energy suppliers, as well as from other references and therefore may not present a consistent picture of the expected costs of introducing smart metering. One reason for the high bandwidth of cost expectations is the enormous development in the markets for electronic components and communications infrastructure. In this study we assume that an increasing deployment of smart metering will lead to continuing strong growth and significantly falling prices in the future. Also the costs of owning and operating the communications infrastructure will further decrease in future years. It seems reasonable to expect an annual costs saving potential of 5-10%. Electronic meters are available on the market at a great variety of prices. These price differences result primarily from different functions (interfaces, data storage, etc.), particularly if the communication unit is already integrated into the meter. Moreover, the number of procured meters has significant impact. As a result, prices of smart meters cannot be easily found on price lists published by the manufacturers, but are rather dependent on the exact specifications of the smart meters and a result of individual negotiations with the manufacturers. Recent KEMA studies showed average costs of for smart metering devices covering a more comprehensive set of functionalities. The costs for a simpler meter (primarily limited a remote meter reading functionality) were estimated at 40-60, for a standard electronic meter 36 was assumed. But these simpler meters have no integrated communication module; hence they have to be connected with a communication unit, for example with a Multi Utility Communication-Controller (MUC-C). The costs of a MUC-C are estimated at about In the medium term it is expected that the hardware of a smart meter may not be much more costly than that of a conventional meter. Price differences between different types of smart meters are however still likely to arise because of differences in the software capabilities of the meters Whereas for example almost all smart meters are able to measure at least some basic parameters of power quality, further specifications in the smart metering software would be required to allow the network operator to monitor power quality at end-user level. Smart Meters Rollout in the Energy Community Page 49/107

51 In pilot projects, the costs of data concentrators were identified as , the installation costs as Using one concentrator for typically meters costs thus per meter. A communication module which is already integrated into the meter (i.e. in most cases this is the electricity meter, theoretically all other meters can also be used) can be acquired at lower costs than a separate communication module, if not several meters are connected to this communication module. Different assumptions regarding the lifetime of the devices, multiutility strategies and different market expectations have resulted in different concepts being applied in practice. GPRS/GSM-enabled devices are more expensive than meters with a PLC communication, because the modulation of a PLC signal is technically much easier than a GSM connection. The price difference is about per meter. For DSL compatible meters, the additional costs are higher, as there is no broad market for this solution yet. It is expected that a significant potential for economies of scale exists, for DSL as well as for GSM devices. The majority of the currently installed units are PLC devices. Regarding the operational costs of DSL, costs of disturbances have to be considered. By using parts of the customer s infrastructure for the DSL based communication, it is likely that more intensive customer support will be required when the customer s internet connection is disturbed, irrespective of the reason for the disturbance. KEMA s observation of the European market is that most utilities consider comprehensive PLC as the most cost effective option, requiring a significant market penetration (30-70% are considered significant). Only for isolated buildings (e.g. in rural areas) does GSM seem to be the most cost effective option. The following table shows the costs of the technical infrastructure for smart metering: Smart Meters Rollout in the Energy Community Page 50/107

52 Min- Variation Max- Variation Procurement costs smart meter ( ) Installation costs smart meter (electricity) ( ) Additional installation costs (selective roll-out) ( ) Annual operating costs smart meter (electricity) ( ) Procurement costs MUC-C ( ) Installation costs MUC-C ( ) 8 20 Annual operating/maintenance costs MUC-C ( ) Procurement costs smart meter (gas) ( ) Installation costs smart meter (gas) ( ) Annual operating costs smart meter (gas) ( ) Procurement costs inhouse display ( ) 5 50 Installation costs inhouse display ( ) Installation costs PLC per smart meter ( ) Annual operating costs PLC per smart meter ( ) 0,7 2 Additional costs GPRS/GSM per smart meter ( ) Annual operating costs GPRS/GSM-modem ( ) 1 10 Installation costs DSL & additional costs per smart meter ( ) Annual operating costs incl. queries DSL-Modem ( ) Table 3: Indication of Potential Costs for a Smart Metering Infrastructure Source: KEMA Major costs associated with smart metering are the purchasing, installment and operating costs of the smart meters and the investment costs for advanced data collection and data communication infrastructure. 5.3 Benefits to Network Operators Smart metering has several benefits for network operators. A wide deployment of smart metering provides a network operator with precise information on the actual consumption and feed-in at specific sites of its low voltage distribution network, offering a range of potential savings directly to the network operator. System-wide benefits arise from optimized distribu- Smart Meters Rollout in the Energy Community Page 51/107

53 tion operations, improved network reliability and the contribution of smart metering towards quality of supply, for example by facilitating the detection of outages and by reducing restoration times. Potential benefits of smart metering for network operators include improvements in the security of supply by a faster fault location and power restoration, improved monitoring of voltage quality, the ability for quick remote disconnection or reconnection of customers and the ability for remote reduction or restoration of power. 46 Smart metering can help network operators to detect and locate faults and power outages more quickly. Reducing the time period between the time the fault occurs and the time the grid operator s control centre receives this information (automatically) via the smart metering communication infrastructure allows the network operator to immediately dispatch the technicians required to restore the fault. By identifying fault locations more quickly, the outage time can be reduced. This provides an obvious benefit to consumers and savings to the distributor from reduced costs by more accurately dispatching crews. When a regulatory scheme for quality of supply is applied linking the actual network reliability (number and duration of outages) to quality standards and penalties or a quality incentive scheme network operators can also benefit from higher revenues following reduced outage duration times. Quality of supply regulation is being increasingly introduced in the European Union in particular for electricity distribution network operators and also considered or developed in a number of EnC Contracting Parties such as Serbia and Macedonia. Smart metering generates real-time, accurate and comprehensive information on the distribution network (e.g. voltage quality, losses), which allows more accurate prediction of electricity flows to be used for improved network and maintenance planning. Detailed information on the current status of the network also provides a basis for sound investment planning. Smart metering together with the application of time-of-use tariffs can provide customers with information on consumption and prices and encourage them to shift their energy consumption into times when energy prices are at a lower level. Smart metering can thus reduce the demand at peak times and thereby reduce the maximum network capacities required to distribute electricity at peak load, which in turn reduces the need for network investments. Integrating smart meters into the IT infrastructure of the network operator can also help to optimize processes and reduce operational costs (process optimization). Further benefits can also be gained from a multi-utility approach integrating gas, district heating or drinking water metering. 46 Remote dis-/reconnection of gas supply is theoretically also possible, however if applied, it is subject to much tighter security provisions. For gas, before reconnection of supply all appliances and valves must be checked in order to guarantee safety. This can be done by the consumer with the necessary instructions provided via smart metering. Smart Meters Rollout in the Energy Community Page 52/107

54 Smart metering can also have a significant impact on the reduction of commercial losses (detection of fraud and energy theft), a topic particularly relevant for the majority of the EnC Contracting Parties. Smart metering allows for an easier detection of previously unmeasured consumption that resulted from bypassing the meter. Furthermore, smart metering also provides more accurate information about the location of losses and theft. Smart meters can also be fitted with anti-tampering devices alerting the DSO automatically when manipulation of the meter is attempted. 47 In most countries as is the case in the EnC Contracting Parties the metering function is also provided by the distribution network operator. Depending on the market model (see chapter 6), the operation and the reading of the meters (metering services) could also be carried out by the supplier or a separate metering company. Potential benefits of meter operators may include reduced costs of manual meter reading and reduced costs through remote disconnection or reconnection. With smart metering, digital meter data are automatically submitted to the metering operator's data centre. Manual meter readings and manual entering of meter data into data management systems are therefore no longer required. Data can be easily processed and evaluated and meter-to-bill operations can be significantly improved. Furthermore, not only the meter reading, but also the disconnection and reconnection of customers can be handled remotely and (partly) automatically, reducing the need to send out technicians to customer sites to suspend and resume electricity supply. Additional benefits can arise in the case of a multi-utility approach, integrating metering for gas, district heat and/or water. Whether these cost savings represent a benefit to the customer as well as the meter operator depends on the ability of the metering service provider to passthrough all efficient and justified meter related costs directly to the customers. Where a large labor force is employed for manual monthly meter readings as in many of the EnC Contracting Parties automated meter reading might however have a substantial negative effect on employment. Also when labor costs are relatively low, benefits from a reduction in labor costs (operating costs) with smart metering might be lower compared to the high capital costs resulting from investments in a smart metering infrastructure. 47 Reduction of fraud is not an immediate economic benefit from a societal perspective, as consumption is not directly influenced and no additional welfare is generated. Nevertheless, if payment discipline can be improved, energy savings are also likely to be triggered. Smart Meters Rollout in the Energy Community Page 53/107

55 Potential benefits of smart metering for network operators include improvements in the reliability of supply by a faster fault location and power restoration, improved monitoring of voltage quality and the ability for quick remote disconnection or reconnection of customers or power. Further benefits can arise from reduced operational costs (through integration of smart meters in the IT infrastructure of the network operator) and improved network and maintenance planning (utilizing the more accurate prediction of electricity flows provided by smart metering). Meter operators (which in most case are the distribution network operators) can benefit from reduced metering costs (reduced costs of manual meter reading and remote disconnection/reconnection). 5.4 Benefits to Suppliers Following the unbundling requirements specified in the internal market Directives of the European Union (and their implementation in the Energy Community), network operation and (end-user) supply have to be unbundled into two separate business activities or entities (in the case of legal unbundling). Therefore separate benefits of a roll-out of smart metering for the supplier could and should also be identified. Smart metering can, for example, reduce the likelihood of incorrectly read or entered meter data leading to faulty invoices, which in turn reduces the number and costs of customer complaints (including reduced customer service centre staff). 48 The integration of smart meters in the IT infrastructure of the supplier and the further automation of the data processing and invoicing process can also result in reduced costs of the meter-to-bill operations (process optimization). The possibility of remote and instant disconnection of customers by the meter operator can also help to reduce the risk of payment default for the supplier. Smart metering also enables suppliers to offer new tariffs and services arising from detailed information on individual end-user's consumption patterns. Such new services could for example help the customer to become more energy efficient (see also chapter 7). Suppliers also have the opportunity to offer customized contracts reflecting individual consumption patterns. These contracts may include time-of-use or more sophisticated tariff elements and might also provide for automatic demand side management. Furthermore smart metering might allow the supplier to use actual load profiles of individual customers rather than standard customer load profiles. Through improved load profiling and forecasting suppliers are 48 Even if metering is provided by a separate entity, invoicing is carried out by the supplier, hence complaints due to faulty readings are generally directed to the supplier. Smart Meters Rollout in the Energy Community Page 54/107

56 able to more precisely predict their customers' demand at specific points of time, which allows them to reduce their wholesale purchasing costs. Smart metering may also provide benefits to electricity suppliers by improved customer satisfaction resulting in a higher willingness to pay and higher customer retention. Customers could benefit, for example, from more frequent and detailed metering and more accurate billing 49 or from easier and quicker customer switching procedures due to real-time metering, allowing customers to change their supplier (at least theoretically) in real-time or at very short notice and on any chosen date. 50 Suppliers can benefit from smart metering through improved invoicing processes (more accurate and frequent billing), resulting in higher customer satisfaction and retention and reduced payment default (via remote disconnection). Smart metering may also allow suppliers to reduce their energy purchasing costs through improved load profiling and forecasting. 5.5 Benefits to Consumers Smart meters can provide consumers with detailed information on their consumption behavior during different periods of the day. Actual and historic consumption data can, for example, be shown on an in-home display or on a computer screen, either provided by a direct data link or on a web page fed with the meter data. Smart metering together with price signals can therefore make the overall costs of electricity consumption and individual consumption patterns more transparent to the customers. Thereby customers are for example able to understand the impact of individual electricity devices or a certain consumption behavior on their energy bill. Such detailed information might also make the environmental effects of consumption behavior, such as the resulting greenhouse gas emissions, more transparent for customers. Constant feedback on consumption and associated costs will increase the consumer s awareness and willingness to save energy. It allows customers for example to decide when and for how long to connect or disconnect some of their electric devices. Several previous studies show a broad range of potential savings. Recent comprehensive studies as well as 49 A higher frequency of actual meter readings provides however a smaller benefit to most of the Contracting Parties of the Energy Community where monthly meter readings are already a common practice, whereas in most EU member states annual actual meter readings are more commonly carried out. 50 Whether improved customer switching is only beneficial for customers and not for suppliers depends on the possibilities of the supplier to gain new customers outside its incumbent service territory. For large incumbent suppliers, easier and quicker customer switching may accelerate the loss of customers in the incumbent territory which is not compensated by new customers in other areas and therefore may result in more costs than benefits. Smart Meters Rollout in the Energy Community Page 55/107

57 KEMA s own observations show that electricity consumption savings between 5% and 10% appear to be realistic. 51 Achieving energy savings with smart metering is however highly dependent on the effectiveness of the feedback on energy use given to consumers and the willingness and ability of the consumers to respond to this feedback. The ability and willingness of customers to realize energy savings also depends on the level of the end-user tariffs and the percentage of the monthly income spent on electricity expenses, whereas higher tariffs or a higher share of income spent on electricity consumption clearly set stronger incentives for energy savings. Also the range of electricity devices used by a customer and the customer's ability to replace old devices with more energy efficient equipment influence the scope of customers to realize reductions in electricity consumption. However not all consumers may be able or prepared to shift or reduce their demand. Accordingly some of them may even face higher energy bills. Consumer education is necessary to achieve changes in consumption behavior. The existence of the smart meter or some sort of consumption feedback itself will not necessarily result in substantial energy savings. The consumer needs to be taught how to use this new information in order to really achieve sustainable energy savings. Customers can further contribute to energy savings if they are offered time-of-use or loadvariable tariffs enabling them to save on their energy bills by shifting certain usage (e.g. dishwasher, heating, cooling) to cheaper periods (requiring less generation capacities and production during peak-load periods). The possibility to offer real-time pricing and innovative tariffs, as well as interfaces between smart metering and household appliances could result in various new types of energy services being available to customers to help manage consumption (and costs) and to promote more energy efficient and green energy networks (such as demand side management, i.e. the direct control of household appliances, see also chapter 7). Smart metering can also facilitate pre-payment options which allow customers to pay in advance and hence to better manage their budgets. In addition to energy savings, customers may also benefit from more frequent and detailed meter reading and more accurate invoices reflecting actual consumption. With smart metering, invoicing is based on real meter data rather than estimated consumption (applies only where manual meter reading does not take place monthly so less relevant for most of the EnC Contacting Parties). Customers would no longer face imposed under/over payments which might require settling at a later date. This could help to improve customer satisfaction 51 Cf. for instance Darby, Sarah, The Effectiveness of Feedback on Energy Consumption, Oxford, 2006, KEMA, Endenergieeinsparungen durch den Einsatz intelligenter Messverfahren (Smart Metering),, Bonn, June 2009 and Van Dam, S. S., Bakker, C. A. and Van Hal, J. D. M., Home energy monitors: impact over the medium-term, Building Research & Information, 38: 5, Smart Meters Rollout in the Energy Community Page 56/107

58 and reduce the number of customer complaints, compared to traditional metering when the settlement occurs after several months or a year. It is also possible for a customer to agree with the respective supplier on how frequently invoicing takes place and to receive an invoice on demand (e.g. when moving from one home to another). Smart metering can also have a strong impact in simplifying customer switching procedures as smart meters can be easily read at any time on request. Automation and simplification of data exchange through smart metering should speed up the process for changing suppliers and simplify the action required from the customer to make the change. The transparency of individual electricity consumption patterns and costs provided to the customer by smart metering also allow the customers to make more informed decisions on the selection of the most convenient supplier, further facilitating customer switching. Customers may furthermore benefit from reduced metering and operational costs through remote meter reading and remote dis- or reconnection of customers, if the cost savings made by the meter operator (or network operator) are passed on to the customers. Depending on the location of the conventional meters (whether located outside a building or inside) smart metering may have also the additional benefit that it requires no more home intrusions by meter readers. Smart meters can provide improved information and/or price signals, making the costs of energy consumption more transparent to consumers, resulting in reduced consumption and/or shifting of load to periods with lower tariffs. Consumers may also benefit from more accurate meter reading and invoices and easier switching procedures. 5.6 Benefits to Society Depending on the type of smart meters, the tariff schemes offered and the market environment, smart meters can facilitate energy savings, demand response and direct load control and thereby reduce demand at peak (and off-peak) times, resulting in lower wholesale prices and reducing the need for investments in generation, transmission and distribution capacities (avoided costs). With a contribution to increased energy efficiency and reduced carbon emissions, through reduced consumption and the facilitated integration of distributed generation, smart metering can also play a role in mitigating the effects of climate change. A large investment program, such as deploying a full-scale smart metering infrastructure might also have a positive impact on economic development and employment. Regulatory authorities can use smart metering to improve quality of supply regulation, in terms of reliability and voltage quality, as smart metering provides the regulator with more precise and detailed statistics on reliability performance (number and duration of outages). Smart Meters Rollout in the Energy Community Page 57/107

59 With smart metering all outages can be recorded, regardless of the outage duration and the voltage level(s) affected. 52 The data gathered with smart metering can therefore also be used to design more advanced incentive schemes for quality regulation resulting in higher levels of quality of supply. Additionally smart metering enables improved monitoring of voltage quality (voltage levels, dips). To perform this task meters (or a certain share of them) should be equipped to carry out such monitoring functions. Depending on the type of smart meters, the tariff schemes offered and the market environment, society as a whole may also benefit from reduced peak demand (resulting in lower wholesale prices), lower investments needs in generation, transmission and distribution capacities (future avoided costs) and from reduced carbon emissions. Regulatory authorities and electricity users may also benefit from the improvements in quality of supply regulation that smart metering is facilitating. 5.7 Set-up of Cost-Benefit-Analysis The decision for a nationwide smart-metering roll-out should be based on sound economic analysis of the costs and benefits. If such an assessment results in a positive net benefit of smart metering deployment, EU Member States are obliged to ensure that electricity meters are rolled-out in a period of ten years and that a schedule for the roll-out of smart metering in the gas sector is decided upon (Annex I of Directives 2009/72/EC and 2009/73/EC). 53 The major element of such an economic assessment is a social cost-benefit analysis. Within the Energy Community all Contracting Parties are expected to carry out such an assessment by January 1, The EnC Contracting Parties have therefore slightly more time to carry out a cost-benefit analysis for smart metering than EU member states. 55 According to the Ministerial Decision however no extension for the roll-out of smart metering deploying at least 80% of customers with smart metering by 2020 after a positive assessment of costs and benefits is given to the EnC Contracting Parties. They face therefore a tight timetable for an assessment and possible roll-out of smart metering. 52 Traditionally, automatic fault monitoring extends only to the medium voltage level, whereas faults on the low voltage level often go unnoticed until consumers alert the distribution network operator. 53 One of the main tasks of such economic assessment might be not only to identify costs and benefits but also which configuration, deployment strategy and schedule will provide the highest benefit. 54 Decision of the Ministerial Council of the Energy Community, D/2011/02/MC-EnC: Decision on the implementation of Directive 2009/72/EC, Directive 2009/73/EC, Regulation (EC) No 714/2009, Regulation (EC) No 715/2009 and amending Articles 11 and 59 of the Energy Community Treaty. 55 Directive 2009/72/EC states that a cost-benefit analysis has to be conducted by September 3, Smart Meters Rollout in the Energy Community Page 58/107

60 A cost-benefit analysis is a tool used to provide criteria for investment decision making by systematically comparing the benefits and costs over the life span of a (investment) project. It is widely applied at societal level as well as at company (i.e. the investor's) level. A financial analysis of costs and benefits from the perspective of a private investor maximizes the net benefits of the company carrying out the investment, i.e. the party investing in smart metering. It only analyses the costs and benefits arising to the investor and therefore also includes all taxes and subsidies to be paid and received by the investor, but does not include any external effects and wider economic benefits incurred by other parties. The social costbenefit analysis focuses on the overall long-term costs and benefits for society as a whole, that is all possible stakeholders directly and indirectly affected by a smart metering roll-out (including externalities such as environmental impacts, and costs and benefits to third parties 56 ). Assessing the social net benefit of a project does not include taxes and subsidies as these are only regarded as transfers. This gives the social cost-benefit analysis a wider economic character with the objectives of maximizing the welfare of a society (country or region) as a whole. A social cost-benefit analysis (CBA) 57 generally consists of the following parts: The selection and definition of input data and model parameters Assumptions on the future development of input data and definition of expected minimum, maximum and base (average) values of the input parameters Definition of alternative (smart metering roll-out) scenarios (e.g. regarding the deployment strategies and type(s) of smart meters) Definition of costs and benefits for other stakeholders (external effects) Assessment of the monetary effects (financial and monetized indirect (external) effects) of a smart metering roll-out on other stakeholders (economic analysis) Assessment of (further) macroeconomic effects (e.g. on employment, GDP, etc.) Calculation of the total net benefit for different scenarios discounting future costs and benefits with an appropriate rate Sensitivity analysis of the results in order to determine critical input variables 56 Third parties in this context include parties who are neither the network operator/meter operator carrying out a smart metering roll-out, nor the end-user on whose premises a smart meter is installed. Such third parties could be, for example, other network operators, end-users or generators, who could also benefit from a smart metering roll-out (as regards for example an increase in system stability), but who could also face costs from a smart metering roll-out (for example if the installation costs of smart metering are to be covered by all network users or if a potential reduction in consumption leads to losses for the electricity generators). 57 A cost-benefit analysis from the perspective of a private investor would assess the financial effect (net benefit) of a smart metering roll-out in a similar way except that all external and macroeconomic effects would clearly (as pointed out above) not be included in the analysis, i.e. the fourth, fifth and sixth bullet in the above list. Smart Meters Rollout in the Energy Community Page 59/107

61 The decision for a nationwide roll-out of smart-metering should be based on an economic analysis of all costs and benefits in a social cost-benefit analysis. All EnC Contracting Parties are obliged to carry out such an assessment before January 1, 2014, according to the Decision of the Ministerial Council of the Energy Community Definition of Input Parameters and Assumptions The selection and definition of input data to be considered in the assessment of costs and benefits and the assumptions on their future development can already predetermine the outcome of a cost-benefit analysis (CBA). It is therefore of particular importance that no bias is shown in data selection and definition at this stage of a CBA, a task for which an independent regulatory authority should be particularly well positioned to carry out. Assumptions on the future development of the input parameters determine the future occurrence and extent of the costs and benefits of smart metering discussed in the previous sections. Assumptions on input data often used in cost-benefit analysis for a smart metering roll-out may for example be made on the: Development of procurement costs for smart meters and smart metering infrastructure Future number of households, buildings and metering points Future development of network tariffs Future development of average, peak- and base-load end-user tariffs for households and small commercial customers and of wholesale prices Future average electricity consumption of households and small commercial customers Expected future customer switching Future average metering charges Future development of taxes on electricity end-user tariffs Future average carbon emissions per household and per small commercial customer Future development of CO 2 prices The assumptions on the future development of input data should also include the definition of maximum, minimum and base (average) levels for each parameter, so that their impact on the final outcome of the CBA can be assessed in a sensitivity analysis. Smart Meters Rollout in the Energy Community Page 60/107

62 A CBA should (ideally) assess all possible future costs and benefits of a smart metering rollout. However, some costs and benefits of smart metering may have immediate effects, but others may be partial, or only take effect in the long-term. A further parameter to be decided ex-ante is therefore the length of the period considered in the CBA model. Some projects assessed in a CBA may require a fairly long period of time to repay their initial investment in order to first start seeing net benefits. Cutting off all further long-term costs and benefits after a certain date may therefore lead to incomplete results. The modeling period should generally be long enough to encompass all major benefits and costs occurring during the economic lifetime of the asset assessed in the CBA, i.e. at least the economic lifetime of the smart meter and the smart metering communication infrastructure. Costs and benefits beyond this time horizon could then be approximated by fixed values based on the results achieved in the modeling period. Future benefits or revenues and costs of smart metering may not have the same value as present benefits and costs. Future values have therefore to be converted into their value today (their present value) by an appropriate discount rate, so that they can be meaningfully used for comparison/evaluation purposes. The discount rate represents the minimum return that an investment project must earn to be economically feasible. In other words, selecting a high discount rate expresses a higher demand to the profitability of the investment. High discount rates can also be applied to express that benefits and costs achieved in periods closer to the (smart metering) investment have a higher value to the stakeholders than those occurring further in the future. Whereas a private financial investor would select a financial discount rate that considers the actual cost of borrowing and actual returns on alternative investments in the market (financial analysis), a social CBA would require a social discount rate (reflecting society's point of view). Such a social discount rate could be derived from the predicted long-term growth in the economy, considering the preference for benefits over time (taking into account the expectations on increased income, or consumption, or public expenditure (social time preference approach). A social discount rate commonly applied is the one calculated and published by the Directorate General for Regional Policy of the European Commission for individual European countries European Commission, Directorate General Regional Policy (2008): Guide to Cost Benefit Analysis of Investment Projects. Smart Meters Rollout in the Energy Community Page 61/107

63 The selection and definition of input data to be considered in the assessment of costs and benefits and the assumptions on their future development can predetermine the outcome of a cost-benefit analysis to a significant degree. A careful selection of model inputs and assumptions therefore has to be made Definition of Smart Metering Roll-Out Scenarios A sound CBA should not assess the net benefits of a single roll-out scenario but compare different scenarios regarding their total net benefits. The scenarios should assess the incremental impact of the roll-out against a continuation of the status quo, i.e. not carrying out a smart metering roll-out, but continuing to use conventional meters. It is however important that the status quo reference case (continuing use of conventional meters) is not regarded as a static case, but that it also based on the same assumptions on future development of the input parameters made in the roll-out scenarios of the CBA. The scenarios should be based on technically and legally feasible alternative options. Furthermore, it should also be considered that some of the benefits of a smart metering roll-out might be achieved by other alternative measures that do not require a roll-out of smart metering; that is for example carrying out other measures to increase energy efficiency or to improve the metering process. Scenarios for a roll-out of smart metering should include at least a realistic base case, an optimistic best case and a pessimistic worst case. They can be characterized by several parameters: Assumptions related to cost benefit items Start and end date of the roll-out for smart metering A smart metering penetration rate (e.g. 100%, 80% etc.) Specification of the smart meters (in particular regarding the ability to support additional services and the purchase costs) Deployment strategy (mandatory, voluntary roll-out) 59 A sound cost-benefit analysis requires the definition and comparison of several feasible alternative scenarios, and should measure the incremental impact towards a continuation of the status quo without rolling-out smart metering. 59 For a further overview on different deployment strategies see chapter 6.2. Smart Meters Rollout in the Energy Community Page 62/107

64 5.7.3 Capturing of Costs and Benefits As mentioned above, a CBA would require capturing and assessing all direct and indirect costs and benefits of smart metering for the respective participants (network or metering operator, consumers, suppliers). Furthermore a complete assessment of all possible costs and benefits would also include the external effects on other stakeholders. Direct effects include for example the costs for the smart metering infrastructure or the savings for manual meter readings which are no longer required. Indirect effects result from other effects triggered by smart metering, such as a reduction in the individual energy consumption due to the feedback provided to consumers with smart metering. External effects of smart metering take account the indirect effects on other parties such as reduced electricity wholesale prices or a reduced need for investments in generation, transmission and distribution capacities, resulting from reductions in peak demand and increases in energy efficiency (facilitated by smart metering). Direct effects can be quantified relatively easily, for example the cost of installing a smart metering infrastructure. However indirect and external effects (such as a reduction of greenhouse gas emissions) may not be directly observable in the market prices. Moreover they may depend on the expected response and behavioral changes of the affected parties. Many of the external costs and benefits (positive and negative) generated by a project (such as a smart metering roll-out) may be easy to identify. It can however be challenging to quantify and monetize precisely these effects. When external benefits or costs (such as the carbon emissions resulting from fossil-fueled electricity generation) are already internalized (i.e. included in the investment costs and in the operation and maintenance costs), eventual benefits (such as reduced carbon emissions) would automatically be accounted for via the avoided costs. When this is however not the case, the social and environmental impacts of a project on a particular stakeholder have to be assessed. Besides major environmental effects, the macroeconomic effects of a smart metering investment on the gross domestic product and employment may also be considered. To improve the quantification of costs and benefits (and also to test available technology) pilot projects can be conducted. Such pilot projects have already been carried out in most EU countries as well as in the EnC Contracting Parties. However, given the particular country and project specifics, results of pilot projects in other countries should be carefully assessed before transferring them to another country. It is therefore advisable to support a social CBA with the results from a country s own pilot projects. Quantifying benefits from pilot projects, e.g. from energy savings or load shifting, can be difficult as pilot projects may not cover enough smart meters and customers and a sufficiently long timeframe to comply with statistical minimum requirements for a long-term assessment of these effects. Smart Meters Rollout in the Energy Community Page 63/107

65 Attributing a monetary value to the quantified benefits and costs should be based on competitive market prices where possible. Where no reliable market prices are available, monetization of costs and benefits may be either based on: The willingness of a stakeholder to pay for or to accept (e.g. noise pollution), determined by observing consumers behavior in a similar market or a using surveys, or Environmental impact assessments (e.g. for carbon emissions), based on an estimation of the cost of potential damage or the costs of preventive measures. The assumptions used to quantify external and indirect effects may often be arguable. Therefore their impact and the resulting uncertainties should be carefully assessed through a sensitivity analysis. To keep the effort and the costs of a CBA reasonable, only costs and benefits with material impact may be explicitly studied. Potentially small effects could be approximated by a fixed value in the same way as effects after a certain point of time are assessed on an aggregated basis. The level of detail up to which indirect and external effects of smart metering may be considered, should also be based on the objectives of the CBA as well as the volumes of evaluated investments. The degree of complexity, comprehensiveness, efforts and costs of a CBA for a limited number of meters and low investment volumes should be adequate to the project scale. Ideally all costs and benefits of a smart metering roll-out should be assessed and incorporated into the cost-benefit analysis, including those affecting other parties. To keep the effort and the costs of a CBA reasonable, only costs and benefits with material may be explicitly studied. Pilot projects on smart metering may contribute to a better quantification of costs and benefits Calculation of Net Benefits The social CBA applies dynamic investment appraisal methods commonly used in the financial analysis of an investment project, such as the Net Present Value (NPV) and (less commonly) the Internal Rate of Return (IRR). 60 The calculation of the economic NPV of all costs 60 In the investment analysis the NPV takes all cash flows associated with a project and reduces (discounts) them to a common denominator (present value) by using an appropriate interest rate (sometimes called the cost of capital or the cost of finance) to take into account the time value of money. The assessment of an investment project is positive if the NPV is positive (NPV > 0), i.e. it has a return which is greater than the interest rate (cost of capital) applied. The IRR calculates the rate of interest (discount rate) at which the expected future cash flows must be discounted to equate them with the initial project cost, i.e. to produce a NPV of zero; in other words the interest rate at which the project will exactly break even. The assessment of an investment project with the IRR is positive if the opportunity cost of capital (also known as hurdle rate) is less than the calculated internal rate of return. Smart Meters Rollout in the Energy Community Page 64/107

66 and benefits includes the monetary costs and benefits incurred by the (smart metering) investor, other sector stakeholders (network operators, generators, suppliers, customers) and society as a whole. The economic NPV is then the difference between all discounted social benefits and costs of smart metering over the project s life time. The economic assessment of a smart metering roll-out (for society as a whole) would be positive if the NPV is positive (i.e. if the NPV > 0). When comparing different scenarios for a smart metering roll-out in a CBA, as discussed above, the scenario with the highest NPV should be selected (which could also be the scenario not to invest in smart metering as pointed out in section 5.7.2). 61 The calculation of the NPV only includes monetized costs and benefits. If other positive or negative impacts are identified which cannot be easily monetized, an estimation / assumption on their monetary value has to be made. Adding additional qualitative assessment criteria to the NPV might result (at least if they are not reasonably quantified) in subjective or biased results. The cost-benefit analysis should use dynamic investment analysis based on the assessment of the Net Present Value (NPV). The economic assessment of a smart metering roll-out would be positive if the NPV is positive (i.e. if the NPV > 0). When comparing different scenarios for a smart metering roll-out in a cost-benefit analysis, the scenario with the highest NPV should be selected Sensitivity Analysis When selecting and defining input parameters and cost and benefit categories realistic minimum and maximum (as well as average) values should be defined. In the sensitivity analysis NPVs are (re-)calculated assigning the minimum and maximum values for individual input parameters of the model. Following this approach the sensitivity analysis assesses the sensitivity of the NPV results on variations in the input data and therefore allows the determination of critical input variables or parameters of the model. The results can be pictured in a Tornado-Diagram showing the sensitivity of single parameters. The sensitivity analysis should not be used to calibrate the input parameters in a way that a "preferred" outcome (such as the highest NPV for a specific preselected smart metering scenario) is achieved. 61 The financial feasibility of a project (i.e. a positive financial NPV) is an essential condition for the viability of a project. In principle the NPV of the selected option should therefore also have a positive value. If, however, the scenario of keeping the status quo is also included in the CBA as should always be the case and this scenario as all other scenarios provides a negative NPV, then the scenario with the smallest negative NPV should be selected. Smart Meters Rollout in the Energy Community Page 65/107

67 The results of a CBA might however be questionable if they only depend on strong assumptions of a single input parameter of the model. Ideally all input parameters should be assessed in a sensitivity analysis in order to avoid a bias towards specific model parameters and outcomes. In most cost-benefit analyses however, only selected parameters have been tested. For example the assumptions on the amount of energy savings by customers have been identified as one of the most critical input parameters in the sensitivity analysis. The results of the cost-benefit analysis should be further substantiated by a sensitivity analysis, which allows the quantification of the impact of critical input variables Important Determinants of a Cost-Benefit Analysis Ideally all direct and indirect costs and benefits of smart metering should be monetized and included in a CBA. However, before carrying out the CBA decisions have to be made on the timeframe (modeling period) within which costs and benefits will be considered and the level of detail to which the costs and benefits will be precisely assessed. Costs and benefits that only have a minor impact on the overall results may for example not justify efforts for their monetization. Such costs and benefits could instead be approximated by a fixed value. Further important issues that should be carefully analyzed when deciding on a roll-out are the effects of smart metering on competition in the energy market and customer support for smart metering. As discussed, since for example data protection could be a very sensitive issue for customers, support from customers which is a key requirement for a successful roll-out could be lacking if this issue is not addressed properly in consultations. Also the ownership of the (conventional) meters if assigned to the customers or to different parties than those responsible for the roll-out of smart metering can be a critical factor to be taken into account when interpreting the results of a CBA. The social cost-benefit analysis should study the impact on welfare for all parties affected by the project (i.e. society as a whole). If the total welfare is maximized (i.e. the project will bring maximum net benefits), then society as a whole will be better off as a result of the project. Different stakeholders however typically benefit to a different extent from a smart metering deployment. As pointed out earlier, addressing such welfare distribution effects separately can be crucial for the public support of a smart meter roll-out (particularly if the roll-out is made mandatory). The installation of the smart metering infrastructure for instance is typically paid by the DSO or the party responsible for metering, whereas major benefits may be on the consumers side due to energy savings. Besides the overall net benefits of a smart metering roll-out, the net benefit results for different market participants (e.g. customers, network operators, suppliers) should also be calculated separately. Smart Meters Rollout in the Energy Community Page 66/107

68 The costs of an investment in smart metering must also be recoverable for the company rolling-out smart metering. While some of the costs might be recovered internally through cost savings resulting from smart metering, others have to be passed on to the customers or other stakeholders, as explained in the next chapter. Finally, the result of a social CBA on the roll-out of smart metering does not always need to be positive. A negative NPV of a smart metering roll-out may not be unlikely if the potential for energy savings resulting from a smart metering deployment is expected to be small and / or the current metering system already provides frequent and accurate meter readings at relatively low costs (because for example labor costs are low). In order to achieve meaningful results with a CBA it is important to define and compare several realistic and feasible scenarios and to define the input parameters without a bias towards a specific outcome. Guaranteeing such an independent and open assessment of the costs and benefits of smart metering could be a natural task of the regulatory authority. The results of the cost-benefit analysis of smart metering will strongly depend on the country specifics, the assumed timeframe and the level of detail to be considered. 5.8 International Experiences with Cost-Benefit-Analysis on a Smart Metering Roll-Out By January 1, 2011 eleven EU Member States had already conducted a CBA for electricity smart metering and six had done so for smart metering in the gas sector. From these assessments of costs and benefits, seven provided a positive result for a roll-out of smart metering for electricity, and five for gas. Negative results (i.e. negative NPVs) for a roll-out of smart metering in electricity had for example been calculated for the region of Flanders in Belgium, for Denmark and for Norway (in the first CBA). Several countries have also already carried out or are considering conducting a second (or an update of the) CBA on smart metering, taking into account for example advanced functionalities of smart meters, changes in the procurement costs of smart meters or a wider range of stakeholders. 62 In a number of further EU member states CBAs (at least for electricity) are already underway or planned. However several countries have not yet published the results of a CBA for smart metering as requested by the European Directives 2009/72/EC and 2009/73/EC, in particular for gas. In some countries an alternative CBA has been conducted by the industry or large DSOs. Furthermore, a few countries such as Italy, Spain and Finland have also made a (positive) decision to go ahead with a roll-out of smart metering without conducting a CBA. While several 62 Updates or second CBAs have for example been conducted or are planned in France, Hungary, Poland, Portugal, Denmark, Norway, Austria and the Flemish region of Belgium. Smart Meters Rollout in the Energy Community Page 67/107

69 CBAs have already been carried out throughout Europe, relatively few countries have already defined precise roll-out plans in their legislation. Depending on the national specifics and the competencies of the regulatory authorities defined in the legislation, CBAs are either carried out by the regulatory authority or the respective government department. In Belgium, France, Hungary and Ireland CBAs have been carried out by the regulatory authorities, in the Netherlands, Slovenia and the UK CBAs have been conducted by the respective government departments (ministries). In some countries such as Austria the regulator as well as the government have carried out CBAs following a change in legislation, the responsibility for a CBA on smart metering has been handed over to the Ministry of Economy. Although the Austrian regulator as well as other regulators do not have the authority to assess the deployment of smart metering through a CBA (and to decide on a roll-out), they may still be responsible for the definition of functionalities and data requirements of a smart metering system. As pointed out earlier, regulatory authorities should naturally be in a good position to objectively assess the costs and benefits of a smart metering roll-out in a CBA. Status quo in CEER member countries Number of countries having conducted a cost-benefit analysis Number of countries with a positive result of the costbenefit analysis Number of countries planning to conduct (or carrying out) a cost-benefit analysis (incl. 2 nd time) Number of countries not planning to conduct a costbenefit analysis Number of countries with no cost-benefit analysis, but already a (yes/no) decision on a smart metering roll-out Electricity Gas Table 4: Status of Cost-Benefit-Analysis on a Roll-Out of Smart Metering in Europe (as of January 1, 2011) 63 Source: ERGEG (2011): Summary of Member State experiences on cost benefit analysis (CBA) of smart meters, Ref: C11-RMC Council of European Energy Regulators (CEER) = 27 EU Member States, Norway and Iceland. Smart Meters Rollout in the Energy Community Page 68/107

70 The cost-benefit analysis of a smart metering roll-out carried out in France in 2007 analyzed three different scenarios with two different roll-out periods each (five year and ten year). Based on a model period from 2011 up to 2038 and a roll-out of 35 million smart meters and 420,000 data concentrators, a positive net benefit of 1.7 billion had been calculated for a smart metering roll-out in the optimum scenario. For all three scenarios the five year roll-out timeframe provided greater benefits than a roll-out period of ten years. A positive result of the cost-benefit analysis has also been calculated for the Netherlands (both in the original cost-benefit analysis in 2005 as well as in the review in 2010). In the 2010 update on the cost-benefit analysis a maximum net present value of billion had been calculated among eight analyzed scenarios. The calculation for the Netherlands included a roll-out of 7 million smart meters and a model period of 50 years. As 95% of all customers are supplied with electricity and gas, the cost-benefit analysis was carried out jointly for electricity and gas. The model scenarios included analysis of the different roll-out periods, communication infrastructures and increases in energy costs. The largest benefits in the model resulted from reduced energy consumption (between 3.2% and 6.4% for electricity and 3.7% and 5.1% for gas), increased (retail) competition, consumer management and reduced meter reading costs. To address the distributional effects of smart metering costs and benefits among different stakeholders the net present values for different groups have also been calculated separately in the Netherlands. As the following figure shows, positive and large net present values (net benefits) have been calculated for the consumers (households) and (also positive but smaller) for the metering companies (carrying out the meter reading). Network operators, who have to cover the investment costs of smart metering face a negative net present value. In addition, as less electricity is consumed by end-users, network operators will transmit and distribute less electricity and will therefore receive less network fees. In the same way power producers and retail suppliers will sell less electricity and the government will receive accordingly less taxes, resulting in a negative net present value (net costs) for these stakeholders. In conclusion, as has also been identified in the cost-benefit analysis in Austria, distribution network operators and retail suppliers face a negative net present value, while customers clearly see a large benefit in the roll-out of smart metering Again, it must be noted that the results of a cost-benefit analysis for all stakeholders as well as at individual stakeholder level strongly depend on the model assumptions and country specifics, so that these results cannot be simply transferred to the Contracting Parties of the Energy Community. Smart Meters Rollout in the Energy Community Page 69/107

71 Figure 12: Net present values of different stakeholders in the cost-benefit analysis for the Netherlands (base scenario) Source: KEMA In the UK several assessments on the costs and benefits of smart metering have been made for electricity and gas between 2008 and 2011 (29 million electricity and 21 million gas meters sharing the same communication infrastructure). For a full scale roll-out between 2013 and 2020 and assumed reduction in energy consumption of 1.5% and 4% for electricity and 1% and 3% for gas, estimated costs of approximately 8.6 billion, benefits of approximately 14.6 billion and a total net present value of approximately 6 billion have been calculated. As main benefits besides the reduction in consumption, the reduced costs of site visits, benefits of easier supplier switching and reduced costs with customer management have been identified. In the cost-benefit analysis carried out in Austria by the regulatory authority in 2010 estimated costs between 3.3 billion and 4.4 billion, estimated benefits between 3.6 billion and 4.9 billion and an estimated net present value between 291 million and 556 million have been calculated for a roll-out of smart metering for electricity and gas. Here four different roll-out scenarios with different roll-out periods and penetration rates had been investigated. As model period, the expected technical lifetime of a smart meter was considered (15 years for electricity and 12 years for gas smart meters). It was further calculated that the highest net present value could be achieved in the scenario with the fastest and most widespread roll-out of smart metering in Austria. Smart Meters Rollout in the Energy Community Page 70/107

72 The specifics of the existing electricity system and metering infrastructure in a country can have a strong impact on the costs and benefits of smart metering roll-out. A CBA on the rollout of smart metering carried out in one country can therefore lead to completely different results in another country. A transfer of CBA results from one country to another may therefore be misleading. Possible country specifics that could be important factors for a positive or negative outcome of a CBA are for example: The condition (obsolescence) of the existing meters Replacement / recalibration programs for the existing meters Accessibility of technology options National energy strategies Customer satisfaction with the existing metering and billing system Security of supply during peak load hours Level of commercial losses (energy theft / fraud) The country specifics as well as the model specifics and the range of considered stakeholders can have a strong influence on the results of a cost-benefit analysis. A simple transfer of the results of a cost-benefit analysis from one country to another is therefore not possible and comparisons have to be treated with caution and understanding. The results presented above also show that the cost-benefit analysis of a smart metering roll-out can lead to a negative result, indicating that smart metering should (at least at the time of conducting the analysis) not be rolled-out. Smart Meters Rollout in the Energy Community Page 71/107

73 6. MARKET MODELS AND REGULATION FOR METERING 6.1 Metering Market Models When deciding on a smart metering roll-out, the market model for the metering services has to be taken into account, as the principle market structure of the metering business not only determines some of the deployment options, but also influences the overall costs associated with a roll-out of smart metering and determines which market actor covers the costs. Three aspects have to be considered: Who (which market party) is responsible for (which) metering services? Who owns the metering assets? Should the metering service be a competitive or a regulated market? Basic Metering Market Models Traditionally metering has been carried out as part of the distribution and supply activities of a vertically integrated utility. In such regimes the customer has (had) a single point of contact for network connection, supply, meter (installing, maintaining and reading) and invoicing (full vertical integration model). Following the European Directives for the internal market for electricity and gas and their implementation in the Energy Community distribution network operators (DSOs) are now subject to unbundling provisions separating regulated network activities form competitive supply activities. This as a consequence has also led to different models for organizing the metering sector. In an unbundled environment the metering services could either be carried out by the distribution system operator as part of its regulated activities (DSO-model), by a separate metering company independent of the distribution and supply companies (unbundled metering company) or as an additional competitive service area of the supplier (supplier model). Smart Meters Rollout in the Energy Community Page 72/107

74 Full Vertical Integration DSO Model Unbundled Metering Company Supplier Model Supply Supply Supply Supply Metering Metering Metering Metering DSO DSO DSO DSO Figure 13: Possible Options for the Structuring of the Metering Sector Source: KEMA In addition to this general structure of the metering market, the provision and operation of meters could be further broken down into several tasks that can be carried out by different market participants. Figure 13 shows that all tasks except the invoicing of customers and the handling of customer complaints regarding the billing which are either performed by the supplier or by an independent third party could principally be carried out by the distribution system operator. Meter reading is sometimes carried out by the customer (who could also be the owner of the meter and even be responsible for installation and maintenance as is partly the case in Ukraine and Turkey) 65. The role of the independent metering company could further be separated into a metering asset provider (MAP) and a metering service provider (MSP). The independent metering asset provider carries out the installation of the meters and the installation (and operation) of the smart metering communication infrastructure (as well as the ownership of the meters). Maintenance and reading of meters as well as the handling of customer complaints regarding the reading of meters, and the installation and operation of smart metering communication infrastructure and accompanying management of the meter data are possible tasks of a metering service provider. The latter two tasks as well as the invoicing and billing procedures could also be carried out by an independent third party, for example by a subsidiary of a telecommunication company. It is also possible that different market participants carry out the same task for different customers within a country. 65 Energy Community Regulatory Board (2010): A Review of Smart Meters Rollout for Electricity in the Energy Community; available at: Smart Meters Rollout in the Energy Community Page 73/107

75 Provision and installation of meters Distribution System Operator Metering Asset Provider Metering Service Provider Supplier Customer 3 rd party X X (X) Ownership of meters X X X Maintenance of meters X X (X) Reading of meters X X X Handling of customer complaints regarding meter reading X Invoicing of customers X X X Handling of customer complaints regarding billing Installation and operation of smart metering communication infrastructure Data collector and distributor of (smart) meter data X X X X X X X X X Figure 14: Tasks and Responsibilities of Possible Market Participants in the Metering Sector Source: KEMA Related to the issue of which market participant is carrying out which metering services is the question of whether metering is to be provided in a liberalized metering market with metering services open to competition or in a regulated metering market with designated companies operating under a regulatory framework. In most regulated markets metering is either carried out by the DSO or the supplier, whereas in a competitive market, independent metering asset or service providers (or other third parties) can also carry out at least part of the metering services. When the metering service is a monopoly business carried out by the DSO, it is paid by the customer either as part of a regulated metering tariff or a part of the regulated network tariff (in both cases determined by the regulatory authority). When metering services are provided in a competitive market, the amount paid by the customer for metering services is also determined by the competitive process. Within the European Union two types of electricity metering market models can generally be distinguished: A liberalized metering market, where the metering sector is open for competition resulting (possibly) in new market entry, for instance by specialized metering asset and service providers, suppliers, telecom companies and others, or Smart Meters Rollout in the Energy Community Page 74/107

76 The traditional and still most common model of a monopolized metering sector, where the metering activities belong to the regulated activities of the DSO or to a separate dedicated regulated national/regional metering asset and service provider. So far only a few countries have unbundled the metering business from other DSO or supply activities and opened it for competition, such as Germany, the UK and the Netherlands. 66 The motivation of the German government to open the metering market for competition was for example the hope that competition between metering service providers would help to reduce metering costs. So far though, no independent metering service provider has been able to compete against the incumbent DSOs in these countries. However suppliers have sometimes taken up metering services (also outside the affiliated DSO home territory), trying to strengthen customer relationships by offering for example smart meters. The model of a monopoly metering service provider at regional or national level (independent from the DSOs and the suppliers) had been under discussion (but finally not implemented) in the UK and in Hungary. 67 In almost all EnC Contracting Parties metering services (currently) remain part of the regulated DSO functions, although sometimes consumers, suppliers and even metering companies are involved. 68 The costs of meters are recovered via the regulated network charges, and investments in metering equipment are subject to regulatory approval. Although the regulatory regimes vary from Contractual Party to Contractual Party in the Energy Community, all regulatory regimes known to us apply an ex-ante regulatory review and explicit approval of investments before inclusion of costs in the allowed revenue. From the different metering market models described above three basic options could be applied in the Contracting Parties of the Energy Community, as the full vertical integration of the metering business with distribution and supply would not be in line with the legal requirements of unbundling (at least for DSOs with more than 100,000 or more connected customers): 1. The DSO owns and operates the metering infrastructure and performs the metering services 66 Although in the Netherlands the metering market model has shifted back to a regulated market model again, due, amongst other things, to the smart metering deployment strategy. 67 In the UK such an approach was discussed under the name Regional Franchise Model. The final roll-out decision however put the responsibility on suppliers. 68 The exceptions being: Albania, where the supplier is in charge of the data management; Bosnia and Herzegovina, where the customers not only are partly involved in meter reading but also in meter maintenance; Ukraine, where customers carry out some of the meter installation, maintenance and meter reading tasks and metering companies are also involved in meter reading and data management; and Turkey, where meter installation and maintenance is only carried out by the customers. See: Energy Community Regulatory Board (2010): A Review of Smart Meters Rollout for Electricity in the Energy Community; available at: Smart Meters Rollout in the Energy Community Page 75/107

77 2. An independent metering company performs the metering services. The ownership and operative responsibility for the metering infrastructure could lie with the metering company ( fat metering company combining the roles of the MAP and the MSP) or with the DSO ( lean metering company or MSP). 3. The metering function is performed by the supplier in a liberalized metering environment. In the traditional and most common market model the metering functions are carried out by the DSO. Certain tasks and responsibilities can however also be carried out by metering asset providers, metering service providers, suppliers, customers or other third parties. Before a roll-out of smart metering is carried out, three major questions have to be addressed by the responsible authorities: Which market party should be responsible for (which) smart metering services? Who should own the (smart) metering assets? Should metering services be provided in a competitive or a regulated metering market? Pros and Cons of Metering Market Models In the traditional market model where all metering functions remain within the DSO (DSOmodel), no unbundling of the metering business is required to take place. A roll-out of smart metering by the DSO could therefore be easily implemented into the existing industry structures within the Energy Community. Moreover, as smart metering might be seen as the natural precursor of the smart grid, it makes great sense for the DSO to take up responsibility for smart metering. In cases where legal unbundling (or accounting unbundling if less than 100,000 customers are connected) has not yet been fully implemented, a separation of supply and distribution activities would be even more strongly recommended in the context of a smart metering roll-out. If the DSO remains active in supply, for example as the default supplier, and also carries out the metering services, he would have strong incentive for abusive and discriminatory behavior against any competing suppliers. Other (new) suppliers however would depend on the DSO forwarding meter data for invoicing purposes and transmitting price data or control commands for demand response schemes from the supplier to the customer. Separating the metering business from an otherwise completely vertically Smart Meters Rollout in the Energy Community Page 76/107

78 integrated utility on the other hand would make little sense (except for the reasons discussed in connection with a liberalized metering market or a multi-utility approach below). 69 The DSO model would be more complicated to implement in a multi-utility environment. There usually considerably less benefits of smart metering in the gas, district heating or water sectors compared to electricity. A roll-out of smart metering in these sectors therefore seldom provides a positive business case. The benefits of smart metering for these sectors are however more likely to be positive in a combined and integrated approach, where the smart metering communication infrastructure is jointly used with electricity. If all services (e.g. electricity and gas) are delivered by the same DSO, the motivation for a smart meter roll-out would be even stronger. When the different services are provided by different DSOs, the smart metering deployment would be led by one of the DSOs, most likely the electricity DSO. Other energy carriers would be integrated by separate metering devices and an additional multi-utility communication controller (either as a separate device or integrated into the electricity meter). In addition, communication interfaces would be required to communicate with the other DSOs, with additional suppliers and with the customers. Such a set-up would not only require significant coordination between all stakeholders, but moreover costs would need to be distributed among all parties. If all DSOs are properly unbundled from supply, no incentives for discrimination of other DSOs would occur. If unbundling for some energy sectors (such as district heating) has not been implemented, serious concerns about abusive and discriminatory behavior against competitors (as in the supplier model) could arise if the non-unbundled DSOs are in charge of the smart metering (e.g. a district heating DSO hindering gas deployment). In the DSO model, ownership of the metering infrastructure is generally with the DSO. In a multi-utility environment a decision needs to be made as to which assets should be owned (and managed) by which DSO. The most practicable approach would be to provide the leading (electricity) DSO with ownership of metering infrastructure including the multi-utility communication controller, whereas the other DSOs own only their metering device. The capital costs for the metering infrastructure would then depending on the concrete contractual set-up be partly rolled-into the service fee charged by the leading DSO to the other DSOs, the suppliers or the customer itself. If the supplier is carrying out the metering services (supplier model), meter operation would shift from one supplier to another with every change in supply by the customer. Even with hard- and software standards widely established, it is possible that the new supplier would require a different meter to be installed resulting in inefficiently high transaction costs and stranded investment. A supplier s responsibility would thus pose a barrier to changing the supplier and subsequently hinder competition. The ability to provide smart metering services 69 Vertical integration between distribution and supply is however not a possible option provided in the legislation of the European Union and the Energy Community. Smart Meters Rollout in the Energy Community Page 77/107

79 by the supplier is also determined by the technological choices made. If, for example, the smart metering infrastructure is based on PLC, meter operation cannot be shifted easily from the DSO to the supplier; this would only be feasible using GSM/GPRS or DSL communication. Furthermore meter operation needs on-site resources, such as the personnel for meter installation and maintenance, which the supplier would first have to build up if the responsibility for metering services were to switch from the DSO to the supplier. A fundamentally different approach would be to unbundle (part of) the metering business from the network and the supply functions (the unbundled metering company model mentioned above) and to create an independent metering company. The metering infrastructure could either be owned by the metering company (or MAP) or the DSO. To facilitate the rollout of smart metering, it makes sense however to leave the ownership and thus also installation and maintenance of the meters with the DSO, as the DSO already has the necessary technical resources and personnel in place. If ownership, installation and maintenance of the metering infrastructure were handed over to a newly founded MAP, such resources would need to be relocated to the MAP and contractual arrangements would also be required for use of the DSO s assets (e.g. for PLC). This would not only be time consuming and costly, but may also depending on the existing market structures require the DSO s voluntary participation. To avoid such problems, the metering company could only be responsible for meter reading, data processing and management and provision of data to all authorized stakeholders ( lean metering company or MSP). Authorized stakeholders would have rolebased access to all the relevant data, i.e. the gas supplier would only have the individual gas consumption data relevant for billing purposes, whereas the electricity DSO would only have access to relevant network data, e.g. power quality and outage data or accumulated consumption data. A separate independent metering company (MAP and / or MSP) could either act as a monopolized, regulated, national (or regional) metering entity or as a competitive metering entity, being in competition with other metering companies and / or the metering business of the DSO and / or suppliers. Assigning a single MSP (or a small number of large regional MSPs) with the metering tasks in a country where there are a large number of DSOs could have the advantage of facilitating the roll-out of smart metering and the communication of metering data by reducing the number of entities with whom a supplier (and a switching customer) has to interact. 70 In a monopolized metering market, the national/regional metering asset and service provider (or the DSO) also selects the type of smart meter, which limits customer choice and innovation, but also promotes a roll-out of smart metering to all customers. 70 On the other hand, assigning independent MSPs (and MAPs) in a country where only a single DSO or very few are operating would increase the number of parties involved in the metering process and hamper the roll-out of smart metering. Smart Meters Rollout in the Energy Community Page 78/107

80 When metering services are regarded as a competitive market, a proper unbundling of regulated network and metering services needs to be in place. Otherwise the DSO would have an incentive to discriminate other metering companies operating in its own network area and / or to cross-subsidize competitive metering services with revenues from regulated network services. 71 Competition in the metering market also requires standardized and transparent procedures for switching the MSP. These should be monitored by the regulatory authority as existing providers of metering services (DSOs, suppliers or MSPs) would have an incentive to hinder customers from switching their provider of metering services. Allowing DSOs to provide metering services in competition with suppliers or MSPs would as the example of Germany has shown hamper the development of competition in metering services. As discussed further below, competitive metering markets tend to be more suitable for a voluntary roll-out of smart metering than for a mandatory roll-out, since the coordination efforts to carry out a full-scale roll-out of smart metering to all customers increase with the number of market participants involved (which are likely to be higher on a competitive metering market). A competitive metering market however has the advantage that it enables the customers to select the type of smart meters and that it promotes a selective deployment of smart meters, i.e. smart meters are firstly deployed to those customers with the largest benefits. The metering infrastructure is generally either owned by the DSO, the supplier or a metering company. In some countries including some of the Contracting Parties of the Energy Community individual metering devices are however (at least partly) owned by the consumers. This is for instance the case in Poland, Romania or Slovenia. Such a constellation could be a barrier to the deployment of smart metering, as the consumer owning the conventional meter might be hesitant to invest in a new (smart) meter, bearing the cost of a stranded investment. Such issues would even be more complicated in countries such as Ukraine, Turkey or Bosnia and Herzegovina, where customers also partly carry out some of the meter installation and maintenance tasks. 72 In order to achieve customer support for a roll-out of smart metering in such circumstances the roll-out can only be voluntary. Incentive schemes to promote investment in smart meters and sound technological standards guaranteeing interoperability of smart meters if consumers change their supplier (preventing stranded investment) can further facilitate a smart metering roll-out when the customer is the owner of the meter. High coordination costs and stranded investments when customers switch their supplier generate a high potential for inefficiencies in the supplier model. Also a multi-utility approach seems barely imaginable with the supplier model. The DSO and the MSP models may both 71 This is in fact a further case of the general economic consideration that competitive (electricity) markets require the unbundling of competitive services from regulated services, if provided by a single (vertically integrated) company, for example between the (retail) supply activities and the network activities of a utility. 72 Energy Community Regulatory Board (2010): A Review of Smart Meters Rollout for Electricity in the Energy Community; available at: Smart Meters Rollout in the Energy Community Page 79/107

81 have their advantages in promoting smart metering. The DSO model seems especially suited to a fast track smart metering roll-out focused on electricity. In such a model the DSO would be responsible for smart metering, although standardized processes and communication interfaces between the DSO and suppliers (and the customers) would need to be established to enable customers to switch their supplier. 73 In a multi-utility approach the MSP model might be better suited to ensuring the successful implementation of smart metering. However, as such an entity would need to be newly founded and the required resources would need to be set up, deployment of the MSP model would require more time. Also a rollout of smart metering based on PLC communication infrastructure would be difficult to implement as this requires access to the network of the DSO. For a fast track smart metering roll-out focused on electricity the DSO model with the DSO being responsible for smart metering might have significant advantages. In a multi-utility approach the model of an independent metering service provider (MSP) might be better suited to ensure the successful implementation of smart metering. Alternative options with the supplier or a metering asset provider (MAP) being (partly) in charge for a smart metering deployment may cause high coordination costs and stranded investments and are therefore generally not recommended. As smart metering further increases the potential for discriminatory behavior of the DSO against competing suppliers, sufficient and functional unbundling between distribution and (retail) supply is of particular importance. 6.2 Deployment Strategies The decision on the smart metering deployment strategy should be based on the objectives targeted, the existing market structure and infrastructure, the timeframe (or speed) of the smart metering roll-out and the targeted penetration rate. The two principal deployment strategies (intertwined with the market model) could be: A mandatory roll-out, where all meters are to be replaced by smart meters in a given timeframe, resulting in 100% market penetration A voluntary roll-out, where customers in a liberalized metering market can decide for themselves which kind of smart and conventional meter to use, resulting in an uncertain market penetration Other decisions have to be made regarding the technological choices and the accompanying measures, such as consumer awareness programs. 73 In particular if the DSO is still part of an integrated utility, although in that case such safeguards are always necessary. Smart Meters Rollout in the Energy Community Page 80/107

82 Regardless of the deployment model, cooperation and interaction between stakeholders is crucial to ensuring successful cooperation during the roll-out of smart metering, as numerous stakeholders could be involved in this process. This is especially the case if a multi-utility approach is chosen, or if ownership issues affect private property rights. A further key requirement is to raise the necessary consumer awareness of the possibilities provided by this new technology and to take into account concerns at an early stage to ensure public acceptance. Also the standardization of hardware, software, communication procedures and interfaces, which are crucial for the deployment of smart metering, may be much easier if a cooperative approach is chosen. Competitive metering operators deploying different types of smart meters with different metering and technology standards would hinder the deployment of smart metering and the development of further smart metering applications. A single entity such as a regulator however typically lacks full information and may be unable to make the most efficient deployment choices without consulting the other stakeholders Deployment Speed and Penetration Ratios The timeframe for a smart metering roll-out and the target penetration rate for smart meters may have a significant impact on the costs and benefits. The period of time by which a full smart metering deployment has to be achieved for all customers can have a significant impact on the Necessity to operate two systems in parallel as long as old conventional meters are in existence Peaking demand for qualified personnel installing the smart metering infrastructure Peaking demand for metering and communications hardware and installation equipment, and Stranded investments if old meters are replaced before reaching the end of their economic lifetime. The benefits of smart metering can be roughly divided into benefits achieved with every individual smart meter installed, e.g. reduction in final energy consumption, and benefits resulting from the establishment of a smart metering infrastructure. Individual benefits can be generated from installing the first smart meter onwards and be quantified as a multiplication of the average benefits per meter/household and the number of meters installed. General benefits of a smart metering infrastructure however may only be realized after a certain threshold of installed meters is reached. The timing of smart metering investments and the penetration rate may also influence the benefits of certain technological choices. While for example a connection of the first smart meters via DSL or GPRS/GSM could be less costly, PLC could be the least cost option in urban areas once a high penetration rate has been Smart Meters Rollout in the Energy Community Page 81/107

83 reached. When significant numbers of smart meters are however already connected via DSL or GPRS/GSM, switching to PLC for the remaining meters or replacing the communication devices of the recently installed smart meters with PLC might not be a beneficial option. Such effects need to be considered when deciding on the deployment strategy for smart metering. The timeframe for a full roll-out of smart metering to all customers is also determined by the approach towards meter replacement (regardless of being mandatory or voluntary). Smart meters could, for example, only be deployed in cases where conventional meters are due to be replaced anyway, or (in a more progressive roll-out) conventional meters that have not yet reached the end of their economic lifetime could also be replaced. Running two metering systems in parallel (conventional and smart) would however also require two separate systems for collecting and processing meter data. Two different invoicing schemes would also probably be required and technical personnel would need to deal with traditional and new meters. As long as old and new metering infrastructures co-exist, additional costs are likely to arise that could be reduced if a quick full scale deployment to all customers is chosen. A fast-paced roll-out on the other hand would result in peaking demand for qualified personnel and for metering and communications hardware and installation equipment, which could drive up prices making the roll-out more costly than a slower roll-out pace. Depending on the age pattern of the existing metering infrastructure a fast-paced roll-out of smart metering could also result in significant stranded investment. A more gradual roll-out could also ease the financial burden and could ensure political support for smart metering deployment, as benefits from new technologies and decreasing prices could be more easily accommodated. Furthermore a more gradual roll-out could also allow the targeting of regions and business sectors first where the highest benefits from smart metering are expected. This could, for example, be urban areas with a dense population and in many cases comparably higher percapita energy consumption. Rural areas, sparsely populated and characterized by a lower energy demand from consumers would then be fitted with smart metering last. Also different technological approaches are likely to be required for different regions. While in rural areas smart meters are typically connected via DSL or GPRS/GSM, PLC is often the preferred option in urban areas This would however depend on the existence of broadband internet connections or a well developed mobile communications network in these areas. Smart Meters Rollout in the Energy Community Page 82/107

84 The timeframe for a smart metering roll-out and the target penetration rate for smart meters may have a significant impact on the costs and benefits. They may result in parallel operation of two (conventional and smart metering) systems (slow-paced rollout) or in a high demand for qualified personnel and equipment in metering and communication infrastructure (fast-paced roll-out). The speed of smart metering rollout may also influence the technological choices and/or the likelihood of assets stranding if old meters are replaced before reaching the end of their economic lifetime Voluntary Smart Metering Roll-Out A voluntary roll-out would be based on the individual decisions of metering providers (regardless of who is responsible for metering) to offer smart meters to customers and to build up a smart metering infrastructure. As significant investment in the communication and data processing infrastructure would be required for a roll-out of smart metering, such a decision would be based on an individual positive business case for the company (e.g. the DSO, a metering company or the supplier) voluntarily rolling out smart metering. With a voluntary roll-out of smart metering the trade-off and decision on the roll-out speed would also be subject to internal optimization of the party deciding on the deployment. Depending on the approach decided on by the metering provider, smart meter installation might also require the individual decision of final customers to choose smart rather than conventional meters. The meter provider could, for example, decide to roll-out smart meters to all its customers, or it could offer smart metering services as an additional option besides traditional meters. Deployment of smart metering may be slow if left to the customers; at first only so-called innovators and early adopters among the consumers are likely to opt for the use of a smart meter. A voluntary roll-out of smart metering could either be the result of a decision by the metering responsible party (for example the DSO) to deploy smart meters to its customers or the result of a decision by the government to promote the deployment of smart metering, but leaving the metering providers and the customers to select smart or conventional meters. The deployment of smart metering in Italy for example was at the beginning a voluntary decision by Enel (covering 85% of low-voltage customers) in Important reasons for Enel s decision were the expected savings or revenues in the areas of purchasing and logistics, field operations, customer services and revenue protection. Fraud (i.e. theft of electricity) in particular was a very widespread problem in Italy. Later in 2007 a mandatory roll-out decision was made by the Italian regulator, requiring 95% of all low-voltage customers to be equipped with smart meters by 2012 and also setting minimum functional requirements. In Sweden, where already 100% of customers are provided with smart meters, the roll-out decision was made by the DSOs, who were being pushed by a requirement for monthly ra- Smart Meters Rollout in the Energy Community Page 83/107

85 ther than annual actual meter readings. Based on this requirement however, the net benefit was assessed as positive by the DSOs. 75 A voluntary roll-out of smart metering may increase customer participation and public support for a deployment as it eases ownership, data privacy and security and cost allocation issues. Deployment of smart metering may however be slow. Throughout Europe voluntary roll-out plans have been only followed in competitive metering markets and in countries where the roll-out has been initiated by the DSO(s) Mandatory Smart Metering Roll-Out Under a mandatory roll-out the metering provider (DSO, metering company or supplier) would be obliged to build up a smart metering infrastructure and to replace all existing meters with smart meters within a given timeframe. Such a roll-out decision would not be based on the economic assessment of an individual player; but requires instead a social costbenefit analysis, assessing costs and benefits for the society as a whole. If the overall net benefit is assessed as positive, a mandatory roll-out would follow. A mandatory national roll-out of smart metering for electricity within a given timeframe is also implied by the Directive 2009/72/EC and the 2011 amendment of the Energy Treaty. If a social cost-benefit analysis shows an overall positive net benefit of smart metering, a roll-out would have to be completed ten years after this assessment, with 80% of customers to be supplied with smart metering by The Directive itself does not actually demand that Member States implement a mandatory roll-out scheme, but given the very ambitious timeframe, it is doubtful whether a voluntary roll-out approach would be feasible. The problems of a fast-paced roll-out mentioned in the previous section (stranded investments and peaking demand for resources) would be even greater in the case of a national roll-out. National targets to be reached within a short timeframe would, for example, limit the ability to deploy smart metering to those customers with the largest benefits first and to take the state of the existing metering infrastructure into account (e.g. replacing the oldest conventional meters first). A national roll-out scheme should take this into account and leave either enough degree of freedom, or create the necessary incentives and arguably even subsidies to mitigate the negative impact on parts of the market. A mandatory roll-out of smart metering is likely to result in a much faster deployment, but may result in higher costs. Achieving the 80% deployment targets by 2020 seems however only possible with a national mandatory deployment plan. 75 In some cases, though, the rolled-out meters fulfill only simpler remote meter reading functionalities and might not be considered as smart metering (even when the meter device itself could be considered as a smart meter). Smart Meters Rollout in the Energy Community Page 84/107

86 6.3 Role of Regulation In a competitive (liberalized) metering market, the role of regulatory authorities in promoting smart metering would be limited for example to general market organization and monitoring tasks. However when metering is not liberalized, but provided as regulated national/regional monopoly activities by the metering responsible parties (DSOs, metering companies or suppliers) as is the case in most EU Member States and the Contracting Parties of the Energy Community 76 regulatory authorities can play a crucial role in the efficient roll-out of smart metering. Given also the fact that a high level of market penetration in the short or medium term is a national objective, as demanded by the implementation of Directive 2009/72/EC within the Energy Community, smart metering topics are of high relevance for regulatory authorities. However, the role of regulation and the necessary amendments to the current regulatory framework are highly dependent on the chosen metering market model and deployment strategy. Major tasks of regulatory authorities in connection with smart metering typically include: Ensuring the overall efficiency of smart metering deployment Recognition of the efficient investment costs of smart metering in the price control process Involvement in tariff setting process Protecting the privacy of consumers and their energy usage data by developing a privacy policy and data security standards to ensure customer energy consumption data is not accessed by unauthorized parties or misused 77 Protecting consumers from unduly rate increases caused by time-of-use pricing or other tariffs that increase energy bills when consumers use energy at times of high demand and are unable to shift their load Ensuring the accuracy of smart meter data Depending on the set-up of the regulatory authority and its tasks and responsibilities regulation may also address further areas of smart metering, such as educating consumers in advance about smart meter installation, the changes that smart meters will bring and how to adjust to them. Furthermore regulation may also use the additional information provided 76 Energy Community Regulatory Board (2010): A Review of Smart Meters Rollout for Electricity in the Energy Community; available at: 77 Also the topic of cyber security is gaining increasing attention, especially in the USA. An interconnected smart metering infrastructure with data transmitted through the public internet might be a target for cyber attacks. The increased potential to be informed about the supply system s state and to control the system may come at the cost of increased vulnerability against outside attacks. Security standards need to be set up and enforced to mitigate this danger. Smart Meters Rollout in the Energy Community Page 85/107

87 by smart metering technology to generate a better understanding of the network and to improve the existing regulatory arrangements (for example in the area of quality of supply regulation). In the following sections we address some of these regulatory tasks in more detail Ensuring Overall Efficiency and Security The division of tasks between the regulatory authority and the ministry responsible for the energy sector generally depends on the national specifics and the competencies of the regulator. Whereas decisions on the general metering market model and the final decision on a voluntary or mandatory roll-out of smart metering are made at government level (and specified in the legislation) in most European countries, the cost-benefit analysis and more detailed decisions on the specifications of the smart meters, the roll-out timeframe and procedures are often made by the regulatory authorities. Regulatory authorities may be naturally in the position to assess the different options of a smart metering roll-out and to set up a national deployment plan and should therefore play a central role in the smart metering deployment. As a first step (as described in chapter 5) this means carrying out or commissioning a social cost-benefit analysis, ideally after initial supporting pilot projects testing technical possibilities and assessing expectable costs and benefits. Regulators can play an important role in accompanying and supporting such pilot studies. If the net benefit of a smart metering roll-out is assessed as positive in a social costbenefit analysis, EU Member States as well as Contracting Parties of the Energy Community need to fulfill the roll-out obligation stemming from Annex I of Directives 2009/72/EC and 2009/73/EC (see chapter 3). When devising a deployment plan for smart metering, the specifics of the existing metering system and infrastructure and the policy objectives of a roll-out need to be taken into account. In addition, the following issues need to be decided by the regulator (or the respective government department in charge): The time schedule for a roll-out The final smart metering penetration rate aimed for The type of deployment (voluntary or mandatory) Technical specifications of the smart meters Before pursuing the deployment of smart metering it is also necessary to analyze the potential barriers to a successful smart metering roll-out. Furthermore, while it is a principal task of regulation to ensure the overall efficiency of the national deployment strategy, it is also important to take the distributional effects for different regions or geographical areas and for different stakeholders into account. It is the task of the regulator to assess such differences, Smart Meters Rollout in the Energy Community Page 86/107

88 and if deemed necessary to initiate mitigation measures to minimize the barriers to smart metering deployment. After fundamental decisions on the deployment of smart metering are made, it is the role of regulation to ensure that the targeted objectives are actually achieved. This can include for example: 78 Ensuring interoperability of metering and communication assets by establishing hardand software standards Defining minimum service requirements Accompanying the roll-out project management Depending on the national specifics and competencies, regulatory authorities or the respective government department(s) has(have) to assess the different options of a smart metering roll-out (including a cost-benefit analysis) and to set up a national deployment plan. Furthermore, regulators and government can also be involved in the definition of hard- and software standards and minimum service requirements. Regulatory authorities should pay special attention to data protection and security issues in order to ensure secure data communication and protection of consumers' private data against unauthorized access Smart Metering in the Price Control Process The treatment of (new) investments in the allowed revenue is a classical role of any regulatory authority in the price control process. Any regulatory regime must address the question of which capital and operating costs should be included in the revenue requirements (allowed costs) and whether specific incentives to promote (or acknowledge) investments should be applied. When the metering business is a regulated (monopoly) activity, a key question for the regulatory authority is how much of the smart metering roll-out should be covered by whom. If the network operator (or the meter operator) benefits significantly from reduced commercial losses and improved system information and security of supply, allowing him to improve internal efficiency and reduce his costs, then he should rightly (at least partly) cover some of the costs himself. By determining the allowed costs or revenue levels of the network operator, the regulator is in a position to decide which costs can be included in the regulated net- 78 See also: ERGEG, Final Guidelines of Good Practice on Regulatory Aspects of Smart Metering for Electricity and Gas, Ref: E10-RMF-29-05, Brussels, 8 February Smart Meters Rollout in the Energy Community Page 87/107

89 work and metering tariffs and to which extent. Generally only the net costs of smart metering for the network operator should be included in regulated tariffs. Net costs are the costs of smart metering minus the benefits that arise directly for the network operator (i.e. minus the cost reductions the network operator can make himself due to a roll-out of smart metering). 79 To the extent that customers benefit from increased transparency and accuracy of bills and consumption, easier switching procedures or new services and tariffs, customers should contribute to the recovery of the smart metering costs. In a competitive retail market with a voluntary roll-out of smart metering the suppliers can also have an incentive to make a payment to the meter operator for an increased roll-out of smart metering, if it allows them for example to introduce new tariffs and services that increase the retention of their customers. Furthermore, if the benefits of improved energy efficiency, reduced carbon emissions and improved security of supply following a roll-out of smart metering are regarded as significantly large by the government, subsidies to smart metering investors could be considered by the government. So far regulatory authorities across Europe have taken different standpoints on if and how investments in smart metering infrastructure (and pilot projects) should be accommodated in the metering and/or network usage charges. In many EU Member States, regulators are however hesitant to allow for higher user charges to accommodate investments in smart metering. Regulators in many countries have therefore taken the position that cost coverage should essentially come from existing revenues. Nonetheless, such topics are still under discussion in many countries and often an issue of dispute between the network operators and the regulator. In Spain for instance, where a national roll-out has already been decided, an increase in the monthly metering fee of around 0.3 is allowed. In Austria, the metering charge for smart metering was set equal to that of conventional metering. However, the Austrian regulator also asks all network operators to provide further detailed information on their investments in smart metering and grids as well as on the research and development projects in the cost data templates each year. In Italy a separate metering tariff is used which should cover costs of smart metering deployment. Italian DSOs who do not meet the scheduled roll-out targets are penalized by reducing the allowed metering revenue. In several other countries investment in smart metering was acknowledged in revenue control, as long as these costs were deemed efficient, as for instance in Germany. It could be argued that the costs of smart metering should not be considered to their full extent in the allowed revenues, as the company rolling out the infrastructure (in most cases the DSO) may already benefit from cost savings it generates itself through smart metering, e.g. due to improved processes, obsolete manual meter readings, less theft and improved asset 79 While retail suppliers, producers and other stakeholders in the energy sector also benefit from smart metering (as described in chapter 5), for practical reasons costs are mainly only distributed between the distribution network operator (or the meter asset or service provider) and the end-user. Smart Meters Rollout in the Energy Community Page 88/107

90 management. The extent to which the benefits of smart metering compensate or even outweigh its costs for the network or meter operator should be assessed by the regulator and considered in the revenue control process. Experience so far however shows that benefits for the DSO may not always be high enough to cover the investment costs required to build up the smart metering infrastructure if this is the DSO s responsibility. 80 The inclusion of the investment (and operational) costs of smart metering in the allowed revenues is a key task of any regulatory authority. Normative regulatory principles would require the full acknowledgement of all efficient costs netted of the benefits that arise directly for the network operator (or the metering provider) in the allowed revenues. On the other hand, insufficient possibilities for network operators or metering providers to pass on these costs may hinder the deployment of smart metering. Social or political constraints related to the cost pass-through arrangements and tariff impact should also be taken into account by the regulator Tariff Setting End-user tariffs within the European Union and the Energy Community should be unregulated according to the Directives 2009/72/EC and 2009/73/EC. In order to maximize benefits from smart metering, the market rules, tariff schemes, technical codes, procedures and processes need however to be adjusted to smart metering. The reduction of final energy consumption through increased transparency of actual consumption patterns to customers can be further promoted through the application of new timeof-use tariff schemes. Rather than applying single or two-period (day-night) tariffs, smart metering allows the distinction between three or even more tariff periods during a day. Tariff schemes that raise prices in peak periods and decrease prices in off-peak periods may only lead to substantial cost savings for consumers if price elasticity 81 is sufficiently high. This is more likely the case if end-user tariffs represent a cost-reflective level, which is currently not the case in every Contracting Party of the Energy Community. 82 Rather than continuing to apply flat tariffs for use of networks, also time and load dependent network tariffs might be applied together with time-of-use tariffs for final customers. This will provide an additional leverage for shifting or reducing demand. Implementing such schemes would however will increase the complexity of the network pricing. 80 For a more detailed description of costs and benefits for the different market actors see chapter The price elasticity of demand describes the dependency between a price change and the quantity demanded. High price elasticity means that demand is more strongly influenced by price changes, i.e. if the price increases, demand will decrease noticeably. 82 If customers are however not able to adjust their consumption to such changes in tariff schemes, these might easily lead to higher costs (that not all customers may be able to pay). Smart Meters Rollout in the Energy Community Page 89/107

91 It is not the task of regulatory authorities to set end-user tariffs, when retail supply is deregulated as required by legislation. Regulators may however play a central role in adjusting market rules, tariff schemes, technical codes, procedures and processes to the requirements of smart metering Enhanced Regulatory Performance Smart metering can be especially valuable if a quality of supply regulation scheme is applied or needs to be set-up. Such a regulatory scheme is typically based on an incentive regulation regime but includes quality of supply as one of the output factors measured, either as an integrated part of a general efficiency assessment, or as a separate element of the revenue control formula. Voltage quality and reliability can accurately and timely be provided and monitored when smart metering is widely deployed. 83 Short interruptions on low voltage level for example are often not recorded in the existing systems. Smart metering would thus provide the basis to significantly improve the regulator s data basis and increase feasibility of quality regulation schemes. As full-scale roll-outs have only been completed in a very small number of countries, power quality and reliability monitoring functions of smart metering have not yet been implemented in quality of supply regulation. As smart metering also enables network operators to improve their quality of supply, for instance by faster outage detection, it can also be a regulatory tool for generally improving quality of supply. In Italy, for example, the regulator created an incentive for DSOs to actively use smart metering to improve quality of supply. Subject to the provision that DSOs also deploy smart metering faster than originally scheduled, the network operator receives via the price control regimes an incentive of 15 per customer when using smart meters to record unplanned interruptions longer than three minutes. 84 Smart metering can help to improve and facilitate quality of supply regulation as it can provide the regulator with improved data on voltage quality and reliability. 83 As smart meters basically measure voltage, current and time, and load as well as energy, data is calculated based on these basic parameters. Monitoring voltage quality can be implemented easily into a smart metering system. 84 From 2008 (gradually, depending on the size of the network operators), the distribution network operators are obliged to keep records of all low voltage customers that experienced unplanned interruptions longer than 3 minutes. The network operators may choose information systems (GIS) that comply with minimum standards set by the regulator, or they may choose smart meters. If they choose to install smart meters they receive a financial incentive of 15 per customer. To qualify for this incentive the network operator should deploy smart metering faster than originally scheduled. See Jorge Vasconcelos: Survey of Regulatory and Technological Developments Concerning Smart Metering in the European Union Electricity Market, RSCAS Policy Paper, P. 49 (2008). Smart Meters Rollout in the Energy Community Page 90/107

92 7. NEW SERVICES Many of the benefits expected from smart metering will in fact not be caused by the smart meters themselves but by the consumption feedback provided to consumers as already described and by new services offered to consumers, giving a further boost to energy savings and energy efficiency efforts and providing additional added value to consumers. The technical infrastructure installed when smart metering is deployed will most certainly open up markets for new services. Some services are already offered commercially or tested in pilot projects, whereas other services emerging in the future may not yet have been anticipated. Obvious new services based on smart metering deployment are new tariff designs, demand response and those in the field of home automation. Many projects are already ongoing in the fields of new tariff schemes and automated demand response. The SmartRegions report on the European smart metering landscape includes a long list of various projects all over Europe Innovative Tariff Schemes As previously mentioned, smart metering deployment is often accompanied by new tariff schemes. The smart metering functionality to register load enables innovative tariff schemes. 86 With traditional analogue meters, a separate meter was required for each tariff zone (or one multi-tariff meter with multiple single-rate registers, which comes close to having multiple meters in one chassis). Subsequently, given the required investment, the maximum number of tariff zones offered to domestic and small commercial consumers was two, typically split into a day time and night time tariff. Smart metering allows for plenty of different tariff models and multiple tariff zones or even dynamic tariffs without any additional investment in hardware. 85 Renner et al., European Smart Metering Landscape Report, SmartRegions Deliverable 2.1, Vienna, February New tariff schemes make the most sense for electricity, as the system is more fragile with respect to the exact balance between supply and demand and there is a clear intra-day peaking characteristic in consumption patterns. For gas and heat demand, tariff schemes to optimize the point in time of consumption would not lead to equally useful results, as for instance final consumer demand cannot be shifted to flatten the peak in the annual profile. On the other hand, significant storage capacities are available for gas at economically efficient costs, which is not the case for electricity. Smart Meters Rollout in the Energy Community Page 91/107

93 In the most basic case, time-of-use tariffs with only two time periods are applied. As mentioned above, such tariffs are already commonly applied, even without smart metering. Such tariffs are, for example, very common if electricity is used for heating with storage heaters. As traditionally two separate meters or a two-tariff meter have been required, the investment in an additional meter would appear feasible for households with a comparably high electricity consumption. With smart meters time-of-use tariffs are more easily applied, as first of all a single meter will be available to provide multi-tariff metering. Smart metering based time-of-use tariffs thus often apply more complicated tariff schemes than simple two-zone schemes, e.g. distinguishing at least between peak, off-peak and shoulder periods and also between normal working days and weekends or public holidays (see section 6.3.3). In a further step, dynamic tariff schemes could be linked to real-time spot prices, enabling consumers to benefit even further from an adaption of consumption patterns to the system state. With smart metering, a real-time price signal could be submitted to the consumer, providing full information on costs actually incurred and the savings potential if consumption is adjusted. Assuming a sufficient level of demand elasticity, first of all dynamic pricing will not lead to decreased consumption but primarily to a load shift into more favorable periods (from the system perspective). However, depending on the kind of load shed in high price periods, it will not be compensated necessarily to its full extent in low price periods, resulting in energy savings, for instance an air conditioner deliberately switched off during the day will not use the amount of electricity when switched on again in the evening. The impact of time-of-use and dynamic tariff schemes is highly dependent on the consumer's individual situation, i.e. its demand structure, elasticity, willingness to change consumption behavior and also on structural circumstances in a country. The changes of consumption behavior in line with new tariff schemes are much easier facilitated in cases where consumers can adopt their demand automatically to price changes using home automation systems and smart appliances, as is discussed in the following section. New tariff schemes can cause irritations if as a result consumers face higher energy bills, because of a larger portion of consumption taking place in more expensive peak-times. For a detailed elaboration on this issue c.f. section 0, in particular section 4.1. Together with new tariff schemes, new payment schemes will also emerge. Smart metering can enable or simplify a wide variety of possible payment schemes, such as for instance prepayment schemes. In the Netherlands and the UK, prepayment schemes are already fairly common, especially for bad creditors. As in these payment schemes, feedback on the level of consumption and associated costs is given in a very direct way, and energy saving incentives are comparably strong. Prepayment tariff schemes with a variety of cashless payment options are much easier facilitated with smart metering, in particular in combination with remote switching or load-limitation functionalities. Smart Meters Rollout in the Energy Community Page 92/107

94 New (end-user) tariff schemes are of high practical relevance, as major benefits possibly stemming from a smart metering roll-out cannot be achieved without innovative tariff schemes. Additionally, new (pre-)payment arrangements can be easily implemented based on smart metering. 7.2 Home Automation Services Consumers can further benefit from new tariff schemes if they do not need to manually adapt to price signals. Home-automation technology will provide consumers with the necessary means to automatically adapt to real-time price signals and optimize energy consumption. Possibly such services will be offered by several parties, e.g. manufacturers of household appliances, electricity suppliers or independent service providers. A supplier could for example offer electricity supply at comparably low prices together with the right to curtail customer s consumption by remotely controlling certain appliances. It seems obvious that new tariff schemes are more successful if an individual consumer needs to do less to manage his consumption, while at the same time comfort should not be affected beyond a certain level. As a lack of manual reaction to price signals could lead to higher energy bills if time-of-use or dynamic tariffs are implemented, automated demand response could result in a higher level of acceptance if such tariffs are offered. Home automation services will be offered in a variety of different approaches, involving appliances from several manufacturers and several service providers, supporting energy savings and increased energy efficiency. Smart metering communication and hardware should be standardized to encourage the development and implementation of home automation services. 7.3 Consumers as Active Market Participants If distributed generation facilities are installed at consumers' premises, the electricity generated and injected into the grid will also be metered by the smart meter, saving the costs for additional meters. Moreover, smart metering can also enable the active participation of consumers on the ancillary and system services market, if the ability to shift load created by smart metering is bundled by a service provider for a large group of consumers and then offered on the market. 87 In these cases not only consumer load but also available generation 87 To take availability factors into account, the capacity firmly offered on the market would be in the magnitude of 10% to 25% of the total movable load. Smart Meters Rollout in the Energy Community Page 93/107

95 capacities, e.g. from solar panels or micro-generation can be combined to a so-called virtual power plant, providing for instance reserve energy. Through smart metering direct control of load and generation capacities will be enabled. Households will not only be consumers of energy but also producers, hence the often used term 'prosumers'. All these services concerning tariffs and home automation target some kind of demand side management or demand response in order to reduce peak loads, thus avoiding investments in expensive peak load generation capacity (often gas turbines with low levels of full-load hours) and possibly also network capacity. In the USA, demand response has already been on the agenda for some years, 88 whereas in Europe it is only slowly gaining momentum. The key driver for demand response in the USA is the compared to Europe very tight generation situation, which, among other things, led to the spectacular outages in California and on the East Coast a few years ago. For the USA, the potential for demand response is estimated as 5% of the annual system peak load. For Europe, the estimations are somewhat lower at around 3%. 89 However, in most European countries, demand response was not considered to be so important in recent years, although this is slowly changing with the discussions about the integration of large shares of renewable energies and the need for a smarter grid. An increasing share of intermittent generation capacities (solar, wind) will require a paradigm shift regarding system operations. Traditionally, electricity is produced when required, with the network frequency as the ultimate real-time control signal to system dispatch and generation units. However, it will be extremely difficult to maintain this with a high share of intermittent generation, requiring a vast amount of reserve capacity. In order to utilize the system more efficiently and given the limited options to store electricity, load will have to follow generation patterns. 7.4 Other New Services In addition to the services described, other services enabled by the existence of the smart metering communication infrastructure may be developed. These services which could emerge in a smart metering environment will not necessarily be associated with energy consumption. Instead, they will simply be piggybacked on the existing communication infrastructure, providing a direct communications channel with the consumers' premises. Thus, for many of these services smart metering will not be a crucial requirement but rather an opportunity; probably every other reliable communication infrastructure could also be 88 Cf. e.g. Borenstein, S., Jaske, M., Rosenfeld, A., Dynamic Pricing, Advanced Metering, and Demand Response in Electricity Markets, Vasconcelos, Jorge, Survey of Regulatory and Technological Developments Concerning Smart Metering in the European Union Electricity Market, RSCAS Policy Papers 2008/01, Florence, 2008 Smart Meters Rollout in the Energy Community Page 94/107

96 used. Such services will typically be safety and security alarms, such as fire or burglar alarms. Alarms could also be sent in case of unusual consumption patterns, e.g. identifying possible need for help. Also the smart metering infrastructure could be used to submit home alerts of elderly or disabled people. Smart Meters Rollout in the Energy Community Page 95/107

97 8. ROLL-OUT PLAN FOR SMART METERING DEPLOYMENT If the cost-benefit assessment as required by Annex I of Directives 2009/72/EC and 2009/73/EC results in a net benefit, proving a smart metering roll-out to be positive, Member States are required to prepare a timetable for smart metering deployment. For electricity the timetable must have a horizon of 10 years, ensuring 80% roll-out by 2020 for those consumers where a positive net-benefit has been assessed. In the Interpretative Note on retail markets accompanying the Directives 2009/72/EC and 2009/73/EC, the European Commission clarified that the 80% target applies only to those consumer segments where a smart metering roll-out was assessed as positive during the cost-benefit analysis. Where no cost-benefit analysis is conducted the 80% applies to all consumers. 90 Assuming that smart metering will have at least a positive effect for certain consumer groups, e.g. those with an annual consumption above a certain amount, it seems unlikely that the cost-benefit assessment will justify a decision against at least a partial smart meter roll-out. Furthermore, it can be assumed that if a cost-benefit assessment were to be used as an argument against a full roll-out, the assessment would be closely scrutinized by market parties in favor of smart metering and probably also by the European Commission. Thus, with the assumption that a (partial) roll-out will in any case be required, and need to be well in progress by 2020, Member States will be required to set-up a smart metering deployment schedule for those consumer groups where smart metering has been assessed as positive. The Directives direct the responsibility for the deployment plan to the Member States or any competent authority designated. In many Member States the responsibility is given to the national regulatory authorities. The deployment schedule has two basic starting points, the first is exogenously given by the Directive, requiring 80% fulfillment of the targeted roll-out by 2020 (electricity only, for gas the Commission's Interpretative Note refers vaguely to a "reasonable period of time"). The second is the result from the cost-benefit assessment, i.e. the identification of those consumer groups where a roll-out has been assessed as positive, as for these consumer groups the roll-out is mandatory. These two factors set the framework, which is "how many smart meters in total" and "to whom". If we look at examples of smart metering roll-out plans, we can observe many similarities between these plans and also with the usual approach in large investment programs. The 90 European Commission, Commission Staff Working Paper, Interpretative Note On Directive 2009/72/EC Concerning Common Rules For The Internal Market In Electricity And Directive 2009/73/EC Concerning Common Rules For The Internal Market In Natural Gas, Retail Markets, Brussels, Smart Meters Rollout in the Energy Community Page 96/107

98 following illustrative example is taken from the French DSO and shows the roll-out schedule during an early stage, i.e. from Figure 15: Example of Smart Metering Deployment Plan Source: ERDF If the early schedule is compared with the actual situation, we can see that the original plans have been somewhat delayed. However, France still plans to deploy 35 million smart meters by 2018, with the large-scale roll-out starting in In general we distinguish three main phases during the roll-out: At the beginning, the roll-out plan starts with a preparatory phase. During this phase the necessary legal provisions (e.g. regulations requiring DSOs to roll-out smart metering) are put into place, the technical specifications (hardware, software, communication) are established and detailed planning of the next roll-out phase is conducted. In order to avoid rushing into a large scale roll-out, a 'real' roll-out is typically preceded by a pilot phase. During this phase the technology is tested and consumers' behavior is observed. Pilot projects can also be conducted in parallel during the first phase. They results can be integrated into the specifications, the cost-benefit assessment, and the planning of the full roll-out. Also an integral part of this phase (or the beginning of the next phase) would be the tendering for a supplier or several suppliers of the smart metering infrastructure. Depending on whether there is a single entity responsible for conducting the roll-out within a country or individual entities Smart Meters Rollout in the Energy Community Page 97/107

99 (e.g. DSOs), there will not necessarily be a single tendering process for one country, but a separate tendering process for each area. During the tendering process, it is important that functional requirements are clearly lined out. Additional functions could also be part of the tendering process. The tendering process could be split into several phases, including a pre-selection phase. In order to facilitate an optimal and undisputable outcome of the tendering process, it should be conducted as transparently as possible. The third phase would be the final mass smart metering roll-out with an established and tested technology. During the third phase, the available resources (personnel, hardware, etc) have to be allocated over time to ensure a feasible distribution in line with the overall roll-out target. A schematic example is depicted in the following figure. Figure 16: Example of Mass Roll-out Schedule Source: ENDESA As can be seen from the depicted examples, the roll-put plans include clear and verifiable milestones. Additionally, such a roll-out needs to include clearly defined responsibilities during the project phases, as for instance the responsibilities for the go-decision in the first figure above. Given that the same 80% target by 2020 applies for the Energy Community, even if the deadline to provide the smart metering roll-out plan is postponed to 2014 (see chapter Error! Reference source not found.), the EnC Contracting Parties face a tight time schedule. As can be seen from the two examples of France and Spain, a full smart metering roll-out may easily take up to ten years. Given that resources for hardware provision and installation of smart metering systems may be limited, and that EU countries and EnC Contracting Parties may compete for potentially scarce resources in order to achieve the 2020 targets, it seems of utmost importance to pursue the roll-out without undue delay. Subsequently, it could be considered to achieve the 80% target with as little effort as possible, e.g. starting the roll-out Smart Meters Rollout in the Energy Community Page 98/107