Publications on the internet
Energy and Climate Change - Minutes of EvidenceHC 161-i
house of commons
taken before the
Energy and Climate Change Committee
Smart Meter Roll-out
Tuesday 14 May 2013
Dr Sarah Darby, Dr Gary Raw, Professor Harriet Bulkeley and Dave Openshaw
Dr Martyn Thomas CBE and Alex Henney
Evidence heard in Public Questions 108 - 204
USE OF THE TRANSCRIPT
This is a corrected transcript of evidence taken in public and reported to the House. The transcript has been placed on the internet on the authority of the Committee, and copies have been made available by the Vote Office for the use of Members and others.
The transcript is an approved formal record of these proceedings. It will be printed in due course.
Taken before the Energy and Climate Change Committee
on Tuesday 14 May 2013
Mr Tim Yeo (Chair)
Sir Robert Smith
Examination of Witnesses
Witnesses: Dr Sarah Darby, Deputy Programme Leader, Lower Carbon Futures group, Environmental Change Institute, Oxford University, Dr Gary Raw, Visiting Professor and Professional Research Associate, UCL Energy Institute, Professor Harriet Bulkeley, Professor of Geography, Durham University, and Dave Openshaw, Senior Adviser, UK Power Networks, gave evidence.
Q108Chair: Good morning. Thank you very much for coming in. We have about an hour, so don’t each feel obliged to respond to each question, so that we have a reasonable chance of getting through in the time. Just by way of introduction, as this is the first time you have given evidence to this Committee-certainly during this Parliament-can each of you very briefly outline your own research background or your involvement with trials on energy consumption behaviour, particularly if those trials have involved smart meters at all?
Dr Darby: I am Sarah Darby from the Environmental Change Institute at Oxford. In the late 1990s, I started doing research on the effectiveness of energy advice programmes. In the course of that, I got interested in how useful it was to be able to give energy users feedback on their consumption, as part of making that advice more effective and actually being able to see the effect of adopting certain measures, or changing behaviour in certain ways. I did a review of the research literature on giving feedback to consumers, wrote it up and then largely forgot about it for a few years. At that point people started asking me for this review, because of the growing interest in smart meters and the dawning of the idea that you could use smart meters to improve feedback to energy customers. For the past few years, I have been doing research on the development of smart grids and smart metering, in connection with improving the feedback to customers on their energy use. I was also part of the external evaluation team for the EDRP trials, and at the moment I am involved in the smart metering early assessment that is being carried out by DECC.
Dr Raw: Good morning. Gary Raw. I am a psychologist by profession. I worked for many years at the Building Research Establishment, on various aspects of people in buildings, their behaviour, comfort and their use of buildings. Most recently I have been working as an independent consultant, but with bigger organisations-currently University College London-on energy, particularly householders’ use of energy. I was largely responsible for the final analysis of the Energy Demand Research Project, EDRP, data and the write-up, and the extensive literature review that went alongside that. I have since been involved in some smaller follow-up projects for DECC. I am currently working on a large project on smart systems, more broadly, for the Energy Technologies Institute. I think that will do for me.
Professor Bulkeley: Good morning. I am Harriet Bulkeley and I am at the Department of Geography at Durham University. My work concerns climate change, politics and policy more generally, and I work particularly on how cities around the world are responding to climate change. That work has taken me to thinking about issues of energy efficiency in the built environment and to the public’s response to those kinds of issues, but I am also interested in the processes through which technologies get deployed and implemented.
At the moment, I am working in the UK on one of Ofgem’s funded smart grid projects, the Customer Led Network Revolution, where I lead the social science research team at Durham working on that. CLNR, as we call it briefly, is led by Northern Power Grid with British Gas, EA Technology and Durham Energy Institute. We are conducting a trial of smart grids in the north of England. The social science component of that work to date has involved over 220 interviews with SMEs and domestic customers about their experience of smart meters in a smart grid, as well as a survey, which has gone to about 10,000 people. We have had about 700 domestic and about 180 SME responses. That work is very much in progress, and, unlike your other two participants on the panel, it has not yet been peer-reviewed. I would like to make sure that that is understood by the Committee, and the evidence that I give is very much preliminary findings from that work.
Dave Openshaw: Dave Openshaw, Senior Adviser at UK Power Networks, which is the licensed distribution network operator serving the east and southeast of England and London. My particular focus is very much towards smart grids. We have done a number of pieces of research around smart grids and the benefits that might lead to, in terms of improved network efficiency and ultimately lower prices to consumers. In particular, we have done research with Imperial College on responsive demand, and-perhaps most relevant to this discussion-we are engaged with a similar project to CLNR, which is Low Carbon London. A unique feature of Low Carbon London is we are trialling a day-ahead dynamic time-of-use tariff, which we believe is quite original. It is not a fixed time band tariff. This is a dynamic time band tariff, whereby consumers are notified a day in advance of the price and the time bands at which those prices will occur over the following 24 hours.
The reason we are keen on a dynamic day-ahead tariff is because we think this will reflect a future where we have high capacity wind generation, which of course is zero marginal cost but is intermittent. The availability of that generation, on a day by day or even within day basis, will have a significant impact on the real time price of producing electricity. Therefore, we think in future there would be real value if consumers are able to reflect the availability of wind generation in the way that they use electricity. Hopefully I will have a chance to say a little bit more about the preliminary results we have seen from that trial.
Q109Chair: Thank you. Do you think the results from the EDRP are sufficiently reliable and transferable to give an idea of how the wider population are going to react to smart meters?
Dr Raw: Shall I pick that up? I think they are, at broad scale. I would pose it slightly differently, that you need to understand in detail what came out of EDRP. If you only read the executive summary and expect that is exactly what will happen, I think you will be mistaken. You need to interpret it in terms of the detail of what was done, and I would put it not that they should be seen as representative, but they should be seen as guidance for how to make the roll-out more effective. I see EDRP more as a learning tool than a defining tool as to what is likely to happen.
Dave Openshaw: It is worth reflecting that the EDRP trials did take place some time ago. There wasn’t a great deal of use made of time-of-use tariffs, and I think we have learned a great deal since the results of those trials took place.
Q110Chair: Was the evidence from EDRP fed heavily into what DECC’s plans for the roll-out are, in particular in relation to the benefits that consumers might gain?
Dr Raw: As far as I am aware they were, although clearly the impact assessment was very much in progress throughout that period. At what point they intersected I could not be certain, but certainly the kind of percentage savings that were achieved, or at least were achievable, backed up the sometimes more modest figures that DECC had included in their benefits calculation.
Q111Chair: Have their assumptions about consumer benefits from the roll-out been realistic?
Dr Raw: From that perspective, yes. Purely from the perspective of the possible energy savings that consumers can achieve, yes, I think they have been realistic. Again, I would say not in terms of, "This is what EDRP achieve. Therefore, this is what we expect, end of story," but rather to say, "This is what EDRP learned about how to maximise the benefit. Let’s use that."
Dr Darby: I think the findings from the EDRP have been very consistent with what we have learned from other parts of the world too, and that is worth saying. The savings that were made were actually quite modest, but, as Gary says, this should be seen in the light of a guide as to how to do this more effectively and the lessons that were being learned in the course of that trial. It was an enormous learning process for everyone who took part in it. I spoke to a senior person in Scottish and Southern quite late on in the trials. He said he had been in the industry for 30 years, and had worked on several major infrastructure projects, and carrying out this set of trials was by far the most difficult challenge they had had to do. There was a great deal of learning that went on by the utilities that conducted those trials, which they are carrying forward, and the findings they got, in effect, about how best to go about building a rather new relationship with your customers, have been borne out elsewhere as well.
Dr Raw: I agree with that. There was a massive amount of practical learning as well, and, from the perspective of the energy companies, that was a very important part of their participation. Bearing in mind that when this started, and in the period over which it ran, they were still learning about: how to put a smart meter in; where it is difficult to do it; where you cannot do it at all, at the time; how you create the linkages between the smart meters and the other technologies that are involved; how you liaise with customers; and how it affects even the number of people you have to have in a call centre to handle the process. All of those practical lessons were of immense value. They are somewhat downplayed in the report because they not of so much research interest, but I think they are vastly important for the industry.
Professor Bulkeley: I would add that I think the new trials that are happening at the moment are going further into depth to try to understand what the effects of the smart meters are, why it is that you get those kind of results, and how they combine with other things, such as time-of-use tariffs, which will add detail to our understanding of what a smart meter roll-out might look like.
Dave Openshaw: I certainly think the context that we are facing now, with the electrification of heat and transport and large volumes of renewable-albeit, intermittent generation-means there are some very real opportunities for avoiding unnecessary investment in generation capacity and transmission and distribution capacity, if we can truly engage with consumers in making use of time-of-use tariffs and other mechanisms. The key word that came out of the EDRP pre-trials is "engagement", and continuous reinforcement of that engagement is absolutely crucial. That is certainly what we are finding with our trials at the moment.
Q112Sir Robert Smith: The original EDRP trial suggested a consumer saving on electricity of 3%. Has that been replicated or improved on in more recent trials?
Dr Darby: If you look internationally, there was quite a large scale review carried out in 2010 by the American Council for an Energy-Efficient Economy, where they found a range of savings from 4% to 12% for improved feedback, not always with smart metering but increasingly with smart metering. They looked at the longest-lasting of those studies-so between one and three years-and they found those savings held up over time. In fact, in two of the nine longest-lasting studies, the savings actually increased a little over time. So you are getting a higher range of figures there.
The VaasaETT Global Energy Think-Tank did another big review of about 100 different trials of both of feedback and dynamic pricing. That was in 2011. They found, for the long-lasting trials, they were getting savings of about 5%.
Dr Raw: I think that is right. The review that you referred to, the Ehrhardt-Martinez-
Dr Darby: That is the ACEEE one, yes.
Dr Raw: Yes, which I also included in my review. Yes, that sums it up nicely. The thing to be clear about from the EDRP evidence, you have heard a figure of 3%, and that is correct, but it is a very different 3% for electricity and for gas.
Sir Robert Smith: Yes, I was going to come on to that.
Dr Raw: Most of those trials that Sarah has referred to have dealt with electricity and feedback on electricity. There is quite a lot of evidence, particularly in relation to in-home displays, real-time displays, which EDRP was entirely consistent with, and those savings were also more or less consistent over up to two years of the EDRP trials. It was dependent on the display, rather than the smart meter itself. For gas, as far as the evidence is available, it seems to be the smart meter was responsible and the display did not add a great deal. So one is dealing with quite different circumstances for the different fuels.
Q113Sir Robert Smith: What was the experience with gas, then?
Dr Raw: Much less experience. I don’t know how much there has been. The Irish trials were published recently and had some similar gas savings, but they had a big mix of interventions and all had some impact. I think it is important to understand that those trials did not look at the impact of smart meters. Because everyone in the trial had a smart meter, they then looked at the impact of adding something to people who already had a smart meter.
Q114Sir Robert Smith: How did the smart meter improve the gas-
Dr Raw: It is a very good question and one that we have struggled to answer. We had some ideas, but nothing that could be proved from the evidence. It has to be something not to do with the meter as such but with the experience of getting the meter: the fact that you have a new encounter with your energy supplier; perhaps you’re promised a bright, new, shiny equipment; exciting technology; possibly as basic as a friendly installer explaining something in the process of going through it; possibly some old meters being replaced, the ones where the dials go in different directions and are very difficult to read; for the first time you have a direct feed-out; you can see the rate at which the numbers are clicking over quite clearly. All these are possible. It could be all or some of them.
Dr Darby: I would add to that that, with a smart meter and gas, you are getting an accurate bill every time. There is a rather different dynamic at work with gas and electricity. Most people have gas heating, so it is the difference between heating energy uses and non-heating. With electricity you are switching things on and off all the time. You can look at the display and you can see the numbers go up and down. For example, you can see immediately what the impact is of switching on your kettle. Of course with your heating you are interested in rather longer time periods, so you can get a bill-ideally you would get it every month or two months-and you can look back over that period and you can remember roughly what was happening. "Oh, it seems to be low, but then we were away for a week" or "It seems to be very high, and we had a bunch of people to stay". Of course, accurate billing for gas gives you an idea of seasonality more accurately and it operates over a long time period. So I think the introduction of accurate billing with the smart metering would also have been a factor.
Q115Sir Robert Smith: Is there a breakdown of the type of consumers that respond most successfully to trials and others who perhaps find it quite a challenge to actually get any benefit?
Dr Darby: Some customers do need more support than others in interpreting the new information they are given, certainly. There was a study done by NEA for DECC last year, which brings this out quite clearly, where they interviewed and did group discussions with people who for some reason are disadvantaged-they are on low incomes, they are vulnerable, they perhaps have mental health difficulties-and that certainly brought it out that, particularly for them, you would want some extra support explaining how to use their display and what a smart meter can do for them. When that was forthcoming they found people were very positive about it.
Q116Sir Robert Smith: Did you have a comment?
Professor Bulkeley: Yes, three points that I would like to make in response to the discussion that is taking place at the moment. The first is that, in terms of our evidence, we find the idea of a smart grid trial has provoked within people a sense of a civic relationship with the grid. They treat it not so much as a matter of a consumer relationship, but they are quite interested in their own role in keeping the lights on, in securing energy futures, and in the decarbonisation and climate change agenda. That may speak for why the introduction of a smart meter in the gas network had an effect, because it changes people’s relationship to their energy system. I think that is quite an interesting emerging finding from our work.
The second issue is in terms of looking at how or why people respond to in-home displays. Particularly households, because our work on SMEs-which I can talk about later, if you are interested-is less developed than our household work. We find three different things that are going on in people’s households. One is about budgeting. It is not so much about price reduction, but it is about managing a household budget over a week or over a month. People really enjoy being able to do that, that is vulnerable customers as well as non-vulnerable customers but the budgeting issue is very important. I think we need to separate that from a sense of reducing overall price. However, they are related, people like to be in control of their finances as much as they like them to go down.
We see a second reason is around family management and oversight. I don’t have teenage children but, for families who do, apparently smart meters are a very useful way of knowing what is happening in the family and managing some family tensions and dynamics. It is quite interesting. Then, for a sizeable minority, we also find the idea of gaming a smart meter quite an interesting motivator. People like to try to beat it and beat themselves at things, and that relates to a whole set of smart data that people use now. We have particularly looked at the equivalent in running, where people record their runs and try to beat their next time and so on, and we can see similar aspects happening with the smart in-home display.
That comes on to the idea about which customers respond most to these issues of in-home displays. One of the things that we would strongly recommend the panel to think about is that this appears not to be about customers and their attributes as such, as much as it is about different kinds of things that people do, which are more or less flexible and amenable to intervention. Perhaps unsurprisingly, we find that nobody really minds putting off their chores, or changing when they do chores, but things like cooking and family meal times are less moveable feasts. So, rather than looking at customer attributes, we are interested in the composition of what is going on in a household and how that relates to the in-home display. That might be more important for us to understand, in terms of where customer benefit can be felt for what kind of households.
Dr Raw: I agree with that. It is very easy to look at the observable characteristics: the income, the education level, the number of people in the household. You can find effects of those. You tend to find higher savings in homes with higher incomes, with higher education levels. However, I think that obscures what is really important, which is the kind of needs that people are fulfilling in using energy, and those are complex, the flexibilities they have, the non-negotiable uses that they have, the particular ways in which they personally do it. Do they dry clothes over a radiator, in which case it is not just about heating? All these varying needs and behaviours need to be understood. Not that you necessarily understand every individual household, but you look at typical patterns and make sure that the interventions that accompany smart meter roll-out are suitable for that range of patterns of needs and behaviours. That is part of what I am working on at the moment. It won’t be fully worked out for another year or so, but I think it is important to be thinking along those lines, beyond the simple characterisation and into the household dynamics.
One example of that that we did find in EDRP was, when it comes to the effect of time-of-use tariffs in shifting consumption from peak period to other times of day, households of one or two people seemed more able to do that. You can easily hypothesise why that would be. It is simply easier for them to manage their time. There are only one or two of them. They can make choices together. If there is only one shower, they can both use it off-peak. There isn’t a queue of people and some of them having to use it on-peak. It is those details that are going to be important in understanding how to get the most benefit from the smart meter roll-out.
Dave Openshaw: Certainly it is interesting to see the early observations from our trial. We have probably not yet fully explored just how flexible people can be if they have the right incentives, and the incentives are in the form of a price. Our tariff is a critical peak price tariff, so the peak price is very, very much higher than the normal or the low price. However, what we have seen is quite significant. Although it is early days, we have seen up to a 20% reduction in peak demand. You asked the question about energy saving, but of great importance, going forward, is the extent to which we can persuade people to move electricity away from peak demand, or, as I said earlier, to use electricity when wind generation is highly available and 20% shifts are very, very significant indeed.
Some of the interesting early anecdotal feedback we have had, which was a little unexpected-because we are expecting people with washing machines, dishwashers and tumble-dryers to be able to flex that demand for over a day if necessary-we had one family who have said, "In fact, we decide whether to cook using the electric cooker or the gas hob depending on what the price of electricity will be on a certain day," so we even have an element of arbitrage going on at domestic level.
Professor Bulkeley: We would echo that too.
Dave Openshaw: I think the potential is not yet fully explored. Again, I come back to the continual reinforcement. These consumers get a daily message and they get a monthly report showing how much electricity they have used at the different price periods. Compared with the control group, you can see there is a very, very distinct difference in behaviour, which I think is very interesting.
Sir Robert Smith: Thanks very much.
Q117Barry Gardiner: You said that it was unclear as to why the gas smart meters had had the effect that they had. Could it simply be that somebody had come into their home for the first time and explained to them how to use their gas boiler, and how to turn it down and how to turn it up, and nobody had actually bothered to do that before, rather than anything to do with the smart meter at all?
Dr Raw: That is perfectly possible. That is one of the series of possible explanations that we put. The problem is that the study itself does not provide evidence of that, because we don’t know exactly what the encounters with the installer were like. In theory, the installer was to go in, fit the meter, show the householder how to use the display, and was not charged with helping them with their boiler. They were not boiler engineers; they were meter-installers. If the householder had asked, they might well have said, "Yeah, yeah, push that button and that button. That’s fine." I would turn that around and say, "Since that is a possibility, why not try to incorporate that positively into the smart meter roll-out?"
Q118Barry Gardiner: Absolutely, and more than that, why not do a controlled experiment to see how much saving you would get by simply doing that and not installing a smart meter? We may be going to a lot of expense to install the smart meters to achieve what you could achieve by another means.
Dr Raw: I think the smart meters will have other benefits but, yes, the trial will be the interesting one.
Dr Darby: There certainly is a big issue about the extent to which people understand their heating controls and can operate them.
Barry Gardiner: Absolutely.
Q119John Robertson: How useful are in-home displays to help consumers reduce their energy consumption?
Dr Darby: Pretty useful. There is a first order effect, which is that, for a lot of people, this particularly gives them an awareness of their electricity consumption that they did not have before and it gives them a tool that they can experiment with. They can switch things on and off and see what effect it has. It gives them a feeling of control that they have not had before, so we typically see savings from that. That is the first order effect.
In the longer term, it helps build up an energy literacy, so that they start to be more open to suggestions of the kind coming from people like London Power Networks about belonging to this whole thing, the grid, being active in it and being able to shift their consumption in such a way as to help the grid to function better. Just to go to the experimental evidence that we have, when people have a display they will typically react better to time-of-use pricing and they will produce better peak savings, so you get that second order effect as well.
Q120John Robertson: What about design, does that come into it as well? Are some better than others?
Dr Darby: Yes.
Q121John Robertson: How does a consumer get to know which ones are the best?
Dr Darby: Trial and error, I suppose.
Q122John Robertson: You are our experts. Do you not give that advice?
Dr Raw: You ask a good question, because the consumer is probably not in a very good position to understand, "Should I buy this meter or that meter?" At one level there will be minimum standards for the displays that are to be provided alongside smart meters, so they will have to meet that standard. Once the roll-out starts in earnest, I would be surprised if the suppliers themselves didn’t compete by trying to explain what their display did that other people’s didn’t. Though there is a risk even in that, because I think what people benefit from most is really simple, direct information presented in a very visual fashion.
Q123John Robertson: So we are not tied into who gives the best advertising is who gets the sales?
Dr Raw: That could be potentially counter-productive, but I guess that is true of the way the market operates in general and not just in this specific instance. The important thing is to understand that the evidence shows that when you give people the choice, the complicated device, ultimately it is settled on a few small pieces of information. They are looking at the cost ticking over or the kilowatts ticking over, but all that helps them with the kind of process that Sarah has described. Once they are engaged with that, the kind of display it is becomes less important, and the way in which it has been initially explained to them is more important. When I say "explained to them" I mean two things. One is how to use it: the operation of it, which buttons to push; the other is how to use the information that you get out of it, which I think is probably more important.
Q124John Robertson: That was part of my next question but, having said that, I find that if somebody gets a new item and they don’t really understand how to use it, it just gets pushed to one side.
Professor Bulkeley: Yes. We have spent almost 500 hours now speaking to people about in-home displays and smart meters, which is probably more than anybody could quite bear. What we found is roughly two-thirds of the people that we have spoken to are very enthusiastic about their in-home displays, about one-third of them are less enthusiastic, and about 3% actively disconnect them, so very few actively move away from them.
The in-home displays we are looking at is a traffic light system-red, amber and green-and people find that intuitive. People don’t even ask for it to be explained. People understand that if it is red, something is not quite right, if it is green it is fine, and they like that.
John Robertson: That is probably like what my grandson gets at school, depending on his behaviour.
Professor Bulkeley: Yes, exactly. Red, yellow and green, they understand it and they get on with it very well.
Q125John Robertson: Can I ask you a question? It is something I wanted to ask earlier, but I felt rude so I moved on. Your investigations really only apply to people who know what they are doing. It doesn’t apply to people who are poor, who don’t understand and who need the additional help. You are really only dealing with a certain amount of the community, are you not?
Dave Openshaw: That is not quite true.
John Robertson: Convince me on that.
Professor Bulkeley: I will convince you. You go first and we will go that way this time.
Dave Openshaw: I am sure it is true of Northern Power Grid’s trial as well, but certainly for the trial we are conducting with EDF Energy, who are managing this dynamic day-ahead tariff, we have deliberately chosen a cross-section of Acorn groups, so we are getting a socio-economic balance in terms of the consumers who are participating. There are something like just over 1,100 consumers participating with, as I say, a balance across different socioeconomic groupings. It is not exclusive to any particular type of consumer. Although it is too early to draw firm conclusions, it is certainly true that we are getting different reactions from different socioeconomic groups. I come back to the point about clarity of display. One of the things that EDF Energy do is each month every consumer gets a clear visual report showing how much electricity they have used at high price, low price and mid price periods and how that is reflected in their overall electricity charges.
Q126John Robertson: Do you think a person I am talking about would have a great deal of understanding, or is there difficulty in understanding?
Professor Bulkeley: I will have a go at convincing you instead.
John Robertson: Okay, your turn.
Professor Bulkeley: My turn. I think it is important for us all to understand that people are not ignorant about their energy use. In fact, the people who struggle the most to pay their energy bills are the people who understand energy the best in many ways. Groups of-
Q127John Robertson: Yes, but these are the people who turned it off.
Professor Bulkeley: Yes, they do turn it off but they try to manage it. There is a group that we have worked with in Middlesbrough, called Thrive, a fantastic group of women trying to help their community understand energy use. For those households, energy is absolutely paramount. It is one of the first things they think about. It is what they try to manage their household economy by. So far, our experience is that where messages are straightforward-and I am not going to comment on whether that is the right approach-for those households that can be a very useful way of communicating with them and helping them manage their budget, but it equally means that we need to understand that all households have an understanding of their energy. We all know whether it is too warm or where the sun comes in a window. Those of us who use a clothes dryer to dry their clothes, like I do, will move it around the house. We find lots of people telling us, "If it is a sunny day I will hang the washing outside." People have a good understanding of energy in those terms. They might not have a good understanding of price. They might not have a good understanding of bills, which do affect them, but they have energy knowledge and we should be trying to use that and come from their perspective.
Q128John Robertson: The one thing that I have found, particularly with people who have the least amount of money, is they all have one of these things.
Professor Bulkeley: They do.
John Robertson: They have a smartphone because they cannot afford to have a landline and various things. They understand these items, so where can we link that into the system? Is that possible? Is it something that we are going down that road? I believe that would be a good end.
Professor Bulkeley: Yes, absolutely.
Dr Raw: It is current; you can do it already.
Q129John Robertson: Are you looking at that?
Dave Openshaw: We are doing it as part of this trial. One of the options that consumers have is to have a text message each day, as well as the message to their in-home display.
Q130John Robertson: The feedback from people? Good, bad?
Dave Openshaw: Yes, very positive indeed, as I say.
Professor Bulkeley: Probably an under-utilised function at the moment.
Dave Openshaw: You make an interesting link there, and, while some people express concerns about complication of time-of-use tariffs, people understand with a mobile phone you will have different rates depending on what time of day you use it.
Q131John Robertson: That is the point I am trying to make. They understand this. Whereas something that goes into the house, if they don’t understand it immediately, it has a habit of getting shoved into the background.
Professor Bulkeley: That is where you come back to the idea of the in-home display, because everybody we have talked to about it, if we say, "Can you show us your smart meter?" they show us their in-home display. They don’t show us the bit that has gone into a cupboard somewhere.
Q132John Robertson: Something I felt very strong about is that the IHDs should be fitted right at the start when we are rolling out the smart meters. Do you agree?
Professor Bulkeley: Yes.
Dr Raw: Yes, absolutely.
Q133John Robertson: Why do you think that is the case?
Dr Raw: The point about people’s understanding is important. In EDRP, what we found is that the SSE trial was the largest of the four, and they spread their sample across different MOSAIC groups, different socio-demographic groups, and the intervention did not depend on which group people were in. In the E.ON trials, they broadly split their sample into people who were more likely to be fuel-poor, less likely to be fuel-poor. It is the people who are more likely to be fuel-poor who, if anything, gave the more positive response.
That makes sense when you look at understanding being important. Everyone has a different level of understanding - their starting point, what kind of equipment they understand. You don’t have to know Ohm’s law to make this work. It operates on a different level. It is not just about understanding what to do and having the equipment; it is also about having the motivation. Sometimes people who perhaps don’t have a lot of money have a lot of motivation to save, so those two things going together are important, alongside the resources to make the change, the time, the space, the money. It may be, if you are a pensioner spending a lot of time at home on limited income, you have the motive and you have the time. Even if you start off with not much understanding, with or without someone’s help, in fact, you will get there. I think it is a question of what comes first. Are you motivated to make a change; therefore, you seek to understand, or do you understand and then it is easy to make the change?
Q134John Robertson: Do you think these meters should be optional or mandatory when they are fitting?
Professor Bulkeley: The IHD?
Dr Raw: I think it is very difficult to tell people they have to have one, but I think they should-
Q135John Robertson: If they are going to have a meter anyway, should they not have the top of the range?
Dr Raw: They should definitely be offered one.
Q136Sir Robert Smith: If they have moved on in technology terms, do they need to put one in every house when the installation is going on, because people will be moving houses, so that there is always an IHD there, or does the first adopter decide, "I don’t need an IHD because I am happy with an internet report?"
Dr Raw: It is a concern that you give the first consumer to receive the meter a display. That is good. The next person to come into that dwelling may not receive that display. They could probably go out and buy one, or they might decide, "I will bring mine with me, along with the light bulbs from the house I had before, because I am not sure I am going to have one when I get there." The dynamics of that will not be entirely positive. As you say, at least if we get one into every home, one should remain in every home. People may have to replace them from time to time. They seem to be quite robust, but you cannot count on them lasting for ever. Yes, certainly get one into every home if possible.
Dave Openshaw: I think what will happen over time is that we will see an evolution. The smart metering equipment technical specification already makes provision for consumer access devices, so I think what we will see is more sophisticated means of interacting with the smart meter. It may be that the in-home display is ultimately replaced by something more sophisticated, and ultimately smart appliances that are actually doing that communication on behalf of the customer so the consumer has less need to interact physically, because we will see an increase in smart appliances that are reacting to price change signals on his behalf. It will be an interesting evolution, and it may be that the simple in-home display becomes obsolete in the long term.
Dr Raw: That evolution would not just be about energy display, but integrating that with other smart services, such as security, would be the obvious way to go.
Q137John Robertson: Are you working on that for smartphones? Were you working on it? If not, why not?
Dr Raw: I think that would be for the app developers.
Professor Bulkeley: I don’t think you want us to, actually.
Dr Raw: You don’t want me designing an app. I almost know what one is.
Q138Chair: To maximise the benefits from all this, it will be necessary to enable people to respond to time-of-use pricing, changes in the weather, occupancy of buildings and so on from their smartphone.
Dr Raw: It seems a logical statement, yes.
Chair: Given that is the way we do everything else now.
Professor Bulkeley: It would be necessary for some people, but it won’t suit everybody to interact with their in-home displays and their home systems in that way, but there will be a large majority of people who will enjoy doing that.
Q139Chair: The proportion is very rapidly increasing. I think those elderly people who have grandchildren at home will say, "Look, you can actually do all that stuff now."
Professor Bulkeley: I am not necessarily thinking of those elderly people. It is just some people-
Chair: Well, speaking as an elderly person myself.
Professor Bulkeley: Some elderly people are much more proficient at those things than some young people. It is just a question of how you-
Chair: That is perfectly true but, from my observation of my contemporaries, a lot of them do have a lot of help particularly from their grandchildren in mastering these things.
Professor Bulkeley: They are missing out the middle generation, I guess.
Chair: If you are over 40 you can forget it.
Q140Dan Byles: I want to explore a bit more of this whole demand response and demand shifting. The Irish customer behaviour trials found that time-of-use tariffs and demand size stimuli could reduce overall electricity usage by 2.5% and peak usage by 8.8%. Are those the sorts of figures that we have seen from the UK trials? Would you say they are-
Dave Openshaw: We are certainly seeing a bigger shift than that, albeit we are three or four months into a one-year trial. I am glad you have raised that point because it is important going forward, in particular, that we focus on what is happening to peak demand. You can imagine a future, with electrification of heat and transport, where people are coming home from work in their electric vehicles, plugging in the vehicle as soon as they get home to make sure it is charged the following day. They will arrange the heating controls so the heat pumps are kicking in around that time. Bearing in mind that peak demand occurs now between 5 o’clock and 6 o’clock on a winter weekday evening, the danger is that, while we might see a 20% increase in electricity consumption, because of heat and transport electrification, we could see a much bigger increment in peak demand, which would be very, very costly in terms of the amount of generation peaking plant we need to build, the amount of transmission network capacity.
A real focus going forward, which is why we think time-of-use tariffs or critical peak price tariffs are going to be important, is to make sure we do shift demand. Even if we don’t reduce electricity consumption, but we shift it away from the peak periods, that will have a huge impact on the national cost of energy, and therefore will have a huge benefit on the national economy, make our exports cheaper and will put more disposable income into people’s pockets.
Q141Dan Byles: Is there a danger that if everybody has a smart meter and a smart system that is helping them shift to the off-peak, that off-peak ceases to become off-peak? If everyone takes the shortcut, the shortcut is no longer a shortcut.
Professor Bulkeley: Yes. We would say in our trials so far, we have found a small post-8.30pm peak. A small one, not as much as the evening peak. In our trial, so far we have only looked at the summer data. Again it is provisional, but we see roughly the same kind of 9% to 10% shift in the peak. It will be interesting once we have had the data run through the winter to see whether we achieve that as well.
We have the energy data that tells us one thing and then our work, actually talking to people, which tells us why they are doing it. It is that we are mainly finding that it is some of the most energy-intensive practices of washing, drying and laundry that are most flexible, and also some kinds of cooking. We could get into the details of that. I am not sure you wish to. There are a lot of households who have shift work, who have multiple families in the same house, who are not all eating at traditional meal times. They seem to have quite a lot of flexibility within them. There are also people who do batch cooking and heating it up, all these different kinds of households who have found time-of-use and in-home displays help them manage all of that quite effectively.
The things that seem to be stickier are showering times, which are often associated with other rhythms and routines of the day, going out to work and coming back from work and school; heating, of course, where there is electricity heating and cooking, although we find a decline in electricity use for main weekday cooking As we have talked about before, there is a reasonable amount of flexibility in households and it is about how you tap into that.
This raises the question about, when we are communicating with people about smart meters, what are the messages that we are trying to give them? Is it about demand reduction? Is it about the savings that they can create as an individual, or are there broader messages about their engagement with the grid as a whole, the overall cost of supplying energy, and about shifting from one time of use to another time of use? Those two things can go together, but getting the message, the communication, support and engagement for people, clear on those two different things is going to be important.
Q142Dan Byles: In terms of the different interventions and incentives, and we have discussed a number of them-time-of-use tariffs, in house displays, smart appliances, smart devices, the smart meters themselves-is there a danger that your trial is going to struggle to identify which ones are producing the incentive, for example, to demand shift?
Dr Raw: That certainly was an issue with the Irish trials, the way that everything is mixed up and everyone had a whole bunch of things happening. It is quite difficult to split it out. To the extent that we could, we split it out in EDRP. Some of the trials are very simple: "We’re doing one thing; let’s see what happens." Others were mixtures and you had to do some quite sophisticated statistics.
Q143Dan Byles: Have you managed to draw any conclusions, from what you have seen so far, as to which are the more effective incentives?
Dr Raw: Could you explain what you mean by incentives?
Dan Byles: Perhaps "incentives" is the wrong term. Which tools are best at leading to demand shifting, for example?
Dr Raw: To shift demand, clearly a time-of-use tariff is pretty much the only thing that is being studied. We have critical peak pricing work coming through, which I think will be very interesting to have a look at. It has been done more in relation to air conditioning in hotter regions but not so much in relation to the UK, so that will be very interesting to see what is coming out of that.
The time-of-use tariff itself seems to do some shifting, and the amount of shifting it is very difficult to pin that down to exactly what is the tariff like. The Irish trials used four different tariffs with increasing extreme difference between the peak and the off-peak, with the daytime in the middle. You can just about see a trend that the bigger the extreme, the bigger the peak time saving. It is not statistically significant, even though they relaxed their criterion for statistical significance in that trial.
What is important seems to be the trigger that makes people think, "Aha, it is more expensive right now. Perhaps I could do that some other time," or to be more careful at that time. Because it is both things. People will switch off the light, "It’s peak time, so why has someone left that light on?" They are not going to then switch the light on sometime during the day to compensate, so you get a total saving. Whereas for other things they are shifting. The Irish trials again show the shift tends to be a delay. They say, "No, don’t do it now. Do it later." With more practice they will probably have more anticipation, "It’s the middle of the day, so I’ll stick it on now-I won’t wait until the evening," or "I’ll use a timer to do it while I’m out."
Q144Dan Byles: Am I right that the Irish trials found no evidence of a tipping point in terms of-
Dr Raw: That is the way they put it, yes.
Professor Bulkeley: I would suggest that it is perhaps a little bit of a red herring to think about the financial difference between the time-of-use and the non-time-of-use tariffs. As Dr Raw has just pointed out, in a sense it is more about a changing awareness of the nature of electricity, where it comes from and what is happening to the grid as a whole, rather than the price itself. We find people saying, "If it saves me 5p, well, that’s okay," but one of the things is that they found they can do it. They found that their dishwasher has a timing switch that they did not realise it had. A lot of people talk about how clever their wives are at all of this-I thought it was worth pointing out-how clever their wives are at cooking and how clever their wives are at washing, and so on.
Dan Byles: The original smart meters.
Professor Bulkeley: Yes, they are the original smart meters. The army of the UK’s wives in the north of England have been going about their business in very different ways, and it is that sense of motivation rather than necessarily only the price.
Q145Dan Byles: Joking apart, has anyone looked at whether there is a gender difference in this? Have they looked at male-only households and seen whether there is any?
Professor Bulkeley: We haven’t looked at it explicitly, but I would say, yes.
Q146Dan Byles: It might be that different genders respond better to different prods. Men like shiny things.
Professor Bulkeley: Yes. Say we have talked to roughly 200 households, I would say in 80% of them it is women who are seen to be doing the work of responding to the smart meter in one way or another-
Dan Byles: That is an interesting dynamic.
Professor Bulkeley: -in terms of shifting their chores, shifting their cooking, but that is because that is the way domestic labour still is predominantly achieved.
Dr Raw: Yes. I would say it is not gender as such. It is the split of the household roles.
Professor Bulkeley: It isn’t, it is the way their household is, but perhaps other studies will find it different.
Dave Openshaw: An untapped opportunity is around explaining the carbon benefits as well. Whether they are sceptics or otherwise, most people recognise that there is a carbon agenda, a greenhouse gas agenda out there. I think the extent to which they can help reduce carbon emissions by their behaviour is a useful area of focus as well. A lot of people are putting photovoltaic panels on the roof of their houses. For example, if you could link that to charging their electric vehicles during the day there is a real synergy there, a real opportunity there. Certainly, if those messages can be explained as well, why the generation mix going forward will be what it is and how we can fully exploit that low-carbon energy resource in future I think would be helpful.
Q147Dan Byles: How much opportunity do you think there is for automation? We heard from the Oklahoma trial that programmable thermostats played a significant role in helping consumers reduce electricity demand at peak times. Effectively, it takes the decision away from the individual themselves.
Dave Openshaw: There is no doubt in my mind that smart appliances are the future, because it takes the physical need to interact with the appliances away from the consumer, provided they trust that appliance. What is important, though, is we recognise that this is a consumer choice, buying an appliance that will interact with a smart meter. There has been a certain amount of scaremongering around Big Brother taking control of your domestic appliances. That is not where we are coming from at all. It is very much about consumers being able to purchase-potentially, for a very low incremental cost-a device that will interact with a time-of-use signal and take that need to physically interact away from them. I think there is a huge, huge future there for automation. Electric vehicle charging is an obvious case in point going forward.
Q148Dan Byles: Other than The Mail on Sunday, are you aware of any real people who are concerned about Big Brother in all this? Has there been any push-back in any of these trials from people who have been concerned that this is in any way sinister?
Professor Bulkeley: Again, from our study of the people that we have talked to, no household or business has raised a question about privacy concerns. I think that it is important to put that into the context of the trial, which has a variety of partners, one of which is a university, and it is regarded as about creating new forms of knowledge. We have had some concerns expressed by a sizeable minority-certainly not by the majority-that this will be used as a means of increasing profits for the energy companies. What is important is what this is being used for. It is not necessarily a question of privacy, per se, but what matters is: why is this data being collected? Who is going to benefit from it, and what is going to happen to it in the future?
Q149Dan Byles: Finally and quite briefly, how useful do you think the international studies are in applying to a UK context? You have referred to air conditioning, for example. Obviously some of these are not necessarily-
Dr Raw: You have to be quite careful what they are studying and where they are studying it. If they are studying air conditioning in Oklahoma, there are some quite big differences in both context and the technology. If they are studying smart meters in Ireland, you can see there are some cultural and physical and climate similarities that-
Dan Byles: A degree of common sense, perhaps.
Dr Raw: Exactly.
Dave Openshaw: Also, I would say you need to be careful what the generation mix is as well. We have not a unique but a specific type of generation mix going forward, which will involve very high volumes of wind capacity, as well as potentially a strong nuclear base for base load. That could be quite a different situation in other parts of the world, where there is perhaps more hydro storage, more solar photovoltaic capability and so on, so you have to be a little bit careful about drawing comparisons that they are looking at a similar energy portfolio.
Dr Darby: I do think there is definitely a role for automating some functions in time, but I would want to add the big caveat that people should always be making a really well-informed choice of that automation, otherwise it is going to feel Big Brother-ish.
I ran some focus groups last summer pitching different types of electricity tariff to people and getting their reactions. When it reaches the point at which you pay a tariff and your supplier or network operator will be able to switch off your water heater at certain times, that kind of thing, there was certainly a bit of nervousness about that. You would feel that people would want to have gone along a certain learning pathway to understand why this would be a good thing for them and for the network, and be fully convinced of that before they did that.
Dr Raw: Yes, I think that is right. There is an intermediate step between where we are now and that kind of automated control. What we alluded to earlier is having appliance-specific feedback, so people are able to see and learn more clearly which are the big users, and therefore are more likely to see the benefit, say, of having a button on their washing machine that they push and it won’t come on until it is the minimum tariff for the day.
Q150Barry Gardiner: You have slightly shifted your ground in the last five minutes, haven’t you? First of all, when Dan said about issues of privacy, you said, "No, no, no," and then you said, "Actually, there may be something if people feel that it is more to the benefit of the company, rather than them." When you look at what Consumer Focus have said, they have said that, even though there have been few public concerns voiced about smart meter data or even health, the potential for these to become issues that jeopardise consumer engagement and result in detriment should not be underestimated.
Professor Bulkeley: Yes, absolutely. I would say that that is about how we engage people, and also what we think the data is for and who is going to benefit from it. Effectively, it is the question of ownership and the question of use, and to what use is it being put, by whom and for what purposes? It is not at all that people don’t have concerns. That is not what I meant to say. I did mean to say that people haven’t raised concerns in this trial, and I think they haven’t done that because of the use to which the data is being put. Do you see what-
Q151Barry Gardiner: Let me give you another scenario that may affect people’s response. If they are engaging with their smart meter, taking active steps to bring their consumption down, but at the same time the price per unit of their fuel is going up and they see no material difference to their energy bill-although, of course, you and I both know that their bill would have been higher had they been carrying on as usual-how likely is that to affect their engagement with the programme?
Dr Raw: You can make it a more difficult sell, clearly, if all you can say to people is, "If you take all this effort, your bill will not go up as much as it otherwise would have done." It is not quite as strong as saying, "Your bill will go down." You can probably work out some pictures that show it all in relative terms that make it more convincing, but there is still the basic issue that bills are going to go up, everybody’s bill is going to go up, "If you do this, your bill won’t go up as much as it would have done." It is a more complicated thing to sell. If people are able to see for themselves that they are using less energy and they are able to focus more on the energy consumption, the number of kilowatts that they have used, then it will make more sense. However, I agree, it is one of the more difficult things to try to get across to people.
Dave Openshaw: Certainly one of the things that consumers will be able to do, if they have 13 months of data available to them, is that they can then use that data to ask suppliers, "What would I have saved with your tariffs, or your tariffs, or your tariffs?" There are real opportunities there for that comparison to be made.
Professor Bulkeley: I want to add two brief things. I realise that time is ticking on. I think the first is that people will not necessarily do it because it is an effort anyway. People will use these things because they fit into their everyday life and they will be gaining other things from it, as I mentioned before: budgeting, looking after their family in particular ways, and this kind of idea of gaming. Those are all other motivations that are not about reducing costs. Cost matters, but there are other household benefits.
The second thing that our evidence shows, and where we need to improve our communication, is this sense that people are willing to be part of it. They are willing to be part of the grid. They are willing to be part of an idea of securing Britain’s future energy system.
Q152Barry Gardiner: This was a self-selected group, wasn’t it?
Professor Bulkeley: This is a sample from-
Barry Gardiner: This is the group that responded and said, "Oh yes, I am willing to be part of your trial." It is not the people who said, "No, why on earth would I want to be interested in doing something as stupid as that?"
Professor Bulkeley: That is true, and I will take that on board. However, it is a sample taken from a group of 15,000 people. It is not like 300 or something early, early adopters. It is a reasonably big trial from which we have chosen this. I also think that when we are talking about the value to consumers, people live in very different geographical contexts. The value to consumers who are living on the end of a ropey grid that often suffers blackouts, and they think, "Actually, we would like to be part of a community that can manage this grid better and that can ensure that we don’t have blackouts," so the value is different.
Q153Barry Gardiner: I think it is a good point and you made that one before about engagement, which I think is absolutely right. Let me press you a little bit on the issue of privacy and security, because there are people who have very real concerns that the data that is accumulated on them is not only going to be able to be used by the companies for their own marketing purposes, but, if it is tapped into, could provide criminals with new pathways to get at them and give them information about periods during which the house was habitually empty, which leads to greater exposure to crime. What did you do on the trials to address that, and how important do you think it is to try to address that in a wider roll-out, when it happens?
Dave Openshaw: If I can start, and speaking from a network operator perspective, certainly on the trial we have engaged with a privacy impact assessment to start with. We engaged with an expert on privacy impact assessment to make sure that what we were doing was not only compliant with the Data Protection Act but was also fit for purpose, in terms of the data that we were taking. The purpose of network operators in taking the data is not to look at individual consumption. It is to aggregate that consumption so that we understand what is happening from a network perspective, so immediately you have taken-
Q154Barry Gardiner: Do you seriously think that there won’t be any way in which a company can use this data for commercial purposes?
Dave Openshaw: What is going to be absolutely critical, and the only way that network operators will have access to half-hourly data going forward, is if they can demonstrate to DECC or Ofgem-depending on what time they present their plans-that those aggregation and data privacy provisions are absolutely concrete and robust. The industry is looking very hard right now, for example, the Energy Networks Association is engaging in a study to understand exactly what provisions need to be put in place to make sure that that data is absolutely secure and cannot be accessed or misused in the way that you describe.
Q155Barry Gardiner: That is the substantive argument. Now let’s take it forward into the argument about how do you communicate that substantive position to people on the roll-out, so that their fears are allayed and they are willing to engage. How have you done that?
Dave Openshaw: Certainly as part of the trial we do that, but I think the question relates more to the mass roll-out.
Q156Barry Gardiner: No, what do you do?
Professor Bulkeley: What do we do?
Barry Gardiner: I am saying, when you go into somebody’s home to install the smart meter, what discussion do you have? How do you train your operatives to have a discussion about privacy and security?
Professor Bulkeley: Anybody who is taking part in the active side of our trial, whoever is engaging with them has a whole protocol of things that they talk about: what the data will be used for, what the purpose of the intervention is, what will happen at the end of the trial. It is all the ethical procedure that you would go through to achieve someone’s informed consent to participate. The other part of our trial is an opt-out, so for the vast majority of customers-it is probably the same on yours-it is an opt-out letter that goes out to them to say, "Would you be willing for your data to be shared as part of a trial for a limited time period?" Those are the two mechanisms that are used, and of course we could let you see copies of that documentation if you wished to see it.
Q157Barry Gardiner: What refusal rate did you have on grounds of privacy?
Professor Bulkeley: About 2% on opt-out.
Dr Raw: About two years ago I was reading through all of the huge piles of papers on the EDRP trials, and it was like this. I still feel quite tired from reading it. I cannot recall anything substantive in there about data security or concerns raised by the participants about data security. For most, if not all people, it seemed at that time to be a non-issue. That is not to say it will remain a non-issue or it will be a non-issue for everybody, but, as you say, apart from the substantive perspective on the controls that are put in place, I would simply point to-
Q158Barry Gardiner: You have not come across Stop Smart Meters! (UK), then?
Dr Raw: Of course, yes, that is why it will probably increase. On the other side, my mobile service provider knows who I am talking to and when I am talking to them. They probably know where I am, if they care to look. My supermarket loyalty card is telling people what I am buying over a long period, my typical patterns of behaviour and which shops I go to. For someone to know when I cook my dinner seems to me relatively trivial.
Barry Gardiner: I am not so sure about that.
Professor Bulkeley: Perhaps the lack of response about these data protection issues is because people have a good deal of trust in that side of things at the moment. The question is: what would happen if that trust was violated, if something was to go wrong with it? My sense is that it is an important issue and we need to ensure that that trust is maintained, but in some senses people are reasonably satisfied with the current arrangements about that.
Q159Barry Gardiner: Who is best placed to provide all that reassurance and information to the public? Who has the undivided trust of the British public? You are going to tell me it is the Big Six energy companies?
Professor Bulkeley: It is David Attenborough.
Barry Gardiner: You are going to tell me it is the politicians? You come along there as independent researchers, and of course you are academics, and everybody says, "Oh, well, if you have a white coat on, we’ll do whatever you tell us." We know that from the Milgram experiments. Who is going to be best placed-it is a serious question-because the messenger often determines whether the message is going to get through?
Dr Raw: You are right, the messenger is important but I think it needs to come from multiple sources. It needs to come from everyone involved. It needs to be trusted public figures who have been brought in, who are entirely independent. David Attenborough would be an option. Alex Ferguson is looking for a job. It needs to be Ofgem. It needs to be the industry itself, and I think-
Q160Barry Gardiner: That would leave one side of Manchester without smart meters.
Dr Raw: The other side without a manager. I think that trust needs to be built up in layers, because you need trust in the motive of what people are seeking to do, trust in the credibility of their plan, how they are going to do it, and trust in their competence to deliver. That is why I say I think it needs to come from multiple messengers, all who are trusted in those different categories of trust.
Professor Bulkeley: Quite a bit of the smart meter roll-out at the moment through the trials is taking place with the participation of third parties, so you have social landlords, you have housing associations, you have local authorities, and you have fuel poverty action groups. You have a whole range of third parties, from the civil society sector on the whole or who are locally based, in one way or another, participating in these trials and in the development of other kinds of low-carbon technologies. We cannot underestimate the role that those organisations have had in establishing trust.
Barry Gardiner: Thanks very much.
Q161Ian Lavery: Just on the costs and the benefits or potential benefits to the consumer, I think you mentioned before, Dr Raw, about the potential savings. DECC estimate by 2020, with smart metering, and better than that, the average consumer will save somewhere in the region of £34 per annum. That is on the average bill of £1,496. Is this really the case? Do you agree that that is the case? Do you think that people will be saving money? Do you think there is real benefit that comes from that?
Dr Raw: It is certainly feasible. I am not an economist and I have not studied in detail every aspect of that impact assessment and the economic case for it. The factors that they have taken into account, the way that they have done their calculations all seem credible. I don’t feel I am in a position to second-guess or say that they should be a little bit higher or a little bit lower, but overall it seems the right sort of order.
Dave Openshaw: They have made reasonable use of optimism bias in their calculations. I think the methodology is robust. It can be described as an IT project, and that always carries risk of cost overrun. Certainly, the Energy Networks Association, for example-and I know Energy UK on behalf of suppliers-have engaged closely to make sure that those projected savings look robust. We have had a lot of input, in terms of the network efficiency benefits that will come from smart meters as well. So I think there is no case at the moment for saying that those assessments are overly optimistic, but they do need to be kept under review as the programme rolls out.
Dr Raw: Again, my point would be not that we challenge are they right or wrong, but rather: how do we make them better? How do we get more out of it than currently expected?
Q162Ian Lavery: Do you think that, with the roll-out of smart meters, inevitably consumers will have to pay more than what they will get back in the long run? Basically, they are paying for the burden of a smart meter without receiving any potential benefits in the long run.
Dr Raw: Inevitable that they will pay more? No, I don’t think so.
Professor Bulkeley: There is also a wide set of benefits, which we have come back to a few times in the evidence that we have given today. The benefits of having a working electricity grid, and the benefits in terms of broader ideas of energy security and in terms of climate change. Those are the benefits on which it is quite difficult to put a cost. Therefore, they are usually not taken into account in those calculations. If we were to ask people, "Do you want an electricity grid that works or not?" then their answer would probably be yes. If the question is, "Is there another route to achieve that other than smart meters?" I am not a power engineer; I rely on the power engineers to tell me that they think that it is necessary for managing the network.
Q163Ian Lavery: Then, if you ask them if they are prepared to pay for the smart meter, they are going to say, "What will I get in return?" That is the question. Will the consumers be paying for the roll-out of smart meters without getting enough benefits in return? That is basically it.
Dave Openshaw: Obviously, it really depends on the extent to which they engage with the information and make use of the information. I cannot overemphasise the point that Harriet makes about the overall cost of providing electricity, going forward, including generation, transmission and distribution. If we get the sort of engagement we really need, then we will be able to roll out affordable low-carbon transition, so we will have secure, affordable low-carbon electricity going forward. It really does depend on the extent to which consumers engage. If the behaviour doesn’t change, then clearly the benefits are not going to be so high. If they engage in the way that we hope they will, through time-of-use tariffs and so on, then the benefits are potentially enormous in terms of saved cost of electricity.
Dr Raw: It is a good point, in the sense that the amount that consumers benefit is partly in their own hands, how they make use of the technology that is there. It sounds a little harsh perhaps. However, the more we can help them to benefit through the process of the roll-out-not just the fact of the roll-out-the better it will be, and the more likely that there will be a larger positive balance in their favour.
Q164Sir Robert Smith: Do we need to restore the confidence in the working of the energy market? Obviously, if the market is working, if you do something that makes it more efficient to provide energy, then the consumers should see the benefit.
Dr Raw: You would certainly hope so, yes.
Chair: Thank you very much for your time. It was a very useful and interesting session for us, and we will take careful note of what you have said.
Examination of Witnesses
Witnesses: Dr Martyn Thomas CBE, Chairman, IT Policy Panel, Institution of Engineering and Technology, and Alex Henney, EEE Limited, gave evidence.
Q165Chair: Good morning. Thank you both for coming in. As with the previous panel, perhaps if you can begin by outlining your backgrounds and the expertise that you have on the issue of smart meters?
Dr Thomas: Happily, yes. I am Martyn Thomas and I am a software engineer. I have over 40 years’ experience in the software industry, working on large real-time, security-critical and safety-critical systems. I chair the IT Policy Panel for the Institution of Engineering and Technology. At the beginning of the smart meter programme, the IET brought together its policy panels on energy, communications and IT to make sure that, in interacting with DECC, we were able to provide a multi-disciplinary view of the impact of smart metering and, more widely, of the smart grid. From the beginning of that programme, I have been providing the IT input to that combined group.
I would also like to say that I am also a non-executive director of the Serious Organised Crime Agency, which of course has focused my attention fairly sharply on the opportunities for crime, which were raised earlier, and on security. However, today I am speaking as Martyn Thomas, the chair of the IT Policy Panel for the IET, and I am not representing the views of SOCA here.
Alex Henney: Good morning. My name is Alex Henney. In the early 1980s, I was the customer representative on the board of London Electricity. I got to meet Bob Peddie, who was chairman of Seeboard, and he was the joint patentee of the world’s first smart meter. We tried to commercialise it but it was too early. Also, in 1987, I was the first person to propose a competitive restructuring of the electric industry in England and Wales, through a report that the Centre for Policy Studies published, and I was involved with Cecil Parkinson and officials in the early days of the restructuring.
Subsequently, I have worked from the Nordic market to New Zealand, via North America and places in Europe. I did a study of electric smart metering and I would emphasise I know nothing about gas, other than one puts it in gas turbines and cooks my food with it, so I focus entirely on electricity. I did a study of electric smart metering in 14 countries, which took in total-together with the review-about a man year. I was very impressed by the big Italian company Enel, which in the year 2000 designed its own smart meter, pilot tested it and, by 2008, it had rolled out 32 million of them at a cost of £65 a pop all in, including IT systems. I was also impressed by Iberdrola in Spain that pioneered open source power line carrier system and meters.
I wrote a book called The British Electric Industry 1990-2010: The Rise and Demise of Competition and it included a chapter titled "Smart Metering Provided Unsmartly". I tracked through the numerous papers, White, Green and any other colour that you can think of, which Her Majesty’s Government and various other bodies have produced. It struck me that it was complicated. It was based on suppliers and every other country with a mandated roll-out bases that roll-out on the DNOs. It was expensive. Quite clearly, DECC did not understand the power line carrier, which is cheaper than wireless. On that you can see the Irish regulator’s comments, and I also provided the Committee with the comments of the person heading the roll-out in Iberdrola. When, under pressure, DECC looked at a DNO roll-out they got the sums wrong for two reasons: first, they did not use PLC; and second, they did not reduce the cost of capital for the meters.
In December 2011 I met Professor Ross Anderson, who is Professor of Security Engineering at Cambridge, and he was concerned that the IT project would be yet another Government screw up. We put together a paper, and last year we met Charles Hendry in February: Ross Anderson did the IT bit, I did the economics.
I noted that the net present value of the benefit proposed for a smart meter roll-out increased from minus 4 billion, when it was undertaken in 2007 by Mott MacDonald, to plus 4.9 billion when it was undertaken by the Civil Service. If that does not ring a bell then you can believe pigs can fly. It was quite clearly manipulated, and Ms Vicky Pryce told Professor Ross Anderson that it had been politically manipulated so that politicians could run around and say, "Oh, we’re going to save 5 billion." I did point out in my submission that the British Government’s assessment of smart metering was far more optimistic than any of the other 14-no, sorry, there were not 14 assessments. I think there were 10 assessments undertaken in other countries. So, if I can be blunt, it is a fantasy. It was manipulated by reducing the optimism bias, and I am sure my good colleague here will agree that 10% for an unspecified IT system is absurd, and also for the physical implementation that the Chief Executive of EDF Energy said was a high risk project. Furthermore, there was no attempt-as the Germans are doing-to discriminate between the benefit of customers who take a lot of electricity and customers who take a little. It doesn’t require much grey matter to work out that, in theory, those who take a lot are going to benefit more than those who take a little.
I regard the in-home display unit as a waste of money-
Q166Chair: I think we have established your credentials, and we are going to return to some of these subjects during questions. So, rather than expressing views about the value for money of the in-home display, perhaps we might explore that during the questions we are going to ask.
Alex Henney: Can I mention that, as a result of this paper, we asked the Cabinet Office Major Projects Review Group to undertake an assessment and the initial noises were quite favourable. As one of the members put it, "We looked at DECC’s evidence base and it was flimsy". Subsequently, I have attempted to get a copy under FOI and it has been a charade of specious Whitehall waffle. So we are now at the position where there are two IT companies on the list for the DCC, G4S of Olympic fame and Capita-see Private Eye- neither of which have experience of systems involving the sort of regulation that we have developed to a refined complexity. So my view is we should stop it before we waste more money.
Q167Chair: Well, that is pretty clear. We have had witnesses who have suggested that pressing ahead with the mass roll-out is pretty risky while there are still some technical and systems issues unresolved. In your view, what are the technical issues that still need to be addressed?
Dr Thomas: The roll-out of smart meters is a complex programme, and I would like to start by saying that the smart grid is the real prize-as you were hearing earlier this morning-and we need smart metering to support the smart grid. This is a national scale programme and it is important to look at that larger prize in determining the timescales, the cost benefit and so on. That hasn’t been done yet. We don’t even have a proper architectural design for the smart grid yet to enable that to be done thoroughly. Even so, it is important to view this not just as the roll-out of a smart metering programme, with the current specification, but in terms of its enabling capability for the smart grid.
Of course the dangers of rolling out, before we have a full specification, are significant and I hope the delay that has been announced recently will be used to make sure that the specification is firmed up before there is a mass roll-out. Clearly, it is not optimal to have a lot of meters already installed, which turn out to be incompatible in some way with the specifications when they are produced in their final form.
One of the things that the IET has pressed on consistently is that the roll-out timescales do need to be based on the engineering realities, rather than determined by judgments made by politicians or by senior civil servants for political reasons. So it is important that engineering judgment is used to inform the roll-out timescales. If we don’t do that we will inevitably get it wrong. We will get cost and time overruns and we will make mistakes that have to be fixed later.
We are concerned that currently the requirements are expressed very informally. They are natural language specifications with some diagrams. That means that they can only be checked for consistency and for completeness by human review. From all other complex systems we know that that means that they will contain inconsistencies, ambiguities and contradictions that will turn out to be significant problems in the programme later. We would press very strongly that the time that we have now, before the mass roll-out starts, is used in part to formalise and properly analyse the specifications for the meters, for the overall architecture, for the security properties and to prove that those things are consistent. We know how to do that and it is not expensive. There is time in the programme to do that and that is something that we would strongly recommend. So there are technical issues to be resolved but I think we have time to resolve them.
Q168Barry Gardiner: Of course DECC has recently put back the start date of the mass roll-out by a year, to avoid bashing into people’s homes with these things just before a general election. Although, of course, the reason that they gave was that the Data Communications Company, DCC, had to be up and running with the systems tested. What is your assessment of when the SMETS 2 specifications will be agreed?
Dr Thomas: It is not clear yet when that agreement will occur. There is still quite a lot of work to be done and there are a lot of stakeholders involved. The UK has a uniquely complex market structure and has chosen to roll-out smart electricity and smart gas meters and, therefore, the complexity of this programme is greater than in other countries. Getting these specifications agreed and then analysed, to be shown to be self-consistent and have the required properties, will take time.
Q169Barry Gardiner: How smart is it to announce that you will be beginning your roll-out at a particular point in time, without having any capacity to know when you will have those specifications agreed?
Dr Thomas: As Sherlock Holmes remarked to Dr Watson, "It is a capital mistake to theorise before one has data." It is important to drive the timescales from the engineering, not to try to constrain the engineering to fit in with predetermined timescales.
Q170Barry Gardiner: Thank you. How long do you think the end-to-end testing is likely to take?
Dr Thomas: The first thing I would like to say is that testing can only ever show the presence of faults and never show the absence. So, if what you are looking for is high confidence that the key properties of the system-like some of the security properties-are genuinely there, and you are looking for high assurance of that, you cannot get it through testing. That is well understood in the safety critical and security industries. It is well understood theoretically in the universities and has been for 40 years. Testing is not the way to get high assurance, either of functionality or security or safety or any other key property. You have to do that by analysis, and that requires using mathematically formal specifications and the associated tools to analyse them. As I said, we know how to do that, it is not particularly expensive. Indeed, everywhere it has been used it has reduced the final cost of systems. We would strongly advocate that these tools are employed where appropriate in this programme. It will reduce the testing times, incidentally.
Q171Barry Gardiner: If you were trying to work out how long it might take?
Dr Thomas: It depends very much on what level of assurance you want, and that is not specified. The degree of confidence-
Q172Barry Gardiner: I presume that they want a high enough level of assurance, so that householders are welcoming in sufficient number to make the programme deemed a success.
Dr Thomas: That degree of assurance has probably been developed through the trials that have been carried out. You could repeat those with the real systems and it would not take very long to do that level of testing. The issues that concern me are whether in fact there are vulnerabilities that could be exploited, or combinations of circumstances that might cause a significant failure, which would only appear later on and which would then cause a need for substantial rework. If you needed to be sure that all of those had been eliminated, then you have to do much more than the level of testing that has been carried out in the trials, and indeed, as I say, testing will not get you there on its own.
Q173Barry Gardiner: Let me be clear that I understand what you are saying. You are predicating that, without that level of analysis and corroboratory testing, there is a risk of a major unforeseen failure, which undoubtedly would undermine public confidence in the system and, hence, reduce the Government’s overall goal, which you alluded to earlier, which is the network goal rather than the smart meter goal?
Dr Thomas: Yes, there is that risk. Of course, it is not possible to quantify that risk.
Q174Barry Gardiner: Thank you. Is there also a risk that customers who receive smart meters during the early roll-out will be left with lower functionality, than those who receive the SMETS 2 compliant meters during the latter phase of the roll-out?
Dr Thomas: That depends very much on how the transition is handled, when you have rolled out a number of meters that are not SMETS compliant and you need to cope with the transition process of making those SMETS compliant. Otherwise, yes, clearly there is that risk.
Q175Barry Gardiner: Given that the Government have now announced these dates of autumn 2015, after the general election, to start roll-out, completion by the end of 2020-no doubt a glowing success just before the following general election-what is your view of these new dates?
Dr Thomas: They are better than the old dates, in that they do give us an additional year to make sure that the specifications are sound and to fit things in better to the engineering realities However, since we don’t have the full specifications, we don’t know the details of the bids that have been put in by the DCC and other communication suppliers, we don’t know what their proposals for assurance will be, we don’t know what compromises will come out of the negotiations over those contracts, therefore we don’t know the full engineering reality of the roll-out of that process. On that basis, setting timescales now is simply a mistake. At the very least we need to be flexible, once those things are known, and to be willing to adjust them again if necessary.
Q176Barry Gardiner: I am glad you amended that position because isn’t it unrealistic to ask the Government to have no timescale at all? I am sure this Committee would be saying, "So the Government has an aspiration but it has no timescale for delivery." We would be criticising them for that, if they did not seem to have a clear and logical framework in which to conduct the analysis and the testing.
Dr Thomas: I agree completely, and that is not what I am saying. I am simply saying they should declare that these timescales are provisional.
Q177Barry Gardiner: The essential point of what you are saying about the importance of the engineering driving list, and the analysis and the subsequent corroborative testing then coming into play, is that all timescales should be defeasible?
Dr Thomas: Yes, timescales need to be driven by the engineering realities. A typical IT project of this complexity overruns its declared timescales by approximately 100% and its costs by about the same. You need to recognise those realities.
Q178Barry Gardiner: If one stopped at smart metering, given what you have said about cost overruns and functionality of smart metering in itself, do you think it is a cost-effective exercise if it were to stop there?
Dr Thomas: The important role of the meters is to enable the smarter grid, and a lot of work has gone into SMETS 2 to provide the functionality for the smarter grid. It is not yet absolutely clear-cannot be absolutely clear-that that functionality is sufficient, because we don’t know exactly what the smarter grid will require. For example, there may be a need for more real time data with lower latency for some of the management aspects. We simply don’t have the architecture to know that yet. If this programme were only to be providing in house displays and remote access for billing, I very much doubt that it would be a cost effective way of achieving those goals. You can already buy in house displays and clip them on to your supply wires, and the remote billing issues are probably not of sufficient benefit to the consumer to merit the cost of the whole smart metering programme but the smarter grid really matters.
Q179Barry Gardiner: Do you think that the Government would be wise to backpedal on what you see as the ultimate goal of smart networks, because that is likely to put consumers off? I don’t know if you heard the earlier panel who said that willingness to engage was very much a function of whether they thought they or the companies were going to be benefiting here, so the Government would be wise to backpedal on the whole idea of the smart network now and simply to focus on what you say is the less cost effective part of the scheme, simply in order to get it in place. Orwell told us that we have to love our oppressor, didn’t he?
Dr Thomas: I certainly don’t think that they should backpedal at all on the goal of implementing the smarter network. If they choose, for presentational reasons, to emphasise different aspects of the motivation for this programme then that is a political decision and not an engineering one.
Q180Sir Robert Smith: In your opening remarks, Mr Henney, you made quite clear your concerns about the cost benefits analysis, and you said you had a meeting with Charles Hendry. Have any of your meetings or discussions with the Department assuaged any of your concerns?
Alex Henney: No. When I was putting together the figures afterwards for a paper that I published called Smart Metering - A case study of Whitehall Incompetence and Profligacy, I discussed with the economist, whose name appears on the impact assessment, what I was trying to do. Namely, I made clear I know nothing about gas, I was focusing on electric only, so I wanted our costs for electric only metering so that I could compare them with the costs from other countries, which I had. I was told that was a useless exercise, and I said, "Why? The other countries are doing it much more cheaply," and he said, "Oh, that doesn’t matter, the benefit is so great that the cost is irrelevant." Now, that is not the way I have been brought up in the private sector.
Q181Sir Robert Smith: When you said they should abandon the whole smart metering project, did you-
Alex Henney: The way it is being done now. I have nothing against smart metering. In my view, it should either be done as a DNO roll-out or alternatively, as in New Zealand, people should be given the option to go and buy their own meter. If they want a smart meter, if they want an in-home display, go out and buy one in the shop, get it hooked in to a PLC system and away we go. This is called a market, I think, which is not something that DECC are very keen on.
Q182Sir Robert Smith: Would you share that view, Mr Thomas?
Dr Thomas: No, I wouldn’t, unless all you are interested in is the provision of the in house display information. If you want to go beyond that and to have: first, the management information needed for the smarter grid; and second, the demand-side management, through being able to take remote control of smarter devices and domestic equipment through the smart meter, then clearly you need very tight specifications on the end-to-end security of that system and on the functionality of that system. You are not going to be able to get that simply by getting people to go out and buy something labelled a "smart meter" from Maplins and hook it up to some power line communications. The architecture needs to be planned in much greater detail than that, as we have seen from the thousands of pages of specifications that have already come out, and the man years of effort that have been put in by GCHQ and CESG on ensuring that that architecture has the kind of security characteristics that you need for something that is part of the critical national infrastructure.
Alex Henney: Could I respond? I focused on the smart metering issue. As far as I am concerned, there is another debate that one can have about so-called smart grids. Last year I spent a fair amount of time running around Europe doing a study on smart grids. However, I would emphasise that that, in my view, is another story, and what is a smart grid in Orkney Islands is not the same thing as what is a smart grid in Eifel in Germany, in the national park, and is not the same thing as a smart grid in London.
Q183Sir Robert Smith: You reckon that the power line carrier communication would be a much cheaper-
Alex Henney: The Irish analysis was that the capital cost of the power line carrier was half that of the wireless. The note that I received on 3 May from Bilbao said, "With regard to OPEX, PLC smart meter costs are two orders of magnitude lower than wireless smart meters, and-"
Q184Sir Robert Smith: Isn’t that the same functionality in terms of-
Alex Henney: Sure. Sure. Assuming that the network is a suitable configuration for PLC. It is not always. One suggestion that Ross Anderson and I did make to Mr Hendry was that someone from DECC go and visit Bilbao, Iberdrola. Instead of going to visit Bilbao, they went to visit Iberdrola’s subsidiary, Central Maine Power, and Central Maine Power: first, did not use power line carrier; and second, half their costs were covered by the stimulus programme, so they did not really care about the costs because the other half were regulated and their costs in total matched DECC’s costs. So DECC did not take the offer of the opportunity of going to see a very successful roll-out. To be clear, what I would advocate is a simple roll-out by the DNOs. It took Enel eight years to roll-out 38 million; it is taking us eight years of faffing around to begin to start.
Q185Sir Robert Smith: If you are doing a roll-out by DNOs is that not slightly different from your idea of individuals just going and buying-
Alex Henney: That is the alternative. Two approaches: if you want to mandate a mass roll-out then get the DNOs to do it. Otherwise ask yourself, "Why shouldn’t we let the market provide?" Because not everyone will want them. I personally don’t want one. My consumption is only about 5,000 kilowatt hours per annum. I have better things to do.
Q186Sir Robert Smith: Better things to do in terms of?
Alex Henney: My time; with my time.
Q187Sir Robert Smith: Yes. Would you not appreciate getting accurate bills from your supplier?
Alex Henney: They true themselves up once in a while when they come and read the meter and I am not fussed. I don’t greatly mind having estimated bills.
Q188Sir Robert Smith: Have you put any thoughts into whether a DNO roll-out would be more effective than a supplier roll-out?
Dr Thomas: I believe the decision was taken because the suppliers were seen to have the customer relationships, and that was the key relationship that would be needed in order to get the acceptability and to make it work. If it would help the Committee to have a note on the issues surrounding PLC communications, then I am sure the IET would be very happy to provide that.
Alex Henney: Could I perhaps correct here? The reason it went to the suppliers is because, in 1997 or thereabouts, Ofgem or OFFER, as it then was, devised the idea of the supplier hub, which is a fiction because, unlike bananas being trucked to my local greengrocer, the wires are separate and integral with the delivery of electricity. You cannot have electricity without the wires. Having set this story up, together with competitive mass market metering, because-aren’t markets great, they are wonderful, innovative and all the rest of these storylines, which you can get if you go back and read the relevant documents-having set that system up it just rolled on, instead of someone saying in the late-
Q189Sir Robert Smith: The meters currently belong to the suppliers, don’t they?
Alex Henney: Yes. The suppliers are supposed to supply the meters. That is the theory of the supplier hub, and I think it is a most ill-advised approach for the mass market. It makes a lot of sense for big industry, and for Tesco and all the rest of them. One of the characteristics of the British Civil Service is we love uniformity, both across the country and across all types of customers. Let them all have the same.
Q190Ian Lavery: I think I have a fair idea of the answers to the questions I wish to put, but I will put them anyway. How convincing do you find that the evidence that the in-home displays, IHDs, are integral to helping customers benefit from smart meters?
Alex Henney: Not at all. If you stop and pause and you ask yourself honestly, "What proportion of the well-educated British populace cannot read, let alone count?" That is not a trivial number. "What proportion are, like me, pretty ancient and not tech savvy and are not going to get interested in this?" Then "What proportion, like me, who have thrown away one in-home display because when I tried to fix it, it didn’t work?" and a friend of mine who said, "I got this from British Gas and I threw it away." So you have to ask yourself, first, "What proportion is going to use it; and second, why can’t they use their smartphones or their computers?"
Dr Thomas: The IET would defer to the evidence that you had from your first panel this morning, the people who have actually carried out the trials and analysed them in detail.
Q191Ian Lavery: Do you think the IHDs add lots of cost to the roll-out of the smart meter?
Alex Henney: They add about £25 a pop. Multiply that by 30 million and that is £600 million. If half are thrown away then that is £300 million. However, as I said earlier, I have never seen any evidence that DECC has any concern about any money being wasted on this or other projects. Other projects are not on the table for today.
Q192Ian Lavery: You mentioned before and you probably heard the earlier panel as well discussing the possibility of using smartphones, web portals or other feedback mechanisms. Could you elaborate very briefly on that?
Alex Henney: I have a smartphone and you have a smartphone. If an app were available, then I might look at it occasionally. Again, when I have nothing better to do. I have a computer and I might look at that occasionally. Coming back to my basic point, my consumption is not that much and I am not going to turn the television on and off because of the price of electricity. I don’t mind if there is a direct system that controls my washing machine and my clothes machines. I am not going to turn my lights on and off because the price of electricity is high or low or whatever, and I am not going to do my cooking to suit some non-existent gas price.
Dr Thomas: The benefits of time shifting will undoubtedly come from smart devices, not from human intervention, simply because they can react so much faster to clip the peaks off demand. That is why the national grid has made the approach that it has, asking for the ability to turn off people’s freezers briefly, that has so excited the Sunday Mail. In order to be able to get the real benefits of shifting demand to times when the wind happens to be blowing, when there are peak demands arising elsewhere and you need to be able to get away from those peaks and so on, we do need automation in place. That is the principal reason why the smarter grid is so important. Otherwise, we are going to have to do so much work to strengthen the distribution networks. We are going to end up digging up half of Britain, which may be good for unemployment but it is certainly not going to be good for the consumers or for the countryside.
Q193Sir Robert Smith: If we could deliver the smarter grid part and those benefits, would that tip the equation in the cost benefit analysis?
Dr Thomas: Who are you asking?
Sir Robert Smith: Both of you.
Dr Thomas: The smarter grid will be delivered because it is much too important not to deliver it. One way or another we will find a way to deliver the smarter grid, I have no doubt. The Government cannot meet its climate change targets without it; it cannot meet its international commitments on carbon reduction without it. Ultimately, we won’t be able to keep the lights on without a smarter grid, because the cost of achieving those things other ways would be so much higher. So enabling the smarter grid is key and it will happen. I have no doubt about that. The smart metering programme is a key enabling step in doing that. If it were to be delayed for other reasons that would be a shame, but that will happen in due course in order to facilitate the smarter grid, beyond doubt.
Alex Henney: Could I put in a note of reservation? The meaning of a word is an explanation of what it is, how it is used, and when one talks about "smart grid" one needs to step back and say, "What does it really mean?" It means different things in different contexts. Orkney has a smart grid. That smart grid is there because there is a lot of wind, there are two cables that connect it to the mainland and people want to build more wind. There are two ways of achieving that. One is building another cable and I think they are about £30 million a pop. The other is to improve the sophistication of control of the existing network. I think that cost was about £300,000. That was the route adopted. That is nothing to do with meters. That is all to do with the medium voltage system. Likewise in Italy, where there is a lot of photovoltaic put in at an enormous rate, there is a problem with back feeding through medium voltage transformers. Again, it is absolutely nothing to do with residential customers. So I think you should look on a so-called smart grid as the application of modern technology to solve particular problems. First, we need to define the problems and then we can talk about what has become a marketing concept.
Q194Ian Lavery: I think we should get back to the IHDs. That was the initial question and then we moved on to the grid again. Can I just finish my line of questioning on the IHDs, and I will be very brief? Is there any reason why the IHDs cannot be used in conjunction with the likes of the internet, with the likes of an app on an iPhone? Could they be used in conjunction?
Alex Henney: Why do you want two?
Ian Lavery: Choice.
Alex Henney: Go buy one.
Q195Ian Lavery: So they could be used in conjunction?
Alex Henney: I have no problem with people going into a shop and buying an IHD. I have a lot of problem with £600 million worth of socialised costs, of which a significant proportion will get wasted.
Dr Thomas: The current specification essentially enables that. The information is available on the home area network, in order that the IHD can get access to it, and the market will undoubtedly generate all kinds of other devices that are driven off the information that can be provided over the home area network.
Alex Henney: The New Zealand Institute of Economic Research concluded that an in house display was not economic and noted, "IHDs are subject to damage or loss by consumers. We assume they require replacement every five years."
Q196Ian Lavery: I am not sure, are you opposed to IHDs? Sorry. The last question. Mr Henney, you suggest that it would be cheaper and more effective to let customers use their laptops and use smartphones rather than IHDs. If IHDs were not mandatory, how would consumers, without internet access or smart devices or apps, or whatever we have discussed this morning, be assured of access to the consumption data?
Alex Henney: If you want, you can start to provide IHDs to everyone over 65 or everyone on welfare. What I am saying is the mass roll-out of these things is likely to be a waste of time. There may be some who should, on social welfare grounds, have them provided but that does not justify dishing out 30 million.
Ian Lavery: I think we have your point. Thanks.
Q197Dan Byles: I want to talk a little bit about the Data and Communications Company, the central communication hub and this issue. It has been suggested to the Committee that the communications model being adopted by the UK is overly complex compared with what has been done in some other countries. I am curious to know what your thoughts are on that.
Dr Thomas: It certainly is complex; there is no doubt about that. The complexity is there for reasons of functionality because of the nature of the range of stakeholders that exist in the UK market, the way that it is currently structured. In looking at that architecture, I haven’t seen any redundant complexity although there is substantial complexity there.
Q198Dan Byles: We heard from Professor Ross Anderson, Professor of Security Engineering at University of Cambridge-and I think from you-that Britain is the only country mandating a centralised communication system feeding a centralised database.
Dr Thomas: Yes, Ross hasn’t been keeping up to date with the evolving specifications, as he will readily acknowledge. I have had this conversation with him very recently. When he wrote his seminal paper, Who controls the off switch? it was at a time when it was believed that the DCC would be holding a central database of all the data that came in, all the half hourly readings from the meters. That is not the current proposal. The current specification maintains the half hourly data in the meters distributed around the country in individual people’s homes. Only aggregated information is sent in response to requests for, for example, billing data. So there is no central database.
Q199Dan Byles: Mr Henney, would you agree with that?
Alex Henney: I am agin the central system because I think it should have been done by the DNOs, of whom there are 14.
Q200Dan Byles: Given where we are now with the way the-
Alex Henney: I would stop and get the DNOs to do it. One can see a vast mess coming down the tracks. Is that why it has been delayed until after the election? I don’t know.
Q201Dan Byles: In terms of the home area networks, do you have any data security concerns about having home area networks integrated into the smart meter?
Dr Thomas: The way in which the security architecture has been designed is to use end-to-end encryption to control the security aspects. There are some complexities that arise because of the inclusion of gas meters that, because they don’t have power supplies, have to be driven off batteries. The batteries obviously have limited capacity, and that reduces the amount of processing power that you can put into the gas meters because it uses too much power from the batteries. That then reduces the amount of heavyweight encryption-public key encryption-that you could in fact implement in the gas meters. Therefore the SMETS 2 specification has a mirror for the gas meter data in the electricity meter, so a gas proxy that contains that information, which is then held in a place that can handle the higher computation intensive security that is needed in order to provide the end-to-end security.
There is a less strong but nevertheless adequate level of security for communication across the HAN, using the ZigBee specification and the security enhancements that have been made to that. We don’t have major concerns with the security around the HAN, other than the general fear that because the way in which the specifications will be written, and the nature of the way in which they are implemented, it is not going to be possible to provide very high confidence that they have been implemented in a secure way. So there may be security vulnerabilities that will show up later, which won’t have been detected during the testing process. That is back to my original point that we need to use mathematically rigorous methods to specify and, ideally, to design the software that is used to implement these specifications.
Q202Dan Byles: Mr Henney, do you have a view on that?
Alex Henney: No, I don’t.
Q203Dan Byles: A final question, which is about DECC’s communication strategy and how confident you are that they are on course to achieve 97.5% coverage that they decided they require.
Dr Thomas: The analyses that have been done by the communications providers seem to suggest that they can get that level of coverage for the communication systems. I understand that that data is available; perhaps it has already been given in evidence to the Committee. We have some concerns about other aspects of coverage. There is a wide range of users. It is not clear that it is going to be easy to get access to all premises. It is not clear that the location of meters in all premises is going to make it easy to install the equipment appropriately, and get the right level of communication with access to the meters where they need to be. Given those range of risks, a 2.5% failure rate feels optimistic. However, we simply don’t have the detailed data that would be required to be able to assess what a reasonable assessment of that risk would be.
Q204Dan Byles: Mr Henney, any thoughts on that?
Alex Henney: No.
Dan Byles: Thank you, Chairman.
Chair: Thank you. Thank you very much for coming in. It has been very helpful and interesting.