Not so smart – what has gone wrong with the smart meter programme and how to fix it
24th March 2017
Like the Green Deal, the smart meter programme started out as a good idea, but has been badly implemented. Both were announced with lots of fanfare, with hard targets and overhyped claims about the benefits. The Green Deal was going to create to 250,000 jobs and be carried out on a street-by-street basis. In the case of smart meters, £11 billion costs are supposed to produce £18 billion benefits, and be paid for by the lucky customers who are assumed to get all these benefits.
Customers were going to have energy efficiency houses, fitted with smart meters, and they would be able to shop around for the most competitive prices. Together with the heavily subsidised wind farms and solar panels, as if by magic these various policies were together going to result in lower prices to consumers. They are all products of the fantasies of the Miliband-Huhne-Davey energy and climate policies, and both the Green Deal and the smart meter programme have left a mess for their successors to deal with.
It has not turned out as planned. The energy suppliers are not on course to complete the smart meter roll out by 2020, there are problems with the meters, and consumers are not behaving in the way politicians assumed they would. They are not switching and the prices are going sharply up as wholesale costs are coming down. There is enough evidence of failure already. Now is the time to take stock and reconsider.
The fundamental error
Obsessed with the idea of customers switching suppliers and with minimising the role of distribution monopoly networks, it was decided that smart meters would be the responsibility of supply companies, not distributors. This was a fundamental mistake, which no other major European country made. Meters had always been in distribution, for the very good reason that they are an essential part of the network system.
This fundamental error has had consequences that now haunt the smart meter programme. It meant that meters were no longer part of the regulated asset base (RAB). They could and indeed may well be stranded. The result is that the cost of capital is higher. Customers are paying and will go on paying the difference.
But perhaps more serious still is the mind-set that goes with this supply location. Suppliers are in the business of selling electricity. Distributors are in the business of optimising networks. It is therefore not surprising that the focus has been on the direct claimed benefits to the customers, not those to the system. This turns out to be exactly the wrong way round.
Miliband, Huhne and Davey seemed to believe that if only they could empower the consumers with a smart meter, they would then be able to manage their demand. They could turn off appliances at the peaks, and they could choose tariffs and suppliers to better match their needs. It was an illusion that the Competition and Markets Authority (CMA) fell for too, in its much-criticised inquiry.
Their smart and active customer is an ideal type – the concerned educated citizen, striving to do their bit for climate change, and with the IT and statistics knowledge and expertise to navigate the complex offering of the competing suppliers. No doubt there are such customers, but they are not typical. Minsters and the CMA too often tend to think that customers are like themselves, rather than “ordinary working people” who are “just about managing”.
Most people are not like ministers. They do not want to spend their time checking data from their meters, and they do not want to shop around on the Internet for deals. Lots of people don’t even understand what a percentage is, and their daily lives and understanding is far removed from the more highly educated and highly engaged people who dream up these policies. It shows: even with £200-£300 savings apparently available from switching suppliers with their old meters, 70% of customers did not switch
Though some may, in the fantasy world, be deemed too old, too stupid or too poor to switch, the reality is rather different. People just want a reliable electricity supply at a reasonable price and want to spend as little time as possible in getting this outcome. They don’t actually want to be “engaged” customers. They want the job done for them.
Coincidentally, and for all the hype, that is what the suppliers want too. They may proclaim their addiction to competition and switching, but the Big 6 make their money from the non-switchers – those on the Standard Variable Tariff (STV).
With this happy coincidence of interests between the suppliers and the SVT customers, a central political aim of the smart meter programme, and one encouraged by the CMA, is seriously undermined by putting the meters in supply rather than distribution. The incumbent fitting the meter does not want the customer to switch. Indeed, it would be better for the supplier if the customers were locked in. There is nothing “neutral” about the incentives of the supplier – and neutrality is exactly what distributors bring. Indeed the reason distribution was spilt from supply was precisely so the networks would be neutral between competing generators and suppliers.
No wonder the most enthusiastic early supporters included those who wanted to sell on other household energy services to keep the customers on board. It is also not surprising that early smart meters have proved less than smart once customers switch. Incentives matter and the incumbent supplier is not and cannot be neutral about the competitive possibilities of smart meters.
The system value of smart meters
The mistake of putting suppliers in the driving seat led the Miliband-Huhne-Davey framework to downplay the system benefits of smart meters. Yet this is precisely where most of the potential value lies. It has yet to be reaped.
The value of the information about demand in real time is in enabling the system operators – at the national and the regional levels – to better match supply and demand. It is part of a wider move towards digitalising the electricity industry.
Digitalisation offers considerable gains in efficiency to energy systems, and it is especially valuable when there is lots of small scale intermittent wind and solar power, the need for rapid response back up, new storage options and a very tight capacity margin.
Digitalisation is taking place within the context of more and more zero marginal cost electricity generation. A digital energy system is one in which the system costs are increasingly what matters, and the wholesale market is slowly withering away. (This is explained more in my new book, Burn Out – the endgame for fossil fuels).
The declining importance of wholesale markets is a long way from what was envisaged by Miliband-Huhne-Davy. They saw the wholesale price of electricity rising, on the back of ever-higher fossil fuel prices. With their assumed high and volatile wholesale prices, the meter would reflect this, and induce responses. What never seems to have occurred to them is that the wholesale price would fall, and behind it the fossil fuel process would fall too.
The implication is radical: there is less and less economic value in switching for customers. The costs that matter are the fixed system costs, and therefore the value of the smart meter information is less and less at the customer end, and more and more at the system end. It seems to have escaped these politicians’ notice that the feed-in-tariffs, the capacity payments, the levies and the network costs are where the economic action is increasingly located. This is more the economics of broadband, not wholesale markets.
Energy specific meters versus the broadband hubs
Putting the meters in supply was a fundamental mistake which has dogged the programme ever since. But it is sadly not the only one. The second flaw is in designing an electricity-only programme.
What the government had in mind was that energy supply companies would visit every single house and all other premises, and fit the meters. This would be entirely separate from the role out of (smartish) meters for water, and broadband as a USO service.
This sees the smart meter as a distinct and separate digital interface. But this is a mistake: households are provided with multiple digital services, and the broadband hub is the key central organising point. Instead of thinking of smart meters as a separate activity, they could have been seen as part of the broadband networks, and within the broadband umbrella, multi-metering data could be collected and utilised. This might also have had the effect of pushing down the high margins suppliers were charging.
The single service digital approach was reinforced by a single and separate data controller – the Data Communications Company (DCC). All the meters have to go through this centralised organisation, and meet its technological requirements. It is a monopoly, assigned through a competitively tendered contract. But why? Here we come back to the mistake about supply versus distribution. Since the data is crucial to system operation, and since this is a system network activity, why is a special DCC needed at all?
The single DCC puts the choice of digital communications in the hands of a single company. In some circumstances this may be a good idea – circumstances where there are few technological choices to be made and the technology itself is stable. Smart meters are not one of these.
Digitalisation is a process surrounded by very rapid technical change, and competing technologies. At the more straightforward end, there are different communications technologies – fixed, mobile, radio and so on. The DCC has made commitments that create sunk costs, which could be stranded. Innovation is not necessarily in its interests. It is a bit like BT’s commitment to copper wires: having made the investment, the business interest is in maximising the returns and holding off standing as long as possible. It is an example of the legacy problem in infrastructure.
But these communication commitments are just the tip of the rapidly changing technological scene. Digitalisation is much more radical. There are robots, and 3D printing and Artificial Intelligence (AI) to contend with. Then there is the move to quantum computing. This is the territory within which delivery date is set.
Some go further and question whether in the new emerging technological context, smart meters will be needed at all. Measuring the molecules of electricity may not need the smart meters currently being installed. It may be that what is possible is energy management systems within which this information is monitored and utilised. It may be that measurement can be done down the wires in much more refined and simpler ways.
We may not know where this will all end up, but we do know that this is an area of incredibly rapid technological change. An £11 billon political programme, to a fixed date, located in supply not distribution, with a single DCC separate from the system operator is not the obvious answer.
Security, cyber crime and privacy
Designing any digital information system raises serious questions about the use of the data and the data communications systems. An obvious question in the smart meter case is the extent to which the data is secure – secure to the system, secure to the supplier, and secure to the customer.
This question seems not to have been very high in the list of concerns when the Miliband-Huhne-Davey policy was been put together. This is understandable in so far as the scale of cyber crime, hacking and on-line fraud was poorly understood at the time. Yet there are several general rules that could have been considered. A single centralised system is typically more vulnerable to attack than a serious of systems, even if this builds in some redundancy. The security of a system depends upon what is connected to it. Smart meters will eventually help to connect millions of appliances to the electricity networks, which are now distinct and separate.
Then there is the security of the multiple suppliers who have access to the data, as opposed to the security-dependent networks. Can all the suppliers guarantee the robustness of their systems to hacking? The recent examples of Yahoo and TalkTalk, both significant businesses, is salutary when compared to the numerous small energy suppliers the government is trying to entice into the market.
Even if the smart meter programme had been designed to take these risks into account, there remains the tricky issue of customer privacy. It has been suggested that with the real time information in electricity use, and with increasingly smart connected system to each appliance, the data will reveal a huge amount of information about the personal habits and life styles of the customers. Many people will understandably be concerned about the sharing such details with their energy supplier or anyone else.
There are other ways this can be dealt with. The first is that the data remains owned by the customer. This looks like an obvious solution, but it is simplistic. What exactly does it mean to “own your data”? In practice it means regulatory rules on its use. Does anyone believe that there is sufficient oversight to ensure that this is properly policed?
If politicians want to embed grand plans with hard deadlines, then in willing the ends they need to will the means. The obvious way to do this is to make it compulsory to change meters. Customers would then have to enable the suppliers to get the work done if they wanted to be supplied after 2020.
In a compulsory metring world, the suppliers would have a plan to work through the list of their customers. They would be able to bunch these and add them in a sequential and organised way. This would be much more efficient. Instead they now have to offer the smart meters and see if people take them up. If they don’t, nothing happens, other than more and more costs in trying to persuade them through propaganda.
There is a precedent for this. When Britons switched to natural gas as North Sea natural gas became widely available, British Gas carried out a national programme of conversion of systems and appliances, in a very safety critical context. It was a great success, done in a timely and cost efficient way. It was effectively compulsory.
The efficiency of a compulsory programme would be even greater if it was coordinated by distribution and not supply. Why? Because distributors have all the customers in a specific geographical area, whereas suppliers do not have complete coverage, and some have very widely distributed customer bases. Furthermore, suppliers’ customer bases change with switching. So a customer of a supplier might be targeted to be persuaded to have a smart meter, only to then switch to another supplier.
A better way forward
Though a great deal of money and time has been wasted on this very inefficiently designed policy, there is still time to change tack to get a better outcome. This was accepted in the case of the Green Deal, and it should be accepted for smart meters.
The necessary steps all follow from addressing the first fundamental flaw of the mistake of putting smart meters in supply and not distribution. This should be reversed, and quickly.
The response from many of the vested interests built up around the current programme is that this would take time, cause disruption, and in any event it might require legalisation. All of this is true. But what is the alternative? To condemn customers to pay for the current inefficient system, expose them to cyber security risk, and to limit the benefits to the system.
Suppose however that the legislative hurdles are serious. The government is obsessed with BREXIT, and it has little appetite for almost any other legislation, overwhelmed as it is by multiple BREXIT-related demands. Could the switch to distribution nevertheless be achieved?
The answer is a qualified yes. Suppose all the parties agree to act as if smart meters are in distribution – now. The suppliers would contract out the programme obligations to the distributors, and OFGEM would allow the distributors to include the smart meters in their CAPEX programmes and then in the RABs. This could be done in the 2021 periodic reviews coming up. Between now and then the costs could be logged up.
Once the distributors are in charge, howbeit initially in a delegated and contracted way, the system operators in the networks can interact with the DCC. Eventually they could take this function over. In a decentralised system operator world to which for other reasons we may be moving outwards, this would then lead to a disaggregation of the single centralised model and the DCC could gradually move over across to the distributors.
With the most important flaw addressed, we can then move to address the technological issues. To do this sensibly the 2020 target could be “flexed”. 2020 is arbitrary and is not going to be met anyway. So rather than double its bets and try to achieve the practically impossible, with all the cost that go with this, recognising reality could lead to a rethink and review of the technological commitments. This would include the cyber security issues, the privacy and data protection issues, and the interaction with the broadband hubs and other metering programme like water. Smart meters should be a cross utility activity, not an energy-only one.
It remains to give the power to the contracted distributors to compel customers to install smart meters as a condition of supply. This could either be achieved by having two distinct tariff structures that massively incentivise people to go for smart meters, or it could be simply a legal requirement.
When it comes to legislation, smart meters are but one part of a more comprehensive rethink, taking into account what a broadband-type, increasingly zero marginal cost system looks like. In doing so, the government should think very hard about what exactly is the question of to which smart meters are supposed to be part of the answer.