As the Business, Energy and Industrial Strategy Select Committee meets on 9 January 2019 to discuss the management of the smart meter rollout programme, we publish our second commentary on smart meters. Read our first post, ‘Why the UK’s smart meter rollout needs to be smarter’, here.

Much of the messaging of Smart Energy GB – the body tasked with informing the British public about smart meters and encouraging their adoption – has focused on the household-level financial savings that smart meters enable. The Government has largely followed suit, with the Energy and Clean Growth Minister, Claire Perry, recently stating, “By 2030, this tiny seemingly inconsequential device, tucked away in a corner of your kitchen, will wipe £1.2 billion per year off people’s bills – an average annual saving of £47 per household.”

But how realistic is this predicted saving? Given that it’s an average, to what extent could some households make greater savings – and some save less? And how robust is the evidence behind this figure?

A hard sell

Even if households do ultimately gain from getting a smart meter (see our twin commentary on why this is not a given), most of the benefits take place in the future – albeit probably at an increasing rate in the short term – whereas the costs are more immediate. This disconnect poses a major barrier to the success of the Smart Meter Implementation Programme (SMIP) and has parallels with convincing people on the need to address climate change: as with the SMIP, the upfront costs combined with uncertain future benefits render climate legislation a tough proposition to sell (case in point: the gilets jaunes protests in France). Based on the Government’s own modelling, the net savings per household rise from £11 a year in 2020 to £47 a year in 2030.

So the Government has a tricky pitch with a complicated message on its hands: uncertain costs soon, uncertain benefits at some point in the next, say, five to 15 years (or even longer) – and all in the name of a futuristic energy system related to ‘the smart grid’, a concept with which much of the public will be less than familiar.

Perhaps in a bid to provide a clear and simple message to lure householders on board, the focus of engagement efforts by the Smart Energy GB campaign has been on projected private savings of £50 a year, with a TV advertisement urging consumers to “get ‘Gaz and Leccy’ under control”.

The devil may be in the details

International and domestic evidence informed the Government’s central estimates of households’ saving 2.8% on their annual electricity bill and 2% on gas by switching to smart metering. It is also assumed that these savings will persist over time but in fact there is limited evidence for this. In its recent report on the smart meter rollout, the National Audit Office (NAO) suggests that if the savings do not persist, the value-for-money case could become marginal or even negative.

The international evidence – consisting of a number of review papers (e.g. from the United States, and pilots in the US, UK and Ireland) and large-scale trials in several European countries – indicates savings in the range of 0–19%. The UK Government’s numbers look comparatively conservative. However, these reviewed trials use a range of methodologies – some more robust than others – and were conducted in contexts in which energy markets and households differ, so it may be wise to err on the side of caution.

The UK evidence was in large part provided by results from the Energy Demand Research Project, run by four of the ‘Big 6’ energy suppliers and based on trials with about 60,000 UK households between 2007 and 2010. Again, a range of methodologies was employed. Resultant savings varied from 0–13% but were largely in the 2–4% range.

Based on these results, do the Government’s central estimates look to be in the ballpark? The honest answer is: maybe – but maybe not.

Outstanding issues

1. A smart meter alone will not deliver savings

The meter is purely a two-way information device. As with the costs and benefits of the SMIP, that information needs to be effectively communicated (see our first post).

Key to the reviewed trials’ results was the finding that savings were much greater when accompanied by information such as energy saving advice or data on real-time energy consumption, sometimes in the form of a wall display. None of the studies disentangled the effect of meter provision alone. The Government’s Smart Metering Early Learning Project report of 2015 states that it is “realistic to expect durable energy savings of 3 per cent provided engagement is effective [emphasis ours], and larger savings are feasible in the future.”

The recent NAO report states that although energy suppliers have an obligation to provide advice at the point of installation, large suppliers are still failing to do so for 27% of customers. In fact, survey evidence suggests that of the 6.8 million households with smart meters, about 2.1 million do not recall being provided with advice on how to save energy at the point of installation. This lack of installation protocol could significantly attenuate the Government’s savings estimates.

2. The Government’s estimates are based on trials that are methodologically imperfect

Assumptions have been based on evidence from trials that (i) often lack proper control groups, (ii) are likely to inflate findings due to participants’ knowledge that their energy use is being monitored by their supplier, and – last but not least – (iii) include only households who had opted in to participate.

While (i) and (ii) call into question the savings estimates from these studies, even among the self-selected households in their samples, (iii) suggests that the findings – even if they could be trusted – likely will not generalise to the entire UK population targeted by the SMIP. For instance, recent evidence from California shows that households who chose to opt in to a dynamic pricing electricity plan behaved very differently in terms of their electricity use to those who were put on the same plan by default.

In other words, even if all households were to accept a smart meter, whether a household made savings would depend on its responses to the information the smart meter made available. Responses would differ between someone who does not understand the internet or the energy grid and someone more used to seeking out information, possibly from a somewhat informed position – such as readers of this commentary.

Savings will also depend on whether new technologies or tariffs accompany smart meters to enable households to take advantage of this information in a more sophisticated way (a prospect that has become more likely with the recent EU court ruling that will encourage innovation in this space).

In short, the risk is that more engaged (and perhaps wealthier) customers will take advantage of the benefits, while the remainder of the population will bear the costs. Energy companies will recoup the costs of the installation by charging all customers, whether they have a smart meter or not. Additionally, without a critical mass of ‘smart’ customers, suppliers and innovators may not be motivated to capitalise on the latent benefits of a smarter energy system.

So is a smart meter recommended or not?

Smart meters may save consumers money, but the likelihood and extent of that saving will depend on their response to feedback from the meter, and how this information is conveyed (for example, through an in-home-display or app, or by letter or web interface). That said, smart meters are a key enabling technology in developing a more flexible energy system, which will help the system operate more efficiently and allow for increased renewable energy integration, ultimately reducing carbon emissions (and potentially significantly so).

In an increasingly digital and ‘smart’ world, energy systems are still surprisingly unsophisticated, relying on 100-year-old technology. This dated infrastructure stifles the technological and service innovation that can create beneficiaries on both the supply and demand sides of the energy sector.

Perhaps the real benefit of a smart energy system is information, which can become available to households, firms, energy suppliers, aggregators, innovators, and system operators. If put to good use, such a system could deliver a more vibrant and competitive energy market, reduce the need to invest in additional generation and network capacity, and deliver gains to all parties.

Of course, adoption of smart metering by individual households would facilitate such a system. Preferably one of the new SMETS 2 meters, however, to bypass the technical issues presented by the original technology!


In 2019 Greer Gosnell and Daire McCoy will be conducting work that aims to understand the relative importance of various barriers to and drivers of smart meter adoption in the UK. This research is funded by Horizon 2020 ENABLE.EU and the ESRC Centre for Climate Change Economics and Policy (CCCEP).

 The views in this commentary are those of the authors and not necessarily those of the Grantham Research Institute.

Keep in touch with the Grantham Research Institute at LSE
Sign up to our newsletters and get the latest analysis, research, commentary and details of upcoming events.