Earlier this week I had the pleasure of speaking at the Energy Industries Club dinner on the subject of energy security…here is a copy of my remarks…

Good evening everyone, I’m delighted to be here for the inaugural evening meeting of the Energy Industries Club.

I feel under some pressure with a dinner-time speech to be entertaining. I was scarred a few years ago at a dinner where an EU Commissioner gave a speech between the starter and main course. He wasn’t brief. By the time the charred remnants of the main course were served even the most die-hard Remainers were dreaming of Brexit!

So we successfully avoided that pitfall, but I still feel the pressure. I might be tempted to throw in the odd joke or witty one-liner. I’ll do my best…

As you may know, following the recent Budget, we have a new approach to energy policy: Gaslighting.

It’s kind of you to laugh but that’s not actually a joke. That really is how I feel about large parts of our energy policy at the moment.

It goes something like this: we’re going to have cheap, reliable renewable energy based on wind and solar but we’ll need subsidies to get it going. What’s that? The weather isn’t reliable? Good point, OK right, so we’ll need subsidies for non-renewable energy to come on when it’s not windy and sunny. OK then.

Not OK? Something about grid infrastructure? Oh yes, I guess we don’t have a much of that in the sea. Good point, we’ll have to build some more. But we can delay some of it to keep costs down for consumers. Sorry what? Curtailment? We have to pay wind farms if we can’t use the electricity they want to generate. Oh, that’s annoying. I’m sure it’s still cheap though.

OK now what? Balancing? It’s more expensive to balance the grid when generation varies with the weather.

OK I don’t care. Renewables are cheap and reliable. End of.

Does anyone else feel like they’re being gas-lit?

electricity bills breakdown august 2021

By the summer of 2021, just before the start of the energy crisis, green levies accounted for 25% of end-user electricity bills, with network costs being 20%. Prices had been rising steadily for two decades at a time when wholesale prices had been relatively benign, and these increases absorbed the capacity of consumers to afford price rises. When wholesale prices abruptly stopped being benign in autumn 2021, consumers very quickly began to feel the pain.

bills and wholesale charts

And on top of all of that, the Government has come to a late realisation that we will probably NEVER be able to stop subsidising wind. The original hypothesis was that subsidies were required to support an immature industry but once the technology matured, subsidies could be phased out. Now, if you read REMA, you can see an explicit recognition of the problem of how to recover high capital costs when operating costs and therefore income if that is determined by short run marginal operating costs, are low.

This reminds me of another joke that has been doing the rounds which is rather en point: My energy supplier claims to sell 100% renewable energy. It just updated my direct debit – can someone please tell me when it was that wind and sunshine doubled in price?

In our enthusiasm for renewables, we have rather neglected security of supply. Ironically it is the gas crisis that woke everyone up to this, but we’re actually experiencing two simultaneous security of supply crises: the gas crisis which is well understood and has a recognised solution (ie more gas production outside Russia) and an electricity supply crisis which is only starting to be understood. We have rushed to deploy intermittent renewables, allowing conventional generation to close, assuming that interconnectors will secure our supplies though a diversification effect.

Unfortunately we’re realising late in the day that firstly our weather is more correlated than we had hoped so if the connected country also relies on wind power it might be trying to import at the same time we are, and secondly, there are really only two countries in our geographic region with materially different energy mixes: France which relies on an aging and increasingly unreliable nuclear fleet, and Norway which relies on hydro but has almost no pumping capability, so  its resources can literally run out, as nearly happened last year when reservoir levels reached 20 year lows.

So I’ll just mention here that I like to be efficient. When preparing for this evening I went back to various talks I’ve done in the past year to see what I could re-use, and here’s something which came up a lot. I said this:

“I believe it is entirely possible that Norway will impose restrictions on electricity exports.”

In fact, Norway is now doing exactly that. It is at the forefront of explicitly recognising the need to put its own citizens first in the scramble for scarce resources – and plans to legislate this year to reduce electricity exports if necessary to protect domestic energy security. It has also recently restricted export capacity on NSL in response to the fact that Britain had never implemented full export capacity on our side. This has cut the import capacity from 1.4 to 1.1 GW.

So what do we really mean by security of supply? In developed nations we expect that when we press a light switch, the light will come on, and this will be true regardless of the time of day, how many other people are also pressing light switches, or if there are any disruptions to grid infrastructure or generating assets.

Energy security in the electricity market implies an excess of generation capacity over demand. In Britain, electricity demand ranges from about 20 GW on a summer’s night to around 56 GW on a cold winter’s day.

GB demand variation

This is a wide margin and it’s set to get wider as we progress along the road towards net zero. The expansion of rooftop solar, which acts like negative demand on the grid, will depress summer demand, while the electrification or partial electrification of heating and transport will increase demand.

We could see some unexpected effects that might see summer demand rise as well, despite the increase in solar PV and that is because heat pumps can be used to provide cooling in summer. As there is currently very little domestic air conditioning in use, this could boost summer demand as people use their heating systems to deliver cooling in summer because they can, irrespective of any increases in summer temperatures.

However, we can probably model most of these effects pretty well. The industry has been managing high levels of demand variability for as long as electricity grids have existed.

But the energy transition is also adding unpredictability to the supply side of the equation. Our energy transition is being built on the basis of intermittent renewable generation, that is weather-dependent sources of electricity.

It is obvious that this leads to much less easily predictable variation – weather forecasting is famously difficult to get right. And it creates two distinct sets of problems – energy gaps which are obvious…when it isn’t windy then you can’t generate electricity with wind turbines, and when it isn’t sunny solar panels won’t produce. That includes at night, something people seem to need to be reminded of surprisingly often.

The second problem is more subtle but no less important and this comes down to the basic physics of our electricity grids. Because Tesla won the Battle of the Currents against Edison, we use alternating rather than direct current.

That Battle was actually quite dirty with Edison and his cronies going to extreme lengths including the public electrocution of horses to demonstrate the dangers of ac, and they connived to ensure that the first electric chairs ran on alternating current. That’s not a joke by the way, that actually happened.

But the ease with which electricity can be moved over long distances with minimal losses using ac rather than dc was the decisive factor and from the 1880s onwards, electricity grids were built using ac.

In the early years, there was no consistency over voltage and frequency levels. In 1920 just 6% of British homes had electricity, and those that did had widely varying services – in London alone there were 24 voltages and 10 different frequencies. In 1921, there were more than 480 authorised suppliers of electricity operating in the UK, generating and supplying electricity.

Nationalisation was seen even by the Conservative government of the time as the solution, and The Electricity Act of 1926 created a central authority to promote a national transmission system.

This grid had a voltage of 132 kV and was the largest peacetime infrastructure project the country had ever seen. It was largely completed by the mid-1930s. 4,000 miles of cable was installed by 100,000 men and quite a lot of horses. The first pylon was erected near Edinburgh in 1928.

The original plan was for the country to be separated into a number of independent regional grids. Each could be connected to a neighbouring grid if required, but typically they were operated separately. The Second World War prompted a change of plan, and with the construction of a bomb-proof national grid control centre in London it was determined that a single national grid would be better. In late 1938, and with some trepidation, the inter-regional isolating switches were closed and the regional grids were all connected. As nothing untoward happened a decision was made in spring 1939 to keep them closed, except in the case of an emergency.

Now I’m coming to the part of the evening everyone looks forward to…the moment the speaker decides to throw in some GCSE physics. I can see you all sitting up in anticipation!

So in 1938 we had a single national high voltage transmission system to which lower voltage regional grids connected, all running on alternating current, with a frequency of oscillation of 50 Hz. That is there are 50 cycles per second.

ac generator

These oscillations are literally created by the rotation of the turbines in power stations. Alternating current is produced when one magnet rotates inside the magnetic field of another magnet. The current and voltage vary in a sine wave pattern whose frequency is determined by the speed of that rotation. To get to 50 cycles per second, these turbines all turn at 3,000 RPM. In a 60 Hz system they rotate at 3,600 RPM.

If supply and demand go out of balance, the frequency will change

  • If the demand exceeds supply, the system frequency will fall
  • If the supply exceeds demand, the system frequency will rise


Electricity grids are sensitive to changes in frequency, and typically deviations of more than around 1% will trigger load shedding – this was seen in GB in summer 2019, when a lightning strike caused two power stations and a large amount of embedded generation to trip.

Demand therefore exceeded supply and the system frequency fell outside its stable tolerance levels, requiring selected load shedding in order to avoid an uncontrolled cascading grid failure. The impact of this is so serious that National Grid ESO has a duty to maintain system frequency within a 1% band of 50 Hz as part of its licence conditions.

In a conventional electricity network, generators don’t just establish a nice alternating current waveform, they are large heavy objects that resist changes to their speed hence providing an important support to the system known as inertia. They act as a brake on changes in system frequency.

However, with the energy transition, not only are we replacing generators which naturally produce alternating current with renewable generation which produces direct current that needs to be electronically converted to ac, we are also reducing the amount of inertia on the grid. As a result the grid is becoming a lot more flighty and controlling grid frequency is becoming harder and more expensive.

System operators must not only fill in the energy gaps from intermittent wind and clouds, they must make sure that this intermittency does not result in frequency deviations greater than operational tolerances. At all times. At all points on the grid.

The energy challenge and the frequency challenge are managed using different tools – energy is more of a macro problem, so alternative sources of generation are needed when wind or solar output is expected to be low.

The purpose of the Capacity Market is to ensure that this generation is available – conventional generators do still operate. In fact our grid is still dominated by gas-fired generation, but the more renewables we have with near zero short run marginal operating costs, the lower the utilisation of these assets becomes. If utilisation falls too much they become uneconomic to run and close, making them unavailable in times of low wind. So to keep them open, they need to receive alternative income sources, which is where the Capacity Market comes in. The Capacity Market is dominated by gas plant and a brand new CCGT won a 15-year new-build contract in last month’s T-4 auction.

The frequency challenge is more of a micro problem. It requires fast response to react to those clouds and gusts of wind, and these services are delivered through the balancing and ancillary services markets operated by National Grid ESO.

Last July, National Grid ESO published a preliminary Winter Outlook for this winter, only the second time it had previewed the Outlook ahead of time.  

Predictably, the press jumped on the “increased risk of black-outs” story, not least because of a comment in ESO’s press release to the effect that it was confident “there’ll be enough electricity to keep Britain’s lights on,” literally raising the prospect of blackouts. To stave off this risk, ESO introduced two new measures to support security of supply this winter: the Coal Contingency and the Demand Flexibility Service.

Unfortunately the coal contingency looks to be a one-off… Of the five units in the scheme, two at West Burton A are closing due to extreme old age, one at Ratcliffe is returning to normal commercial operations, and two at Drax could stay open but at the moment the company says they won’t. Whether or not that’s a negotiating tactic remains to be seen – the biomass subsidies Drax receives are under threat so it may be thinking strategically.

Unfortunately by law even Ratcliffe has to close by the end of September 2024, although it is to be hoped that there will be a reprieve. One coal power station is not going to make the difference between climate change or not, but it could make a serious difference to security of supply on low wind days. I’d like it to stay open until, like West Burton A, it simply falls to bits.

I would also like to see the coal units at Drax stay open and the biomass units convert back to coal. It would certainly create less air pollution around the plant, as well as avoiding the ridiculous supply chain emissions in drying, pelletising and shipping all that wood. If people want to plant trees to offset the emissions, there’s nothing to stop them doing that anyway…it’s not some fundamental component of biomass supply that cannot be replicated with other fuels.

We had some small relief with the recent announcement that EDF plans to extend the lives of the Hartelpool and Heysham 1 reactors for 2 years until March 2026 but with new delays to Hinkey Point C, now not expected until September 2028, in the summer of 2028 we will only have 1 nuclear power station running (assuming it’s not offline for maintenance)…Sizewell B.

GB nuclear assets

Despite the Government’s apparent new commitment to nuclear power, it is too little too late. Doubling down on the abysmal failure that is the European Pressurised Water reactor is not a viable solution, nor is hoping that small modular reactors will arrive on time at the end of this decade.

This could still be improved if policymakers had courage and determination: Advanced Boiling Water Reactors were built on time and on budget in Japan before Fukushima in under 5 years. We should just build some of those. Straight away. And the Government should simply pay for them – we don’t expect the private sector to finance physical security in the form of the army or police…there is surely a role for the Government to play in the arena of energy security. It doesn’t need to supply all of it, just a slug of high capital cost baseload that the market otherwise struggles to deliver.

And we should leverage the regulatory frameworks in countries such as Japan, Taiwan, South Korea and the US all of which have reliable nuclear regulators. Re-authorising technologies that have been authorised elsewhere by such reliable regulators is a major waste of time and resources, neither of which we can really afford.

But back to this winter, the other new innovation in the security of supply arena was the Demand Flexibility Service or DFS. There’s an old joke: what did communists use before candles….the answer being “electricity” but that feels a bit close to the bone these days when we hear stories of people sitting in the cold and the dark to save a few pence under this scheme.

Like most industry insiders, I thought people would avoid cooking their dinner or doing laundry during the DFS periods. I really didn’t think they would go to the lengths some of them have, sitting in their cars, or going to bed with everything they could possibly turn off, tuned off. People have had to be warned against turning off their fridges and freezers to avoid giving themselves food poisoning.

Some people have earned very little under the scheme since they were already reducing demand in response to high prices. Others figured out they could game the system by boosting demand during the re-calibration period so even normal consumption during the DFS would look like a reduction and earn them money.

The scheme is likely to continue, but it needs to be improved.

First of all, the day-ahead activation is inefficient. National Grid doesn’t necessarily know the day before if it’s going to be needed – the two times it was activated this winter looked pretty un-necessary when you look into the actual supply and demand balance on those days. In particular there had been no price signals in the forward markets to suggest it was needed, and little response within day in either the Balancing Mechanism or cashout price, suggesting the market actually wasn’t that tight.

Secondly, only about 45% of households are equipped with smart meters functioning in smart mode, so even if 100% of suppliers participated in the scheme – which was not the case this year – fewer than half of households could take part. Suppliers are struggling to meet smart meter rollout plans and consumers requesting them so they can sign up for the DFS are facing installation delays of many months. The 1.7 million household with smart meters that do not work in smart mode must be particularly aggrieved since through no fault of their own they are excluded.

There is actually a risk that more people will get pulled into this category since more than half of households live in the central and southern regions and have devices that rely on 2G and 3G mobile networks which are going to be switched off.

Thirdly, we need to address the issue of consumer welfare. Ofgem already put some thought into this during its network charging reviews but the introduction of the DFS and the evidence of harmful behaviours by some consumers suggests this work should be accelerated. It would also be a good moment to ensure we are all educated about the limits of demand-side response in the domestic market, and stop recommending people do their laundry at night, something fire brigades across the country explicitly warn against. And the people most in harms way are the fuel poor, who are overwhelmingly more likely to own older, cheaper and riskier appliances. As an aside, we should also look at standards to ensure isolation switches for solar panels and car chargers are located with fire safety in mind.

Of course, the high cost of energy has stimulated a strong consumer response to the DFS. Consumers are also receiving support with their bills from the Government: we currently have the bizarre situation that not only are most forms of electricity generation subsidised, so is most consumption. All we need now is to lob a few subsidies in the direction of network operators and we’ll have a full house!

Many people have been inclined to pipe up that the way to reduce costs is the have more renewable generation, since it has a near zero marginal cost. But as I outlined earlier, having near-zero marginal cost of generation does not translate into lower costs to consumers, or at least, it hasn’t so far in 20 years of trying.

People point to the falling capex and opex costs of renewable generation to indicate that these cost benefits will materialise at some unspecified point in the future.

Unfortunately, analysis of the actual capital and operating costs of off-shore wind projects in the UK fails to support this theory either. Professor Gordon Hughes at the University of Edinburgh has studied the audited accounts of UK windfarms – most off-shore wind projects in the UK are incorporated as separate legal entities and therefore are required to file their accounts at Companies House where they can be viewed by any interested party. I have done my own research into this and agree with Professor Hughes’ conclusions.

The implications of this analysis is that many wind projects will be uneconomic once their subsidies expire unless wholesale electricity prices remain high. This completely undermines the argument that these forms of generation will lower costs to consumers.

Of course, arguments can be made that the costs of climate change may be higher, but unfortunately, the government has declined to publish any analysis it has made of the relative costs of mitigating versus avoiding climate change, so it is difficult to know whether this is true or not.

In terms of human rather than simply financial costs, we should also remember that thousands of people die in the UK each winter due to the effects of fuel poverty and being unable to heat their homes. While climate change may cost lives and livelihoods, so too does fuel poverty.

Actually, on average between six and eight thousand people die as a result of fuel poverty each winter, and we can imagine the figure for this year will be higher. We have had literal hypothermia deaths this winter – normally fuel poverty deaths are indirect as people experience fatal respiratory illnesses linked to underheated homes. People don’t normally die because their body temperatures fall below 35oC…that actually kills people quite quickly.

The other aspect of security of supply that is even more overlooked is that of access to raw materials and in particular minerals. We are waking up to some of the geopolitical risks as a result of the Russian invasion of Ukraine, recognising that many of the minerals essential for the energy transition are under Chinese control.

Sky News wrote an excellent analysis of this last year, which I encourage everyone to look at – it was about mega mines in Chile but included some interesting facts about the mineral requirements of the energy transition.

First of all it’s worth remembering that just about everything we use originates from materials that were either grown or dug up. And when we dig stuff up it’s rarely in a form that can be used immediately – it involves a huge amount of not just digging but blasting, leaching, processing and shipping.

As I mentioned earlier, we intend to build more wind turbines and solar panels as well as nuclear reactors, and we need more network infrastructure to connect them all up. We also plan on using a lot more batteries both on the grid and in cars.

Some of the most important minerals we will need are copper and lithium. To build a moderately large offshore wind turbine with a 1 MW capacity, you need about three tonnes of copper in the generator and another in the transformer.

Then you need about half a tonne of copper in the cables carrying the power down the tower and up to five tonnes in the cables connecting the turbine to the shore, depending on distance.

Copper is also required at the substation and in the distribution cables that take that energy on to the transmission system, where, a lot of the cables are actually made from aluminium.

That adds up to more than 15 tonnes of copper per megawatt of electricity.

Extrapolate that across the country never mind the world and we can see that the demand for copper is going to grow significantly – analysts at Goldman Sachs estimate global copper demand from green energy will grow six times between now and 2030 or nine times if the transition is accelerated.

copper demand

The expected growth in the requirement for lithium, the main component in rechargeable batteries is even more extreme. The IEA believes lithium demand could rise by a factor of 40 over the coming years.

If the energy transition is to be implemented successfully, these essential minerals will need to be secured. The largest deposits of both can be found in Chile.

We not only need to worry about availability of these materials, extracting and refining them is a dirty process, consuming vast amounts of water which is left unsuitable for other uses. South America is already experiencing disputes over access to water between mining and agriculture and even human consumption.

Around Chile’s copper mines, cancer levels are higher than in any other region of the country. The soil has high levels of arsenic which is driven into the air by the drilling and blasting at the mines.

Another important mineral is cobalt used extensively in the production of batteries. The main producing nation is the Democratic Republic of the Congo, where child labour is used extensively with children as young as 11 working in the mines.

These are ethically and environmentally dirty supply chains, and they are also politically risky. Local initiatives to clean up the processes could put supplies at risk and or raise prices. And if these processes are not cleaned up, we have to ask ourselves whether polluted water and air, and the use of child labour are acceptable prices to pay for achieving net zero carbon.

The recent increase in energy nationalism resulting from the war in Ukraine has started to re-focus attention on the issues of energy security. In the UK, new policies are emerging which offer some hope, particularly around the development of new nuclear and the renewed commitment to domestic oil and gas production, but are still almost entirely supply-side focused. You could say that we need to rely more on solar energy, but that’s just not going to happen overnight!

Sorry about that!

Reducing demand, and specifically reducing demand due to energy waste needs more focus and more investment, and the best hedge against high energy costs in the future would be to reduce heat losses in both commercial and domestic properties. If the energy transition is to be successful and retain public support it needs to work for consumers – the only thing likely to be less popular than expensive energy is unreliable expensive energy.

It is also positive that, particularly here in the UK, we are seeing a more pragmatic and less ideological approach to energy policy. Some people are unhappy about this, feeling that the climate emergency demands rapid action, but expensive and unreliable energy also costs lives and livelihoods, and we need the energy transition to be fair. Accelerating net zero by a decade, as the IPCC is calling for, can only be achieved at significant cost and most likely with a major reduction in living standards, which is unlikely to be feasible.

We are beginning to see the transition as more of an evolution than a revolution, which is important, because many of the net zero pathways rely on technologies that are yet to be commercially developed, or even invented at all in some cases. Technology will of course play a key role, whether that is new forms of generation, or entirely new uses of energy such as hydrogen, but low tech solutions are also important such as better insulation in homes.

We need to keep the big picture in mind. People have recently started to get very exercised about unabated gas generation, but right now there is no feasible technology to achieve abatement. Rather than worrying about this, it would be more productive to address larger sources of emission, particularly in the transport sector. The single dirtiest activity any one of us can do is to take a flight, yet how many of us are willing to give up our foreign holidays? And that reluctance underlines the difficulty of the challenge ahead.

We should prioritise the things we already know how to do, starting with the ones that will have the highest impact.

The current crisis reminds us of the importance of secure and affordable energy supplies, but it is also important that those supplies are clean, in every sense of the word ie not just low carbon, but genuinely sustainable with ethical supply chains.

So, while policymakers deal with the near term challenges they have an opportunity to re-frame our approach to transition to deliver credible and achievable roadmaps. We can only hope they rise to the challenge.

Anyway, I thought I’d end with another bit of science, this time from the Discworld and the wisdom of Terry Pratchett:

“It’s known that knowledge is power, and power is energy, and energy is matter, and matter is mass, and therefore large accumulations of knowledge distort time and space.”

Thank you!

Subscribe to the Watt-Logic blog

Enter your email address to subscribe to the Watt-Logic blog and receive email notifications of new posts.