Liebreich: Beyond Three Thirds, The Road to Deep Decarbonization

By Michael Liebreich
Senior Contributor 
Bloomberg New Energy Finance

In my BNEF Summit keynote in London last September, I talked about how far clean energy and transport had come over the last fifteen years. Where renewable energy used to be dismissed as “alternative”, I talked about the “new orthodoxy” of what I called the Three-Third World: by 2040 one third of global electricity will be generated from wind and solar; one third of vehicles on the road will be electric; and the world’s economy will produce one third more GDP from every unit of energy.

The fact that we are on track for the Three-Third World is quite extraordinary. It certainly outstrips my expectations when I founded New Energy Finance in 2004. And it is probably unstoppable: wind, solar and battery costs will continue to fall faster than any mainstream energy forecasters expect, and there is nothing that makes me think President Donald Trump will succeed in his attempts to revive coal.

That’s the good news. The bad news is that even though we are on track to achieve the Three-Third World by 2040, it will not be enough to reach the goals of the Paris Agreement. Electricity currently meets only just over 20% of the world’s final energy needs; and even if you add in the proportion of oil going into passenger cars and light trucks and you are still only addressing about a third of final energy consumption.

As I pointed out at the end of my Summit keynote, there is no emerging new orthodoxy on how to decarbonize the rest of the economy – industry, chemicals, aviation, shipping and, in particular, heat.

Heating loads

As I sat down at the beginning of March to write this piece, the “Beast from the East” brought a prolonged blast of cold air from Siberia to the U.K., blanketing the country in snow, with temperatures well below freezing. This doesn’t happen very often, perhaps once or twice every decade, but it’s a real challenge for anyone thinking about the future of energy. Any solution for deep decarbonization of the U.K. economy has to be able to take the Beast from the East in its stride.

Even in a normal year, the U.K.’s winter heating load – which is practically zero in summer months – reaches peaks six times as high as the country’s electricity load, and it can cycle up and down by a factor of three in just a few days. If your plan for zero-carbon heating relies on electrifying it all, you need to take into account, not just the massive cost of swapping all those domestic boilers and commercial HVAC systems with electric heaters and heat pumps, but also the required investment in generating capacity, grid infrastructure and power storage. In 2016, the Policy Exchange estimated the cost at a staggering 300 billion pounds ($420 billion at current exchange rates).

In sunny climes, the combination of solar PV plus batteries should be able to cover a very substantial proportion of power and heating (or more likely cooling) needs – solar PV only needs a few days’ worth of storage, and battery costs are following it down a very steep experience curve, even in the absence of breakthroughs in battery chemistry. Despite enjoying very low-cost shale gas, sunny U.S. states are seeing solar and batteries starting to compete with gas-based power – for the moment after investment tax credits, but surely in due course without any subsidies.

However, anywhere north of around 40 degrees latitude, while solar power may generate affordable power in summer, its output drops dramatically in winter. The solar panels on my roof in London produce one thirteenth of the power in December that they produce in June. You can love domestic solar power as much as you want, but you cannot suggest it will do much to meet the UK’s winter heating load. Of course you could suggest importing solar from Southern Europe or Africa on a vast scale via HVDC connections, but then you have to account for the cost of high-voltage DC links and explain in a grown-up way how to deal with the resilience risk.

What about wind? Its output rises in winter, broadly in line with heating loads – but wind can suffer lulls lasting several weeks, and they can occur at times of low temperatures. Coupling batteries with wind is not like coupling them with solar. For all the growth in battery installations that BNEF is forecasting, the total volume of grid-connected batteries by 2030 will be sufficient to meet the world’s power needs for just 7.5 minutes; add the batteries of every electric car and light truck, and you could meet a few hours of electrical supply – not days, not weeks, not months. And that’s just comparing against electrical loads – and remember the U.K.’s heating loads can be up to six times higher.

Combining different sources of renewable electricity will help. It is theoretically possible to build a system that melds wind and solar power with whatever dispatchable renewable power is available – principally biomass and hydro – and power storage, to meet the U.K.’s energy needs, all of the time, in all weather conditions. However, costs start to escalate alarmingly as you get to very high penetrations of renewable power (over, say, 60%).

Even in the U.S., which enjoys a far greater diversity of renewable resources than the U.K., it’s hard to get the maths to add up, as Professor Mark Jacobson, outspoken champion of 100% renewables, has just found out. Last month he dropped his lawsuit against 21 respected academics who pointed out that his plan for the U.S. rested on an un-costed and wildly improbable 15-times increase in the use of existing hydro resources to provide back-up power on low-wind / low-sun days.

Does that mean that we have no hope of decarbonizing our heating – and therefore no hope of getting beyond the Three-Third World? Not at all. There are plenty of possibilities, once we stop putting all our eggs in the renewable electricity basket.

Building efficiency

First of all, we all need to start treating the energy efficiency of our buildings like it really matters. Mainly that means insulation, air-tightness, and good thoughtful architecture and design. It doesn’t need to add much cost to the building; in many cases nothing at all. Ten years ago, I had never heard of the PassivHaus building standard; in ten years’ time, all new buildings could easily meet it. In fact, there is no reason why new houses shouldn’t produce more energy than they consume, receiving utility revenues instead of incurring utility costs. It’s just a question of applying technologies and techniques we know work.

Retrofits are harder. The important thing is that any time a building undergoes a deep renovation, its energy performance has to be brought up to the highest standard. It is possible – I’ve done it. As long as you are doing deep renovation works anyway, the extra costs are not prohibitive. Even a twenty-year payback would be equivalent to 5% risk-free after-tax – a highly attractive rate of return to most home-owners in a world of persistently low interest rates. Mainstream mortgage providers need to stop colluding in a system that treats the cost of a new kitchen as an investment, but the cost of a low-energy retrofit as an expense.

Once all new-builds and deep renovations are done properly, we will halve our heating challenge over twenty years, allowing a lot more of the heating load to be met electrically, mainly with air-source and ground-source heat pumps. If you think they can’t work in cold temperatures, just look at Norway, or Japan.

Then there are other new technologies. Some of the most intriguing start-ups I come across are working on thermal batteries, using phase-change materials, salts, clever thermodynamics, or just big chunks of concrete or tanks of hot water. Drake’s Landing Solar Community meets over 95% of its winter heating needs from solar energy collected during the summer. It lies just 45 minutes’ drive from the 1988 Calgary Winter Olympic venues. How cool is that – or rather how warm?

There are, however, significant benefits to continuing to power a large proportion of the world’s heating with solid, gas or even liquid fuels. These are easier to store in bulk to cope with seasonality and resilience than electricity, which will always need to balance to within a few days’ of real time, and will also be needed by industry. The question is how to make them zero-carbon.

A significant proportion of the heating load in temperate climates –  in countries like the U.K., Northern Europe, New England, Canada, the former Soviet Union and Northern Asia – could be met by biogas or biomass, most efficiently using combined heat-and-power, or CHP, cogeneration. Though it is hard to add district heating in existing neighbourhoods, it can be done – look at Sweden. Some 10% more Swedish households have been connected to district heating every decade since the 1960s, to the point where over half of all homes are now connected. And here’s a thought – since you are going to have to add more capacity to local grids to charge all those EVs, how about combining new bio-based CHP delivering local heating, with massive battery storage, to provide grid services and improve resilience for energy-intensive industries, all while reducing investment requirements in the distribution grid?


If there’s not enough biogas, you might consider running your CHP on natural (i.e. fossil) gas, which would still be up to 85% efficient, but not zero-carbon. To achieve that, you would need to use CCS (carbon capture and storage), but let’s be clear, that is not happening in the absence of a carbon price. Micro-CHP is attractive until you consider the capital cost, and even with a carbon price it’s hard to see how to capture the emissions from distributed sources.

And that brings us to hydrogen, which can be used anywhere without creating local emissions.

My skepticism about hydrogen vehicles is well known. What real problem do they solve? If you have electricity and you want to drive somewhere, just use a battery electric vehicle (BEV) – they will be fully competitive with internal combustion vehicles on a total-cost-of-ownership basis with no subsidy within five to six years in most markets, according to BNEF forecasts (clients can see more here). Why would you waste half of your electricity electrolyzing hydrogen, compressing and storing it, only to turn it back into electricity in a car?

If you are concerned about how long it takes to refuel, well that is a problem for the few percent of us who actually drive long distances; everyone else will charge their EVs overnight. Most people won’t want to visit a hydrogen station every few days just to avoid a 20-minute charge on the rare occasion when they drive long-distance. Even commercial vehicles, unless they regularly drive long distances – say, over 300 miles – will go electric. Ships, trans-continental trains, long-distance trucking and niches like fork-lift trucks are the only parts of the transport system where hydrogen makes any sense.

In fact, even if you have already produced your hydrogen for some other reason – such as seasonal storage – and you want to drive somewhere, it will make more sense to generate power centrally and charge an EV, rather than to put it in a hydrogen-fueled vehicle. Doing so will be much lower-capex per megawatt, much more efficient, and you can extract value from the waste heat. And that’s before getting into the lack of hydrogen filling stations compared to the ubiquity of the grid, the complexity of fuel cell vehicles versus the simplicity of EVs, maintenance costs, safety and so on.

Nevertheless, I am bullish about hydrogen. It is one of the most promising ways of dealing with longer-term storage, beyond the minutes, hours or days that could be met by batteries, or the limited locations in which pumped storage could work. It can be stored as hydrogen, perhaps blended into the existing natural gas system, or after conversion into ammonia, natural gas (so-called power-to-gas, or P2G), methanol or some higher-value synthetic liquid fuel. It can help provide the huge pulses of reliable power needed by some energy-intensive industries like ceramics. We need to stop fooling ourselves about hydrogen as a transport fuel, and explore its pervasive use throughout our energy, chemical and industrial system.

Whether your hydrogen is clean depends, of course on how it is generated. Currently, the cheapest method is steam methane reforming (which applies heat and a catalyst to methane to produce hydrogen, and confusingly shares its initials with small modular reactors). Like natural gas CHP, however, unless you capture the CO2 released during reforming, and sequester it, SMR hydrogen has no place in a deeply decarbonized system. Any system using SMR-based hydrogen might as well just use natural gas – it would be more efficient and require less new infrastructure – and anyone promoting SMR-based hydrogen as a climate solution deserves their own circle of hell.

The second main way to generate hydrogen is via electrolysis (there are other ways, biologically, or directly from solar power, but they are in their infancy). While it is fashionable to posit electrolysis as the perfect way of using up surplus wind and solar power, this is probably wrong. Electrolysis, plus the associated equipment to compress, store and transport hydrogen, or to turn it into higher value feedstocks, will be capital-intensive. Even when the cost of equipment comes down – and there are developments promising considerable reductions – it will always make more sense to run it 24/7, rather than intermittently. It is the same if you want to use surplus renewable energy for desalination – on the surface another way to use excess wind and solar power – desal plants are capital-intensive, so you want to run them flat out.

Most likely, the cheapest way to produce hydrogen using renewable power will be from solar PV, buffered through batteries to provide 24/7 power, or offshore wind hubs, such as the artificial island being proposed by the Netherlands, which will benefit from very high capacity factors. Solar thermal is also promising because you can use waste heat to drive high-temperature electrolysis, which lowers the electrical power requirement.

If there is a carbon price, high-temperature electrolysis provides an elegant solution indeed. You can combust natural gas with the oxygen which is co-produced from electrolysis, and use the resulting exhaust stream of near-pure CO2 to generate power directly using an Allam Cycle Turbine before being sequestered. The waste heat is used to drive the high-temperature electrolysis. So you have natural gas, in, hydrogen, electricity and water, out – and CO2 sequestered at the lowest possible cost.

Nuclear power

There is of course one other technology that could contribute massively to deep decarbonization: nuclear power.

At present, of all the zero-carbon electricity currently being generated in the world, 28% is nuclear. French electricity production had an estimated gross carbon intensity in 2017 of 65 grams of carbon-dioxide-equivalent per kilowatt-hour, compared with Germany’s 488 (excluding biomass). Despite its famous Energiewende, Germany is on track to miss its 2020 climate targets, largely because of its decision to shut existing nuclear power stations. The U.K., which made no such decision, is on track. If the world is serious about climate change, we should be keeping existing, safe nuclear power stations open, not shutting them prematurely.

Whether we should be building new nuclear power stations is much more questionable. The estimated 30 billion-pound subsidy awarded to Hinkley C, and the extreme delays and cost over-runs of Flamanville, Olkiluoto, Vogtle, VC Summer and other new-build nuclear projects do not bode well. To paraphrase Oscar Wilde, if that’s how the nuclear industry manages its projects, it doesn’t deserve to have any. Even in China and India, where they are building nuclear plants as fast as their supply chains allow, renewable energy output is growing faster than nuclear.

The current generation of nuclear technology almost certainly has no place in a deeply decarbonized future. However, in 2015, think tank Third Way found that there were almost 50 companies in the U.S. and Canada alone, backed by more than $1.3 billion in private capital, developing plans for new nuclear plants. Almost all are working on small modular reactor designs that are flexible, modular and failsafe – in other words even in the worst scenarios, with complete loss of power and personnel, they shut down safely rather than melting or exploding. There are many other competing designs around the world.

What remains to be seen is whether these so-called IV-Generation designs can generate cost-competitive power: dispatchable zero-carbon power at 5 U.S. cents per kWh will find a market; at 15 cents per kWh it won’t. There are simply too many other ways of meeting power needs more cheaply, and the whole point of the Three-Third  World is that it is not going to wait and see whether the next-generation nuclear bet pays off. There will be so much flexibility in the system, in the form of storage (cheap) and demand response (pretty much free), that there is no way to stop “base-cost renewables” – large volumes of variable power generated at 2-4 cents per kWh, or maybe even lower – from penetrating the power mix. “Ecomodernists” whose main plan to promote nuclear seems to consist of attacking wind and solar, will fail. They would do much better to think about unique roles that only nuclear can play in the Three-Third World and beyond.

One such role might be in the production of hydrogen. Nuclear power has two great advantages over wind and solar as a power source for electrolysis: first, like electrolysis itself, it likes running flat out, not subject to intermittency; and second, it produces waste heat, which can be used to drive high-temperature electrolysis. Nuclear power’s ability to produce heat is its secret weapon as it seeks its role in the Three-Third World: no other zero-carbon source of electricity can match it at scale – not geothermal, not biomass. In fact China has been looking into using small nuclear reactors for district heating. Opponents of nuclear will always voice loud concerns about proliferation, waste disposal and decommissioning costs – these are legitimate concerns that will only ever be soothed by highlighting contributions to a zero-carbon future only nuclear can offer, not by attacking wind and solar.

Pivotal moment

The purpose of this piece was not to propose a blueprint to get beyond the Three-Third World and address deep decarbonization. Anyone who says they know which technologies will be dominant in thirty years or more is fooling themselves. My point is about the variety and scale of opportunities we now face.

In the past fifteen years we have witnessed several pivotal points along the route towards clean energy and transport. In 2004, renewables were poised for explosive growth; in 2008, the world’s power system started to go digital; in 2012, it became clear that EVs would take over light ground transportation. Today I believe it is the turn of sectors that have resisted change so far – heavy ground transportation, industry, chemicals, heat, aviation and shipping, agriculture. One after the other, or more likely as a tightly-coupled system, they are all going to go clean during the coming decades.

Astonishing progress is being made on super-efficient industrial processes, connected and shared vehicles, electrification of air transport, precision agriculture, food science, synthetic fuels, industrial biochemistry, new materials like graphene and aerogels, energy and infrastructure blockchain, additive manufacturing, zero-carbon building materials, small nuclear fusion, and so many other areas.

These technologies may not be cost-competitive today, but they all benefit from the same fearsome learning curves as we have seen in wind, solar and batteries. In addition, in the same way that ubiquitous sensors, cloud and edge-of-grid computing, big data and machine learning have enabled the transformation of our electrical system, they will unlock sweeping changes to the rest of our energy, transportation and industrial sectors. To paraphrase Marc Andreesen, Software Will Eat the Inefficient.

In the end, clean technologies will outcompete fossil-fuel-based approaches. They won’t even need a carbon price, though one would clearly speed up their adoption and it is heartening to see some momentum among U.S. Republicans for Carbon Fee and Dividend. All they really need is an internally consistent set of government policies – across research and development, standards, taxation, trade policy and government procurement – in order to reduce friction, eliminate coordination risks, and help the most nimble players get through the singularity: the point when they become competitive on their own costs and merits. What Marianna Mazzucato calls “mission-oriented” innovation, and the U.K. government calls a Clean Growth Strategy.

So whether you are working flat-out towards the Three-Third World, or whether you are trying to figure out what comes next, this is a moment of unparalleled opportunity. Fantastic rewards await those countries, businesses and investors who place the right bets. For those of you attending the Bloomberg New Energy Finance Future of Energy Summit, on April 9 and 10 in New York, I won’t be joining you this year, but you are in good hands.

As Buzz Lightyear almost said: “To the Three-Third World and Beyond!”

About BloombergNEF

BloombergNEF (BNEF) is a strategic research provider covering global commodity markets and the disruptive technologies driving the transition to a low-carbon economy. Our expert coverage assesses pathways for the power, transport, industry, buildings and agriculture sectors to adapt to the energy transition. We help commodity trading, corporate strategy, finance and policy professionals navigate change and generate opportunities.
Sign up for our free monthly newsletter →

Want to learn how we help our clients put it all together? Contact us