Michael Liebreich is the doyen of energy change analysts. He founded New Energy Finance which was bought by Bloomberg to become BNEF.
I've taken extracts from this piece published on BNEF's blog. (The emphases are mine) If you don't want to read the whole piece, see my summary at the end.
I believe the new decade will see us hit peak energy-related greenhouse gas emissions and start to see a modest but meaningful decline. Just to be clear, we will not see the sort of decline demanded by the Intergovernmental Panel on Climate Change – a 20% cut by 2030 to keep temperature rises to 2C, a 45% cut to remain under 1.5C – but I would guess at a drop of around 5%.
Clearly that is not enough to put us fully on track to avoid appalling climate change impacts, and by 2030 we will have to admit 1.5C is out of reach. But it will be a game-changer: it will demonstrate to even the most pessimistic that we can bend the arc; it will end the feeling of helplessness and impending doom that has taken over our public discourse; and it will set us up for much more decisive reductions in the subsequent decades.
Optimism, therefore, but not without limits. It’s hard to be an unbridled optimist. Over the past decade, global emissions have risen by 15%. For one brief three-year period, between 2013 and 2016, they were flat – for the first time ever outside major recessions – but in 2017 they took off again, as the global economy boomed and China reverted to pumping cheap money into high-carbon infrastructure and building. Since then, emissions have been growing by around 1.2% per year.
[But,] while emissions grew 15% over the past decade, the global economy grew by 45%. On average, economic growth outstripped emissions growth by 2.4 percentage points per year. [I.e, energy intensity fell by 2.4% p.a.] If that gap can be made to increase by just 1.4 percentage points, emissions peak, even in a growing economy. Increase it beyond that and we are over the hump: emissions will start to fall. Impossible pipe dream? Or something we will see in the course of the next decade?
First of all, the maths. In 2017, a report by the World Resources Institute (WRI) showed that there were 49 countries, representing 36% of global emissions, which have already passed peak emissions [even though they have growing economies]. Almost the whole of the OECD is reducing its carbon footprint, even when you adjust for imports (something the ‘degrowth’ brigade pretend is impossible). The WRI expects a further eight countries, representing another 23% of emissions, to peak within the next decade.
Last week, in Paris, I attended the first meeting of the International Energy Agency’s Global Commission for Urgent Action on Energy Efficiency, of which I am a member. The Commission’s goal is to raise the rate of improvement in global energy intensity from its current 1.5% to 3% per year.
I have three main takeaways from the meeting: 1) energy efficiency is finally gaining recognition as a national priority in many countries, whether for climate or energy security reasons; 2) there is as much ‘low-hanging fruit’ today as there ever was; 3) there is a far better understanding of how to deliver improvements than a decade ago.
As we close out the decade, BNEF has concluded that around two-thirds of the world’s population now live in countries in which wind or solar are the lowest-cost ways of generating power. The world records for low-cost wind and solar are both now down to around $17/MWh. That is around a third of the cost of new gas-powered generation – even in the U.S., where there is a glut of cheap gas. [Averages are of course higher, for both renewables and coal]
By 2030, I have no doubt whatsoever that the world record for low-cost onshore wind and solar will be below $10/MWh. It will probably be set in China, Morocco, Mexico or the Gulf states, who have been vying for leadership for the past decade. However, there is a chance it could be in India, Brazil, the U.S. or even Australia.
As we approach the final days of 2019, wind and solar are generating around 8.5% of global electricity. BNEF estimates that figure will be nearer 25% by 2030.
The biggest unanswered question, as renewable penetration grows, is whether the cost of managing intermittency will drop – with cheaper storage, growing demand-response capacity, business model innovation and smart policy design – as claimed by fans of renewables – or soar, as claimed by their opponents.
It’s a vital question, which will decide whether wind and solar can maintain their historical growth rates, or whether their penetration must soon saturate. All the main energy models are designed around the idea of renewable energy saturation: growth rates decelerating into the future – whether abruptly or slowly, but always markedly decelerating. The IEA’s central scenario, Stated Policies or SPS, has the combination of wind and solar reaching only 24% by 2040. BNEF’s NEO model, always more bullish, shows them at 39% in 2040 and 48% by 2050.
What if these models are wrong? What if learning, innovation and the co-evolution of demand-side industries continue to win the race against the physics of intermittency, and allow historical growth rates to continue for a few more decades?
Two decades ago everyone assumed that the cost of managing intermittency would soar after the first 5% of wind and solar entered the power mix; a decade ago we thought the inflection was 20%; now we know it is not this side of 40%. Modelling exercises around the world suggest that it is not until you reach 80% or more in any decently-connected grid that the cost of managing intermittency really starts to go vertical. [But of course, by the time we get to 80% renewables, (a) storage costs will be much lower, and (b) overcapacity will be much cheaper]
To believe in a renewable energy singularity, the first thing to do is to extract all the latent flexibility in our current power systems, and then build more, in the form of power storage, demand response, long-distance interconnections and linkages with transport and heat. We’ll see a lot of that in the coming decade.
The second thing we need to do is learn to love overcapacity. As I said in a keynote in 2014, in a high-renewables system, overcapacity is not a bug, it’s a feature.
The average capacity factor of the world’s hydro plants is 42%; gas peaking plants 15%. Even so-called baseload coal plants run on average only 54% of the time. If technology is cheap, and demand or supply are intermittent, we overbuild. Wind and solar are no different.
Will we build many weeks’ worth of power storage, or hydrogen electrolysis, just to capture peak renewable electricity that would otherwise go to waste? In a word, no, because curtailment will be cheaper. Think about it: if your $20/MWh wind or solar suffers 33% curtailment, you know what happens? It turns into $30/MWh wind or solar – still half the price of power from any other source.
The third thing you would need to see if you want wind and solar to sustain their current growth rates is significant electrification of transport, (which I don’t think anyone doubts is on the cards – Daimler Benz clearly things so, for the first time in 135 years it is not working on the next generation of internal combustion engine) and heating (which, with global heat pump sales growing at 12% per year for the last decade, might finally be kicking off).
The fourth thing would be the electrification of industry and the generation of green fuels, be they hydrogen, ammonia or liquid fuels like methanol. It is hard to believe it was just 18 months ago that I wrote about this in Beyond Three Thirds: The Road to Deep Decarbonization, because there has been so much progress since.
BNEF’s seminal work on the cost of electrolysis suggests that green hydrogen (based on renewable power) will start to be competitive with brown hydrogen (from steam methane reforming of natural gas with no carbon capture) by 2030, and that by 2050 it will have a clear advantage. That means that, even in the absence of a carbon price, green hydrogen has the chance to eliminate the 5% of global emissions that currently result from fertilizer production and oil refining. A $20 carbon price would see it eat into the 2.2% of emissions from the global shipping industry. A $50 carbon price pushes green hydrogen into the 13% of industrial emissions from steel and concrete; and a $100 carbon price would take it into space heating, glass and other sectors. So the combination of cheap green hydrogen and a $100 carbon price will create an addressable market by 2050 of nearly 30% of global emissions. Neat.
Let’s talk about the fifth driver of a potential renewable singularity: batteries. By 2030, EV batteries will cost around $65/kWh at the pack level. [Assuming a compound 15% p.a. decline, $65/kWh will be reached in 2024] That’s $6,500 for the battery in a full-sized vehicle with a range of 300 miles; $13,000 gets you a 600-mile range – certainly bigger range than my bladder can handle. [And the batteries in small 'city cars' with 40 kWh will cost just $2600]
All the pinch-points in the mineral supply chain will have been long ironed out, and by 2030, all end-of-life batteries will be recycled – if there are any.
Yes, you read that right. Solid-state batteries may or may not have hit the market, delivering four-times the energy density, and launching swarms of electric planes. But there will certainly have been continued progress in lithium-ion technology towards the “million-mile EV battery”, which can deliver 10,000 charge cycles. It will make possible either cars with 50-year lives, ubiquitous vehicle-to-grid business models, or second-use applications at scale – or all three. Mindblowing.
Oh, and by 2030, you will not even remember about range anxiety – the same way you don’t remember that there were once insufficient modems to connect to the internet, or insufficient bandwidth for online video.
Those who doubt the value of renewable energy in addressing climate change always point to negligible impact so far. Even at 8.5% penetration into power demand – and after investment of $2.7 trillion – wind and solar have only reduced global emissions by only around 2.5% from where they would otherwise have been. They have so far failed to absorb growth in energy demand.
That is to miss the point. As leading energy economist Professor Michael Grubb has pointed out in Conditional Optimism: Perspectives on Deep Decarbonisation, the key clean technologies are growing according to the dynamics of logistic curves, penetrating into incumbent technologies. Professor Grubb uses lots of fancy economics to forecast what might happen next. I’ll paraphrase: in a logistic curve penetration, the first 1% takes forever; from 1% to 5% is like waiting for a sneeze –it is going to be explosive, you just don’t know when it will happen; 5% to 50% happens much faster than you think – that is when the restructurings and bankruptcies happen.
No single “sneeze” will wipe out fossil fuel use across energy and transport; It will occur sector by sector, country by country. Over the past six years, LED light-bulbs have gone from less than 5% global market share to over 40%; coal power in the U.K. from 40% to a couple of percent; plug-in vehicles in Norway from less than 5% to over 50%. In each case, there was a slow start, an agonizing wait, and then the sneeze. Bless you!
What does all this mean for coal consumption? According to Global Energy Monitor (formerly Coalswarm), in the final 2.5 years of this decade, global coal capacity grew by an average of 56GW or 2.8% per year – which hardly looks like a harbinger of peak emissions.
India and China alone have between them a pipeline of 280GW of new plants, bigger than the entire current U.S. fleet and equivalent to 15% of current global capacity. However, this does not begin to tell the full story.
First, what really matters is not capacity, but how much coal is actually burned. Over the past decade, capacity factors for thermal generation have been falling around the world, in China’s case to record lows. Globally – not that you would know it from the mainstream news – coal consumption in the power sector has been flat since 2012; preliminary figures for 2019 show a drop of around 3%.
Just this month it was announced that over half of the power plants operated by China’s Big Five state-owned utilities are running at a loss. The government has plans for up to one third of them to shut by 2021, removing 15% of the country’s coal capacity. As for India, despite its 85GW pipeline, on average it has commissioned less than 10GW per year for the past three years. This September, Prime Minister Narendra Modi announced a push for 450GW of zero-carbon generation by 2030.
In the EU, eight out of 28 countries have already committed to phasing out coal by 2030; it will be entirely gone in the U.K. by 2025. Germany, having prioritized the closure of nuclear over coal to date, will be off coal by 2038. The EU’s Green Deal, announced last week by new President Ursula von der Leyen, included 35 billion euros of support for Poland and other countries to get off coal.
In the U.S., despite promises to end the so-called “war on coal”, more coal capacity has been shuttered under President Trump’s first term than during any three years of the Obama administration. Every publicly-quoted coal company has gone through Chapter 11 since 2016, as has privately-owned Murray Energy, whose CEO, Robert Murray, wrote the blueprint for the president’s energy policy. Not one new coal plant has been built since 2015. None are being built today, and it looks like none ever will be again.
Of course, coal is used outside the electricity sector, notably in heating and industry. That source of demand looks likely to carry on increasing for a few years at least.
Before 2000, the orthodox view of oil demand (and energy analysts always default to orthodoxy) was that by 2030 it would grow to around 130 million barrels. By 2000, oil demand was still expected to grow forever, but it would reach only 120 million barrels by 2030. By 2010, the accepted wisdom was still endless growth, but only 105 million barrels of demand by 2030. See the pattern? Oil demand growth consistently undershooting the growth predicted by experts.
At no point did it cross the experts’ minds that maybe the same trends that they kept missing would see oil demand peak, and then start to fall. When I first suggested it in 2015, it felt like a transgressive act.
Today, there is not an oil company in the world that is not talking about peak demand. Even Saudi Aramco’s recent IPO prospectus predicted “a levelling-off around 2035”. BNEF expects demand from light and heavy vehicles to peak in 2030; this is one of the rare times I depart (slightly) from its view. I see peak road transport demand around 2025; add in the growing areas of air transport, shipping and petrochemicals, and I think we’ll see peak oil this side of 2030.
So far, you will notice I have barely mentioned policy. The picture I have painted is a bit like the IEA’s Stated Policies Scenario: these trends, which should see emissions from fossil fuels peak by 2030, are what I see happening even in the absence of significant further policy in favor of climate action.
But, of course, there will be further climate policy – and lots of it.
In the U.K., the new Johnson government is committed to achieving net zero by 2050. At this September’s Climate Action Summit in New York, 77 other countries, 10 regions and over 100 cities announced their intention to follow the U.K.’s lead. As I write this, the EU has just unveiled its Green Deal, which is going to enshrine a 2050 net zero target in law, as well as a plan to reduce greenhouse gases by “at least 50% and towards 55% by 2030, in a responsible way”.
In Canada, Justin Trudeau hung on to government, albeit not his parliamentary majority, in this year’s General Election. The country is therefore the first to implement the sort of tax-and-dividend scheme that could be a model for carbon pricing in political economies where new taxes are all but impossible to impose.
In the U.S., climate is a key battleground for the Democratic Party Primary, but what is more interesting is that Republicans with an eye on political life after President Trump are finally conceding they too need a climate policy.
Even the international climate negotiations should be expected to deliver some level of positive mood music over the coming decade, despite the potential withdrawal of the U.S. from the Paris Agreement and the failure of the latest COP conference in Madrid, as I describe in the sister-piece to this article, Climate Wars Episode IV – a New Hope for the 2020s?.
So there you have it – the reasons why I believe we will see peak fossil fuel emissions during the coming decade.
 |
The small inset chart shows atmospheric concentrations of CO2, the larger chart annual increases in CO2 concentrations.
RCP means representative concentration pathway. RCP2.6 would be consistent with a 1 degree C rise in temps, RCP4.5 1.8 degrees C. Liebreich's analysis suggests something between RCP2.6 and RCP4.5. If emissions only peak in the late 2020s, policy shifts as panic about climate change increases could lead to a much steeper decline thereafter.
|
To sum up:
- Renewables are going to get insanely cheap.
- Thermal coal demand will peak before 2030
- Peak oil is within sight
- The green hydrogen/methane economy will be cost competitive by 2030, and at $100/tonne carbon price, will replace 30% of emissions
- energy saving still has plenty of low hanging fruit.
- renewables overcapacity will be a design feature in electricity generation.
- CO2 emissions will peak before 2030, and by 2030, will be 5% lower than they are now
I've made a lot of these points before. Liebreich makes them better, and with greater authority.
No comments:
Post a Comment