By Michael Liebreich
The word resilience has become increasingly fashionable in climate circles. This week the IPCC published the second part of its sixth Assessment Report (AR6), focusing on climate impacts, vulnerabilities and adaptation. The idea of climate resilience featured prominently.
What does resilience mean in practice? What are the risks against which we need to protect ourselves? What does being resilient entail? How much public or private money should be spent? Are there trade-offs between resilience to climate and other types of risks?
In January 2022, the World Economic Forum’s Global Risks Report identified the top three “most severe risks on a global scale over the next 10 years” as biodiversity loss, extreme weather and, in top place, “climate action failure”. Seriously?
As I write, Russia has begun its invasion of Ukraine. The U.S. and its allies appear on a path to confront an increasingly authoritarian China. And the worst pandemic for a century is still taking a grisly toll on people and the economy. Any one of these threats presents a more severe global risk than climate change over the next decade. It is worth noting that previous editions of the Global Risks Report failed to warn of the 2008-09 Great Financial Crisis and the pandemic.
Activist and climate scientist Professor Michael Mann claims that “wildfires, droughts, floods, heatwaves, coastal inundation – climate change is already costing far more lives than Covid 19”, but he’s wrong by at least two orders of magnitude. The confirmed toll from Covid so far is 5.8 million, and the best estimate of the actual total, including deaths before testing was in place globally is nearly 20 million. Weather-related deaths, by contrast have averaged 18,500 a year for the past decade – each one a tragedy, but down by a factor of 95 over the past century. Once you normalize for increased global wealth over time, there is a clear downward trend in economic damages too.
Now, before you jump all over me, I’m not saying we should not be acting on climate change – far from it. Last year’s 6th Assessment Report of the IPCC was perfectly clear, confirming as an “established fact” that climate change is already increasing the frequency and violence of extreme weather, and things will only get worse without adequate investment in both mitigation and adaptation. Longer term, while we may no longer be tracking the disaster scenarios that drive the rhetoric of Greta Thunberg and Extinction Rebellion, we are far from the 1.5C warming goal that guarantees stability of the climate for the long term, or even the less safe 2C.
What I am saying is that those who seek to recruit the concept of resilience purely in support of climate action are short-changing other, and potentially more catastrophic, near-term risks. In addition, in our haste to transition to net zero, we may even be rendering our energy and transport system more fragile, rather than more robust.
The future net zero system
This discussion is playing out in the early stages of a complete transformation of our energy, transport, industrial and infrastructure systems. Let’s start by taking a look at what it is we are building.
The electricity system embarked on the transition a couple of decades ago. The old system of baseload-and-peak, with a relatively small number of large generators connected to a transmission grid, is being replaced by one with enormous numbers of smaller generators, connected at all points from the center to the grid edge. Significant proportions of supply and demand will vary at will, with balancing provided by demand response, storage (power, thermal and chemical), overcapacity and interconnections with neighboring systems.
Transport, next out of the blocks, is going electric. Not just urban transport and not just cars, but all of it, with the exception of aviation, deep-sea shipping and (perhaps) long-distance freight.
Electric vehicle chargers will end up as ubiquitous as Wifi. The shockwaves, as BloombergNEF’s then chief editor Angus McCrone and I wrote in 2016, will cascade through every sector of the economy, from insurance to dealer networks and from city layouts to finance ministries
Clean hydrogen will be essential for the de-carbonization of fertilizer, plastics and chemicals, as well as perhaps steel and long-duration storage – demanding the development of an entire new distribution and storage system. There may or may not be a substantial e-fuels sector. And, as I explained in Part II of Separating Hype from Hydrogen in October 2020, the bulk of domestic, commercial and industrial heating will be electric.
So we are talking about a massive increase in electrification. Over the last half-century, electricity went from delivering just 10% of final energy to just over 20%. Over the next half-century we will see a far more dramatic increase. There is no version of a net zero economy in which electrification will serve less than 60% of final energy.
In addition to being vastly more efficient, the system will need to be managed and optimized in real time. This means 100% digitization: pervasive sensors, low-latency communications, big data, machine learning, distributed ledger accounting, digital twins and so on.
The real risk of Net Zero
A decade or so ago, it was fashionable to talk about the Energy Trilemma: the need for energy to be simultaneously affordable, clean and secure. The term fell into misuse as the cost of wind, solar and lithium-ion batteries plummeted.
As the future of limitless, clean, affordable energy hove into view, the security aspect faded into the background. I was as guilty as any: in my BloombergNEF Summit keynote in 2015, I described the new “Age of Plenty” – cheap oil, cheap gas, cheap renewables and cheap energy efficiency – but I did not talk about resilience.
Some argue that the new system will be inherently more resilient because it will be less dependent on a small number of large suppliers. De-carbonized, Decentralized, Democratized, runs the mantra. It is wishful thinking. It may be relatively easy to design a wind and solar-based system that meets power needs for much of the year, but it is orders of magnitude harder to design one that can get through monsoon and rainy seasons, winter weeks with low wind and freezing weather in northern climes, or shocks of an unprecedented scale.
The new German government has announced its intention to shut the country’s final three nuclear power stations at the end of 2022 and all of its coal-fired power stations by 2030, while at the same time reducing the country’s CO2 emissions by 65% – effectively ruling out any increased dependence on gas. What is the plan to keep the lights on?
Germany’s wind and solar resources are counter-correlated, with more wind in winter and more sun in summer. However, research shows that around once per year there is a shortfall period of five days or more when their combined output falls to less than 15% of average. Batteries are not going to fill these periods (the Dunkelflaute, or “dark doldrums”). Under even the most optimistic views on experience curves and raw material availability, batteries big enough to power the country for five days but used only once a year would be prohibitively expensive.
Inevitably, Germany is going to become even more dependent on its neighbors than today. Setting aside the fact that one of its key energy suppliers has invaded Ukraine, there are two problems: first, it is impossible to be sure that your neighbors will have excess power (or hydrogen) just when you need it, particularly in the light of their own transitions. And second, even your most friendly neighbor today may not be quite as friendly tomorrow. During the early stages of the pandemic, we saw ugly outbreaks of PPE and vaccine nationalism. Which country will export power if its old people are dying of hypothermia?
Plan for the worst, hope for the best
In the U.S., coal’s share of power generation fell from 33% to 19% between 2015 and 2020. Great news from a climate perspective, not so much from a resilience perspective. In 2015, the country held enough stocks of thermal coal to meet its entire power needs for 31 days. By 2021, the figure was down to 17 days. Meanwhile, over the same period, natural gas rose from providing 32% of U.S. electricity to 40% but didn’t provide any increase in storage. Wind and solar’s contribution grew from 5% to 10% – also without providing an iota of additional storage.
This can only mean that the U.S. power system has become less resilient to supply shocks lasting for more than few weeks. The point is not that piles of coal at power stations are the only way to provide resilience, as President Trump’s first Secretary of Energy Rick Perry tried to claim in 2017. The point is that if you remove safety margins in one place without adding any elsewhere in the system, don’t be surprised if your chickens one day come home to roost.
There are plenty of energy modelers who will point to their models to claim there is no problem. They remind me of financial analysts in the run-up to the Great Financial Crisis, using their huge computers to analyze historic trends every which way and find profits, but blind to fat tail risks that were not in their data sets.
In Asia, many national grids are insufficiently robust to allow particularly deep penetration of renewables, be that internally within Japan, China and India, or between many smaller countries
and archipelago nations. China has for years been promoting the Global Energy Interconnection Development and Cooperation Organization (GEIDCO), using HVDC technology to even out variability in renewable resources from the Arctic to the tropics, from Europe to East Asia. But would Japan – or any other major Asian nation for that matter – be wise to become dependent on China for power?
We all celebrate the growth of solar rooftops in Africa and the building of wind and solar plants. For the next few decades, however, it is natural gas that will keep the lights on throughout the G20 when there’s no wind or sun; the EU has even included natural gas in its green taxonomy. Meanwhile Africa is being denied the funds to develop its own natural gas resources. As a result, Africa risks being rendered permanently uncompetitive by an unreliable power system.
Shutting down investment in fossil fuels before you have a plan to replace their role in the energy system is neither resilient nor just.
There are no easy answers. If you think nuclear power holds the only hope for a truly robust system – setting aside the obvious risks of proliferation, accidents and the lack of long-term waste storage – it faces its own, very specific Catch-22. To have any chance of achieving economic competitiveness, dozens or even hundreds of identical plants need to be built. Each time one experiences a safety scare, the risk is that all will need to be shut down. Unlikely? Recently, 10%
of France’s nuclear output was taken offline when cracked pipelines were detected in multiple plants. And it turns out even swarms of jellyfish can shut nuclear plants – they have done so on multiple occasions worldwide.
The push by nations to achieve energy independence is not a resilient strategy – energy commodities are traded globally and shocks are transmitted by markets. Fracking in the U.K. will do little or nothing to protect consumers from price spikes – the chance of extracting enough gas to significantly move European prices is zero. Nor is relying on imported LNG a resilient strategy – you cannot be sure there will not be times when it is unavailable or extortionately expensive.
With great power comes great power cuts
Power cuts today can be catastrophic. Last year’s Texas outages left 4.5 million people without heating, water and power for days and were responsible for between 250 and 700 deaths. Sick babies in need of oxygen and warmth start to die 20 minutes from the beginning of one of the frequent power cuts across the developing world.
Future power cuts may be much worse. We are building a system of systems – power, transport, telecommunications, health services, logistics, payments, emergency services, public information – dependent on each other, with tight linkages and no circuit breakers. How do you protect the energy system from looting if you can’t charge the vehicles of your first responders? How do you procure spare parts for power plants if the payment system fails?
Studies of complex systems provide a theoretical basis for concern. It is possible to model systems that are stable but which, as you extend them and link them to other stable systems, reach a bifurcation point where they become unstable. Cascading voltage collapses have been behind a number of continent-scale power outages. This is the dark underside of interconnecting power systems over wider areas to deal with renewables’ variability: in January last year, a fault in a Croatian substation came close to taking down the entire European electrical system from the West of Portugal to the East of Turkey.
To make things worse, as we have seen from the pandemic, there is hysteresis in crisis. Things don’t just bounce back to the way they were.
In 2011, Germany’s Office of Technology Assessment published a report on what would happen in the event of a prolonged power outage. Within a few days the country’s communications, including those used by the emergency services, would fail. Rail and air transport can’t run without electricity; neither can the road system without signaling and emergency services; ports would grind to a halt. Fuel quickly becomes unavailable, fresh water and sewage stop flowing. Without refrigeration or transport, farming, food processing and food distribution all stop. Healthcare and financial payments fail. Military and civil emergency services swing into action, but what can they do without transport or communications? Civil disorder breaks out as people starve and, if it’s winter, freeze.
The authors conclude: “it would be almost impossible to prevent a collapse of society as a whole.” And this is Germany, not a fragile developing country.
Countries are no longer at peace until they are at war. Today, they maintain a persistent state of cyber-hostility against their rivals, which they dial up or down in line with international events.
In 2007, Estonia was the first ever victim of a concerted cyber-attack from entities based in Russia; Baltic republics’ cyber-defenses are now constantly being probed. In the 2020 SolarWinds hack, attackers affiliated with the Russian state were found to have penetrated multiple U.S. government websites and up to 18,000 corporations. In January this year, Russia hacked 70 Ukrainian government websites.
Infrastructure presents a particularly juicy attack surface. In a 2019 survey by German engineering powerhouse Siemens, 56% of utility company IT executives reported that cyberbreaches were responsible for at least one incident per year, causing “outages, damage, injury, and even environmental disaster”. Last year, one such attack took down the Colonial Pipeline in the U.S. What began with a ransomware attack on the company’s payment system led to a fiveday shut-down; fuel deliveries were disrupted, which in turn triggered panic buying and real fuel shortages. Last month the same thing happened to German fuel distributor Mabanaft.
Panic buying is one of the biggest non-linear vulnerabilities of our connected world. In the U.K. last September, the mere rumor of fuel shortages led to lines of cars at forecourts until fuel supplies ran out. A month later, stocks were back to normal.
Don’t look up
Are you worried yet? You should be. But we’re not done yet: nature still has some surprises up her sleeve.
Meteor and comet impacts are enjoying a moment in the public gaze due to the success of the Netflix climate spoof “Don’t Look Up”. The reality, however is that NASA judges the probability of a city-killing impact on the scale of the 1908 Tunguska event at less than 0.1% per year; that of an extinction event is once in 100 million years, and practically ruled out before 2175.
A far bigger threat is posed by solar storms. On September 1-2, 1859, the most intense solar event in recorded history hit the planet, in what is known as the Carrington Event. Northern Lights were visible as far south as Colombia and southern Japan. This was 20 years before Thomas Edison created the world’s first electric utility, but the telegraph system was knocked out: operators reported electric shocks from their equipment, pylons threw sparks, and messages could be sent and received without generators being connected.
Solar storms less severe than the Carrington Event hit the Earth in 1921 and 1960. In 1989 a geomagnetic storm knocked out power in Quebec. In 2012 a storm of a similar power as the Carrington Event missed Earth by a few degrees. It is not a question of whether this sort of storm will hit us in future, but when. One scientific estimate puts the likelihood at just over 1% per year, with the danger peaking at solar cycle maxima. The next one of those, since you ask, occurs in
If a Carrington-level solar storm were to hit Earth today, it would knock out satellites, radio and electrical equipment, and create powerful currents in conductors such as railway tracks, pipelines and, of course, cables – frying transformers, switchgear and connected appliances. No GPS system. No electrical system. No communications. Probably for months, and with less than a day’s notice. In a 2013 report, Lloyds of London estimated the potential damage at $0.6-2.6
trillion, but that feels like an underestimate.
Of course, developed countries have contingency plans: the U.K.’s 2015 Space Weather Preparedness Strategy, the U.S. Department of Homeland Security’s Strategy for Protecting and Preparing the Homeland Against Threats of Electromagnetic Pulse and Geomagnetic Disturbances, and so on. However, if the pandemic taught us anything, it is that those who think they are well prepared often are not.
The do’s, don’ts and four R’s of Resilience
So what’s to do? First, if we are serious about resilience, we need to identify the risks we actually face, not just focus on ones in the political or civil society spotlight.
Next, we need to plan infrastructure and systems that can withstand the shocks identified. There is an extensive literature on how to do this, much of it structured around the four principles first described by the U.S. National Infrastructure Advisory Council in 2020, which the U.K. government borrowed in 2011 and restated as the “four R’s”: Resistance, Reliability, Redundancy, Recovery.
• Resistance means assets must be designed to withstand the forces and disruptions they are likely to encounter. In the context of climate change, anything we build needs to take into account higher sea levels, higher temperatures, more violent storms, more frequent droughts and wildfires. In some cases, this will also prepare them to face non-climate related risks, but that must not be assumed.
• Reliability means that assets must be able to operate within a wide range of conditions and not degrade over time. The power grid needs to continue to serve the bulk of demand – and certainly critical demand – even in the event of parts of it failing or of failures in other related systems. We need to be building a grid which, like the internet, cannot be brought down in its totality under any circumstances. We also need to invest dramatically more in the maintenance of aging infrastructure – a huge resilience issue with our without climate change.
• The importance of Redundancy cannot be overstated, as we saw when global supply chains were disrupted for months by one ship stuck in the Suez Canal. In a world governed exclusively by profit motives, redundancy is often squeezed out of the system and regulators may need to step in. As I first said at my 2014 BloombergNEF Summit keynote, overcapacity is a feature of the future energy system, not a bug.
• Recovery governs the ability of the system to adapt after it is challenged, either in real time – as when power utilities seamlessly shifted output from offices to homes during the pandemic – or by learning and adapting after incidents.
Finally, though, if we want resilience, we are going to have to pay for it. And, as New Zealand’s leading futurist, Roger Dennis, put it in the context of the pandemic: “resilience is expensive in the short term, but cheap in the long term”.
The stakes are high. The upcoming IPCC report on adaptation will rightly point out that climate change is a source of near-term risk. But we must bear in mind that it is only one source among many. We need to do clear-headed analysis, roll up our sleeves, get our checkbooks out, and start building.
Michael Liebreich is the founder and senior contributor to BloombergNEF. He is also the CEO and chair of Liebreich Associates, founding managing partner of EcoPragma Capital and an advisor to the U.K. Board of Trade.