Five Times Faster takes a systems approach to understanding why we haven’t made nearly as much progress on climate as we need to over the last 30 years of trying. Sharpe — a former UK climate diplomat — looks at the problem from 3 different angles: the way we assess and communicate the potential impacts of climate change; the way we think about the economy in the context of climate action, and finally how diplomacy and international agreements around climate action have been structured and negotiated. In each of these contexts he highlights the need for more consideration of non-linearity, feedbacks, and the potential for changing the underlying dynamics of the climate, economic, and political systems.
Assessing Climate Risks
Rather than taking the typical scientific approach of trying to identify the most likely outcomes, and quantifying the uncertainties on those predictions, he suggests that we should be using risk assessment tools that are common in the world of insurance or national security. We shouldn’t focus on trying to avoid or adapt to the most likely outcome, rather we should consider the worst plausible outcomes, and if they would result in unacceptable losses. This would mean communicating climate change risks and scenarios to policymakers very differently than has historically been done in the guarded, conservative language of the IPCC reports.
This felt a lot like how I came to climate change work. In 2008 I happened to read Nassim Taleb’s book Fooled by Randomness and Richard B. Alley’s The Two Mile Time Machine at the same time. Taleb’s book is about epistemic uncertainty and the inherent difficult of learning about the probability of rare events. Alley’s book is about paleoclimate and what we know about the last few hundred thousand years based on Greenland ice cores. Together they were pretty disturbing! The climate is a complex system filled with feedbacks and tipping points, capable of rapid and potentially irreversible state changes. Our GHG emissions are pushing the climate system into a domain that we don’t understand because we have no experience of it — the last time Earth was as warm as we expect it to be by the end of the 21st century, there were no humans, not even any other members of the genus Homo. We have no ice core records that old.
In this paradigm, decisions are driven by outcomes that may be unlikely, but would have very high impact, such as the rapid collapse of the West Antarctic Ice Sheet, or the irreversible collapse of the AMOC.
Economy as Ecology
In the second part of the book, he sharply criticizes the prevailing equilibrium economics that treats the economy as relatively mechanical, stable, and efficient, and discourages interventions beyond correcting “market failures” such as externalized costs. Instead, he encourages thinking about the economy more like an ecology — an interconnected web of dynamic relationships that can change dramatically when disturbed due to feedback loops and other nonlinear dynamics.
He draws on the work of Mariana Mazzucato (author of The Entrepreneurial State) and others who study how innovation can be catalyzed to radically and permanently alter the structure of the economy. The now canonical example of this dynamic is the dramatic reduction in the cost of solar (and batteries) as overall deployment has scaled up. Greg Nemet chronicles the policy interventions made by the US, Japan, Germany, and China that led to this outcome over decades in How Solar Energy Became Cheap.
The point being, we can intentionally change the nature of the economy not just through brute force, but by nudging it in new directions with the development and early deployment of technologies that we want to have around. If we choose well, then a small push can result in huge movement in a new direction, if it becomes self-sustaining. Potentially to the point of becoming an accelerating, runaway process. I mean:
China alone now has the manufacturing capacity to pump out 1 TW (1000 GW) of solar PV every year. Total global energy consumption is only about 20 TW. In terms of that capacity, we’re probably past the peak of the growth curve. There will never be 10 TW of solar PV manufacturing capacity.
But this kind of dynamic likely doesn’t apply to every technology equally, so it’s worth trying to understand which ones are susceptible to it and why. In 2021 some folks at Oxford looked at the current trajectories and learning rates of solar PV, batteries, wind, and electrolyzers in this paper, and compared them to fossil energy sources. interview with Dave Roberts here). Abhishek Malhotra and Tobias Schmidt have tried to understand why some technologies behave this way and others don’t. (another interview with Dave Roberts, and their paper in Joule).
80/20 Sectoral Diplomacy
He notes that the massively multiplayer negotiations between the ~200 parties to the UNFCCC, aimed at adopting economy-wide, global emissions targets that are often decades in the future, he suggests that narrower initiatives are likely to be more tractable and productive. He wants narrower focus along at least two dimensions.
First, rather than looking at the entire economy, which results in very diffuse responsibility and doesn’t tend to lead to any specific policies (other than economy-wide carbon pricing schemes which… read Making Climate Policy Work by Danny Cullenward and David Victor). Sharpe suggests focusing on individual high-emitting sectors like electricity generation, building HVAC, road transport, shipping, aviation, cement production, steel production, petrochemicals, or ammonia synthesis… Negotiating within a particular sector makes it possible to be much more specific, about the processes, technologies, incentives, bottlenecks etc. that need to be addressed or developed to reduce emissions. And to develop concrete short and medium term goals that are actionable now.
Second, he suggests negotiating among a smaller group of countries, companies, and industry groups that aren’t literally everyone, but that do cover enough of the sector that if they were all to agree on a path, it would be hard for the rest of the sector not to follow. What that threshold is and how many actors it takes to get there will depend on the sector, and he thinks it’ll be easier to start with those that are more concentrated and potentially subject to nation-level interventions. The two he looks at in some detail are electricity generation and road transport.
The goal of all this policy making and negotiation is to try and set up what he calls tipping cascades — large and varied positive consequences that result from a couple of relatively small interventions. For example, in the case of supporting the early deployment of EVs, you get cheaper batteries, which also helps further decarbonize the electricity system economically, and might result in oil companies choosing to diversify their businesses once they see the writing on the wall:
In contrast to that plausible happy story, he also looks at how hard it will be to prevent deforestation in the service of producing agricultural commodities. While there are a small number of countries that account for a large share of both the world’s remaining forests and all agricultural commodity production and consumption, there don’t seem to be the same kind of advantageous technological feedbacks to take advantage of. So maybe there are some domains where Sisyphus does just have to push the boulder up the hill forever.
Utilitarian Corporatism?
The decision to focus on entities like electric utilities and car manufacturers because they’re huge and centralized and thus susceptible to focused pressure is understandable, but man it sure leaves a lot on the table. Zoning, building codes, transportation that isn’t cars, eating (much) less meat, etc. are mentioned but written off for the most part because they are personal or local government decisions. This makes them harder to influence with nation-level interventions, and also more likely to be salient to citizens & consumers who are considered resistant to change.
I would like to think that there are opportunities for desirable tipping cascades in these systems too. Maybe it’s not for everyone, but for many people once you’ve experienced Barcelona’s Superblocks, or lived in a city built for biking, or a co-housing community, it’s hard to imagine going back to “normal.” If these experiences were available to more people, and the laws and financial regulations that functionally prohibit them could be swept away, couldn’t those changes also be self-reinforcing?
Focusing on actions that will move and appease the biggest corporate (and national) actors seems likely to further entrench or at least replicate their powerful positions in whatever systems come next. Which I guess is part of the idea. They have the power now. We either have to pay the ransom they demand, or take their power away. And that latter option doesn’t seem to be going well at the moment.
Nonlinear Race Conditions
I appreciated the focus on dynamical systems, and the potential for much more rapid change than you might expect because of extremely non-linear behavior. This is great in the context of economic and technological change, and terrifying when it comes to the climate system.
Some of the diplomacy chapters dragged and felt overly self-congratulatory, with shallow descriptions of recent agreements and initiatives whose ultimate outcomes are yet to be determined and (if history is any guide) may not come to much. More concrete, successful examples that have had longer to bear fruit would have been nice, even if they had to come from other domains.
By the end of the book, I felt like we are struggling with dueling complex systems, with the outcome totally unpredictable. On one hand, we have the climate system that is purely reactive — it has its own internal dynamics that we don’t fully understand, especially as we reach further and further into a new and unfamiliar regime. (See Zeke Hausfather’s NYTime OpEd about the climate freak show that was 2023-2024, or this NASA Earth Observatory piece). This system is responding unsympathetically to our forcing.
On the other hand, we have our own complex sociotechnical & economic systems that, at least in theory, we can influence with the goal of minimizing our perturbation of the climate system. We’re not all pulling in the same direction, which makes this a mess. One advantage of technology driven change is that once a new thing is cheap and easy, it can be hard to put it back in the box. If PV and batteries are dirt cheap, then price-sensitive consumers will flock to them. Witness Pakistan’s totally unexpected emergence as the world’s 3rd biggest importer of solar panels in 2024. They imported 13GW of PV in the first 6 months of 2024; the whole country only had ~50GW of installed generation capacity in 2023.
This is good because another giant non-linear thing we have to deal with is skyrocketing electricity demand — and I don’t mean because of EVs, heat pumps and AI data centers in the US and Europe. Most people globally don’t have access to enough energy to get good quality of life outcomes (2-3kW per capita, far less than the 5-10kW that rich countries use). Over the remainder of the century they’re going to do whatever they can to get it.
But economics alone also isn’t destiny. We’ve been able to build self-perpetuating systems that are economically insane for many of the participants trapped in them. The same land use and transportation systems that Sharpe thinks are too difficult to shift in the time that we have left make no economic sense from many different angles, but the US is culturally committed to them anyway, to the tune of trillions of dollars, and has successfully exported them around the world as an aspirational model. Close to half of all US soy and corn go toward uneconomic biofuels with no end in sight.
The interlocking interests of construction contractors, sprawl developers, car manufacturers, oil companies, agribusiness, unions representing auto workers, and marketing companies selling an American Fever Dream on credit have kept it all humming along, and as a result US burns a huge portion of its national income supporting the system, even though other options exist, and are cheaper, safer, healthier, and more enjoyable to live in. So I’m apprehensive about relying entirely on technological change and learning rates to make change happen.
But it’s clear there are no linear futures available. The only way to avoid horrendous climate change is with non-linear change in the systems causing it. If we fail to create that change, we’ll explore the non-linear response of the climate system at our peril.
