What deal do we need for 2030?
How fast do we need to move? How fast could we move?
Last week I introduced my plan to explore the energy social sciences based on the Green New Deal. I want this newsletter to help us understand what needs to be done, and know what we still don’t know.
Now, the Green New Deal is ambitious. And you don’t have to take my word for it:
The Green New Deal is the most ambitious climate proposal ever brought to Congress. (The New Yorker)
How ambitious is the Green New Deal? Incredibly ambitious, both on climate change and with its reimagining of society. (The Guardian)
Is the Green New Deal too ambitious? (The BBC)
The bulk of my missives here will concern the substantive dimensions of its ambitions: what it sets out to do, and what that may imply. For this week though, I would like to take one step back, by dwelling on dimension of its ambition: the speed at which it wants to reach its emission targets. A 40 to 60 percent reduction from 2010 levels by 2030 and net-zero emissions by 2050. It will be the first in a two-part series on the social science of forecasting!
***
For their demanding targets, Ocasio-Cortez and Mackey draw on the latest IPCC report, with which they also open the resolution. If we want to keep global warming below 2 degrees Celsius (and please, pretty please below 1.5 Celsius), the report’s authors insist, then we need to stop pumping out greenhouse gasses over our blue planet.
OK, fine. But do we have to do it so fast? I mean 40 to 60 percent reduction in 12 (now 11) years? Can’t we, like, warm up a little (excuse the pun)? Get the proverbial starting blocks ready, draw the lines on the race track, know how to go, and then make a sprint for the finish line? As long as we make it by 2050, right?
The problem with that (not imaginary) objection is that greenhouse gasses tend to stay in the atmosphere for a long time and thus accumulate. You could say we have a ‘budget’ of total emissions for this whole century. That means that if we dilly-dally on reduction of emissions now, for instance because we want to lay the groundwork for some greater and better transition in the future, the logic of accumulation implies that the amount of work we’ll need to do later is disproportionately greater.
This common-sense reasoning, like sunscreen before it, has been proved by scientists.
You see, a few years ago, Riahi et al (2015) found themselves a little sceptical about the pledges made at the Copenhagen Accord. If you’ll remember, the Copenhagen Accord was a non-binding agreement to keep following in Kyoto’s footsteps and in particular cool off global warming to 2 degrees Celsius by century’s end. So they wanted to figure out: could these pledges actually do the job?
To answer that question, they created a few scenarios for models to run through. The main variable in these different scenarios was short-term ambition: either high ambition with stringent short-term emission targets for 2030, or low ambition one with permissive short-term targets. Then they threw in this variable with the by then conventional cocktail of population growth, economic growth, and energy demand…er, growth on the one hand, and different energy supply side developments on the other (for instance, how much nuclear, with or without carbon storage, strong or weak energy efficiency rules, etc.). Then they looked at what the implications would be for 2050 and thereafter.
Well…
“Following [a permissive scenario] to 2030 requires an acceleration of the global rate of emission reductions to 2050 by almost a factor of two compared to the optimal policy scenarios” (14).
In different terms, the share of low-carbon energy production would need to quadruple between 2030-2050, if we don’t go fast now (p. 16). Also, crucially, the finish line mentioned above is not 2050. We’re in this fight for the long haul. If you don’t go fast now, you’ll also need to keep reducing faster after 2050. It will be a more intense marathon. (I’ll stop with the running metaphors now).
These calculations have some immediate policy consequences. Firstly, as also mentioned by the authors (p. 14), is that you do not want ‘lock yourself into’ fossil-fuelled energy infrastructure by building more of it the coming years. Not only can we not afford the extra output of greenhouse gasses, but these plants last a long time, and so we’ll be stuck with them well beyond 2050. Secondly and conversely, it means that anything that can prevent more carbon going up in smokes now is more valuable that something that can do it later. This is one of the arguments against adding more nuclear plants – yes, they are super-efficient at producing zero-carbon energy, but it takes a long time to build them – time precisely that the models say we don’t have. Better in that sense to take the many billions they would cost to build and spend it on rapidly multiplying wind turbines. Finally, if we do not heed these warnings, we run a greater risk of having to rely on experimental, potentially dangerous and mos’ def more expensive technologies such as carbon capture and sequestration later on.
***
So far, so theoretical. But can we actually do this? Given the complexity of the energy system and the scale of its overhaul, is there any reason to assume we could pull this off?
Now, most forecast models do try to factor in some “feasibility” indicator, but that’s feasibility defined more as “if we do a and b, can we get to x endpoint?” It’s not the kind of feasibility defined as “can we get our shit together on time?” The social science of transitions would actually stem pessimistic on this count. As Benjamin Sovacool (2016) points out, most of the (historical) transition literature converges on the idea that transitions take a long time – two to three generations long. So… we’re doomed?
We may be. However, Sovacool does want to cast some doubt on this common conception. A first reason for doubt is the way the models are built. They tend to build on more or less linear development rates in technology and policy adoption. The famed s curves are difficult to predict (things can go fast). Also, feedback loops (for better and worse) are consciously excluded for the same reason.
Then there are the cases that the transition literature has overlooked. Sovacool introduces a broad range of instances where new energy technologies were rapidly adopted. Efficient energy lightning in Sweden, cooking stoves in China, gas stoves in Indonesia, ethanol vehicles in Brazil, and A/C in the US (all examples of end-user technology), or crude oil-to-electricity in Kuwait, natural gas in the Netherlands, combined heat and power in Denmark, and coal plant retirements in Ontario, Canada (in terms of energy supply technology): all these technologies were adopted at rates of 40% to 95% within 9 to 30 years.
The World Bank also got involved with gas stoves.
A picture emerges from Sovacool’s paper in which transitions occur (simultaneously) on many scales, in many domains, and always in locally specific ways. Transitions likely won’t ever be revolutionary transformations, but that does not mean that the interaction between these ‘partial’ transitions can’t be significant. (see p. 211f.)
These partial transitions do make them harder to predict. Sovacool does think we can do better though, by learning from previous transitions, the big and small ones, the slow and fast ones.
In other words, we need more social science in our forecasting models. And that, children, will be the subject of next week’s edition! For now, sleep tight and don’t let nightmares of uncontrollable warming wake you up tonight.
If you do wake up and want to tell me about it, write me a letter!
Sources
Riahi K., Johnson N., Krey V., McCollum D.L., Riahi K., et al. 2015. "Locked into Copenhagen pledges - Implications of short-term emission targets for the cost and feasibility of long-term climate goals". Technological Forecasting and Social Change. 90 (PA): 8-23. http://dx.doi.org/10.1016/j.techfore.2013.09.016 (Open Access)
Sovacool, Benjamin K. 2016. "How long will it take? Conceptualizing the temporal dynamics of energy transitions". Energy Research & Social Science. 13: 202-215. http://dx.doi.org/10.1016/j.erss.2015.12.020