Thread by @jgkoomey

I want to add a subtle clarification to this thread, in which people try to answer the question on whether we should "let go of [1.5 C] as a policy goal".

https://twitter.com/dhofstetter_x/status/1385705462075625472

Dana and I agree that we need to cut emissions as much and as fast as possible, starting now, but that's not the end of the story.

https://twitter.com/dana1981/status/1385730921911836674?s=20

The nature of path-dependent systems dominated by increasing returns to scale, learning effects, network externalities, and spillovers, like our economic/social/technical society, is that we can't actually know what is possible until we try to take action.

Here's a classic summary of this way of thinking, which points to the difficulties/inability of economic modelers to accurately assess systems like these. Arthur, W. Brian. 1990. "Positive Feedbacks in the Economy." In Scientific American. February. pp. 92-99.

The most widely known reflection of this reality are so-called "learning curves" for mass produced technologies, often shown for solar PV, wind, and batteries, as per this BNEF graph, courtesy of @solar_chase.

There is no better reflection of the idea that our "choices now create options later" then technologies with big learning rates. Each doubling of cumulative production experience yields cost reductions between 10 and 30%, typically.

This insight goes back to the early days of aircraft manufacturing: Wright, T. P. 1936. "Factors Affecting the Cost of Airplanes." Journal of the Aeronautical Sciences. vol. 3, no. 4. 1936/02/01. pp. 122-128. [doi.org/10.2514/8.155]

The great Kenneth Arrow also studied this topic: Arrow, Kenneth J. 1962. "The Economic Implications of Learning by Doing." The Review of Economic Studies. vol. 29, no. 3. pp. 155-173. [doi.org/10.2307/2295952]

More recently: McDonald, Alan, and Leo Schrattenholzer. 2001. "Learning Rates for Energy Technologies." Energy Policy. vol. 29, no. 4. March. pp. 255-261.

More recently: Rubin, Edward S., Inês M. L. Azevedo, Paulina Jaramillo, and Sonia Yeh. 2015. "A review of learning rates for electricity supply technologies." Energy Policy. vol. 86, 11//. pp. 198-218. [sciencedirect.com/science/articl…]

If the US, then Germany, then China (along with other countries) hadn't invested in the deployment of these technologies, their cost reductions wouldn't have been nearly as steep, and our options now for reducing emissions would be more costly.

Solar is the best studied of these technologies. See this, from @GregNemet: Nemet, Gregory. 2019. How Solar Became Cheap: A Model for Low-Carbon Innovation. New York, NY: Routledge. [howsolargotcheap.com]

What is the main implication of "our choices now creating options later" for the question that started this thread?

This is the key: The future is dependent on human choices. That means assessing what's "likely" is a fool's errand. Economic and social systems that depend on human choice are not the same as physical systems, where likelihood has meaning.

Nobody knows what's likely or even possible until we actually start down the path of aggressively reducing emissions by deploying technology and institutional innovations. If we choose to do so, many things will become possible that wouldn't be possible if we didn't.

So nobody knows if 1.5 C is still possible, and analysts should not ever say that it isn't, because it discourages efforts to reduce emissions. It's a subtle form of "doomism", as @MichaelEMann calls it.

The fact is, we don't actually know if 1.5 C is impossible now, and in fact we CAN'T know (because "our choices now increase options later"). In addition, assessments that claim 1.5 C is impossible embed assumptions that are unverifiable and probably wrong.

It is fine to say, as IEA does, that "each year of inaction makes 1.5 or 2C goals increasingly difficult to achieve". It is not fine to say "1.5 C is impossible". We won't (and can't) know until we try. So let's try.

That's why I advocate choosing an aggressive goal, and 1.5 C is as good a goal as any. If we fail to meet it, so be it, but better to aim high and fail than to aim low and do less well than we might have because we mistakenly believe the the high goal is impossible.

These learning effects (among other important factors that lead modelers to underestimate emissions reduction potentials and overestimate costs of mitigation) will open up possibilities that wouldn't exist if we don't start aggressively reduce emissions.

So, we do need to reduce emissions as much and as fast as possible, starting now. But we also need to shoot for an aggressive goal, because we won't know what's possible until we try, and we haven't really started trying.

And if we start down the path of emission reductions and discover it's easier than we think, we need to do even more. It's virtually impossible right now to move too fast on reducing emissions at this point.

I explore these issue more here: Koomey, Jonathan G. 2012. Cold Cash, Cool Climate: Science-Based Advice for Ecological Entrepreneurs. El Dorado Hills, CA: Analytics Press. amzn.to/2eiZE2C