Emissions Rising

25 November 2024

Peak emissions cannot be forecasted by tracking the growth of clean energy. For a number of years now, however, that appears to have been the game many were playing. Worse, some have attempted to call peak emissions in real time, and while that can’t be done either, it must be said that doing so is quite foolish. The challenging variable in all these attempts of course is not renewables growth, or fossil fuel growth, but economic growth itself. As a result, we will only know when emissions have peaked in retrospect, when their downward trend is sustained. Everything else is just throwing darts, and using hope as a strategy.

Incentives for making peak emissions forecasts are many. Think-tanks want to demonstrate they have a competitive edge with the data, the audience wants to hear good news, and journalists are understandably eager to report that progress is being made. Indeed, one way to describe our current moment is that the storyline of amazing growth in renewables has overshot actual emissions progress on the ground. This divergence can be explained pretty easily, at least on one level: peak emissions forecasters get renewables growth right, but get economic growth—and the demand it represents—wrong. If you’re a think-tank that has forecasted for a couple years now that global power sector emissions have peaked, or that China emissions have peaked, or that the US is still on track to hit emissions targets by 2030, some advice: stop doing that. Don’t put out those forecasts anymore. The repetition of failed attempts (revealing the lack of a sound method) has become embarrassing. Also, if you’re a publication covering this area, try being less credulous.

Some well needed sobriety on this subject was delivered last week, however, by Umair Irfan, writing for Vox, with a very on-point headline: We were supposed to hit peak emissions. Why won’t they stop rising? The article keys off the latest assessment from the University of Exeter, and its Global Carbon Project science team, showing that emissions rose again this year, this time by 0.8%. This is a touch better than 2023 growth, which saw emissions rise by roughly 1.1%. And we are becoming more successful, it must be said, at lowering the growth rate.

But peak emissions forecasting has another problem to contend with: reaching the peak offers no promise or even likelihood of declines. That declines follow peak is a human intuition, not how the reality of fossil fuel consumption actually plays out. More often than not, consumption plateaus, often for years, and we should expect the same when emissions finally stop growing.

Energy transition may have a modeling problem, when it comes to the displacement of legacy energy sources. Consider the following statement:

We'll build so much wind, solar, and storage that we'll cut into fossil fuel in global power, sending emissions into decline.

This has been a reasonable expectation for over a decade, but the thesis has run into a hurdle. The statement needs to be adjusted.

We'll build so much wind, solar, and storage that we'll start covering all marginal growth in global power. But then we’ll settle into a pattern where legacy fossil-fuel power generation takes an overly long time to retire.

How did we get from the first statement to the second? One explanation may seem like a paradox, though it’s quite straightforward and not exactly a surprise: we’ve been very busy retiring the oldest fossil fuel assets first, which over time has made the remaining assets far younger as a group. And that may be why a standoff has developed between the new, and the old. Over the past 15 years, the US for example has closed down a bunch of ancient dams in hydropower, older natural gas plants, and very old coal plants. These types of older plants were barely economic, or worse, entirely uneconomic. Only a few incentives were required to tip them over the line into closure.