This is something almost everyone has now heard about.
To be part of the fight against climate change, companies should engage in a three step-process: they should measure their GHG emissions, reduce them, and simultaneously increase carbon sinks.
While there is some intense debate on the latter (especially on net-zero frameworks and on the claims companies should be able to make -contribution or offset?), there is according to me much less discussions on the emission reduction theory, although this theory faces serious issues.
I already exposed some of these issues (as others may have done it too).
I ended up suggesting a couple months ago that companies should be able to take into account the avoided emissions from their sold products and services and deduct them from their absolute scope 1 to 3 emission reduction targets.
Back then I thought this idea could help solve the issues I had mentioned, mainly the fact that absolute scope 1 to 3 emission reduction targets are unreachable and not relevant (as well as unfair) for companies that sell low-carbon products or services and are about to strongly grow in a decarbonizing world.
As I thought more deeply about the suggestion I had made, it turned out that what I proposed is not theoretically satisfying. Indeed, this may still lead to growing net emissions (scope 1 to 3 emissions minus avoided emissions from sold products and services) for companies which help decarbonize our societies.
I recently came up with the idea that we’re actually facing a trilemma that will eventually seriously jeopardize the way we’re currently approaching the emissions reduction theory.
Here is the trilemma.
One simply cannot have simultaneously:
(1) (attributionnal) emission reduction targets per company that ensure compliance with global carbon budgets once aggregated,
(2) unpredictability of the future (an ex-ante inability to know exactly what will happen),
and (3) an efficient transition (the cost of the transition is as low as possible).
Indeed, let us consider the different possible cases:
if (1) and (2) are true, then this implies identical absolute emission reduction targets for all firms (variable targets that ensure global carbon budgets compliance would imply that we know in advance which scenario is going to happen -amongst all possible scenarios we face). As already mentioned, identical absolute emission reduction targets for all firms lead to a transition that is highly inefficient: (3) is therefore not possible.
if (1) and (3) are true, then in order to guarantee the economic efficiency of the transition, companies set variable intensity targets (convergence of their physical or monetary emissions intensity towards a target value at a given horizon, as proposed by the SBTi for instance). However, these customized targets are obtained by using models (such as models from the International Energy Agency) that have as input the volumes of activity per sector up to the target horizon (quantity produced of electricity, steel, etc.), which is equivalent to assuming that future is predictable (and the point there is not to comment the likelihood or relevance of such scenarios). Thus (2) is impossible.
If (2) and (3) are true, then efficiency implies that we no longer care about the evolution of a company's attributed emissions from one year to the next, but rather about the evolution of their carbon impact, in a consequential perspective (the emission variation induced by a company activity). What matters in such a case is no longer that a company's attributional emissions (such as GHG Protocol scopes 1 to 3) fall by X% per year, but rather that a company's counterfactual scenario emissions (without the company) fall by X% in absolute terms. In such a case, there is therefore no longer a target value for emissions (in absolute terms or intensity) that can be quantified for a given timeframe. With the current SBT approach, and knowing the emissions of a given reference year (let's say 2019), we can translate the reduction target in terms of emissions to be reached for a given horizon, for example the year 2030 (and the target would be of the type: the 2030 emissions must be 50% lower than those of 2019). With the proposed change of perspective, such a translation is no longer possible, since we are no longer comparing emissions with respect to a fixed and static value (the emissions of the reference year), but with respect to a moving and dynamic scenario (the emissions of the scenario in the absence of the company under study, which vary from year to year).
A very simple implication of the above is that either efficiency, emission reduction target per company with carbon budgets compliance, or non-predictability of the future must be sacrificed, if the other two are to be satisfied.
As it is impossible to sacrifice the "non-predictability of the future" component (despite all the progress in terms of artificial intelligence, modelling, etc.), this implies that we must -there is literally no alternative- either sacrifice the efficiency of the transition, or the attributionnal emission reduction targets per company (which once aggregated respect the global carbon budget).
The Science-Based Targets initiative (SBTi, which is the main international framework for emission reduction for companies) is all about setting emission reduction targets for companies.
According to the trilemma we just exposed, this means this framework (and the way it is designed) will either lead to economic inefficiency or to global carbon budgets that are not satisfied (and so to a global warming that may be higher than that targeted).
Either way this is very problematic.
And what we’ve just shown is a theoretical justification: this means one can only hope to limit the issues but not to solve them by keeping this approach.
The upcoming articles will be about deep diving into how we can reasonably assess the carbon impact of a company and setting a proper reduction framework.
Bonus
Accounting avoided emissions of a company (from its sold goods and services) in an attributionnal framework (such as done in the pillar B of the Net Zero Initiative -NZI- from Carbone 4) does not solve whatsoever the trilemma that is exposed above.
Avoided emissions are based on a consequential framework (you need to do some counterfactual reasonning to assess these emissions) and more or less reveal the carbon impact of a company.
Whereas scope 1 to 3 emissions (also known as induced emissions) rely on an attributionnal framework and solely reflect ... the emissions we decided to attribute to the company.
Thus you cannot combine them in a relevant manner because of their different meanings, nor you can set compatible targets on avoided and induced emissions.
I'll talk more deeply about that in further articles.
Comments