The shaky logic of emission reduction
While I was recently thinking and writing about the issue of additionality (or lack thereof) in renewable electricity purchases, I came across this clarification in the GHG Protocol’s scope 2 guidance that makes it clear that this widely accepted corporate GHG accounting framework is based on attributional accounting. All that the scope 2 accounting — and the GHG Protocol’s corporate standard in general — is aiming for is accurate allocation of emissions without double counting between end-users. But this framework is not designed to quantify the impacts or consequences of specific actions taken by a company which may change the overall size of the emissions pie.
This issue is rapidly coming to the forefront as major companies are using non-additional, unbundled renewable energy certificates (RECs) to greatly “reduce” their reported scope 2 emissions without actually reducing the amount of GHG emissions entering the atmosphere. The Science Based Targets initiative, with its focus on emission reduction, is beginning to recognize the problems with scope 2 reporting but any solution will depend on changes to the underlying accounting framework.
I want to step back a bit and point out that this problem is more general than just scope 2 emission reductions using renewable electricity. But first, a synopsis of attributional vs. consequential modeling in LCAs and corporate GHG inventories.
Attributional modeling is used in nearly all LCAs and probably 100% of the GHG inventories for its simplicity. The circle on the left represents overall emissions, and attributional modeling simply allocates a slice of it to a particular product or company. The circle on the right shows what might happen under consequential modeling, which tries to answer the question of how global emissions would actually change as a result of manufacturing a product, running a company, or changing how a company is run (for example, by purchasing renewable electricity, switching to recycled materials, changing the product recipes, etc.).
When a company purchases existing RECs, additionality doesn’t play a role in attributional accounting because the emission reduction reported by this company is supposed to be balanced by increased emissions that other companies must report under the market-based approach (by calculating residual emissions). This acknowledges that there is no immediate climate benefit when purchasing from the existing supply of RECs. The area of the circle on the left is constant.
But it gets worse in practice. Those “other” companies can (and do) avoid reporting higher emissions using the various loopholes in the reporting standard. So, the sum of the emissions reported by all companies could decline while the area of the circle on the left (representing actual total emissions) is unchanged.
This issue is more general than just scope 2 accounting and applies to scope 3 as well. Imagine a company replacing its purchase of virgin raw materials with recycled materials. The recycled materials are purchased from the existing supply of these materials in the market, just like the existing RECs. Replacing a unit of virgin material with a recycled substitute simply means that there is one less unit of that recycled material available to other companies, but they can buy one extra unit of virgin material. At best, this is just moving around the ownership of materials (just like the ownership of RECs) without actually doing anything to reduce emissions in the short term.
The argument in support of this type of accounting is that increased demand for RECs or recycled materials would eventually spur more production of those goods and less of fossil energy and virgin materials. But companies claiming those reductions today stand on dubious ground. Much of the blame for this unhappy situation should reside with the standard setters who have allowed attributional accounting to be used for assessing the immediate impacts of climate actions by companies. Many of these actions may have a longer-term positive impact depending on scale and market alignment.
A full consequential modeling approach to assess the impacts of climate actions can get intractable very quickly due to system expansions. A possible solution is some type of hybrid approach that uses attributional accounting to establish a baseline (as in the circle on the left) and then uses a limited consequential model to evaluate the impact of a change.
Very importantly, the solution should differentiate between actions that cause immediate climate benefit (such as additional renewable electricity production, energy efficiencies and electrification within the company boundary, etc.) for which companies can take credit in their ongoing emissions inventories, and potential long-term impacts (such as purchase of existing unbundled RECs, substitution of recycled materials for virgin materials, etc.) for which companies should be recognized but not rewarded immediately.
Without a better and more nuanced standard for how to report the impacts of a company’s climate actions, the logic of emission reduction will remain shaky and questionable.
