Myth busting – Are corporate Scope 3 emissions far greater than Scopes 1 or 2?
Are corporate Scope 3 emissions far greater than Scopes 1 or 2? The short answer is that the question is meaningless because the Scope 3 does not measure anything that is comparable with Scopes 1 or 2.
Now for the detailed answer. How many times have you seen someone claim “Scope 3 accounts for the vast majority of corporate greenhouse gas (GHG) emissions”?1 Frequently, they will cite a CDP questionnaire summary report stating that Scope 3, on average, accounts for 75% of reported corporate GHG emissions from all Scopes.2 This belief is used to argue that greater attention should be placed on Scope 3 emissions in corporate GHG inventory reporting. I argue that this justification is flawed. To understand the exact nature of the flaws, we have to understand the differences between what Scopes 1, 2, and 3 are attempting to quantify and how.
- Scope 1 addresses direct emissions (i.e., emitted by the company’s assets), which are exclusively allocated to be reported (as Scope 1) by a single company and released to the atmosphere within a single year.*
- Scope 2 addresses indirect emissions (i.e., from consumed electricity, heat, steam, or cooling), which are exclusively allocated to be reported (as Scope 2) by a single company and released to the atmosphere within a single year.*
- Scope 3 are indirect emissions that are “duplicatively” assigned to be reported by many companies as Scope 3 emissions. Scope 3 emissions may occur over many years—including past years, the reporting year, and future years—yet are summed and reported as if they are released in a single reporting year.
* In practice, exclusive allocation will not perfectly occur when companies use different organizational boundary consolidation approaches.
To express more simply…no other company is to report your Scope 1 emissions as their Scope 1. Similarly, no other company is to report your Scope 2 emissions as their Scope 2. However, many companies report your Scope 3 emissions as their Scope 3 emissions. Of course, if I double count something repeatedly and compare it to something that is not counted multiple times, then the former will tend to appear larger.
As I explain in my paper on the problems with applying concepts from life-cycle assessment (LCA) to GHG accounting, Scope 3 also mixes annual and lifetime (or life cycle stage), cumulative emissions into one aggregate emissions total that is reported as if it occurs in a single year. For example, emissions associated with the use of sold products (i.e., GHG Protocol Scope 3 category 11) are totaled over the lifetime of the product, which often spans many years and reported as if they occurred all in one year. Yet, indirect emissions from employee commuting (i.e., GHG Protocol Scope 3 category 7) are reported by the year in which emissions physically occur. What is the rationale for not reporting cumulative employee commuting emissions over the tenure (i.e., life cycle) of an employee with a company? What is the principled justification for the aggregation of emissions that occur over many years and reporting them in a single year for some Scope 3 emissions but not for others? Instead of any lifecycle aggregating, I would argue that the proper emissions inventory reporting approach is to assign and report emissions—such as those assigned to a product, capital item, or investment3—separately for each year in which emissions physically occur. For example, emissions from product disposal in landfills will physically occur over many years and could be reported according to the years in which emissions actually occur rather than as a cumulative number that is reported under the year that the reporting company conducted a transaction that connected it to the value chain of the product disposed.4
This problem of duplicative counting across companies and temporal aggregation is one of the many problems Scope 3 inherited from LCA, but it is not inherent to the reporting of indirect emissions in GHG inventories. Indirect emissions reported as Scope 2 do not entail temporally misleading emission totals because electricity and other forms of energy (e.g., steam, chilled fluids) are approximately produced and consumed simultaneously; if they’re stored, it’s only for far shorter periods than an annual timeframe. Further, the estimation boundaries of Scope 2 intentionally include only emissions physically produced by the process generating the electricity or other forms of energy (i.e., Scope 2 implicitly excludes indirect emissions upstream or downstream of energy production processes). Scope 2 is an example of a clearly delimited approach to indirect emissions reporting (excluding the problematic market-based approach).
So, the ‘Scope 3 is bigger’ claim compares a class of emissions that is both massively double counted and cumulatively totaled over an undefined span of many years, to two other classes (i.e., Scopes 1 & 2) that are counted once and cumulatively totaled over just a single year. Then, this comparison of all the oranges grown by a tree during its lifetime to a single apple (i.e., the ‘Scope 3 is bigger’ claim) is used to convey a grand insight which should inform our interpretation of GHG accounting results and guide our policy and decision making.
So then, what is the intellectual root of this erroneous belief and refrain about the relative size of Scope 3? What do those that repeat it think they are conveying? At this root is the concept of assigning “responsibility” for GHG emissions to a company, which is the purpose of GHG inventory accounting. The logical error resides in the fundamentally different conceptualization of responsibility that is applied to Scope 3 compared against the conceptualization that exclusively assigns responsibility for emissions in Scopes 1 and 2.
To repeat this erroneous refrain is to instead conceptualize responsibility as an indicator of a potential influence. There are many more physical emission sources in other companies that a company could indirectly influence than the number of sources it owns and/or operates. Yet, as I explain here, conceptualizing responsibility in terms of a “potential to influence” for GHG inventories (i.e., physical allocational GHG accounting) is unworkable.
Unfortunately, the current approach to Scope 3 is based on this unworkable approach to assigning responsibility—it has companies report GHG emissions from all sources anywhere in a fuzzily defined value chain that they are seen to have a potential to influence. For example, a company’s decision to produce, use, or sell a product is expected to influence the emissions occurring upstream and downstream of its operations. In other words, by equating responsibility with influence, the Scope 3 inventory boundaries are improperly treating all corporate activities as interventions. Based on the theory that everyone in a value chain can intervene to influence anything, the Scope 3 accounting and reporting structure attempts to assign everyone responsibility for everything, which effectively assigns no one responsibility for anything.5
In summary, emissions reported as Scope 3 by companies appear deceptively large relative to other Scopes because of duplicative counting across companies and the reporting of cumulative multi-year lifecycle emissions under a single reporting year. Yet, there is still value in the underlying exercise of a Scope 3 type of investigation as a form of canvassing to identify the physical processes that are emissions “hot spots” (i.e., are likely to release relatively larger fractions of emissions) within a company’s value chain. The results of such an investigation can provide useful information to then target mitigation interventions for proper consequential (i.e., intervention) GHG accounting and reporting. What is then needed—and missing in existing corporate GHG reporting and disclosure—is not more focus on Scope 3 reporting, but instead, a set of accounting rules and a new reporting framework for quantifying and aggregating the impact of a company’s meaningfully ambitious value chain interventions. We will have more to say about what such an intervention-based GHG accounting framework should look like in a future post.
Endnotes
1 See (Blanco et al., 2016; CDP, 2021; Downie and Stubbs, 2012; Gopalakrishnan, 2022; Hertwich and Wood, 2018; Klaaßen and Stoll, 2021; Yang and Chen, 2014).
2 CDP also reports that this value varies by sector, as discussed here: https://www.wri.org/update/trends-show-companies-are-ready-scope-3-reporting-us-climate-disclosure-rule
3 The GHG Protocol guidance describes the exercise of estimating Scope 3 emissions for category 15 (investments) as a snapshot of emissions reflecting the product and supplier mix for an investment portfolio composition at one moment in time (e.g., on 31 December each year). Such an exercise is not really an accounting of aggregate emissions allocated to a company or its investments over a year. A company wishing to game these rules could simply choose to sell off its high emitting investment on the day the snapshot is taken and then repurchase the next day.
4 For example, the minimum boundary for GHG Protocol Scope 3 category 12 for product end of life disposal is specified as the Scope 1 & 2 emissions of the waste management companies handling of product waste. These Scope 1 & 2 emissions will be from the single applicable reporting year. Yet, most disposal emissions (e.g., landfill decay) occur over multiple years as the product decays.
5 Unless one has a magic wand that eliminates all frictions inhibiting collective action.
References
Blanco, C., Caro, F., Corbett, C.J., 2016. The state of supply chain carbon footprinting: analysis of CDP disclosures by US firms. Journal of Cleaner Production 135, 1189–1197. https://doi.org/10.1016/j.jclepro.2016.06.132
CDP, 2021. CDP Global Supply Chain Report 2020. CDP.
Downie, J., Stubbs, W., 2012. Corporate Carbon Strategies and Greenhouse Gas Emission Assessments: The Implications of Scope 3 Emission Factor Selection: Implications of Scope 3 Emission Factors. Bus. Strat. Env. 21, 412–422. https://doi.org/10.1002/bse.1734
Gopalakrishnan, S., 2022. The why and how of assigning responsibility for supply chain emissions. Nat. Clim. Chang. 12, 1075–1077. https://doi.org/10.1038/s41558-022-01543-x
Hertwich, E.G., Wood, R., 2018. The growing importance of scope 3 greenhouse gas emissions from industry. Environ. Res. Lett. 13, 104013. https://doi.org/10.1088/1748-9326/aae19a
Klaaßen, L., Stoll, C., 2021. Harmonizing corporate carbon footprints. Nat Commun 12, 6149. https://doi.org/10.1038/s41467-021-26349-x
Yang, J., Chen, B., 2014. Carbon footprint estimation of Chinese economic sectors based on a three-tier model. Renewable and Sustainable Energy Reviews 29, 499–507. https://doi.org/10.1016/j.rser.2013.09.013
Cover image via Unsplash: Anne Nygård.
Hello Michael, It has been a long time since we spoke. Hope you are doing well. I read your article with great interest as we here at Princeton are trying to understand, quantify, and reduce our scope 3 emissions. Our initial focus is on new building construction. We are building much of the building structural frame and floor decking from wood instead of concrete and or steel. This should have direct reductions to our scope 3 emissions that no others should be able to claim credit for – right? Next we need to study our food and supply purchases. Are there any double counting issues that we should be aware of?
Tom,
First, Scope 3 is in theory about reporting emissions that are assigned to something (i.e., a GHG inventory). Not for estimating the avoided emissions resulting from an intervention/action (e.g., replacing steel with wood). If we are referring to the emissions inventory reporting under scope 3, then yes many other may be reporting the same emissions…any entity that deems themself to be part of a value chain that is associated with that framing and decking (or the building it is going into, or the business occurring in that building, etc.). If we are referring to a claim on avoided emissions due to some action of choosing wood then it is possible that anyone involved in making that intervention could try and claim. It would be something that the parties involved should agree on. Luckily this tends to be a smaller number of parties.
Michael, I found your article interesting, but I’m not sure it goes far enough. You say that ‘no other company is to report your Scope 1 emissions as their Scope 1,’ but, in fact, the GHG Protocol allows this – as a company that chooses an ‘operational control boundary’ will duplicate Scope 1 emissions of a company who is a partner in the operator’s joint venture reporting on an ‘equity share’ basis. Likewise, an energy company selling natural gas to a power producer may find emissions from the same molecules of natural gas appearing in both his Scope 3 ‘sold products’ and his ‘Scope 2’ ‘imported electricity’ emissions. When you look at scope 3 in more detail you will find that most Oil & Gas companies and their suppliers burn oil and gas in their operations and so to the extent that the oil and gas used is from the reporting company’s own production, double counting of emissions from the same molecules of oil and gas is possible in Scope 1, 2 and almost every category of Scope 3! What is missing from these ‘standards’ is a clear definition of what the objective is – is it (a) to replicate a life cycle analysis; (b) estimate financial climate risk; (c) to encourage emissions reductions along the supply chain (rather than value chain) or (d) compare companies in an industry? Different methodologies would be required for each. The danger comes when companies set targets for Scope 3 emissions – as multiple companies could end up buying offsets to cover the same hard-to-abate emissions in order to meet a target – this may sound like good news for the environment, but not for the consumer!
Malcolm,
Thanks for your excellent comments. I noted this consolidation approach potential overlap issue in the in my essay with the “*” point:
“* In practice, exclusive allocation will not perfectly occur when companies use different organizational boundary consolidation approaches.”
And as you point out, there are all sorts of smaller exceptions where double counting happens because value chains are far more complex than the simply linear model they are presented as in the Scope 3 standard. I go into these issues in detail in this previous essay.
https://ghginstitute.org/2023/12/19/what-is-greenhouse-gas-accounting-turning-away-from-lca/