Carbon Emissions: Under the MicroScope3

Scope 1, 2 and 3 are three connected but distinct lenses that can help investors conceptualise and calculate carbon emissions at company and portfolio levels. However, subjective interpretations remain an issue – one we seek to clarify in this new research paper.

The application of carbon constraints has become common practice in ESG portfolios. But is that enough to achieve our goals for a cleaner, greener world? In particular, what does it take for a portfolio to be aligned with the Paris Accord and net zero?
 
Read Article

Introduction

An objective lens is an integral component of a microscope. Essentially a powerful magnifier, these objective lenses are typically layered to create the clearest possible picture of whatever is being examined.

Responsible investors have historically had to rely on rather more subjective lenses, although improvements to the quantity and quality of available ESG data are helping move the market towards greater objectivity.

In this paper, we discuss how carbon emissions can be viewed through three connected but distinct lenses known as Scope 1, 2 and 3. These lenses can help investors conceptualise and calculate carbon emissions at company and portfolio levels, but – as we will show – subjective interpretations remain an issue. The lenses nevertheless improve the ESG microscope, so we can more clearly understand an investment strategy’s environmental profile.

To understand a company’s emissions profile, one must account for all sources of corporate emissions.

Motivation

Climate change is one of society’s greatest challenges. To have any hope of reducing the earth’s rising temperature, we must set aggressive targets to reduce Greenhouse Gas (GHG) emissions. To accurately set those targets and monitor our progress towards achieving them, we must first be able to accurately measure the emissions generated. This is a complex task but an essential one, as comprehending the data is crucial for companies and their stakeholders to manage net-zero commitments and for investors to incorporate that data in their portfolio’s carbon emissions.

To understand a company’s emissions profile, one must account for all sources of corporate emissions. This has begun in earnest, but most of the analysis focuses on emissions produced directly by company-owned facilities (Scope 1) and emissions from purchased energy in those controlled facilities (Scope 2). However, we feel there should be a greater focus on Scope 3 – the emissions attributed to a company’s value chain. Scope 3 often dwarfs the more commonly used Scope 1 and 2 emissions, and we believe it will become increasingly important and necessary for accurate GHG accounting. Regulators seem to agree. In March 2022, the Securities and Exchange Commission (SEC) announced a proposal for Scope 3 emissions to be reported for large companies with either material Scope 3 emissions or with Scope 3 emissions targets. Furthermore, the EU Taxonomy will require Scope 3 reporting commencing in 2023.

Given the rising relevance of Scope 3, we will focus in this paper on understanding these historically difficult-to-measure and consequently overlooked emissions. We discuss the current GHG Protocol accounting guidelines in the context of current data quality and how Scope 3 differs from Scope 1 and 2. We will also provide some estimation of the impact of incorporating Scope 3 in portfolio construction.

Definitions

The standard GHG carbon reporting mechanism is broken down into different segments of a company’s operations: direct emissions, indirect emissions and indirect emissions from the value chain. These are called Scope 1, Scope 2 and Scope 3 carbon emissions respectively, as detailed in Figure 1 below. Scope 1 is the direct emissions from a company’s owned operations, such as its trucks or facilities. Scope 2 and 3 are similar to each other in that they are indirect emissions, but differ in their proximity to the business. Scope 2 refers to the indirect emissions generated by the company’s purchased energy. However, Scope 2 is still within the owned operations of the business, which makes Scope 1 and 2 the easiest to measure and most frequently reported. Scope 3 emissions are those in the upstream or downstream value chain specifically not reported in Scope 2. As Scope 3 emissions come from sources outside the company’s directly owned operations, they are more difficult to estimate but can be very impactful to the overall firm’s carbon footprint.

Figure 1. Overview of GHG Protocol Scopes and Emissions Across the Value Chain

Source: Adapted from GHG Protocol,
Corporate Value Chain (Scope 3) Accounting and Reporting Standard.

Measuring Scope 3 is a challenging problem; the emissions must be estimated either by the company itself or a third party. While the GHG Protocol supplies accounting guidance, the methodologies across companies may not be standardised. The GHG Protocol breaks Scope 3 down into upstream and downstream emissions, and more specifically into 15 categories.1 Upstream emissions include those from the production of product inputs, such as purchased goods and services. In contrast, downstream emissions refer to emissions that occur from sources such as the use of a company’s products.

It is important to account for Scope 3 to ensure that Scope 1 and 2 are not being reduced at the expense of increasing Scope 3 emissions.

While one might initially think that Scope 3 is out of a company’s control, companies can make efforts to mitigate these emissions. For instance, they can use less emissionsintensive materials to build their products, thus lowering upstream emissions, or they can create a product that uses less carbon throughout its product life cycle. None of these examples would be captured in Scope 1 and 2, but nevertheless are decisions that companies can make. In addition, a company can outsource all or part of its manufacturing process, effectively reducing its Scope 1 emissions, without truly lowering their emissions footprint. Thus, it is important to account for Scope 3 to ensure that Scope 1 and 2 are not being reduced at the expense of increasing Scope 3 emissions, or vice versa.

Furthermore, Scope 3 can be a significant part of a company’s overall emissions. As Trucost data suggest, Ford and Tesla both have very low Scope 1 and 2 emissions intensity, for example, but Scope 3 accounts for the majority of their emissions (Figure 2). Incorporating only Scope 1 and 2 when evaluating such companies can miss a significant portion of their carbon emissions.

Figure 2. Ford and Tesla Carbon Intensity (FY 2018)

Problems loading this infographic? - Please click here

Source: Trucost; data for FY 2018.

Data Options

To first get a sense of the reported versus estimated breakdown of Scope 3, we turn to Sustainalytics. Focusing on the MSCI World benchmark, in FY 2019 we see a reported Scope 3 percentage in the range of 20% to 40% which is dramatically lower than that of Scope 1 and 2, which range from 50% to 80% of index constituents (Figures 3 and 4). This observation holds across sectors and regions. There is some variation by region, with a higher percentage reporting Scope 1, 2 and 3 in Europe and UK while the United States has the lowest reported Scope 3 disclosures. From a sector perspective, some of the highest-emitting sectors (utilities, materials, and energy) also have some of the highest reporting percentage across all Scopes.

Figure 3. Reported Scope Disclosures by Region (MSCI World, FY 2019)

Problems loading this infographic? - Please click here

Source: Man Group, Sustainalytics; data for FY 2019.

Figure 4. Reported Scope Disclosures by Sector (MSCI World, FY 2019)

Problems loading this infographic? - Please click here

Source: Man Group, Sustainalytics; data for FY 2019.

Across the MSCI Emerging Markets, Europe, and World indices, Scope 3 has low reported coverage (Figure 5). Due to Europe’s early advancements in ESG and climate issues, the region is leading the reporting of emissions, with just over 40% of companies reporting Scope 3 emissions. For most of the unreported Scope 3 names in the benchmarks, the values are estimated by the data vendor. Estimated Scope 3 names can account for up to 80% of the benchmark, such as in MSCI Emerging Markets.

Figure 5. Sustainalytics Scope 3 Coverage by Benchmark (FY 2019)

Problems loading this infographic? - Please click here

Source: Man Group and Sustainalytics; data for FY 2019.

Since Scope 3 data vendors may be estimating a large percentage of Scope 3 values, it is important to understand the estimation methodology.

Since Scope 3 data vendors may be estimating a large percentage of Scope 3 values, it is important to understand the estimation methodology. The remainder of the analysis in this paper focuses on emissions data from Trucost, which provides information on Scope 1 and 2 emissions intensity as well as Scope 3 intensity broken down by upstream and downstream activities. Upstream and downstream emissions intensity coverage begin in FY 2002 and 2017, respectively. For the upstream model, Trucost uses an environmentally extended input-output model; relationships between sectors are used to estimate carbon intensity in a company’s supply chain. Downstream emissions are estimated either through a bottom-up approach for the oil and gas, coal, and automotive industries, or are imputed at the subindustry level using reported emissions. As Scope 3 can be difficult to measure, there are some limitations in using largely estimated data. For instance, we do not find large variation in downstream Scope 3 intensity within subindustries, which may be the result of estimation at the subindustry level for the majority of companies. Trucost also notes as another potential issue that the estimated values may be lower than the true Scope 3, as the companies that report might be those that have lower emissions intensity.2

Relative Importance of Scope 3

To get an idea of the level of upstream Scope 3 by sector, we look at a recent cross-section of the data which highlights that energy and consumer staples have the highest emissions, although largely due to outliers (Figure 6). Real estate and communication services have the lowest emissions.

Figure 6. Upstream Scope 3 Emissions (Million Metric Tons of CO2 Equivalent)

Problems loading this infographic? - Please click here

Source: Man Group and Trucost; as of 1 January 2021.

The first question that comes to mind is whether or not Scope 3 would add additional intra-sector information to Scope 1 and 2, or if it is just a linear combination. To get a preliminary view on this, we run a contemporaneous regression of upstream Scope 3 intensity on combined Scope 1 and 2 intensity (Figure 7).

Figure 7. Regression of Upstream Scope 3 on Scope 1 & 2

Problems loading this infographic? - Please click here

Source: Man Group and Trucost; data from 1 January 2013 to 31 December 2021.

The relative importance of Scope 3 can depend on a company’s industry and business model.

The first thing to notice is that the beta is relatively low for most sectors. This highlights that the magnitude of Scope 1 and 2 intensity is quite different from Scope 3, even within sectors. Additionally, the R2 values are also very low, apart from the utilities sector. Therefore, trying to use Scope 1 and 2 to estimate the values of Scope 3 even with sectors may not provide very accurate estimates on what the actual Scope 3 would be, except for utilities. As reporting standards develop and investor demand for these values increases, so will the accuracy of Scope 3 data.

The relative importance of Scope 3 can depend on a company’s industry and business model. To examine this, we show the average percentage breakdown of Scope 1, 2, upstream and downstream carbon intensity by sector (Figure 8). At the sector level, Scope 3 accounts for around 80% of total emissions intensity in consumer staples, but less than 50% for utilities, where Scope 1 is on average the most significant contributor to carbon intensity. These variations can suggest that the incorporation of Scope 3 may even paint a different picture of what sectors or industries are actually more or less energy intensive than Scope 1 and 2 alone would show.

Figure 8. Average Percentage Breakdown of Carbon Intensity by Sector (MSCI World, FY 2018)

Problems loading this infographic? - Please click here

Source: Man Group and Trucost; data for FY 2018.

To provide more colour on this, we compare United Airlines with Chevron (Figure 9). United Airlines’ Scope 1 upstream emissions intensity accounts for 85% of its total emissions intensity, while Scope 3 intensity accounts for just 14%. This high Scope 1 proportion could be due to the company’s consumption of fuel to operate flights, while the emissions attributed to the upstream production of planes and other inputs is a smaller percentage of its emissions. In contrast, downstream emissions intensity accounts for 79% of Chevron’s total emissions intensity, as one might expect given the company’s customers’ use of oil and gas. More notably, without consideration of Scope 3 upstream and downstream, United Airlines would initially appear to be the less carbon efficient company relative to Chevron. Incorporating Scope 3 data flips this assessment.

Figure 9. Carbon Intensity Breakdown of Chevron and United Airlines, FY 2018

Problems loading this infographic? - Please click here

Source: Man Group and Trucost; data for FY 2018. Chevron’s total emissions intensity was 3,946.3 Metric tons of CO2 equivalent / Million USD revenues; for United Airlines it was 955 Metric tons of CO2 equivalent / Million USD revenues.

Contrasting Scope 3 with Scope 1 and 2

Using estimated values from Trucost, we can see quite a difference in the emissions profile of the Scope categorisation by sector for upstream emissions. Plotting Scope 1 and 2 versus upstream Scope 3 emissions intensity, we see the relationship is fairly sector dependent (Figure 10).

Figure 10a. Scope 1, 2 and Upstream 3 Carbon Intensity (Metric tons of CO2 equivalent / Million USD revenues) by Sector

Problems loading this infographic? - Please click here

Source: Man Group and Trucost; data as of 1 January 2021.

By clicking on consumer staples in the chart above, we see the sector has very low Scope 1 and 2 scores, with relatively higher Scope 3 emissions. The outliers in this sector with higher Scope 1 and 2 numbers are the packaged food and meats/agricultural subindustries. Clicking on real estate, on the other hand, reveals very small values on all three Scopes, with the more extreme emissions from hotel and diversified REITs that have buildings used for high-emitting activities such as data centres.

We can also get a good view on the movement of the various carbon Scopes over time through a deep dive into Walmart (Figure 11).

Figure 11. Walmart’s Scope Emissions (top ) and Carbon Intensity (bottom )

Problems loading this infographic? - Please click here

Problems loading this infographic? - Please click here

Source: Man Group and Trucost; data from 1 Jan 2013 to 31 Dec 2021.

While companies may have ambitious plans to reduce carbon emissions and report successes based only on Scope 1 and 2, we need to hold them accountable across all three Scopes.

An examination of Walmart’s carbon emissions illustrates why it is important to account of all sources of emissions when evaluating a company’s GHG reduction programmes. Walmart pledged to reduce the carbon emissions in its supply chain with ‘Project Gigatron’ in 2017.3 Walmart has been successful in decreasing its Scope 1 and 2 since the project started, which we see in its decreased total emissions and carbon intensity. However, its upstream Scope 3 total emissions and intensity have been rapidly increasing. While companies may have ambitious plans to reduce carbon emissions and report successes based only on Scope 1 and 2, we need to hold them accountable across all three Scopes.

Potential Problems with Scope 3

In the above sections, we have focused on highlighting the importance of Scope 3 emissions, the issues with reporting levels, spotty estimation techniques and how regulations should improve the data over time. We now turn to the issues faced when using this data to accurately compare a company’s total emissions across all three Scopes or perform aggregated group emissions levels.

Scope 3 captures indirect emissions from the supply chain and product use. For business-to-business firms, one company’s Scope 3 can make up another firm’s Scope 1 and 2. Let’s dive into the nuances of this, which can be both problematic (from a total emissions perspective) and desirable (on a comparison basis). Take, for example, a grocery store that outsources delivery of its goods to a trucking company. The trucks’ emissions would count as upstream supply-chain emissions for the grocery store, and thus be reported in Scope 3. However, the same emissions would count to the trucking company’s Scope 1. Therefore, summing Scope 1 and Scope 3 over both companies would overstate total emissions. The matter is further complicated since the trucking company is carrying goods for other entities, so not all those emissions should be attributed to the grocery store. One potential solution to understand the degree of double counting would be to use detailed supply-chain data to see what percent of the trucking company’s revenues are from different grocery store chains and use that as a proxy for allocating its Scope 1 emissions to that chain’s Scope 3. To be clear, despite an overstatement of total emissions of the grocery store and trucking company, we believe that we need to account for Scope 3 emissions not only to understand the extent of the grocery store’s carbon footprint, but also fairly compare it with potentially more vertically integrated competitors. For instance, in the case of a competitor grocery store that transports its own goods via company-owned trucks, these emissions would count towards their Scope 1. If we were only to compare the Scope 1 emissions of the two grocery store companies, the store that outsources may appear more carbon efficient as we have not accounted for the full impact of outsourced upstream emissions.

There are also clear cases where emissions overlap would not be an issue. A simple example would be a car company producing vehicles for personal use. Since the end user is not a business, these cars would not be counted in another company’s emissions. However, it is not always that clear. The auto emissions incurred by Walmart’s 2.2 million employees commuting to work are included in Walmart’s Scope 3, but the personal use of those same cars is not. However, for the manufacturer, 100% of the auto use is included in its Scope 3. In an estimation by MSCI, approximately 80% of Scope 3 emissions are counted towards another firm’s Scope 1 and 2.4

The double counting becomes even more complicated by an allocation effect. How do you allocate the secondary components of the employees’ commute to carbon emitted by recently purchased new tires, or an oil change on the way home from work? Computer usage becomes even more complicated. How do you allocate the Scope 1 and 2 GHG emitted from Amazon, Apple and Microsoft to Man Group when an employee downloads some data from Amazon Web Services to her Apple Mac using Excel?

One final consideration of double counting is the group of stocks that are being aggregated which might have a significant impact on the amount of double counting that would be present. If an industry-level analysis on carbon emissions was the goal, there may be significantly more overlap than in a diverse portfolio of 100 stocks.

Portfolio Implementation

The most practical way to implement portfolio level information on Scope 3 emissions, in our view, is either through trend analysis or relative company comparisons. These approaches compare the same emissions methodology and thus would not suffer from double counting or a requirement to know the total emissions of a group of firms, either at portfolio level or sector/industry level.

A typical approach we use in quantitative portfolio construction is to constrain the total emissions of a portfolio to be less than a certain percentage of the benchmark, or the long side less than the short side of the portfolio. This is a relative comparison that does not require a certain level of emissions to be approximated. For Scope 1 and 2, this has been seen to impact some sectors more than others, as highlighted in our previous paper. This same effect occurs when we add upstream Scope 3 into the carbon-to-market-cap constraint. The largest upstream Scope 3 emitters predominantly appear in energy, consumer staples and materials, which we can see as our exposure in these sectors changes the most. However, due to outliers in the data, we observe that a 20-40% reduction in carbon to market cap is attainable.

Below, we look at the 1,500 most liquid names in our global equity universe and constrain a hypothetical core long/short portfolio in two ways:

  1. Varying levels less than the benchmark for the long exposure
  2. Long side varying levels less than short side exposure

We show first the portfolio exposure with just Scope 1 and 2, followed by three levels of Scope 1, 2 and upstream Scope 3 constraints at just less than the benchmark or long side less than the short side and finally 60% less (Figure 12). We can see that the sectors with high emissions are hit the hardest, namely energy, utilities and materials. Energy actually goes from positive (overweight) to negative (underweight) exposure. We also see that financials and real estate go from essentially zero exposure to positive, given they have relatively low upstream Scope 3 emissions.

Figure 12. Portfolio Exposure with Increasing Carbon-to-Market-Cap Constraints

Problems loading this infographic? - Please click here

Source: Man Group and Trucost; data from 1 Jan 2013 to 31 Dec 2021. Please see the important information linked at the end of this document for additional information on hypothetical results.

Scope 3 is essential to capturing a full view of a company’s carbon emissions.

Conclusion

While there are currently issues with Scope 3 emissions data – mainly that the current reporting standards are quite loose and the data are not reported broadly enough – we do expect these to improve through increased regulation and market demand. There are nonetheless ways we can gain insight through relative comparisons across companies and sectors, as well as trend analysis. While more cumbersome to gather and interpret, Scope 3 is essential to capturing a full view of a company’s carbon emissions. Yet given the challenges when implementing portfolio analytics and constraints, we believe managers who take the time to understand the scope of the issue will have an advantage.

Scrutiny of environmental credentials and responsible investments is only likely to grow in the years ahead, so being able to analyse the total emissions of a company or a portfolio through clear and powerful lenses will be essential for investors, corporations and policymakers alike. It is this microscopic detail that gives us the truly holistic picture.

“Where the telescope ends, the microscope begins. Which of the two possesses the larger field of vision?”

So asks Victor Hugo in Les Misérables. For responsible investors, the answer is surely that both are necessary to improve our field of vision – but neither is sufficient on its own. We will keep carbon emissions under the microscope’s objective lens.

Further Reading

Wee, J., Xiang, V. (2022); “Is Your Portfolio’s Carbon Constraint Paris-Aligned?

 

1. Source: GHG Protocol, Corporate Value Chain (Scope 3) Accounting and Reporting Standard.
2. Source: Trucost, “Methodology - Trucost Scope 3 Carbon Emissions Data”.
3. For more, see https://www.theguardian.com/environment/2021/sep/13/walmart-climate-change-plan-can-it-work
4. Source: MSCI, Scope 3 Carbon Emissions: Seeing the Full Picture.

User Country: United States (237)
User Role: Public (Guest) (1)
User Access Groups:
Node Access Groups: 1