Last year, I published Carbon in Reverse, forecasting that global industrial carbon emissions would plateau at 36.90 ± 0.61 GtCO₂ by 2027 and begin structural decline. New global numbers for 2025 were released late last year by the Global Carbon Project1 they estimated emissions hit another record: 38.1 GtCO₂.

That’s 1.2 GtCO₂ above my June forecast for 2025. So what went wrong?

In this newsletter, I go deeper into the new data but have significantly expanding my carbon forecasts from a single approach to multiple independent forecast methods and the results suggest we may heading for a longer plateau of global carbon with reversal by 2030 - depending entirely how long and deep the Iran War goes on (see my last post on this).

Global Fossil Fuel Emissions

So why was my prediction for 2025 off by 1.2 billion tonnes?

Baseline revision: A meaningful share of the gap is bookkeeping, not a forecast error. The Global Carbon Budget revised historical emissions across all years, not just the most recent one. My forecast was trained on the GCB 2024 series, so the entire trajectory was anchored lower.

US weather rebound: The US had a cold winter and 58% higher gas prices reviving coal, so my forecast of a decline reversed to a 1.7% increase.

Natural gas underestimated: gas emissions grew 1.3% globally on LNG expansion.

It’s also important to recognise that global industrial emissions carry ±5% uncertainty (±1.9 GtCO₂ on 38.1), so despite the corrections from GCB, my June forecast of 36.9 sits inside this range.

The Most Important Carbon Signal

The most important signal in the new numbers isn't the headline - it's the growth rate. Since the early 2000s, annual CO₂ emissions growth has been declining at approximately −0.12 percentage points per year, from ~3.7%/yr in 2005 to ~0.8%/yr in 2025. Excluding COVID years (2020–2021), that structural deceleration is unbroken. On current trend, the growth rate turns negative around 2027. China's clean energy revolution is the primary driver, and it's moving faster than I had modelled.

New Forecasts using Four Methods

I’ve introduced four different forecast methods as they represent genuinely contrasting philosophies about how to make a projection from empirical data. Understanding the differences matters for how much weight you put on their outputs.

Growth Rate Extrapolation is the most explicit of the four. It doesn’t model the emissions level directly - it models the growth rate and integrates that forward. It fits a linear decline to the annual growth rate from 2000 to 2025, identifying the −0.12 pp/yr structural deceleration as the key signal. Uncertainty is quantified via bootstrap resampling across 2,000 regression resamples.

The strength is transparency as the assumption is stated explicitly. If you believe the the decarbonisation trend continues at its recent pace, you believe the projection. The weakness is the same - it assumes the linear decline in growth rate is the right functional form indefinitely. If clean energy deployment were to slow and emissions growth stabilised at some positive level, this method would overshoot downward.

Structural Deceleration (Gaussian Process) is the philosophical opposite. It makes no assumption about functional form. It learns the shape of the emissions time series directly from the data, using a kernel function that controls how rapidly the function can change between years - fitted with recency weighting that gives 5× weight to 2022–2025 data and 2× to 2010–2019. The result is a smooth, flexible forecast that captures the curve’s shape without committing to a parametric story. Uncertainty comes from the GP posterior predictive distribution, widening with the forecast horizon.

The two Recency-Weighted Quadratic methods sit deliberately in between. They use Bayesian polynomial regression, fitting a curved trajectory directly to the emissions level rather than its growth rate, weighted so that recent data dominates the fit. The difference between them is a single decision: how far back to look.

The 25-year window trains on 2000–2025. It sees the full arc of China’s coal boom and the subsequent deceleration, a pronounced growth-then-flattening shape that produces meaningful curvature in the polynomial and bends the forecast downward relatively early, peaking around 2028.

The 20-year window starts in 2005, after much of that early surge. Without the steep early-2000s slope, the model perceives weaker curvature and projects a shallower, later-peaking decline - the most conservative method by design, peaking around 2033. It is a deliberate upper bound on the ensemble.

Both variants use Bayesian Ridge regression, which provides a posterior distribution over the polynomial coefficients. Uncertainty bands widen with forecast horizon as parameter uncertainty propagates forward.

What forecast should you trust?

• Growth Rate Extrapolation if you have conviction in the −0.12 pp/yr trend

• Gaussian Process if you want the data to speak with minimal assumptions

• RW-Quad 25yr if you believe the full 25-year transition arc is the informative signal

• RW-Quad 20yr as a stress test: built to be conservative, and it still puts peak emissions no later than 2033

What matters most isn’t which individual method wins. It’s that all four, built on different statistical philosophies and trained on different slices of the same data, converge on peak emissions between 2027 and 2033. That range of six years — not twenty — is the signal. The spread between 27.7 and 35.4 GtCO₂ by 2050 is the honest uncertainty around it.

What the IEA says

The IEA’s World Energy Outlook 2025 Stated Policies Scenario takes a completely different approach - modelling the global energy system from the bottom up, projecting economic growth, policy implementation, and energy mix shifts country by country. It arrives at below 30 GtCO₂ by 2050.

My observation-based ensemble median for 2050 is 30.7 GtCO₂

It’s very interesting that two entirely independent methods point to the same number. It means that the underlying decarbonisation trajectory of the global economy is robust enough to show up regardless of how you look at it. The near-term paths differ but by 2050 it seems like a robust result.

My Prediction for 2026: with Big Iran Disclaimer

The GCB 2026 will be released this December and I forecast a likely estimate: 38.1 GtCO₂ (range: 37.6–38.3).

Big Disclaimer: The Iran War has up to 1GtCO₂ of reductions depending on how long and deep into their energy infrastructure. See my last post for details.

So this year is looking flat (or marginally lower depending on Iran) - the first year without meaningful growth since the pandemic.

If correct, it will be the first evidence of peak emissions visible in the annual data. If I’m wrong - if 2026 comes in above 38.5 despite the Iran War - it means the structural forces I’m tracking are weaker than the data suggests, and the conservative end of my forecast range deserves more weight.

Take-Home

The turning point I described in June is still here. Four fundamentally different empirical approaches converge on peak emissions between 2027 and 2033, declining to an ensemble median of 30.7 GtCO₂ by 2050 - with an honest range of 27.7 to 35.4.

We’ll get the first real test this December and who knows what happens in Iran. I’ve put a number on it: 38.1 GtCO₂. Let’s see.

Until next time,

Ben

I’d be grateful if you could hit the like button ❤️ below! It helps more people discover these ideas and lets me know what’s connecting with readers.

Global private equity (PE) assets are now over $8 trillion, according to some estimates. Using GDP as a scalar of global carbon emissions, that would mean the footprint for private equity would be about 3 billion tonnes globally. The vast majority of unlisted companies simply do not report carbon emissions - so there is no way of knowing the carbon risk of this hidden asset class for investors.

Previously, the only way to get a footprint for PE investments was through ‘Industry Factor Methods’. But as I show, those traditional methods are wildly inaccurate and biased. So, we at Emmi have pioneered a new, more accurate, unbiased approach to help investors uncover their hidden PE footprint and climate transition risk exposures.

In this week’s newsletter, we take a deeper look into the old and new ways of estimating emissions from private equities..

Share

The Traditional Industry Factor Method

The most common approach to estimating private company emissions is to use Spend-based Industry Emission factors, as they are somewhat easy to apply across many different companies. These directly translate financial values into CO2e per dollar spent or earned. They rely on Environmentally-extended Input-Output (EEIO) models that attempt to map the flow of goods and services in the global economy.

The Exiobase model1, for example, includes 163 sectors across 44 countries and five ‘rest of the world’ regions, generating industry emission factors based on the economic output and total emissions for each combination of sector and country. This has become the default method, but it introduces significant problems for investors:

1. All companies within a sector are treated the same

EEIO models are at a sector-level, so companies within as sector have the same emission factor.

Carbon emissions grow non-linearly as companies get larger. Using an average industry emission factor means that all companies are treated equally, no matter what suppliers, processes, or efficiencies they have in place. This introduces huge biases.

Steel Example

Taking an example of the United States steel manufacturing sector, let’s assume the EEIO model suggests a sector revenue of $100 billion, which produces 500 million tonnes of CO2e. Therefore, the resulting industry emission factor is 5 tonnes CO2e per $1000 revenue.

So for investors, to estimate emissions for a specific steel company (Company A) you multiply their revenue by the factor (e.g. $2 billion x 5 tonnes/$1000 = 10 million tonnes CO2e).

But let’s say you are also invested in Company B in the United States, which has the same revenue as Company A, but uses electric arc furnaces for more efficient production. Using EEIO industry factors means both companies will have the same emissions estimate, even though Company B may likely have far less emissions since they have invested in better technology.

2. Most material downstream emissions aren’t covered

For EEIO models, they typically only cover upstream and production emissions, they don’t cover the emissions used by the product.

Our research, published in 20232, shows that downstream Scope 3 emissions categories ‘Product Use’ and ‘Product Processing’ are by far the most material, covering 60% of emissions, while ‘upstream’ categories only contributed ~10%. Some data providers use an ‘uplift factor’ to account for the large missing downstream component. However, this likely introduces even more systematic biases. Using industry factor models doesn’t allow the breakdown of Scope 1, 2 or 3 emissions required to understand carbon risk and reporting mandates.

3. No understanding of uncertainties

There has been no easy way to assess the accuracy of using EEIO industry factors at the company level.

The output of EEIO factors is taken as given, which means there has been no systematic quantification and understanding of emission uncertainties. Therefore, this black-box approach makes assessing the sensitivity and materiality of holdings within a portfolio difficult.

A New Machine Learning Approach

Since 2021, Emmi has partnered with pioneering climate finance researchers from the University of Otago and Griffith University to build and develop machine-learning methods that give better accuracy and coverage of Scope 1, 2 and 3 emissions for listed companies that don’t report emissions3. We have now extended this machine-learning approach to cover private equities, releasing our full methods here.

The approach utilises the reported emissions for over 10,000 listed companies globally from FactSet ESG Solutions. Then, it optimises machine-learning models based on typical financial data-input constraints observed for private equities (eg Revenue or Total Asset Values). The machine learning framework is designed for quality, flexibility, coverage, and timeliness.

This new approach explicitly breaks down Scope 1, Scope 2 and all of the most material categories of Scope 3 for each company. Aside from that, other significant advantages over the traditional industry factor method include:

1. No more sector averages!

The ML method uses actual emissions patterns from reported company data, not sector averages. The algorithm more intelligently learns that emissions may not always increase at the same rate as revenue within and across sectors and countries.

For instance, in the steel manufacturing industry, while EEIO emission factor may assign a universal ratio of 5 tonnes per $1000, Emmi ML might learn that emissions per revenue increase slowly up to $500 million before increasing at a faster rate between $500 million and $2 billion, with diminishing increases beyond $2 billion. These growth patterns would vary among countries and sub-classifications within the steel industry.

To demonstrate this concept, we present the Scope 1 emissions as a function of revenue for Utilities, which has the highest emissions (around 2.3 billion tonnes) of any industry. We show the reported 'actual' emissions compared to the Industry Factor Estimate and the Emmi ML estimate.

The Utilities chart shows how non-linear the relationship is between revenue and reported emissions. While the ML estimates accurately follow this non-linearity, the average industry factor method wildly overestimates the carbon footprint for most companies, particularly between $100 million-$1 billion.

One limitation of the ML method is for companies with less than $100 million in revenue, as there is not enough training data available to make accurate estimates; therefore, these footprints tail-off ~1,000 tonnes. However, at an investor level, these smaller footprints have minimal impact on overall financed emissions calculations.

2. Up to 7 times more accurate

Overall, when comparing methods across the reported universe, we find the Emmi ML approach outperforms industry factors by nearly seven times, with an MDAPE of ~70% for the ML compared to ~480% for industry factors (see chart). We demonstrate that Emmi ML models have the ability to use reported data more intelligently, learning the complex relationships between classification, geography, scale, and emissions.

3. Transparent uncertainties for PCAF or auditors

The benefit of using real world data and models is a clear understanding of the uncertainties for each company. This is important for PCAF alignment, but also for the first time to assess materiality given those inherent uncertainties. Previously, EEIO factors gave no understanding of uncertainties, making it impossible to assess materiality. Using our ML method now gives investors (and auditors) a clear pathway to assess uncertainties and materiality within a portfolio.

The Way Forward for PE Investors

We present a new way to estimate Scope 1, 2 and 3 emissions for any unlisted company in the world, which is shown to be up to 7 times more accurate than traditional investor methods. The 3 billion-tonne hidden private equity footprint is important for investors to understand their climate risk, as well as to meet their new reporting requirements. You can download the full analysis here or listen to a 10-minute podcast summarising the method below:

Emmi

Company Emissions Estimates for Private Markets

Private Equity represents as much as 20 per cent of a multi-asset portfolio and reached a record-high value of $8.2 trillion in 2023…

Listen now

a year ago · Emmi Solutions

If you’d like me to talk about your PE footprint and climate risk, just reply to this email.

Until next time,

Ben

Back in 2009, I wrote a book called ‘The Clean Industrial Revolution’ which argued that the world will inevitably price carbon emissions and create a new transformative phase of low-carbon economic growth. Back then there was no macro indications of a decoupling between carbon and the economy.

Fifteen years on, in todays newsletter, I wanted to investigate whether the world has in fact started to decouple it’s economic growth from carbon. This is important for investors since it provides one piece of evidence that shadow carbon pricing is operating in markets across the world, generating new carbon alpha opportunities.

Global GDP and Carbon Emissions

Since the year 2000, GDP and global carbon emissions seemed intimately linked.

Absolute numbers can obscure the trends so lets dig a bit deeper into the carbon intensity of GDP over time (CO2 emissions per unit of Inflation-Adjusted GDP). To visualise the trends and quantify progress, I’ve calculated the average emissions intensity during the 2000-2008 period. This baseline carbon intensity was then used to project what emissions would have been in subsequent years if the economy had maintained the same carbon intensity in the 2000s (see chart below).

Global Industrial Carbon Emissions (Billion Tonne)

The results reveal a significant decoupling of economic growth from emissions growth over time starting in about 2012. Today, emissions should be 45.2 billion tonnes under carbon intensities from the 2000s - which is nearly 20% higher than actual emissions (37.8 Billion Tonnes).

The cumulative emissions savings of this decoupling is substantial. Between 2010 and 2023, actual cumulative emissions were 500 Billion Tonnes CO2, while the projected emissions based on the 2000-2008 relationship would have been 552 Billion Tonnes CO2. That means the world has saved ~52 Billion Tonnes CO2 which is nearly 1.5 years of emissions.

Global Carbon Intensity (Billion Tonnes per trillion of GDP)

Inflation-adjusted carbon intensity gives you a much clearer perspective. During the early 2000s, the intensity remained relatively stable, hovering around 0.43 Billion Tonnes of CO2 per trillion dollars of GDP (in constant 2015 USD).

The transition begins around 2011 when emissions intensity starts a more consistent downward trend. This marks the beginning of a decoupling phase where economic growth became increasingly detached from proportional increases in carbon emissions. Between 2010 and 2019, there was a steady improvement in carbon efficiency, with the global economy requiring progressively less carbon to generate the same economic output.

The COVID-19 pandemic in 2020 appears as a brief disruption in the trend, but the overall trajectory of improvement continued afterward. By 2023, the emissions intensity had fallen to about 0.13 billion tonnes per trillion dollars of GDP, a 70% reduction from the 2000-2008 average.

Global Carbon Intensity of Growth (Billion Tonnes per trillion of New GDP)

The carbon intensity of new growth clearly shows the new normal post 2011, where carbon intensity of growth drops by over 50% from prior to the 2010. Several years (2015, 2016, 2019) even show near-zero or slightly negative values, indicating economic growth with no additional emissions.

The post-pandemic period (2021-2022) shows a partial reversal of this trend, with the average rising to 0.311 Billion Tonnes CO2 per trillion USD. However by 2023, the lower carbon intensity of growth (0.130), suggesting that recent economic expansion may be returning to a more sustainable pattern after the post-pandemic rebound.

What is causing the carbon decoupling?

During the 2010s, a number of factors played a role across major markets:

1. Companies setting emission reduction targets, deploying new tech and improving carbon efficiencies

2. Consumers demanding and switching to low carbon products (ie renewable energy)

3. Energy markets shifting away from coal to lower carbon forms of energy

4. Government schemes, incentives and mandates (particularly in China)

5. Many major banks internalising carbon pricing for new project finance

   ---

Take-Home

Despite all the headline news you read, the world’s economy is shifting away from carbon. Just one example of the new low carbon growth era on the local level is how Microsoft have signed a 20yr deal to reopen Pennsylvania's Three Mile Island nuclear plant, which plans to use it’s carbon-free energy to power its AI data centers1. Although the nuclear plant was uneconomic, carbon pricing makes it economic.

The next phase of low carbon growth is already here, so it’s time to integrate carbon into your investments.

Until next time,

Ben

Heading into 2025, many people are wanting to understand the climate implications of a new Trump administration. I was interviewed recently about this and I have a different, more optimistic, view from what you have likely heard. Here’s a clip from me about it.

During Trump’s first term he did a number of things including withdrawing from the Paris Climate Agreement, rolling back various Environmental Protection Agency regulations on emissions, promoted fossil fuel development, reduced restrictions on drilling and reduced fuel efficiency standards for vehicles. Sounds like a lot, so what impact did those measures have on carbon emissions? In this week’s newsletter I want to go much deeper into why I am more optimistic than others about the next four years.

The USA is Decarbonising, Trump or not..

Between 2007 and 2022, total emissions in the USA dropped from ~7,500 to ~6,300 million tonnes CO2 equivalent per year, representing a reduction of approximately ~16%.

USA Total Greenhouse Emissions (Million-Tonnes)

USA Sector Greenhouse Emissions (Million-Tonnes)

Looking more into the most material sectors, the electric power industry has decarbonised by a whooping 36% since 2007 (see above). What has caused this decarbonisation and what happened during Trumps 1st term?

Electricity decarbonisation (-36%)

The huge decarbonisation of electricity in the USA is from a shift from coal to natural gas and renewables. According to the Global Coal Plant Tracker database¹, there have been more than 300 retirements of coal-fired power stations since 2007. New generation capacity has been gas and renewables, which have a much lower carbon intensity.

Number of US Coal Power Retirements

Despite his ‘pro-coal’ policies, about 29 coal-fired power stations were retired per year under the first Trump administration. The total of 115 retirements during this four year period accounts for about one-third of all coal-plant retirements since 2007.

What is clear from this is that politicians aren’t dictating trends in the low-carbon transition within electricity supply - the massive decarbonisation has been driven by economic and technology factors, state policies and consumers demanding more low carbon energy options.

Annual % decarbonisation for each sector since 2007

Each year the electricity sector in the USA is decarbonising -2.7% in Trump years and non-Trump years. Politics isn’t driving the energy transition.

Transport decarbonisation (-8%)

Transportation emissions in the USA have proven more difficult to reduce. They peaked at 1,967 million metric tonnes of CO2 equivalent per year in 2007 but, outside of the Global Financial Crisis and COVID, they remained around 1,800 million metric tonnes, a reduction of 8.4% between 2007 and 2022.

The power generation sector has achieved significant emissions reductions by switching from coal to natural gas and renewables. However, transportation emissions have remained stubbornly high because there has been a much lower market penetration of low-carbon alternatives. Transportation has now overtaken electric power as the largest source of emissions in the USA, highlighting the challenges of decarbonising a sector that remains heavily dependent on fossil fuels despite improvements in vehicle efficiency.

Transportation now the largest Footprint

That said, despite ‘pro-drilling’ policies, decarbonisation of transport still occurred under the first Trump administration.

Decarbonisation under Trumps 2nd Term

The Shift from Coal Continues

Number of planned coal retirements

Looking ahead to Trump’s second term (2025-2028), according to the Global Coal Plant Tracker, there are 55 coal plants currently scheduled to retire in the period, irrespective of Trumps posturing around coal. These retirements would prevent approximately 229.6 million tonnes of lifetime CO2 emissions. The retirements are distributed across the term, 15 plants in 2025, 12 in 2026, 10 in 2027, and reaching a peak of 18 plants in 2028.

These data suggest that the transition away from coal-fired power continues to progress, driven by economic and technological factors.

Decarbonising Transport - The Musk Factor

Tesla CEO Elon Musk has become one of the most influential figures for Donald Trump. Although Musk has softened his stance on climate change, electrification of transport is still very much close to his heart.

Donald Trump expressed opposition to electric vehicles during his campaign, but his new alliance with Elon Musk has raised doubts about his stance on EVs once he officially becomes president in January 2025.

It looks like government tax subsidies for EVs put in place under the Biden administration will be removed by Trump. In addition, his threatened tariffs on Chinese imports would create serious barriers to the lowest cost EVs entering the US market.

However, given Musk’s global popularity and influence, along with his electrification of transport agenda, I would argue there could be more not less investment in low-carbon fuels, battery technology, EV infrastructure, and even backflips on government incentives or tariffs in transport in Trump’s 2nd term.

The net impact of these factors is difficult to predict. If, as a consequence of Musk’s prominence, electrification of transport accelerates under the 2nd Trump term, then that could create a larger decarbonisation pathway than under any other President, because transport has the biggest sectoral emissions footprint in the USA.

Most people are assuming there will be no climate progress in the US over the next four years. The carbon data suggest otherwise - markets and technologies are driving decarbonisation trends irrespective of politics. That’s very important for investors who are considering the carbon exposure of their portfolios.

Thanks for reading. Until next time.

Ben

The annual human fossil fuel CO₂ footprint is about 36.8 billion tonnes1. This includes direct emissions from all listed companies, unlisted companies, state-owned enterprises and our own individual direct emissions from using fossil fuels. But how much does each segment contribute to the global footprint? And are they decarbonising?

Today I’m excited to share with you our white-paper on estimating the listed universe corporate carbon footprint, showing corporate emissions may have peaked in 2022. You can download the paper here.

The white-paper goes into detail about our unique approach to estimate Scope 1, 2 and 3 emissions for up to 45,000 listed companies globally. It’s quite technical, so we’ve created a new podcast channel that distills it down in 10minutes. You can listen to that on various channels here:

Emmi

Estimating The Corporate Carbon Footprint: A Meta-Model Machine Learning Approach

Our new research suggests corporate emissions peaked in 2022. In this 10-minute AI-enhanced breakdown, we explore the machine-learning approach behind our findings, examining how we analyse emissions across 45,000 listed companies. Using Google's NotebookLM, we distil complex methodology into key technical insights about corporate decarbonisation trends…

Listen now

a year ago

Listen on Spotify

Listen on Apple Podcasts

The Goal: Fixing the Carbon Gap

Quantifying a baseline footprint for the corporate universe has been difficult given the lack of quality reported data. The number of companies disclosing however is growing, yet some 83% of publicly listed companies don’t report their Scope 1, 2 & 3 emissions2.

Over the past few years, we’ve collaborated with leading climate financial researchers to develop a unique, robust method to estimate corporate emissions for investors across the listed universe. Today we share this methodology and results in the white-paper.

Key Takeaways

• We introduce a new, robust machine learning ‘Meta-Model’ approach to estimate Scope 1, 2 + 3 emissions for up to 45,000 publicly listed companies.

• Our results are validated by independent reported data, with a median error rate of 6-10% for the most carbon-intensive sectors. For a typical investor, we demonstrate portfolio-level uncertainty of ~15% for Scope 1, ~17% for Scope 2  and ~19% for Scope 3.

• We estimate the cumulative footprint for the listed equity universe to be 11.4 billion tonnes in 2023, a reduction of ~2.5% from 2022. This implies the listed corporate universe accounts for ~31% of the industrial fossil-fuel footprint.

• Although we find the listed universe to be decarbonising overall, it’s decline is within the uncertainties of our approach, therefore this decarbonisation trend will need to be confirmed in subsequent years.

• If however the decarbonisation trend was validated and continued, then the corporate listed universe will have reduced its overall footprint by nearly 20% by the year 2030.

• Given our results, it’s likely that the listed corporate universe footprint may have reached peak emissions, implying that the growth in industrial emissions must be driven from either private companies, state-owned enterprises and/or individual fossil-fuel use.

Reach out if you have any questions. Until next time.

Ben

In my last newsletter I shared an analysis of what companies and sectors were most exposed to Scope 3 footprint mandates. Although emissions exposures can provide a guide - they don’t tell us anything about the financial exposure of those emissions. Including Scope 3 into climate mandates is something that is being debated by some regulators like the US SEC, whereas some companies are actively avoiding disclosing Scope 3 targets.

So what is the financial impact from excluding Scope 3 in climate transition risk? What sectors have a material financial liability from Scope 3 emissions? In this newsletter I investigate more with examples of Amazon and Airbus.

Share

Scope 3 carbon liabilities by sector?

At Emmi, we calculate the climate transition risk for any asset across multiple different climate scenarios with our Carbon Diagnostics product. We can also include or exclude Scope 3 emissions (which some regulators are debating). For this analysis, I’ve run climate transition risk across all public companies including and excluding Scope 3 emissions under the IPCC Net-Zero scenario.

The differences are huge depending on the sector. To make it easier to show, I’ve calculated the average % of enterprise value by 2030 saved by not including Scope 3 carbon liabilities. So those companies with largest Scope 3 exposures as a function of their Enterprise Value save the most. Sectors like Motor Vehicles, Building and Construction and Airline Manufacturing with large Scope 3 exposures will save more than 70% of their Enterprise Value if Scope 3 is not included in climate transition risk.

On the other hand, Steel and Electric Utilities have far less to save by not including Scope 3, since most of their carbon risk comes from Scope 1 or 2 emissions.

You may be wondering why last week I showed that the Scope 3 exposure index was largest with the banks and financial institutions, yet above demonstrates the least savings when including Scope 3? This shows exactly why larger emissions footprints don’t necessarily translate to larger carbon financial risk. For banks, Scope 3 is by far their largest footprint, yet the financial risk as a function of their enormous enterprise value is low. That means even when including Scope 3, banks carbon transition financial risk is low relative to other industries.

Should Scope 3 be included for climate risk?

Scope 3 emissions are difficult to accurately measure while companies have little direct control over their reductions since those emissions occur upstream and downstream of a company’s products. However as carbon pricing and regulation get sharper, the market is shifting away from carbon intensive products towards lower carbon products/alternatives. So including Scope 3 is financially prudent to truly understand carbon risk.

To demonstrate how important Scope 3 is, lets take the examples from my earlier newsletter using electricity utility Southern Company, aerospace manufacturer Airbus and chemical company Linde. I’ll include Amazon, a large consumer trade business for comparison too.

Using Emmi’s carbon financial risk models, by 2030 under IPCC Net-Zero, there are some huge divergences across those companies. For Southern Company whose emissions are particularly high in Scope 1, including Scope 3 doesn’t change their large financial exposure with 100% of their enterprise value being potentially eroded by 2030.

The largest impact is for Airbus, where virtually all of their emissions are Scope 3 - the downstream emissions from the airlines who create the emissions when using the Airbus airplanes. The change in climate risk goes from virtually no impact when not including Scope 3 - to a situation where 100% of their enterprise value is at risk when including Scope 3.

Omitting Scope 3 for Airbus doesn’t make sense because the carbon pricing/regulation will hit the Scope 1 of their customers (ie all the airlines). That will then theoretically cause a drop in airline demand for Airbus without a low carbon option - so including Scope 3 allows investors to capture that risk.

A company like Amazon, also has little climate exposure without Scope 3 - yet over 20% of their enormous Enterprise Value is at risk when accounting for their Scope 3 emissions by 2030.

Accuracy vs Materiality

Let’s not obsess over accuracy vs materiality. Measuring Scope 3 emissions doesn’t need to be accurate - it simply needs to be at a level whereby it allows a materiality check on their climate risk.

Even if Scope 3 emissions are >70% uncertain (yes they are), knowing those uncertainties still gives us the ability to assess the materiality of an assets climate risk. At Emmi we predict Scope 3 for all companies and know the uncertainty. That allows us to quickly and efficiently inform companies or investors whether they should invest into more accurate Scope 3 reporting.

Scope 3 doesn’t have a material financial risk for most companies, so they do not need to invest heavily into better reporting or consultants to measure their Scope 3. But for others like Amazon, Airbus and Linde - we find Scope 3 is hugely material in contributing to their future climate risk. They do need to invest into understanding their Scope 3.

Including Scope 3 emissions is critical for investors and regulators to provide more robust climate insights for decision-making, risk management, and sustainable investment strategies. However, to avoid lots of wasted resources, companies should only invest into measuring Scope 3 if they have a material exposure - since that is what the regulators require. How to do that in an efficient way?

If you are a company and want to quickly assess your Scope 3 emissions liabilities and whether they are material to your climate exposure, we are happy to help - just reply to this or contact me here.

Thanks for reading

Ben

There is a perception that corporate disclosures of carbon emissions are far more accurate than those that come from model estimates. Is there an accuracy-gap? For investors, this perception limits carbon integration into investment decision making and clouds our understanding of carbon exposure in markets.

So in this newsletter I want to explore how much emphasis we should place on data that is reported vs estimated. In reality, it’s a false choice. As I argue, both have limitations but are indispensable for investors in the foreseeable future. So rather than choose a side I provide perspective on how the next generation of carbon coverage can evolve using both.

Share

What are ‘Reported’ Emissions?

Under the GHG protocol, companies measure and report carbon emissions using two methods: spend-based or activity-based. It’s very important to understand that reported emissions are also estimates and rely on emission intensity factors (which are also subject to large uncertainties).

Spend-Based Estimates

The spend-based method uses the value of goods and services purchased by an organization to estimate the amount released into the atmosphere. This method involves multiplying the organization's spending in an area (eg. fuel spend) by emissions intensity factors of the goods or services purchased (eg. CO2 per $ of fuel burnt).

Activity-Based Estimates

The activity-based method uses the units of the product or material to estimate the amount released into the atmosphere. So it involves quantifying the units or volume of a product used (eg tonnes of fuel) and using emission intensity factors (eg CO2 per tonne of fuel) to estimate the emissions associated with that activity.

Spend-based estimates are easier to use and get better coverage across a business while activity-based estimates are more accurate but impractical for most companies to use given the complexity of their products and supply chains. So in reality, the approach taken is mostly spend-based estimates with activity-based where they can.

Limitations of Reported Data for Investors

Accuracy

The biggest misnomer in the financial industry is the notion that reported data is very accurate. Even if reported emissions are audited, it’s very important to understand that:

Audits cover process not accuracy

The process for corporate carbon accounting has a significant potential uncertainty depending on the methods used, the data from millions of emission factors and the accuracy of categorising a product.

For many enterprises the errors in their reported emissions is significant. In 2021, a Boston Consulting survey of 1,290 executives found they estimate a 30-40% error in their reported carbon footprint calculations. In reality, errors will vary significantly depending on the company - unfortunately there is no systematic way to assess them across all companies.

Under-Reporting

Even if companies report emissions, they often don’t report the most complex and/or emissions intensive activities. We analysed all reported data between 2016 and 2021 for over 12,000 companies. For scope 3, companies often reported ‘Business Travel’ but didn’t report on far more intensive categories like ‘Use of Sold Products’. You can read more about that research here.

Portfolio Coverage is Small

Only 10% of the worlds public companies report emissions. For a typical multi-asset portfolio, listed equities and bonds make up about one-third of asset allocation. So relying only on reported data would typically cover less than 30% of a typical multi-asset portfolio.

Old Data

Typically, reported data is 12-18 months old.

What are ‘Estimated’ Emissions?

When a company doesn’t report emissions, what can analysts and investors do? Typically analysts and investors don’t have access to spending or activity-based data from within a company, so revenue is typically used.

Emission-Factor Estimates

Since public companies report their earnings to the market, spend-based methods can be used using revenue. This method involves multiplying the organization's revenue by emissions intensity factors revenue in similar companies. Emission factors are typically sourced using EEIO (Environmentally-Extended Input Output) models from organisations like the US EPA or the EU.

Machine-learning Estimates

A new approach that we have helped pioneer at Emmi is to use the reported data-set to train machine-learning models that predict emissions for any company from standard financial and business information. You can read more about this approach here. This approach doesn’t rely on static emission factors, but rather trains and adapts with the new reported data each year.

Limitations of Carbon Estimates for Investors

Accuracy

Relying on revenue and business information for a company is likely less accurate than spend or activity-based estimates. In reality, it’s hard to assess the difference in accuracy between reported and estimated. What we do know is that the machine-learning based approach is up to 30% more accurate than the simple revenue-based approaches.

Overall for larger companies as an example, the accuracy of our machine-learning predictions are within 5-12% for Scope 1 and 15-30% for Scope 2 (see below). Interesting this is within the errors suggested by companies themselves in how they estimate reported emissions.

A better-hybrid approach for investors

For investors, there is no choice between reported and estimated emissions: you need to use both. For us at Emmi however, we use the reported numbers to build the next-generation of data prediction models in order to cover your portfolio.

Investors aren’t the government - so getting accuracy down to the last possible tonne of emissions per holding is not a priority. For investors, greater portfolio coverage and knowledge of the uncertainty allow for the best decisions to be made.

In reality, 20% of your portfolio will have a material exposure to carbon, and with new machine-learning estimates you now have a systematic way to check virtually all of your portfolio - then make decisions to go deeper or take actions when needed.

When reported emissions aren’t useful?

Take COVID-19 for example. The shutdown and financial collapse of airlines was immediate. Emissions fell off a cliff in a matter of months - but given reported emissions are typically 12-18months lagged - reported emissions were too old to be useful. On the other hand, our quarterly emission prediction models captured these timely drops well (see below).

Beyond better carbon information for investment decisions, greater coverage and timeliness of carbon data is important since carbon reporting standards like the ISSB are seeking carbon to be synchronised on a reporting schedule with finance. That isn’t possible with reported emissions and unlikely for many years.

But using the reported data-set as training for better machine-learning models is the only approach best equipped to allow for carbon disclosures to occur on a timescale relevant for finance/ISSB goals. Once reported emissions are eventually disclosed 12-18months after the fact, our models are recalibrated and learn from the new data - becoming even more accurate and useful for investors.

Given the uncertainties and complexities of reporting carbon across the market, reporting gaps in carbon are huge and will continue to be huge for many years. In the meantime, next-generation machine learning estimates use the growing reported data in order to fill the gaps and can do so within similar uncertainties than reported data. If you focus just on reported emissions, you miss a big opportunity to get better information to make investment decisions.

Thanks for reading. Until next time.

Ben

With all the buzz around ChatGPT, what about tools for climate finance? Can machine learning help financial markets get better data to make decisions?

As ChatGPT shows, the application of machine learning to complex problems is immensely valuable - and at Emmi we believe it will transform our understanding of carbon in markets to help solve for big gaps in coverage, timeliness and consistency of data to make decisions.

So we’ve been collaborating with leading climate financial researchers from the University of Otago and Griffith University to research and develop machine learning algorithms to better predict corporate carbon emissions. We’ve found that machine learning can dramatically improve methods for estimating corporate carbon emissions in comparison to traditional methods.

In this newsletter I wanted to share what we have learnt and give you a working example of these machine learning predictions with the top 50 European companies. If you are interested in testing out our models for yourself- just hit ‘reply’ to me in your inbox or on our website here.

Share

BACKGROUND

The Corporate Carbon Data Gap

For publicly listed companies, the vast majority (>90%) don’t report their operational carbon emissions (Scope 1 or 2).  For unlisted companies and SMEs, carbon reporting is virtually non-existent so far. Global coverage of carbon emissions for companies does not exist.  Predictive approaches for estimating carbon emissions are critical to fill the large gap for financial institutions and investors. 

To overcome the carbon gaps for investors, traditional approaches tend to use sector averages or linear regression to fill carbon gaps. This is done by calculating the carbon intensity of revenue for sectors, then using revenue as a scaling factor to estimate carbon for missing data. This approach is the most widely used in the market and one is important to compare with.

MACHINE-LEARNING PREDICTIONS

A new approach to estimate company emissions

Machine learning is a catch-phrase for mathematical algorithms that ‘learn’ by classifying patterns in data for a given set of inputs and predictor variables. This algorithmic approach has the potential to find and extract the most predictive information from a given data-set. Overall machine learning has the potential to improve the quality of estimates with broader coverage. Is that true?

In partnership with our university partners, our data science team at Emmi have developed novel new machine learning methods to estimate corporate carbon emissions. If you want details about our approach, I’ve shared a summary at the end of this newsletter and you can also read two published research papers: 

1. Predicting corporate carbon footprints for climate finance risk analyses: A machine learning approach

2. Scope 3 Emissions: Data Quality and Machine Learning Prediction Accuracy

Key Finding from our Research:

Many of our  best performing models (linear tree, gradient boost and meta-model) predict Scope 1+2 emissions for the world's largest companies within ~10%. This is a significant improvement from traditional methods that take sector-based averages and use revenue as a scaling factor. So let’s test it out with a case study.

CASE STUDY

Predicting emissions for the top 50 European companies

To help you understand our methods, we have applied our machine-learning algorithms to estimate emissions for the largest 50 companies in Europe for the year 2020. We can compare our predictions with reported numbers1 . Here’s what we found.

Scope 1 Emissions Predictions

Using our best models along with financial data for 2020 we predict Scope 1 carbon emissions and compare it to reported. I’ve charted our predictions and reported values for the top 20 largest emitters in Europe.

Across the 50 companies our models predict Scope 1 to within 11% MdAPE (median absolute percentage error) - similar to our overall research. Machine learning estimates tend to be slightly higher than reported values, (Figure 2), however breaking it down to the most influential sectors makes an important difference (Figure 3).

We find that sectors with highest aggregate emissions tend to have significantly higher predictive skill than sectors like IT or real estate that have a much lower magnitude of Scope 1 emissions. For example, prediction errors within the energy, utilities and materials sectors were within 2-6%.

Scope 2 Emissions Predictions

For Scope 2 emissions, the distribution of across the top 50 companies in Europe are quite different (Figure 4), with chemical companies contributing most due to their dependency on electricity and combustion for their processing.

Similar for Scope 1, we find predictive skill to be higher amongst sectors with highest proportional emissions. For Scope 2, the materials sector contributes nearly half of the overall Scope 2 footprint, while our models accurately predict this sector to within 11%. This is a very encouraging result.

Beyond the materials sector, our Scope 2 predictions seem to underestimate reported emissions ~39% (Figure 5). Since 2020 was a unique outlier in terms of the global economy and energy use - this may partly explain this since patterns in the past may not be sufficient enough to predict trends in such an anomalous year. Despite this, we continue to improve the accuracy of all our models each year.

Most Influential Sectors

Our models work the best where it matters. The energy, utilities and materials sectors make up 90% of Scope 1 emissions and 60% of Scope 2 within Europes top 50 companies. For these most influential sectors, our machine learning models predict Scope 1 to within 6% and Scope 2 within 16% (Figure 6). These are encouraging and gives us confidence in applying our machine learning approach to both fill carbon gaps in coverage and timeliness.

Details of our Machine Learning Methods

What carbon data-set did we use?

Any data model requires a clean, robust data-set to ‘train’ the models. For our training data-set, corporate carbon emissions data between 2018-2021 were used for over 5000 companies via data providers (CDP and ISS). We sourced financial information for these companies via FactSet.

Often financial data reported from companies is incomplete. For example, revenue or employee numbers are missing. Ignoring this missing data limits the training data-set. To maximise the amount of training data used for our models we impute input financial data - which simply means we replace this missing data using more advanced imputation methods such as Multiple Imputation by Chained Equations (MICE).

What models were explored?

We applied a range of different machine learning models and found the best models include, Gradient Boosting, Random Forest, Linear Tree and a meta-model median of our best 5 models. We also optimised the hyperparameters of these models to obtain the most robust models. Here’s a quick summary of how the best models work.

Random Forest

A random forest is a type of classification algorithm made up of a collection of several decision tree models that work together to make predictions. Each decision tree in the random forest is built from a random sample of the data, and at each step of building the tree, a random subset of the features is considered for splitting the data. This means that each tree in the forest is different and is able to capture different aspects of the data. The final prediction is made by averaging the predictions of all the trees in the forest. This combination of many decision trees results in a model that is more robust and less prone to overfitting than an individual decision tree.

Linear Tree

Linear Trees combine the learning ability of decision trees with the predictive and interpretive power of linear models. Data is split using decision rules, similar to tree-based algorithms, and the quality of the splits is measured by fitting linear models at each node. This means that the models in the leaves of the tree are linear, as opposed to constant predictions as in traditional decision trees. Here’s a great write-up of this method.

Gradient Boost

Gradient boosting attempts to improve existing models by minimising a loss function that iteratively tests different sets of model input parameters known as hyperparameters, such as how deep a tree should go. The optimiser then runs the model many times with a different set of hyperparameters, and evaluate the performance of the model using a validation set. The optimizer will return the best set of hyperparameters that will be used to train the final model. This process can be computationally expensive, but it can lead to a significant improvement in the performance of the model.

What is a Meta-Model?

Our meta-models take the mean, median or max of the best 3-10 well trained models such as those described above. By combining the predictions of multiple models, the overall prediction accuracy and model stability is improved. By using the median of several predictions, the meta-model can effectively reduce the impact of outliers or individual model biases that may negatively affect the final prediction. This approach is better and more robust than relying on a single model, as it can help reduce overfitting and increase the generalisation and performance of the model. Furthermore, using multiple robust models can also help capture different aspects of the data, leading to a more comprehensive understanding of the underlying patterns in the data.

What predictors did you explore?

We explored 64 different financial predictor sets to test the skill of estimating Scope 1, 2 & 3 footprints for listed companies. Some predictor sets included Revenue, Country, City, Employee Number, GICS, Energy Use, Total Assets, Net Property, Plant and Equipment (NPPE), Capital Expense, Gross Profit, Operational Expense, EBIT, EBITDA, Net Income, Intangible Gross Debt, Capital Intensity, Asset Age, Costs of Goods, Total Debt, Current Liability, Current Asset.

How do you test the skill of each model?

We used a training, testing and validation data split to ensure that 20% of the data was quarantined in order to independently validate the results. Here’s a visualisation of the carbon training reported versus predictions using three different models.

How do you choose the best models for a company?

Each model and parameter set then competes with one another in a battle of the models. One of our data scientists Dr Nic Pittman has coined it ‘Fight Club’ for carbon. Here’s a look at our internal model setup showing ways we can assess model skill like r2 and Median Absolute Percentage Error ( MdAPE).

Want to test out our predictions?

We’re helping a number of financial institutions with their financed emissions and carbon transition pathways. If you are interested in talking to me and our team - just reply to this message or on our website here.

Thanks..

Ben

As COP 27 kicks off in Egypt, quantifying the impact (and therefore financial exposure) a business has on climate is changing globally. New climate disclosure rules that include Scope 3 emissions are in law or planning to be across the developed world. This has big implications if you’re a bank or financial institution.

In todays newsletter, I want to share some data-driven insights on the magnitude of Scope 3 emissions for banks and financial institutions, the state of disclosure for the largest 200 banks and some challenges ahead.

Share

Corporate Carbon Disclosures: Refresher

For corporate climate disclosures, authorities have traditionally only cared about Scope 1 & Scope 2 emissions (eg. emissions created via use of buildings, vehicles and electricity). However authorities across the developed world are increasing mandates on company climate risk disclosures that go upstream and downstream emissions for companies.

Known as Scope 3 emissions, there are 15 different categories (see chart above). They include emissions created from upstream activities like purchased goods, business travel and employee commuting - to downstream activities like processing/use of sold products and investments.

We recently analysed Scope 3 emissions disclosures globally with the University of Otago and found that over 70% of global Scope 3 emissions exposure come down to just 3 categories - ‘processing of sold products’, ‘use of sold products’ and ‘investments’. So in general material Scope 3 carbon exposure generally comes to companies that produce, export or finance a carbon-intensive product.

For example , oil & gas companies Scope 3 emissions are dominated by ‘use of sold products’ when millions of consumers startup their cars or turn on their gas-stoves. Since the product is carbon-intensive, oil companies have a material Scope 3 exposure of billions of tonnes.

But what about the financial industry? What’s the magnitude of Scope 3 when finance is the product?

What are financed emissions?

When a financial institution issues investments and loans through a financial product, it’s the Scope 1 & 2 emissions of their customers that make up the vast majority of their Scope 3 emissions exposure (see chart below1). This is known as financed emissions or ‘investments’ in the down-stream Scope 3 category.

Although there are 14 other Scope 3 categories for a financial institution (see chart above) - they are all either negligible or not relevant. Yes, even business and employee travel is insignificant in comparison to their financed emissions exposure (more on that later). Financed emissions is therefore interchangeable with their ‘Scope 3’ when talking about a financial institution.

How are financed emissions calculated?

Quantifying financed emissions is complex, since financial institutions provide investments and loans to potentially tens of thousands of different companies and assets within their financial portfolio. And in reality the vast majority of those investments have no carbon data. Even in listed equities, we know that just 10% of the 55,000 companies report their annual carbon emissions globally. For the millions of unlisted companies, carbon reporting is non-existent.

From fixed income products, infrastructure assets, mortgages and project finance, you can see how complex quantifying financed emissions is for a financial institution. Because of this, our research found just 9% of financial institutions disclosed their financed emissions by 2019. Even for those who have disclosed financed emissions, data gaps and incomplete coverage is also the norm, with some disclosures covering only a fraction of their exposure.

Latest financed emissions disclosures

Due to the complexity and importance of understanding the magnitude of financed emissions for financial institutions, the Partnership for Carbon Accounting Financals (PCAF) was recently formed to help provide guidance and standards for financial institutions to assess and disclose greenhouse gas emissions of loans and investments.

PCAF provide ongoing methodologies and a database of financial emissions disclosures for financial institutions. I took the world’s largest 200 banks and compared it to the PCAF disclosure database.

The vast majority of big banks (~80%) have not yet disclosed their financed emissions. However we can see a big trend in disclosures during 2022, with disclosures more than doubling over the past year.

What’s a typical financed emissions footprint for a bank?

It’s encouraging to see banks taking on the task of quantifying financed emissions. Of the banks who have disclosed their financed Scope 3 emissions under PCAF, the majority have so far disclosed a fraction of their portfolios. The largest banks have focused on estimating their financed emissions in carbon-intensive portfolios like oil, gas, power and auto. Because of this, it has been difficult to understand the scale of a banks total financed emissions footprint given those sectors make up a fraction of their total portfolio.

Three recent disclosures from larger banks however, give us much greater understanding and granularity for a typical banks financed emissions:

• UK bank NatWest disclosed financed emissions that covered 52% of its total on-balance sheet gross lending and investment exposure.

• Australia’s largest bank, Commonwealth just disclosed their first financed emissions covering 84% of its portfolio.

• Meanwhile, the Japanese mega-bank Mizuho recently disclosed emission estimates for nearly all its over $2 trillion in lending and investments.

Using these recent disclosures, I’ve compared Scope 1, 2 & 3 footprints and compared them with some other equivalent companies across technology, airlines and diversified mining. You can see pretty clearly just how important financed emissions are to the total footprint for a bank - the operational footprints don’t show up, particularly in comparison to other companies in tech, airline or mining.

Taking into account financed emissions (ie Scope 3) for these larger banks makes them comparable in footprints to other large companies. For example Mizhuo’s financed emissions (~154million tonnes) are much larger than American Airlines total footprint (~100million tonnes) and even larger than miner Anglo American (~135million tonnes). These large financed emissions numbers for banks are expected given how important lending is to the global economy - which for now is carbon intensive.

Financial Institutions will be hit hardest with new mandates

As shown, financial institutions generally have very small Scope 1 & Scope 2 footprints relative to other sectors. If authorities only care about those operational-type carbon emissions, then financial institutions would have negligible emissions and financial exposure.

Scope 3 mandates change that. By lending or investing in other companies who do have carbon-intensive physical products via their Scope 1 & 2 emissions, they now have a much greater carbon exposure via their financed emissions. A good way to show this jump in potential carbon exposure is by comparing the footprint ratio of Scope 3 / Scope 1+2 across sectors.

For a big miner like BHP, the Scope 3 ratio is about 40x while for big tech like Amazon, their Scope 3 ratio is ~4. For banks and financial institutions however, their Scope 3 ratios are much higher at ~200x (for Mizuho Bank it’s ~900x). Inclusion of Scope 3 disclosures by authorities means carbon financial exposures would increase orders of magnitude (>200x) more than those implicated from just Scope 1 & 2.

What next for financial institutions?

Most banks haven’t yet disclosed financed emissions but are in the planning phase and are likely facing large carbon data gaps that limit their coverage. For the 20% of the worlds largest banks who have disclosed financed emissions, the challenge is to improve coverage, data quality, efficiency and scaleability that allow consistent reporting of financed emissions.

At Emmi we’ve built award winning machine learning models that predict emissions for virtually any company in the world to help fix data gaps and quality at scale. We’re helping a number of financial institutions with their financed emissions and transition pathways. If you are interested in talking to me and our team - just reply to this message or connect with me on LinkedIn here.

Thanks..

Ben

There’s been a lot of activity from the world’s financial regulators on mandating carbon emission disclosures for businesses. For those that missed it, in March the US SEC submitted a draft rule mandating disclosure of climate-related risks, including Scope 3 emissions.

That means for investors who manage funds, your portfolio carbon emissions will need to be calculated and disclosed. Beyond government mandates, portfolio carbon foot-printing is the critical first step for investors to understand financial risk and reward as the world moves to net-zero. 

So how can you calculate the carbon footprint of a portfolio? And what are best practices? In this newsletter, I wanted to provide a helpful guide to allow any portfolio manager to easily calculate their carbon footprint and understand what it means. 

Share

Explainer: What are financed emissions?

The Taskforce for Climate-related Financial Disclosures is clear on adopting an equity ownership approach for financed emissions. The higher equity percentage holding, the more emissions the investor ‘owns’. For disclosure purposes however, what type of emissions are those ‘financed emissions’?

Carbon disclosure reporting is guided by the GHG Protocol which divides greenhouse gas emissions into three scopes:

• Scope 1 emissions are emissions which come directly from a company and it’s controlled entities.

• Scope 2 emissions come from the generation of purchased energy. 

• Scope 3 emissions are all indirect (value chain) emissions - like your investments or how customers use a product. 

Financed emissions are defined as Scope 3 emissions under the GHG protocol, however most financial institutions are unaware of its magnitude.

A recent analysis from the Carbon Disclosure Project (CDP) estimated that on average, only 25% of financial institutions report their Scope 3 financed emissions. Meanwhile, those Scope 3 financed emissions are over 700x larger than reported Scope 1 & 2 emissions for financial institutions.

So as the starting point for any multi-asset portfolio, how do you calculate this for equities?

Step 1: Source company-level emissions for your portfolio

You will need to source carbon emissions data for the companies you invest in. If you have private companies, you can ask them directly and for public companies, you can do this yourself with some research. And for those companies that have never reported emissions, you can use our predictive models at Emmi.    

If you don’t have many companies in your portfolio, a simple way to source publicly disclosed information yourself is to look at each company’s annual and/or sustainability reports. For example, BHP has a detailed report online. Using a spreadsheet like the example below, you can manually add the Scope 1, 2 and 3 emissions:

Step 2: Calculating your total portfolio carbon emissions

Total carbon emissions associated with a portfolio, expressed in tons of CO2e, are the proportional sum of the carbon emissions of each stock within an investment portfolio.

For example, if a stock represents 5% of a portfolio and its latest carbon footprint is 1.25 million metric tons, then its contribution to the portfolio carbon footprint is 5% of 1.25 million, or 62,500 metric tons. The equity contribution for each stock is added up to calculate the total portfolio carbon footprint. 

This is a meaningful metric to track over time for investors who are specifically targeting emission reduction goals. This is because it provides a gauge of an investor’s responsibility for the emissions of each portfolio holding. However, this metric doesn’t allow portfolios to be compared on a like-for-like basis as portfolio size skews the results. As such, additional calculations are needed.

In the example below we calculate the total carbon emissions for a portfolio of three stocks. The carbon emissions of each stock’s portfolio weighting are added up in the portfolio carbon emissions column to a total of 6,933 tonnes. This represents the Scope 1 and 2 emissions this portfolio is responsible for. If we then include the Scope 3 component for each company, we have a total of 16,166 tonnes of financed carbon emissions.

Total portfolio emissions is a useful metric for investors who are comparing portfolios to identify which are the most carbon intensive, and vice versa. 

Step 3: Calculating your portfolio carbon footprint

Your carbon footprint is the total carbon emissions for a portfolio normalised by the market value of the portfolio, expressed in tons CO2e / $M invested:

In the example below, we calculate the Carbon Emissions to Value Invested for the same portfolio of three stocks. In simple terms, it’s the Total Carbon Emissions calculation from above divided by the current portfolio value (in millions). The Carbon Emissions to Value Invested for each stock’s portfolio weighting are added up for the three stocks in the last column.

The total of 5,389 tonnes CO2e/$M invested represents this portfolio’s carbon footprint. 

Step 4: Calculating your portfolio WACI

The Weighted-Average Carbon Intensity (WACI) is the portfolio’s exposure to carbon-intensive companies via revenue, expressed in tons CO2e / $M revenue.

It’s a popular portfolio metric because it’s a simple calculation which can be applied across asset classes as it doesn’t depend upon market cap or enterprise value as key inputs. It also provides a clear measure of a portfolio’s exposure to carbon-intensive companies and thus transition risks. But, it’s useful to keep in mind that there are downsides to using WACI that I have written about before here.

In the example below, we calculate the same portfolio’s Weighted Average Carbon Intensity by multiplying the portfolio weighting by each stock’s carbon intensity as a percentage of revenue. The Weighted Average Carbon Intensity for each stock’s portfolio weighting is added up for the three stocks in the last column. 

The total 2,694 tonnes CO2e/$M in revenue represents this portfolio’s weighted average carbon intensity. 

NB: This example takes you through carbon footprinting using revenue or market capitalisation as the relevant financial metric for simplicity. However there are problems in using revenue or market cap, since it neglects other important financial fundamentals like debt or dividends. The EU is moving towards EVIC (Enterprise Value + Cash), while at Emmi we include EVIC and 11 financial metrics to give a comprehensive breakdown. I’ve written more about those here.

What if I don’t have the data?

Data availability and quality are usually the biggest challenges when calculating a portfolio footprint.

If you only hold large cap index funds, you’ll probably find some data via company reports or existing data providers. But there’s always gaps, particularly for Scope 3 which can be 90% of the footprint for certain portfolios.  

At Emmi, we’ve made it easy for you by building predictive models that give you emissions coverage across +40k equities. We’ve also recently opened up our sandbox - and that’s all you need to calculate your footprint, net-zero alignment or even value at risk.

Request Sandbox Access

What does the portfolio carbon footprint tell me?

Calculating your portfolio footprint is merely the first step in understanding carbon risk and opportunities for investors. Questions like:

• What warming scenario is my portfolio aligned with?

• Does my portfolio have material financial value at risk?

• What companies are the most important drag on this carbon risk?

• Do my portfolio company targets match up with committed targets?

• What can I do to reduce risk or seize on new opportunities?

These questions cannot be answered with a carbon footprint alone. Another reason to request access to our sandbox to explore and test the most comprehensive tool in the market to understand carbon.

Many thanks,

Ben

As the world accelerates towards net-zero, investors are needing to quantify portfolio-level carbon risk from the bottom-up across their equities. But there are still big problems in sourcing high quality, timely carbon data that covers enough of the global market. But don’t worry - there is a way forward for investors.

In this newsletter I describe some of the data challenges that exist and the approach we have taken at Emmi to help you navigate the net-zero world of finance.

Big carbon data challenges exist

Old data

To make the best decisions, investors need the latest data. For carbon however, the data is almost always more than 12-18 months old, which makes it less useful to drive decision-making. Since the latest carbon data is tied to when a company voluntarily discloses, it can sometimes be an after-thought and it’s not uncommon for data to be even two years old.

Information must be timely for any market to function efficiently- and current carbon data fails to provide this to all market participants.

Lack of Coverage

The majority of global equities have never reported on carbon emissions - EVER - creating another big gap in the carbon information market. The lack of reporting is particularly evident for small to mid-sized companies and in the developing world. Coverage is a major challenge for investors, with most portfolios needing wider and more granular data outside the large companies and indices.

The Scope 3 problem

Carbon emission reporting for companies is guided by the GHG Protocol which divides greenhouse gas emissions into three scopes.

• Scope 1 emissions which come directly from a company and its controlled entities.

• Scope 2 emissions come from the generation of purchased energy. 

• Scope 3 emissions are all indirect (value chain) emissions - like how customers use the product. 

Annual carbon emission disclosure and reporting is slowly getting better via initiatives such as the Carbon Disclosure Project, but is still voluntary. Each year, the number of companies who voluntarily disclose grows and currently sits at ~7,000 globally. 

However it’s important to understand that Scope 3 is often mis-reported or unreported and that can make a big difference.

For Scope 3, there are 15 different categories to report under the GHG Protocol. Employee and business travel tends to be well reported, but quantifying emissions from ‘Use of Sold Products’ is more difficult to calculate so often goes unreported. 

For example, Chevron recently came under fire for only reporting Scope 1 & 2 in setting their net-zero targets. The company did not report Scope 3 emissions which are generally >10x higher in oil & gas. 

Only a fraction of companies disclose Scope 3 emissions and of those that do, many are selective about the categories they report on. Given its importance, how can investors get better insights into carbon risk and temperature alignment without it?

Trust and Uncertainty

Some companies put resources towards quantifying their carbon emissions data. Some even get third-party carbon auditors to verify their annual emissions. But most companies don’t have resources or do the minimum. This creates a quality and trust problem across the market. How much can we trust the reported data?

Black-box models

To overcome some of the problems listed above, carbon data vendors often predict emissions for Scope 3 as an example - but often the methods are secretive and opaque. It’s extremely difficult to trust any model that lacks transparency.

Blackrock recently disclosed its carbon footprint for its entire global portfolio. The world’s largest fund manager could only estimate its carbon emissions for 65% of its portfolio. The other 35% lacked the data quality and/or methodology to quantify it. This highlights the overall challenges existing today due to sub-optimal carbon data and or models. 

A different approach

For us at Emmi, carbon data is a the first critical element to building the financial infrastructure and tools required for the net-zero world. So I thought it would be helpful to others to learn our approach to overcome some of those data problems.

1. Better Research to Predict Carbon

To solve some of the carbon data challenges, we need to discover new ways of predicting carbon emissions. So a couple of years ago I went on a mission to find innovative teams in academia and found a novel piece of research that used a machine learning approach to predict corporate carbon emissions. It was written by Dr Quyen Nguyen and Associate Professor Ivan Diaz-Rainey at the Climate and Energy Finance Group (CEFGroup) at the University of Otago.

We reached out immediately and have been fortunate to collaborate to expand their techniques allowing better predictive models that help us solve some of the carbon data challenges.

2. Aggregate, expand and compare carbon data

For company carbon emission data, it’s best to break it up into three segments which depend on what data is available at the company level.

Reported data are carbon emissions provided voluntarily by the company via external data providers like CDP and others.

Estimated data are emissions where the company has historically reported carbon emissions in two or more years, but with gaps in recent years or over time. We then use models to estimate the latest year of data or fill gaps in older data.

Predicted data are emissions where the company has never reported carbon emissions. In this case we use predictive models based purely on financial data.

2.1 Reported Carbon Data

Instead of relying on one data vendor, our approach is to aggregate carbon data from three different providers for comparison: Bloomberg, Refinitiv and ISS. Each have different methodologies and coverage of carbon data and they do rely somewhat on republishing emissions from the Carbon Disclosure Project (CDP), which have the largest dataset of reported emissions: currently ~7,000 companies.

But that is not wide enough to cover the +40k companies across global public markets. How can we expand the coverage to make it useful for portfolio managers? We have developed a 5 step process that encompasses our approach to building better carbon data for investors.

2.2 Estimated Carbon Data

Estimated data are emissions where the company has historically reported carbon emissions in two or more years, but there are gaps in recent years or over time. We then use regression models to estimate the latest year of data or fill gaps in older data.

The first regression model uses either Revenue & Employees or Revenue or Employees dependent on data availability along with historical emissions. The data is split into training, testing and validation and uses categorical predictors for each GICS sector. It is used for gap filling across companies within a GICS sub-sector within a single year.

The second method we use is for a single company which has reported historically two or more times. Our research with the University of Otago, has found that estimating carbon emissions for companies that have reported more than twice provides the most robust estimates of any method we have researched. 

Overall, estimating carbon emissions has a much higher confidence than predicting emissions when companies have never reported.

2.3 Predicted Carbon Data

Predicted data are emissions where the company has never reported carbon emissions previously. In this case we use predictive models based purely on financial data and GICS sector information.

For our initial data release in September 2021, we used an industry fill method. Our next data release in 2022 will include our machine learning predictive models that we have developed with the University of Otago. This research will be published in an academic journal.

The industry fill method takes a bottom up approach to predicting carbon emissions for companies that have never reported. We cluster reported carbon into sub-industries, GICS and sectors to find their closest peers. We then calculate the Scope/Revenue Ratio of their peers. Using company financial data we then multiply this ratio to predict emissions.

The accuracy and confidence of this approach is low, but varies widely depending on the sector. Some sub-industries have tightly defined distribution of data, while others (like materials which have diverse business activities) have a wide distribution of carbon data. There are limitations especially on small firms in industries dominated by large producers, as this method assumes that all companies within their respective sector are as Revenue : Emissions efficient as one another.

Given the various sources of reported data and the models we use to estimate and predict emissions, it is important to be open and transparent about Emmi’s methodology, while also providing a data-centric approach to gauge the trustworthiness of the carbon data across the +42k equities we have assessed. 

3. Build a Carbon Trust Framework

Why is carbon trust important?

Whether reported or estimated, corporate carbon emissions are inherently uncertain, particularly for Scope 3. Despite this, companies and data providers cant easily disclose the uncertainty of this reported data. Is it +/-10%, +/-50% or +/-100%?

To build a global standard on assessing carbon risk, it is essential we can quantify the trustworthiness and uncertainty of corporate emissions. At Emmi we assess the accuracy of reported and our modeled emissions by combining statistical outlier analysis with a range of clustering and/or temporal volatility analysis. Without a single source of truth from corporate carbon, the premise of our approach revolves around comparing independent sources for data, both reported and modeled.

How does Emmi Calculate Carbon Trust?

Each company can have multiple sources of carbon data and estimates using our models. We combine this ensemble to understand the trustworthiness of the data we use to assess carbon risk.

We assign trust by asking ourselves:

• Do different data providers report similar emissions?

• Does the reported value appear to be under-reported compared to its industry peers or the ensemble of Emmi modeled estimates?

• What is the range of the Emmi emission estimate ensemble?

• Are estimates relatively consistent between years?

After a number of checks, we assign a level of trust to a company as a whole, their scope 1&2 and also their scope 3 emissions data. Company emissions are ranked with a trust score between A and D, where A has high confidence in the reported data, and D should be taken as very uncertain.

Using our method, we generate Scope 1,2 and 3 emission predictions and trust scores for over 40,000 publicly listed securities. See below a histogram of our trust scores within the Emmi carbon data universe.

Our process is generally conservative when it comes to the high level rankings, with around 12,000 companies receiving an A or B. The lowest trust companies ranked D is larger (~18,000), however this is due to the volume of companies that misreport or don’t report emissions. Scope 3 is generally penalised due to it’s difficulty in reporting and predicting.

4. Embrace the Uncertainty Using Data!

Corporate carbon data is always going to be uncertain, particularly when predicting across tens of thousands of companies. However what is most important to investors is whether that uncertainty materially impacts the carbon financial exposure of a company and portfolio. Traditionally, it has been difficult for investors to know whether their portfolio is exposed even if the data is uncertain.

But by integrating finance with carbon within a data science framework, we can for the first time propagate carbon data uncertainties that allow investors a data-driven perspective on whether a company or portfolio is truly a carbon risk. 

Below is the example output from 6 companies, where we build Monte-Carlo simulations to understand if carbon risk materially changes because of carbon data uncertainty (high risk =0, low risk =100). All the dots are independent simulations - and you can see that for most companies, the uncertainty in carbon data doesn’t change the overall risk band (Box range)..

So instead of not knowing the uncertainty of carbon data and hiding from it like in the past - it’s of critical importance for us at Emmi, since it provides a complete picture of the carbon risk exposure, giving our clients more certainty in their decisions.

Carbon data will always be challenging and there will always be uncertainty. But fortunately for investors, these challenges are manageable if using predictive models with known uncertainties that can then be integrated into financial risk for their portfolios. This will give investors a more certain understanding of carbon transition risk so they can make the best investment decisions in deploying their capital.

If you are stuck on access to carbon data and portfolio-level carbon risk for your own universe of companies and want any help - just reply to this message to email me.. I’m keen to help.

Ben McNeil

Co-Founder, Emmi

The politics of COP26 in Glasgow is over and the financial world continues to accelerate towards net-zero. Investors are looking to better identify, quantify and solve for the carbon transition risk in their portfolios. So what are the best carbon metrics to do that?

Timely and accurate carbon emissions data for companies is the first obvious step. But as I have written previously, carbon foot-printing alone is virtually useless to assess the carbon risk of a portfolio.

To be efficient and compete in a carbon constrained world it’s not about smaller carbon footprints, it’s about optimising for the value you create per unit of carbon. So beyond just carbon foot-printing, what metrics are used to understand carbon risk for companies and portfolios?

In this week’s newsletter, I explore the evolution of carbon metrics and the pros and cons of using some like WACI (Weighted Average Carbon Intensity) via a case study of FedEx.

Share Carbon Alpha

The evolution of carbon financial metrics

‘Carbon Intensity’ was the first carbon financial metric used, introduced back in 1997 by Professor J. Timmons Roberts in a paper that explored national carbon economic productivity. Defined as “the carbon dioxide countries emitted per unit of their gross domestic product”, it introduced the concept of benchmarking carbon emissions relative to economic output.

Over time, carbon intensity became a good metric for many countries, since it allowed easy comparisons between countries while it could help inform national policies that were aligned with their economy. Some countries still prefer using carbon intensity to set targets. In Dec 2020 for example, China announced a target of reducing carbon intensity by more than 60% by 2030.

But using carbon intensity presents some problems. Firstly, national carbon intensity declines naturally via technological advancements, industrial efficiency and GDP growth. Ever since industrialisation, every dollar in GDP has become more carbon efficient, every year - without doing anything - even in China.

China recorded a 81% decline in national carbon intensity since the late 1970s (see below). Yet during that same period, China increased its absolute carbon emissions by nearly seven-fold, reaching 10.7 billion tonnes in 2020. Carbon intensity hides what is happening with total carbon emissions which is why most political leaders tend to set absolute emission targets (eg 50% by 2030 etc).

Despite the drawbacks The ability to benchmark carbon emissions on a financial indicator was the first step towards a useful metric in finance - since companies could be easily compared with each other.

Carbon metrics for Investors: from WACI to Emmi

Over the years, carbon intensity has been applied to financial markets as the leading metric to assess the carbon exposure for a company or portfolio. Instead of benchmarking GDP, carbon intensity in finance is the measure of carbon emissions relative to the economic output of the companies in a fund or portfolio. Typically, revenue has been the financial metric used to benchmark carbon intensity of emissions for companies, expressed in tons CO2e / $M revenue.

1) The Weighted Average Carbon Intensity (WACI)

For investors, the Weighted Average Carbon Intensity (WACI) has become a key metric that uses the carbon intensity by revenue for companies. WACI measures a portfolio’s exposure to carbon-intensive companies by revenue, expressed in tons CO2e / $M revenue. Emissions are allocated based on portfolio weights (the current value of the investment relative to the current portfolio value). WACI has been endorsed by the Task Force on Climate-related Financial Disclosures (TCFD).

2) Enterprise Value with Cash (EVIC)

There are big problems in using WACI alone to assess carbon financial risk (see below), and some progress has been made. In 2019 the European Commission made a significant shift away from WACI. Under the EU Benchmark Regulation, they instead mandated the use of Enterprise Value Including Cash (EVIC) to calculate carbon intensity for companies and portfolios.  Enterprise Value is the total value of a company when you take all stock shares, debts, and cash reserves into account.

3) Emmi combines WACI, EVIC and 10 other carbon financial metrics

Instead of using one carbon financial metric however, at Emmi we consider WACI, EVIC + 10 other financial metrics. Why does our algorithm do that? Because it gives a more complete and robust perspective on carbon risk. From debt, liquidity, dividends, cash and company value - it allows us to give a comprehensive assessment of the carbon efficiency of a business or portfolio.  

The limitations of using single carbon metrics

Imagine if investors only used one decision point as their source of intelligence when making buy/sell-side decisions? For example a company might have great revenue or a high enterprise value but has uncontrollable costs, debt issues, a short term liquidity crunch. One metric alone would miss all these pieces of the carbon valuation puzzle, potentially leading to the erroneous decisions being made on incomplete information. 

For example, although WACI is being used by many to assess carbon risk, it has some major problems. The most obvious is that revenue is not an all-inclusive gauge of the financial health of a company.  

For carbon risk evaluation, just using revenue and WACI will also lead to significant pieces of the carbon exposure puzzle being missed. In a carbon-constrained world, revenue is only one of many financial fundamentals that will be impacted as carbon is restricted and eventually priced in the market.  

Focusing solely on revenue presents major problems for investors to understand the real risks and therefore the opportunities. WACI for revenue misses all the other financial fundamentals from valuation and assets to debt and liquidity.  Looking beyond a single metric of value creation we can give greater certainty of how a company might perform in a future constrained world. 

FedEx: How WACI alone can hide carbon risk

Federal Express is a US multi-national in transportation and e-commerce, with huge revenues of $183b. In 2020, combining their reported scope 1 & 2 carbon emissions with Scope 3 estimates from our models, FedEx’s total carbon footprint is about 43million tonnes.

If using only carbon intensity of revenue for FedEx, they seem to be highly carbon efficient in comparison to other big businesses. In fact, we find that they get a perfect score on the revenue metric of 100/100, implying great carbon efficiency and a very low risk exposure to the climate transition. The problem is that although FedEx’s huge revenue reduces its revenue exposure to carbon - its operating margins are tight and debt relative to those margins is high. So, when accounting for these broader financial metrics, FedEx carbon transition risk exposure increases substantially.  

Instead of looking at just one metric like WACI (revenue), Emmi covers 11 other core financial metrics. The results for FedEx are less optimistic in terms of their overall carbon risk - with an Emmi Score of 42/100 – suggesting greater risk and incompatibility with a carbon constrained world. Very different from the perfect score as judged by just revenue (WACI).  

FedEx Case Study: Emmi Carbon Risk Analysis

Using the Emmi platform, we can investigate the carbon exposure relating to each financial metric, from Book Value, Dividends, P/E to Interest and Cash. For FedEx - you can see that the revenue scores are 100/100, while for most other metrics the scores are low (between 0-50).  

Digging deeper we can see that there may be challenges ahead for this organisation in preserving their value, servicing their assets, and supporting their debt and dividends when carbon is constrained. These areas drag down the overall Emmi Score - meaning those areas of the business have significantly more carbon risk exposure than just focusing on revenue/sales.  

To understand who will survive and thrive in a carbon constrained world we have to look at all pieces of the puzzle; if we focus on only one or two data points we risk missing half of the story.  

Stay safe and well,

Ben

Share

Hope you are safe and well. One long-time frustration of mine is how people like to lump climate change in with deep-green ‘environmental’ issues. From deforestation to anti-GMO and protecting polar bears, climate change often gets thrown into the broad tranche of politicised issues that are not systemic to the financial system.

Our entire society and global economy was built around a stable climate. As such, any changes in the climate system fundamentally impact individuals, families, businesses, cities, and nations. This means that the impacts of climate change are not about far-away polar bears, they are fundamentally part of the global financial system.

Fortunately, the zeitgeist that climate change as a financial issue has shifted. Mark Carney, Governor of the Bank of England and Chairman of the Financial Stability Board, gave his now-famous 2015 speech ‘The Tragedy of the Horizon’ suggesting climate change could cause the next financial crisis. 

Since then, it’s clear that key players from central banks to large institutional investors have begun attempting to integrate carbon risks into their decision-making given the seismic reallocation of capital taking place. 

However, they remain hamstrung by traditional metrics. Thanks to simplistic carbon-footprinting measures, carbon risk has traditionally been diluted with the broad scope of ‘Sustainable Investment’ and the qualitative non-financial metrics that come with it. That’s a problem for all investors.

In this month’s newsletter, I explore trends in Sustainable Investing, the utility of ESG scores to assess carbon risk, and some ways to better integrate carbon risk into financial decision-making. 

Share Carbon Alpha

The evolution of ESG: sustaina-bubble?

Companies operate to create value for society. With that comes a social license that continuously evolves. As social progress is made in areas from human rights to gender diversity, companies must evolve too, integrating these new social norms into how they operate. 

If a company is exploiting child labor, then it rightfully doesn’t have a social license to operate. It’s obviously not a great investment either, no matter what your opinion of child labour is.

A myriad of metrics have been formulated by ratings agencies over the past two decades to assess companies’ performance across this broad range of evolving environmental, social, and corporate governance (ESG) issues. In fact, over 50 ESG metrics can now be assessed, with this number growing.

Companies can be now be rated on things like anti-competitive behavior, human rights, workplace safety, gender diversity, water, biodiversity or recycling. These give us diverse insights across important issues - but a challenge often associated with them, is how much can we trust them and how can we use them?

Sustainable Investing, sometimes called socially responsible investing or ESG investing, use these ESG metrics to rate stocks, exchange-traded funds (ETFs), and investment funds, as a proxy for sustainable investing. 

There has been a surge in ESG investing - with COVID-19 only fuelling the market. By some estimates, some $35.3 trillion - or ~36% of total assets under management - are now ‘sustainable investments’. 

Scrutiny is increasing on ESG investing. The EU this month cracked down on ESG fund greenwashing, causing a $2 trillion fall in on-paper assets that can call themselves “sustainable”. Aside from the debate on whether the ESG-bubble has arrived, the question I want to explore is whether ESG metrics are helpful for investors on climate and carbon?

Carbon risk is not binary

Traditionally, ESG metrics have been used for screening purposes (either negative or positive) for investment decisions. Certain sectors, companies, countries and activities can be easily screened by activity as measured by ESG metrics (e.g. weapons, tobacco, coal) or via norm-based standards (e.g. human rights, labor rights).

ESG-based screening has also been used for fossil-fuel divestment (eg any coal, gas or oil activities), which has gotten alot of media attention over the past few years. But for carbon risk, ESG-screening doesn’t work as it takes a sledgehammer to a very complex problem.

Unlike activity-based ESG screening (eg child labor or thermal coal), carbon transition risk is not a binary equation. Carbon emissions don’t need to be 0 tomorrow or even in the next 10 years - it’s a carbon transition. Over-reliance on metrics that don’t account for future transition risk is a problem. Not all metrics are created equal and we need deeper insights to allow us to take well-considered action and ensure we are not destroying (rather than creating) value in the process. 

Over the transition period, a company can still emit carbon emissions so long as they are generating the requisite value. As the transition to net zero continues, a company should only be emitting carbon emissions if the marginal benefit of the carbon emissions is greater than the marginal cost. That means coal and oil will still play an important role in the shorter term and basing it on simplistic carbon footprinting ESG metrics is not helpful for investors.

Qualitative, Non-Financial ESG metrics can hide carbon risk

Climate change and carbon metrics are included in the suite of ~50 ESG metrics from rating agencies. Typically, a company is scored based on a much broader mixture of sustainability indicators. Alot of these are qualitative judgments on how strong governance structures are or whether a company has issued an aspirational target for net-zero. Some companies get good ESG sustainability ratings as a result - whereas the real carbon risk is buried. Here’s an example…

Example: Carbon risk for major Sustainability Index

To help you understand how good ESG ratings can hide real carbon risk - I took a portfolio of the top 10 securities from one of the largest Sustainability indices in Australia. I then built an equally-weighted portfolio of those top-rated ESG Sustainability companies and uploaded the portfolio to Emmi to better understand the carbon risk. Here are the results.

Top 10 Listed Australian Securities in a large Sustainability Index:

Emmi Carbon Risk Portfolio

Emission reductions needed to maintain carbon financial risk bands

What we found was no clear correlation between those companies that rated highly in ESG-investing via a top Sustainability Fund and carbon risk metrics via Emmi. On average there is a medium level of carbon risk while only 2 out 10 companies listed are currently net-zero aligned based on their existing carbon finances.

Learn more about Emmi here

Idiosyncratic vs Systemic Risk

The other problem with lumping climate & carbon in with the 50 other ESG issues is that climate change is fundamentally different to other ESG issues. It presents a systemic risk to the financial system as highlighted by Mark Carney, while other ESG issues have idiosyncratic risk. Systemic risk is the likelihood of damage being done to the health of the system as a whole. Idiosyncratic risk, by contrast, is the likelihood of damage being done to one or small-selection of securities.

By including climate within the scope of ‘Sustainable Investing’, traditional ESG downweights and keeps climate change external to integrating with investment decision-making. As a result, hidden risks remain throughout the system. This is why we hear talk of ‘Green swans’. 

It’s easy to see why this has occurred given ESG carbon metrics are based on carbon footprinting and qualitative analyst judgments. How can investors convert a companies carbon footprinting or carbon intensity rankings into risk or pricing? Traditional ESG metrics work entirely in a financial vacuum, scoring carbon emissions. But the fundamental question for investors is how best to re-position capital markets to prosper from the low carbon transition.

Traditional carbon footprinting ESG metrics haven’t been built to integrate into financial decision-making. Ultimately they can only be used for screening or more likely, sales and marketing departments.

Integrating carbon with finance unlocks opportunities

Whilst ESG metrics have their place in decision making, traditional ESG-investing doesn’t solve for carbon risk. We need a new approach for carbon risk, an integrated quantitative financial approach to help investors make decisions that maximise alpha opportunities. The ultimate goal is to find the lowest-cost carbon pathway and there are many alpha opportunities in all sectors that ESG screening isn’t built for. 

Source: Global Sustainable Investment Review, 2020

That is why over the past five years, institutional investors have started to shift their approach. A surge towards ESG integration has occurred (see plot above) while ESG screening as a strategy has declined, according to the Global Sustainable Investment Alliance.

Yet traditional ESG metrics are incapable of providing a solution for investors that can solve for and easily integrate carbon into their financial decision-making. That is the reason we built Emmi. By providing quantitative risk management tools for companies and investors our mission is to re-position capital markets to prosper from the transition to net-zero emissions.

Using the Global Carbon Efficiency Rating, we integrate corporate finances to help the market to understand whether a company is generating enough value overall from their carbon emissions - and whether stated strategies are really making an impact on this measure. 

Our approach is to focus on quantitative, objective carbon financial data today - then track that over time. As the carbon constraint enters the marketplace, it should be directly quantified and tracked as carbon risk over time, without qualitative judgments. Whether a company is transitioning towards lower carbon risk is demonstrated by data - not by press release.

Emmi does this by integrating 12 financial metrics to understand the complete financial risk for investors. Each business unit (profits, dividends, revenue, debt, liquidity) is carbon benchmarked to give a complete risk profile of a company to the carbon transition.

By combining climate benchmarks with financial metrics, optimal carbon pathways can be projected for companies and their portfolios. This allows investors and companies to fundamentally benchmark carbon risk and invest in different strategies to ensure long-term alpha while demonstrating impact.  

I hope you found this post useful and if you’d like to learn more about how Emmi can help you move beyond qualitative ESG metrics to capture carbon alpha - just reply to this in your inbox.

Next newsletter I want to go a bit deeper into optimizing carbon pathways to prosper from the transition to net-zero emissions.

Stay safe and well,

Ben

Share

You’ve probably been hearing a lot in the news about countries and companies committing to ‘net-zero’ carbon emissions. President Xi Xingping for example has committed China net-zero emissions by 2060 while US President Joe Biden, has recently committed the USA to achieve net-zero by 2050. In Australia, Prime Minister Scott Morrison seems close to setting a net-zero target too, along with countless other leaders.

A recent analysis shows the ‘net-zero’ momentum has surged, with over two-thirds of the world’s economy committing to net-zero emissions. The world is moving towards net-zero. But what does it mean scientifically and practically at the company level? and how does that translate to investment risk? We explore those questions this week…

Before we get into it, this week the Australian Financial Review published a good overview of the wider carbon risk problem and our mission to eliminate carbon risk.

What does ‘net-zero’ mean?

The scientific definition

The Intergovernmental Panel on Climate Change (IPCC) defines net-zero as the point when:

anthropogenic emissions of greenhouse gases to the atmosphere are balanced by anthropogenic removals over a specified period.

‘Anthropogenic’ means human-caused and the ‘specified period’ is typically on an annual basis, but this can be longer as we will find out.

So ‘net-zero’ emissions simply mean that over a year, humans do not incrementally add carbon dioxide to the atmosphere. Any human-caused carbon emissions must be compensated by negative human-induced carbon emissions (otherwise known as carbon sinks).

Carbon neutral electricity isn’t enough to reach net-zero. If for example, we converted all the world’s electricity to renewable energy, the atmospheric loading of CO2 will still increase because of emissions from industry, agriculture, and transport. To reach net-zero would still require negative emissions that sucked carbon from the atmosphere.

When does the world need to reach ‘net-zero’?

In late 2018, the Inter-governmental Panel on Climate Change (IPCC) published a special report looking at what the world needs to do in order to give us a chance of limiting climate warming by 1.5°C beyond pre-industrial times. The take-home was that the world must halve carbon emissions by around 2030 and reach net-zero emissions by mid-century.

For a 2°C warming limit, the take-home was that the world must reduce carbon emissions by around 25% by 2030 and reach net-zero emissions by 2070-20801. The net-zero timeline is pushed out by ~20years under the 2°C scenario.

So what does ‘Net-zero’ for companies mean?

On a scientific basis, ‘net-zero’ is pretty clear, and this definition equally applies at the company level. That is, a ‘net-zero’ company is defined as one that does not incrementally add carbon dioxide to the atmosphere over one year. Any carbon emissions that do occur must be negated by removal from carbon sinks (like large-scale forestation, biochar, BECCS (bioenergy with carbon capture and storage), DACCS (direct air capture and carbon storage), and enhanced weathering).

Caution needed over net-zero claims

There are some important questions each of us should consider around ‘net-zero’ claims from companies.

What emissions aren’t included?

The challenge for companies and investors alike is that unfortunately, there is no commonly agreed definition of what constitutes net-zero emissions for companies - so companies can claim net-zero status even when they aren’t. For example, Facebook recently announced it ‘has reached net-zero’2, but reading the details, the company only accounts for emissions from their data centers and electricity, not their supply-chain or customer usage (they plan for those emissions to be net-zero by 2030).

For some companies, the upstream and downstream emissions (known as Scope 3 emissions) can dwarf any other emissions. And they can have a big impact on translating the carbon financial risk for individual companies.

Hundreds of companies from JBS, the world’s biggest meat company, to BHP, the world’s biggest mining company are pledging to be net-zero. Each company can choose to include or exclude Scope 3 emissions. Those details are very important as they can have a big impact on translating the carbon financial risk for individual companies. At Emmi, when Scope 3 emissions aren’t reported, we’ve built in-house models to estimate those for any company. This allows us to take a consistent approach across thousands of companies.

What are your 2030 targets?

As the race to claim net-zero status grows, the time horizon is often not very meaningful for investors. Company targets to 2050 for example, can easily be avoided given the average tenure of executives is around 5 years. If a company has material carbon risk, it’s the next 5-10 years where the big decisions and gains need to be made.

At Emmi, we focus on assessing carbon risk as of today and then calculate scenarios needed for the company to meet over the next 5-10 years in order to align with carbon risk bands. Like a credit rating, Emmi allows investors to understand how companies are tracking across all sectors on time horizons that suit.

Does the company have a financial carbon risk in the first place?

From a financial perspective, it’s important to assess whether a company has a material carbon risk anyway? Do they need to act on carbon emissions in the first place?

All the global technology giants including Google, Amazon, Microsoft, and Facebook have announced big ambitious plans to be net-zero in the near future. Some like Google, even go a step further and have claimed to become net-zero across the entire history of their company3. Microsoft wants to beat them and go ‘Carbon Negative’4. That is, they want to take more carbon from the atmosphere since the beginning of their business.

The tech giants are in a race to eliminate carbon risk for their businesses. The thing is, they were already very low carbon risk from a financial risk perspective. Facebook, Microsoft, and Alphabet all received near the highest Emmi Score possible (99/100) - meaning they are the lowest possible financial risk on carbon.

Facebook as an example, can actually INCREASE emissions by 60% and still be aligned with a net-zero 1.5°C climate (see below). The green area is low carbon risk over the next 10 years.

Facebook Inc: Carbon Risk Profile from Emmi

Emissions changes needed for Facebook by 2030

What companies are currently aligned with a net-zero 1.5°C climate?

Our approach at Emmi is to combine the latest carbon data with company finances to allow investors to understand the carbon risk of any company. We’ve built scenarios that allow us to quantify what companies are aligned with a net-zero today and those that are not.

Under the net-zero carbon constraint aligned with a 1.5°C climate warming scenario, we have analyzed over 2900+ companies worldwide to understand who is already on track and doesn’t need to achieve emission cuts over the next 5-10 years. The results may surprise you but we have found that 36% of the world’s largest publically-listed companies don’t need to reduce emissions (like the Facebook example above) - they are already aligned with a net-zero 1.5°C world.

However, different sectors and countries have very different carbon exposures. For the ASX100 as an example - we find only 13% of the companies are aligned with a net-zero 1.5°C world.

Net-zero claims in the media can be misleading. You can take the guessing out of it by using Emmi. If you want to learn more and get a free 1-month trial to the full suite of tools on the Emmi platform - hit reply from your email or you can contact us here.

Get a demo or free trial here

If you have any thoughts on today’s post or would like to suggest topics for me to investigate in future newsletters - feel free to comment below.

Ben

Emmi in the News

Super funds back carbon startup
The Financial Standardreports on Emmi, a new carbon risk management startup backed by Aware Super, Energy Super and Future Super has launched with former AMP chair Catherine Brenner on the board. It further mentions that Melior, Hyperion Asset Management and Perennial will back the startup. 

A new Aussie startup is scoring companies on their carbon risk, and big super funds have given it their support
A Business Insider Australiaarticle also reports on Emmi, mentioning Hyperion Asset Management, Melior and Perennial as fund managers that have been joined by three super funds – Aware Super, Energy Super and Future Super in backing Emmi.
NB. This piece was syndicated toMSN.

Ex-AMP chair joins ESG start-up
An Investor Daily article reportsformer chair of AMP Catherine Brenner has joined the board of an ESG start-up, Emmi - targeting major institutional investors wanting to understand more about carbon risk in their portfolios. It mentions Hyperion Asset Management chair Tim Samway and Patrick Sieb of start-up incubator Startmate, who will join Ms Brenner on the board of Emmi, with major super funds Aware Super and Energy Super already among its clients.

Super funds-backed carbon risk startup Emmi lands high profile company director Catherine Brenner as investor
A Startup Daily article also reports that Tim Samway, chair of Hyperion Asset Management, Coca-Cola Amatil and Boral director Catherine Brenner and Startmate investor Patrick Sieb have all chipped in, alongside leading super funds, with the trio also joining the Emmi board.

Share

Welcome to the first edition of Carbon Alpha newsletter! My name is Ben McNeil, long-time climate scientist and part of the founding team at Emmi.

The goal of this newsletter is to help you navigate the world of carbon risk to make better investment decisions. Why is carbon risk important? As Larry Fink, CEO of Blackrock shared in his 2020 letter to clients, “there is a fundamental reshaping of finance” taking place where “climate risk is investment risk”.

What is climate risk versus carbon risk?

It’s important to understand that ‘carbon risk’ is not the same as ‘climate risk’, so let me explain. ‘Climate risk’ often combines two different risks: physical climate risk and carbon transition risk. Here’s a schematic to help.

Physical Climate Risk

Physical climate risk relates to physical asset damage associated with climate warming - things like the frequency and severity of extreme weather events, heat waves, storm surges with sea-level rise, droughts, floods, fires, etc. Aside from temperature, heatwaves, and sea level, however, the science of predicting shifting climate patterns and extreme weather events is very uncertain, particularly at the small-scale asset level.

Physical climate damages (eg sea-level rise) also occur on long timescales(50-100yrs). That is good news since even the most climate-exposed companies (say coastal infrastructure or agriculture) can adapt and absorb much of the risk over that time frame by investing in risk abatement. For example, investing in storm barriers, seawalls, or flood mitigation technologies will all help climate-exposed companies absorb and survive the worst parts of physical climate risk.

Carbon Transition Risk (‘carbon risk’ for short)

The other type of climate risk is called ‘carbon transition risk’ - which refers to the business impact caused by the transition from a high-carbon economy to a low-carbon economy needed to address climate change.

Each year about 36 billion tonnes of carbon dioxide are emitted from global industry. The Paris Climate Agreement—ratified by over 190 countries—aims to ensure that the increase in average temperatures above pre-industrial levels is kept to ‘well below’ 2°C by 2100. To achieve that, science suggests carbon emissions must go to near zero by mid-century. Most countries (even China) have now committed to achieving net-zero carbon emissions by 2050-60.

This global transition from a high-carbon to a low-carbon economy is enormous and this carbon risk has a much broader impact than physical climate risk since all companies have exposure to energy and transportation for example. A recent major UN-backed climate finance analysis suggested that carbon transition risk could impact “as much as 13.2% of overall portfolio value which would represent a value loss of USD10.7 trillion”.

Carbon risk will also occur on a much shorter time scale than the physical climate risk. Why? Because the legal, regulatory, market, or reputational risks are driven by policy and market behavior which can be imposed very quickly. Glaciers melt slowly, but humans act quickly. As the global response to COVID-19 showed, market demand and governments can effectively shut down entire economies to reduce risk. The next decade is where most of the carbon risk will be played out - since that is when peak carbon emissions are likely to occur and financial markets demand lower carbon reliance.

What is Emmi? and why did we build it?

If a business is heavily reliant on producing carbon emissions, then in a world moving to net-zero emissions, it has a high carbon financial risk. But how can we score that risk for any company? Each company has wildly different carbon exposures.

Currently, there is no easy way to assess and track carbon transition risk in a standardized way. Emmi solves that problem for investors and companies.

Emmi was built to make it easy to assess and track carbon financial risk for any company without needing a Ph.D. in climate science or finance. Using Emmi, institutional investors & corporates can understand the financial impact of carbon across any portfolio or company. This will allow them to avoid stranded asset risk and financial loss by using Emmi to understand, anticipate and accelerate the market changes required for a net-zero economy.

Who am I? From climate science to climate finance

My name is Ben McNeil and was a climate scientist for over 20 years. My academic expertise was in building predictive data-driven models to understand carbon dioxide flows around the world. While in academia I became frustrated with the politics of climate change, so in 2009 I wrote a book called ‘The Clean Industrial Revolution’ which argued the economic case for reducing carbon emissions.

I was fortunate to get positive reviews from a number of people including leading economist Ross Garnaut and Nobel Prize winner Professor Peter Doherty. The core thesis I argued was that those who adapted to the emerging low carbon opportunities would be the ones to prosper. On the flip-side, backing carbon-intensive industries didn’t make financial sense in a world moving to low carbon.

Over the past decade, much of these low-carbon trends have accelerated. Gas and renewables have grown to replace coal across the developed world. Electrification of transport on the back of companies like Tesla is surging. In sectors like energy, it’s pretty obvious to observe the shift away from carbon-intensive fuels like coal and oil. But beyond just energy, what about at the company level across ALL sectors? It’s much harder to assess their carbon exposure in a standardized way. You need to combine both climate science and finance to make sense of the exposure - which isn’t easy or available until Emmi.

Although allowing markets to better understand carbon financial risk was critical, I didn't have a finance background. That’s where I was introduced to Michael Lebbon in 2019, the founder of Emmi. With deep financial expertise in carbon finance, Michael had already made good progress towards building out the Emmi platform. I was excited to join and as part of the founding Emmi team.

Over the past 18months, we’ve been busy building out the core Emmi risk algorithm and platform and we’ve also just finished being part of the Startmate accelerator cohort.

Has it launched yet?

Yes! We’ve soft-launched the Emmi platform to our foundation clients and we are in the process of opening it up to those interested. If you would like to book a free demo, you can simply hit reply from your email or you can contact us here.

What to expect from the newsletter?

The Emmi platform covers 2800+ companies across the globe. From Apple to Xero and everything in between, there is a lot to navigate and understand in carbon finance. My goal for these newsletter posts is to share our own learnings/insights in carbon risk whether from the company level, across countries or sectors or even across fund managers.

Welcome and if you have any questions, by all means, let me know!

Ben McNeil