EU Climate Benchmarks: The row over carbon intensity metrics explained

A key part of the EU’s sweeping sustainable finance strategy, which it introduced two years ago, is the creation of concrete expectations for indices that promise climate outcomes. With more and more passive products coming to the market making big green claims, the “benchmark legislation” will establish two new regulatory categories, defining what “climate transition” and “Paris-aligned” indices need to offer if they want to be credible and truly aligned with Europe's climate ambitions. The criteria won't be mandatory – providers can still use any methodology they want to when devising green indices – but will seek to provide a trustworthy and transparent standard for those who want one. 

At the heart of the plans is a requirement for indices to reduce their emissions annually, across both high- and low-carbon sectors, to demonstrate meaningful progress on climate transition. But the carbon intensity metric being proposed to calculate the cuts has faced fierce market pushback from some corners. As the plans go out to public consultation, RI takes a closer look at what the plans might mean for investors, index providers and decarbonisation.   

Carbon intensity measures exist because variations in size and output prevent absolute emissions from different companies being directly comparable. At its simplest and most intuitive, carbon intensity describes emissions per unit of economic output (emissions per tonne of steel, for example) but, due to the issues around heterogeneity, metrics such as ‘weighted average carbon intensity’ (WACI) are derived using a proxy for output that is comparable across the wider economy: revenue, market capitalisation, employee numbers and so on. 

The Technical Expert Group on Sustainable Finance (TEG) that was asked by the European Commission to, among other things, help it develop the Climate Benchmarks criteria, recommends using a WACI based on Enterprise Value, which encompasses both market capitalisation of equity and debt. This suggestion raises eyebrows, because it diverges from the more common and simpler revenue-based WACI endorsed by the Taskforce for Climate-related Disclosures (TCFD).

The TEG says that using revenue to calculate intensity tends to benefit “sectors such as coal, which are exposed to potentially discontinued assets [stranded assets]” because of their chart-topping earnings. Last year, the oil & gas sector reported the highest revenues of the world’s largest companies, so expressing carbon emissions per unit of revenue would be likely to result in a lower intensity value. TEG's preference for Enterprise Value, then, is to “avoid greenwashing in favour of polluting sectors". 

But critics say that using market valuations creates a needlessly volatile and complicated carbon intensity measure. Indeed, none other than the TEG noted that “market effects can significantly affect this indicator and create misleading results”; giving the example in its interim report of a rally in share prices having the effect of reducing a company’s carbon intensity while absolute emissions stayed the same.

The recent slide in oil prices due to COVID-19 – which would have the effect of boosting intensity – provides a useful case study. Data shared by MSCI shows that between January and March, both its provisional EU-aligned Climate Benchmark products recorded a carbon intensity increase of just under 25%, driven by fluctuations in Enterprise Value. For perspective, Climate Benchmark indices are required under the new plans to reduce their intensity by 7% every year.

These figures should be interpreted with caution – Enterprise Value is updated annually and benchmark administrators are allowed to adjust for inflation – but they indicate just how acute the effects of price movements can be.

Kelly Hess, Project Director at Swiss Sustainable Finance tells RI that the recent market volatility exposed the “shortfalls” of using Enterprise Value for benchmark providers, who would have seen a “huge spike in their carbon intensity”.

“If these portfolios are measured against a specific target intensity, they all of a sudden perform horrendously,” she observes. “And this can be very confusing to communicate.”

Unstable carbon intensities may also have implications for the ability of benchmark providers to “engage and encourage companies to improve” – one of the aims of the Climate Benchmarks programme from the EU. Faced with a fall in share price and a corresponding increase in carbon intensity, for example, index providers using the framework could be forced to reweight constituents, potentially breaking off engagement with those that fall out of the index. Similarly, portfolio companies may be discouraged from pursuing aspirational emissions reduction targets in a bear market if they know they will achieve big recorded reductions by default.

A benchmark provider who requested anonymity says that while optimisation would allow annual decarbonisation requirements to be met if share prices fluctuated, it risked becoming “purely a mathematical exercise” and could result in vastly differing indices from one period to another.

"Data shared by MSCI shows that between January and March, both of its provisional Climate Benchmark products recorded a carbon intensity increase of just under 25%"

In addition, measuring emissions as a proportion of Enterprise Value may result in a significant bias toward ‘growth’ stocks, points out Vitali Kalesnik, Partner and Head of Equity Research at US index specialists Research Affiliate . These refer to shares which are anticipated to grow at a rate above the market average, often contrasted with ‘value’ stocks which are shares under-priced by the market despite strong fundamentals and due for a future price correction over the long term. For administrators of Climate Benchmarks, growth stocks will look more attractive as steadily appreciating share prices offer a way to keep carbon intensity in check and meet annual decarbonisation targets.

Although he agrees that using Enterprise Value means “more punishment for companies with stranded assets”, Kalesnik notes that a strong growth tilt has been historically associated with depressed returns over the long term. “The performance effect may take time to notice, but in the long term it may lead to worse performance,” he says.

Adopting a separate, “more aggressive” policy for polluting industries would allow for stringent thresholds while avoiding much of the drawbacks associated with tweaking the intensity indicator, he said.

But despite the criticisms, choosing the right proxy for output to replace EV is far from straightforward.

According to Jennifer Steding, Head of Strategic Initiatives at Solactive – which, along with MSCI and S&P Dow Jones, have released provisional EU Climate Benchmark indices – all options have flaws associated with them, and it is up to benchmark providers and investors to understand the assumptions and limitations of each. 

She said: “We have spent a lot of time internally discussing this and what we recognise is that whatever you use as the denominator, there are pros and cons to it. The important part for us is being transparent and highlighting these in client conversations.”

The TCFD made the same point in its final report, pointing out that a WACI derived by revenue “tends to favour companies with higher pricing levels than their peers”. A luxury carmaker, for example, will tend to have a lower intensity per unit of revenue as they will have less input, and therefore lower emissions, compared to a mass car manufacturer selling relatively cheap cars.

Using revenue to derive intensity may have resulted in similar spikes over the current period of turmoil, says Viola Lutz, Associate Director at carbon and environmental data provider ISS ESG, the responsible investment arm of Institutional Shareholder Services.

“Companies also have fluctuations in sales, so a bad year would have the effect of raising intensity. Similarly, I would assume for this year that revenue would have been affected by the crisis together with enterprise value,” she says. 

“However, it is true that there will always be a certain link between a company’s real economic activity and the carbon intensity, which is perhaps why the TCFD chose to endorse revenue, while enterprise value could theoretically be driven purely by speculation.”

To prevent misinterpretation, Lutz emphasises the use of multiple indicators to “fully capture a company’s strategic alignment with a two-degree world”.

“I certainly find it key to look at a company's forward-looking commitments or target-setting combined with absolute carbon intensity. I think this is important as it indicates intention and a clear strategy which goes beyond exposure.”

To provide feedback on draft legislation for the Climate Benchmarks – including the carbon intensity metric – click here. Consultation ends May 6.