EU’s new carbon-scoring metric bedevils investors

Buy-side risk survey 2021: Evic could have harmful consequences for green investing

Click here to download the full survey results.

The devil – in carbon-scoring, as in so much else – is in the detail. What might look like an arcane, even insignificant tweak to the evaluation of carbon scores by European Union policy-makers, could have an outsized effect on carbon metrics, influencing billions of dollars of future investments.

Investors, however, aren’t buying it. Experts say the EU’s policy switch to using enterprise value as the denominator in calculating carbon emissions scores will favour fast-growing firms over the rest – distorting green credentials and bringing undesirable consequences to both portfolio construction and to efforts towards greening the economy.

“The debate here may seem academic. But the choice can influence significantly how capital is allocated, as well as a fund’s factor exposures and transaction costs associated with turnover," says Vitali Kalesnik, director of research at Research Affiliates.

Before the shift, carbon intensity was most often calculated by the common practice of dividing a company’s emissions by its revenue. The new way – written into EU law by a revision of the Benchmarks Regulation that took effect in December 2020 – is to scale emissions by enterprise value including cash, or Evic.

But green investors say the Evic measure jumps around too much – and for the wrong reasons – impeding efforts to gauge a company’s actual progress in reducing emissions. Businesses with an accelerating market value will see their carbon intensity fall quickly, regardless of their true credentials in the field.

“What you get is a metric driven by the market performance of companies, as opposed to their climate performance,” says Frederic Ducoulombier, ESG director at Scientific Beta, a provider of quantitative indexes.

Portfolio managers, meanwhile, complain the approach will skew portfolio construction. It will arguably force investors to load up on growth stocks – to satisfy what ought to be emissions-driven targets – and will likely increase turnover, thereby adding to costs.

“It changes the relative attractiveness of a company, based on market sentiment or the market environment,” says Kalesnik. “If the price goes up, the company becomes more expensive and [therefore] becomes more attractive. That’s an undesirable feature.”

Research Affiliates, is researching how the EU’s choice of metric will affect portfolio construction, including any unintended consequences.

The industry verdict on Evic already seems to be in though. Andrew Lapthorne, head of quantitative research at Societe Generale, sums up the view of many market participants who spoke with Risk.net: “If the best way to cure a carbon problem is for the share price to go up, clearly that’s not solving anything.”

The scores on the doors

The purpose of carbon-intensity or footprint scores is to allow investors or regulators to compare emitters of different sizes. They normalise emissions relative to a company’s size and output – so a large firm with a high total volume of emissions might still score better than a smaller firm with less total output, thanks to the relative scale of its emissions.

Of the 124 investors quizzed for Risk.net’s buy-side risk survey, 50% said they calculated a carbon intensity score for their portfolios, expressed in terms of tons of CO2 emitted per $1 million invested.

 

 

And until last year, the standard methodology for calculating a stock’s carbon intensity was to divide its emissions by its revenues – but that’s when the EU changed tack.

The EU Low Carbon Benchmark Regulation, an amendment to the existing Benchmarks Regulation, created two new index classifications: EU Climate Transition Benchmarks and EU Paris-Aligned Benchmarks.

For both new benchmarks, the regulation mandates a minimum 7% per year decarbonisation standard – far more than most firms are currently targeting.

 

 

Critically, it measures progress to this end by using emissions divided by Evic rather than revenues.

Firms are under no obligation to use the Evic version of intensity in selecting individual investments, says Ducoulombier, but in practice they will do so because of the regulatory metric. “If you have a regulatory restriction couched in Evic-based intensity, it makes sense to use it in all steps of portfolio construction.”

Critics say inserting Evic into carbon-intensity scores will accomplish a sort of accidental greenwashing.

Certainly, some of the numbers created by the new approach seem topsy-turvy.

E.on is a prime example. Its Evic-scaled carbon score went down by a fifth for the year to October 2021, indicating a lower carbon intensity, according to data from Research Affiliates. Yet the company’s Scope 1 and 2 emissions increased by more than a quarter during that time.

Nestlé sits in the 76^th percentile of consumer-staple stocks by emissions-to-revenues, UBS researchers calculated, but only in the 42^nd percentile by emissions-to-Evic.

Increasing revenues is difficult. Increasing debt is easier

Carmine De Franco, Ossiam

There are also some distorted effects at the sector level. Energy companies, which this year have seen valuations soar by more than 40% amid the Covid recovery, suddenly look much improved under the newer measure.

Beneficiaries of the EU’s decision are companies with high valuations relative to sales, which include online retailing and real estate development, both of which are contributors to higher emissions.

Cynics also say companies will be able to arbitrage their way into indexes by issuing bonds to increase enterprise value and pushing down their carbon scores in the process.

“Increasing revenues is difficult,” says Carmine De Franco, head of research at asset manager Ossiam. “Increasing debt is easier.”

The problems don’t end there.

Judging whether companies are cleaning up over time becomes problematic because Evic-scaled numbers are likely to be volatile.

Between 2013 and 2020, more than 6% of Evic figures from companies in developed markets showed a year-on-year increase of half or more, according to data from Scientific Beta.

“You can go from a good carbon intensity to a very bad carbon intensity in the space of a month, simply because stock prices have moved a lot,” says Georgios Oikonomou, a quant analyst at Societe Generale. 

From October 2020 to 2021, BP’s Evic-scaled carbon score improved by 18%, for example, despite the firm’s Scope 1 and 2 carbon footprint’s remaining level, according to data compiled by Research Affiliates’ Kalesnik.

The EU added a mechanism in the Benchmarks Regulation to adjust scores for big market moves. But critics say the ‘detrending’ mechanism has little effect on individual security selection.

Ducoulombier found that nearly a third of companies that improved Evic-based scores by 7% or more after applying the adjustment had nonetheless increased emissions. For revenue-scaling, the cadre of firms that improved by this degree was 18%.

Currency and inflation effects muddy the picture too.

If the currency of a company’s stock gains versus the currency of its carbon reporting, its Evic increases – again making the company appear to get greener.

Evic-scaled carbon intensity scores calculated by the Dutch central bank show Dutch pension funds and insurers cutting emissions by 34% and 31% from 2012 to 2019. When the bank adjusted for inflation and currency fluctuations, the figures fell to 24% and 23%.

Practically speaking

In building portfolios, meanwhile, Evic-scaling will magnify factor bias, says SG’s Lapthorne.

“The change pushes portfolios towards more expensive companies because they have higher valuations. It pushes them into lower-risk companies because, typically, more expensive companies are lower risk, with higher market cap,” he says. “You bring out a much stronger large-cap bias and a negative-value bias.”

Lapthorne and Oikonomou have constructed factor-neutral Paris-aligned portfolios. But they see the task getting harder as the requirement to cut emissions by 7% a year squeezes investors’ choices in stock selection. “The need to reduce carbon every year is going to be a big challenge for investors,” says Lapthorne.

Value funds are likely to suffer especially badly from the EU switch. European value funds invest heavily in energy and utilities, De Franco points out. “In value, you are buying a lot of the real economy. And the real economy does pollute. Your carbon scores compared to the cap-weighted benchmark will look horrible.”

Andrew Lapthorne

The growth bias in an Evic-scaled measure will only hurt more. “Lowering carbon scores for value will tilt the portfolio towards a point where you are no longer a value fund, but neither are you a growth fund,” he says. “You’re in the middle of the distribution, where typically there’s no signal.”

Evic volatility will also lead to higher turnover in portfolios. “Turnover is often under-appreciated,” says Kalesnik. “Investors could be losing millions of dollars cumulatively.”

Studies by Research Affiliates show smart-beta indexes that over-trade can incur tens of basis points in unnecessary costs, he points out.

Evic-based measures could even dilute the alpha in certain investing strategies.

Research conducted by Gerald Garvey, Mohanaraman Iyer and Joanna Nash in 2018 found that companies producing less greenhouse gas by sales use their resources more efficiently and should be a better bet for investors.

Indeed, analysts at UBS found that a simple strategy of buying stocks with stronger emissions-to-revenue scores beat its benchmark by about 60 basis points a year. The same strategy using Evic in the signal beat the index by a more modest 40 basis points a year.

But could the problems with using Evic be resolved? Scientific Beta advocates a return to revenue-scaling for metrics used in portfolio construction. “It’s not perfect, but it’s better,” says Ducoulombier.

Fixing the problem further down the line may add complexity that reduces transparency and increases costs

Vitali Kalesnik, Research Affiliates

Kalesnik suggests normalising emissions scores with a five-year-average Evic number would smooth some of the volatility and prevent needless turnover.

“Fixing the problem further down the line may add complexity that reduces transparency and increases costs,” he says. “If you could do something simpler, it would be cheaper and would benefit investors.”

But Evic’s advocates say the criticisms might be balanced out by other advantages.

Revenues can also be volatile. In industries such as copper mining, for example, where prices change quickly, volatility can lead to choppy intensity scores – much as with Evic.

Companies that spend on developing cleaner technology might suffer under the old measure because their revenues drop in the short term.

Also, policy-makers have made no secret of wishing to penalise energy firms, which can earn high revenues in the near term, while economies slowly transition to cleaner fuels.

Even Evic’s critics acknowledge the effects of the switch may remain narrow to begin with.

In the EU, the metric will be a binding constraint only for portfolios that specifically seek Climate Transition or Paris-aligned status. Investors might use both calculation methods depending on the task in hand.

Sweden’s AP2 pension fund, for example, tracks emissions-to-sales for the purposes of stock selection in some funds and engagement with companies. But it uses emissions-to-Evic in its reporting and as a constraint in the optimisation of certain factor funds.

How differently stocks will rank under each measure can perhaps be overstated. UBS found that nearly half of stocks fell within the same quintile by sector and region.  

Equally, the metrics are likely to be reviewed and amended in coming years.

And, in time, other more forward-looking measures will become more important.

Practitioners describe carbon-intensity scoring as a stepping-stone between counting Scope 1 and 2 emissions and calculating implied temperature rise scores.

But the industry may be moving to implied temperature scores faster than expected – 52% of respondents in Risk.net’s latest buy-side survey said they already calculate temperature ratings for at least a portion of their portfolios.

 

 

BlackRock, the world’s largest asset manager, with around $9 trillion in assets, has committed to publishing implied temperature scores for its public funds that invest in markets with reliable data by the end of the year. The firm, which struck a deal to acquire consulting firm Baringa’s climate-modelling software in June, will enable rival fund managers to calculate temperature scores for their funds through its Aladdin risk management platform.

Third way

A global consensus on how to formulate carbon intensity scores may yet diverge from the EU line.

The Task Force on Climate-related Financial Disclosures (TCFD), which issues guidance aimed at forming an international industry standard, recommends asset managers calculate a portfolio-level score built up from measures for individual stocks – a so-called weighted average carbon intensity, or Waci – which uses revenue as a scalar. The organisation published updated recommendations in October that left the door open to alternative approaches that use market capitalisation – but not Evic – to normalise emissions across companies. 

That said, emissions-intensity scores using Evic are locked into the EU’s hard rules already, and practitioners see the new standards driving further adoption of the EU approach in time.

“These kinds of metrics will not only drive the construction of low-carbon portfolios. They will also drive the construction of a much larger slice of investment portfolios,” says Ossiam's De Franco.

Sarita Gosrani, director of ESG at investing consultancy bfinance, says the “vast majority” of passive investment managers she has spoken to have either launched an EU benchmark-aligned product, are developing one, or are applying the EU criteria internally as part of monitoring carbon intensity. 

An increasing number of institutional investors wish to see the Evic-type metric reported

Frederic Ducoulombier, Scientific Beta

Ducoulombier fears the chance of an alternative course has passed. Even in the TCFD guidance, he notes, an Evic-based measure was introduced alongside Waci for the first time.

“The EU has ruled,” he says. “An increasing number of institutional investors wish to see the Evic-type metric reported. Some people that thought something else was better will just go with it and won’t be so vocal anymore. I hope I’m proven wrong and we don’t incentivise a metric that has some unfortunate unintended consequences.”

The asset managers interviewed by Risk.net acknowledged the risk of policy-makers suffocating under the detail of setting metrics. Any single measure is “bound to be imperfect”, says Lucian Peppelenbos, climate strategist at Robeco. “Action matters more than measurement when it comes to climate change.”

Yet concerns aren’t going away about unwanted side-effects to a complex policy that’s changing fast.

“We might find we tried to push capital toward companies that provide solutions to fight global warming and mitigate climate change,” says De Franco. “And we ended up simply piling up on artificially low-carbon companies, such as tech and healthcare.”

• Tweet  
• Facebook  
• LinkedIn  
• Save this article
• Send to  
• Print this page  

• Topics
• Benchmark
• BlackRock
• BP
• Climate change
• E.on
• Emissions
• Nestle
• Portfolio management
• Quantitative analysis
• Robeco
• Societe Generale
• UBS
• Volatility
• Investing
• ESG
• Buy-side risk survey 2021