The carbon counters

In Hard Times, Charles Dickens’s ‘industrial’ novel, smoke belched from the factory chimneys of the fictitious Coketown day and night. Today we would shudder to think of the amount of carbon dioxide emissions that would mean. The anti-hero of Hard Times, Thomas Gradgrind, would surely know. “Facts. Facts alone are wanted in life. Nothing else will be of any service,” he tells the children of Coketown.

Yet even the obsessive Gradgrind might struggle to marshal the salient facts that determine which companies are the 21st-century heroes and which the villains of the race to rid the world of greenhouse gases. Of facts about the race there are plenty. The challenge is to distinguish which are the fluffy kind and which are the ones so hard as to give service to the Gradgrinds of the investment world.

Granted, this may not matter to some. The UK arm of broker Charles Schwab recently published the results of a survey which showed that only 44 per cent of investors regularly consider environmental, social and governance (ESG) factors when making a new investment. Simultaneously, two-thirds of those asked did not mind whether their investments were ‘sustainable’.

Such results are not rare in the marketing battle for the minds – and wallets – of investors. Much more numerous, however, is the sort of result that Pimfa, a trade association for wealth managers, produced from another recent survey of UK private investors. This said eight out of 10 considered ESG factors ‘very important’ or ‘important’ when making investment decisions. True, among those aged between 56 and 75 – probably the ones with the most capital – less than a third thought their investments should have a positive impact on the environment, while almost three-quarters of those aged 18 to 25 – a small cohort among investors – felt they should.

No matter. Like it or not, environmental, social and governance-focused investing isn’t going away, least of all its most important component, the ‘E’ part of the ESG acronym, which chiefly – though not exclusively – stands for combating global warming and its feared effects.

The goal is well known. Based on the targets of the 2015 Paris Agreement on combatting climate change, companies must cut their carbon emissions to half their 1990 levels by 2030 and be so-called ‘net zero’ by 2050. It is easy to quibble about what that actually means, whether it will have the desired effect and even whether it is that vital. In practical terms, discussion is largely irrelevant. The Paris Agreement summarises a new orthodoxy that cannot be challenged, least of all by company bosses and investors, who are more likely to be seen as part of the problem than the solution. Wanting desperately to be on the virtuous side of the narrative, bosses embrace the climate-change imperative with the fervour of the converted.

Thus quoted companies must be seen to be reducing their carbon footprint. Do that better than the others, so the argument goes, and companies will be rewarded with a lower cost of capital and – as part of the same market mechanism – their shareholders will be rewarded with higher returns.

Of course, the virtuous can only be rewarded if they make their virtue known and that can only be the result of disclosing the relevant information. This notion is hardly new and was investigated in the 1980s by, among others, the British-born economics Nobel laureate, Oliver Hart, whose theory of voluntary disclosure suggests a company will only reveal information voluntarily if its bosses are confident their company will benefit. It has been tested using various aspects of corporate disclosure and, by and large, the findings vindicate the idea – companies that freely disclose do benefit via a higher share rating.

Last year, two academics from London’s Imperial College, Patrick Bolton and Marcin Kacperczyk, applied the idea to carbon disclosure on a big scale, using data from 14,400 listed companies spread over 77 countries and the period 2005-18. Their basic conclusion was emphatic: “Carbon emissions disclosure significantly lowers the cost of capital as reflected in the stock returns required by investors. These results are robust and hold across all 77 countries in our sample.”

So far, so conclusive. Yet the conclusion also highlights an interesting question, which had already been discussed in Hart’s earlier work. If disclosure is so beneficial, then why don’t more companies do more of it, and why are there ‘disclosure laggards’?

One simple explanation is that companies may be reluctant to reveal commercially sensitive information. But the UK’s regulatory regime concerning companies’ carbon emissions helps to shed light on the matter. Since 2014, UK-quoted companies have been obliged to report their Scope 1 and Scope 2 emissions (see box on page XX). Even before mandatory disclosure, a significant proportion of companies voluntarily revealed such information. Therefore, the academics were able to compare the effect of the new rules on those that already disclosed and the disclosure debutants. After the rules took effect, about 20 per cent more companies revealed emissions information and, on average, their cost of capital fell (ie, their share prices rose).

However, Bolton and Kacperczyk made “an additional striking finding” – among the first-time disclosing companies, the ones with the highest emissions’ levels saw their cost of capital rise. “In other words,” say the academics, “the worst carbon performers were penalised by investors who demanded higher returns after they were surprised to find that these firms had higher average emissions”.

Put another way, badly-performing companies have an incentive not to disclose information about emissions, or to say as little as possible. That possibility was reflected in a paper published in January which suggested that “investors might want to be cautious about assuming that carbon emissions are priced by equity markets”.

The research, by academics from the London School of Economics and Columbia Business School and Fordham University both of New York, used data for over 2,700 US firms for the period 2005-19. But the snag with data about the emissions of US companies is that much of it is estimated by data providers – about 80 per cent, say the academics. As a result, emissions tend to be a function of company size and say little about ‘emissions intensity’ (the amount of carbon a company emits for a specified unit of production). “Using unscaled emissions rather than emissions intensity to measure carbon risk is analogous to using net income rather than ratio-based measures, such as return on assets, to measure a firm’s financial performance,” say the academics.

True enough, but – as the UK’s experience shows – would that it were so simple. It would be wonderful to compare companies’ emissions performance just as analysts compare their financial performance. Granted, depending on what companies do, it may be silly to compare, say, their profit margins (food retailers have thin margins, capital goods makers have fat ones). But it is often fair to compare companies operating in the same sector. Besides, measures for return on capital employed (ROCE) will tend to equalise across sectors since ROCE can be measured by profit margins multiplied by capital turnover (sales divided by capital employed). Thus, for example, thin margins and high capital turnover at supermarkets operator Tesco (TSCO) could well result in an ROCE comparable with diversified engineer Smiths Group (SMIN) with its fat margins and low capital turnover.

Want to do a similar exercise for companies’ CO₂ emissions? Forget it. Sure, many companies show data for their emissions intensity, a measure of the CO₂ emitted for each unit of output. But even at groups with similar operations, such data is almost never comparable. Take Table 1, which shows emissions intensity for two of the UK’s major public transport operators, Go-Ahead (GOG) and National Express (NEX). Each group shows its emissions intensity in a completely different currency. This is understandable, since National Express is primarily a bus and coach operator while Go-Ahead’s revenues are split between buses and rail. Nevertheless, comparing intensity measured in passenger kilometres (National Express) with vehicle miles (Go-Ahead) is obviously meaningless.

Even where the currency of comparison is the same, the data can look odd. Both mining groups Antofagasta (ANTO) and Anglo American (AAL) use emissions per tonne of copper produced as their intensity measure. However, as Table 2 indicates, something seems wrong. How is it that Anglo American emits twice as much CO₂ per unit of production as Antofagasta? The clue may be that Antofagasta is almost exclusively a copper producer with just some revenue from silver and gold. Meanwhile, Anglo American is every bit the diversified miner, with diamonds and iron ore its major revenue generators. Quite likely, therefore, when Anglo American speaks of emissions intensity in terms of ‘copper equivalent’ it is not talking the same language as Antofagasta.

Yet it may be possible to take refuge in comparing companies’ emissions intensity using their revenues or, perhaps more likely, their gross profit (ie, revenues minus cost of goods or services). The exercise is fairly simple to do but, as Table 3 shows, the ratios still look odd. Take the comparison between the two cigarette makers, British American Tobacco (BATS) and Imperial Brands (IMB). Both are in pretty much the same lines of business, so why is it that Imperial emits much less CO₂ per £mn of revenue than BATS (14.5mn tonnes in 2021 versus 19.3mn for the latter) but emits much more per £mn of gross profit (43.1mn tonnes compared with BATS’ 27.8mn)? Is it all about the greater profitability of BATS, or is it perhaps that their definitions of gross profit are not the same?

Similar points might be made about the comparison between National Express and Go-Ahead in Table 3. Happily, however, the comparative emissions metrics for Antofagasta and Anglo American look as if they are in the same ball park. Back in 2017, both groups emitted similar amounts of carbon per £mn of revenues (794 tonnes at Antofagasta and 883 tonnes at Anglo American) and both were still roughly level pegging in 2021 (440 tonnes at Antofagasta and 490 at Anglo).

So there may be virtue in keeping it very simple. Even so, the data that UK-listed companies spew out in vast quantities really has just one purpose alone – to show that their CO₂ emissions are falling. That’s the chief signal that companies want to make. Falling emissions shout out their virtue louder than anything else.

No surprises therefore that almost all companies achieve this. It is rare, indeed, to find a five-year record where a company’s carbon footprint is getting bigger. True, that happened to the transport companies during the lockdown years of 2020 and 2021 (see Table 1), but that was surely exceptional.

Yet falling emissions do not necessarily mean companies are getting more virtuous. Rather, it may indicate that emissions are easy to cut in the early stages of the process, especially as companies can measure their Scope 2 emissions using market-based rather than location-based data and show instant results.

It also indicates we are reduced to judging companies’ comparative performance by the pace at which emissions fall and nothing else (see the percentage change column in Table 3). Does that make British American Tobacco the most virtuous company in the sample because its emissions have fallen by the biggest fraction since 2017? And what do we make of the fall in emissions in 2020 and 2021 at National Express even while its emissions intensity was rising?

At the close of Hard Times, Dickens speculates that Gradgrind, by then a broken man, questions his utilitarian convictions, “making facts and figures subservient to Faith, Hope and Charity”. More prosaically, we might hope that the facts and figures of carbon accounting become subservient to good investment analysis. There is still plenty of work to be done.