Can reporting on carbon save you money?

19 August 2015

Obese pig

They say you don't fatten a pig by weighing it. But apparently, according to a new report, reporting your company's carbon footprint can save you $1.5 million a year.

Those of us who aim to promote corporate responsibility are always on the look out for rigorous business case arguments that may persuade the sceptical senior executive. But it's important to be able to tell these from the various flavours of wishful-thinking-research and / or greenwash that are out there. Why? Pragmatically, because senior executives are pretty sharp cookies who can see through limp arguments reasonably easily. Ethically? Well, that should be obvious.

So what about these recent claims? According to researchers at Maastricht University and Oxford University, companies can save millions on the cost of debt interest by volunteering data on their carbon emissions.

There have been plenty of pressures growing to encourage companies to report on carbon anyway. Apart from the focus of environmental campaigners, in Europe an EU directive is due to come into effect in 2017 that will require large companies to publish - amongst other things - how they perform on climate risks. But the unpublished paper, "the Carbon Spread", argues that early movers will gain considerable financial rewards. 

Research that is lazy about the claims of proof that it makes is not helpful to the cause

This is based on the spread level of bank loans, which shows how much companies have to pay for debt. It suggests that firms that voluntarily disclose their greenhouse gam emissions "enjoy more favourable lending conditions, with lower interest costs, than companies that don't." The second factor is companies' actual emissions levels. The researchers found that the poor performers paid significantly higher debt costs. Information on CO2 emissions came from the Carbon Disclosure Project, as you might expect. Data on the cost of debt came from Dealscan.

According to one of the authors Michael Viehs, this reflects "financial markets' increasing interest in accessing more granular information about the environmental performance of companies." If you view the slides produced by the European Centre for Corporate Engagement, there are all sorts of complex looking equations and graphs showing how the data was processed. It concludes with a slide summing up the implications of the research, which leads with the message "managing environmental performance can lower cost of debt" - by voluntarily disclosing environmental information and, separately, reducing emission levels.

The problem that undermines this apparent good news is, unfortunately, the same phenomenon that makes so many of the studies in this area little better than junk. It is this. All the research that is described so far (unless there is some lurking in the unpublished report that has gone thus far unmentioned) demonstrates a correlation, not causation.

All the authors are really able to claim is that companies who openly disclose their emissions tend to also have lower costs of debt. That's it. Correlation. As soon as you start suggesting that it's the act of disclosing that creates the lower debt cost, then you need data that currently isn't on the table.

It's not that the claim is impossible. It could be, if financial institutions are becoming more progressive and enlightened on such matters, that they are beginning much more efficiently to factor in levels of climate risk into how they cost debt. That would be an excellent finding, that might really carry weight with top business decision makers. But to make that assertion, you would need - in addition to levels of disclosure and cost of debt - some direct research with the lenders themselves that would record to what degree those institutions had consistent policies to build those factors into their debt pricing. That is, they would have to tell you that it is their intent to reward disclosure - it is part of their decision-making formula.

After all, there is a perfectly likely alternative scenario. That is this. Well-managed companies tend to be successful, and the evidence of their success leads to a lower cost of debt based on their track record. Well-managed companies also tend to take seriously some of the key non-financial risks, so they're more likely to publicly disclose on their emissions. The causal factor? Good quality leadership and management.

There's still a more nuanced business case argument in there - at least to fend off the suspicion some might have that running around measuring all this carbon may be a cost with no benefit. But it's a different one. Not tied to one single thing that you do.

A few days ago, I contacted the researchers directly to ask them if they had covered anything to answer this correlation-not-causation objection. At the time of writing, there has been no response.

Want to judge for yourself? You can see the slides here (pdf).

In general, the principle is this. We are all very keen to get whatever extra ammunition we can use to keep momentum going in the direction of promoting more responsible behaviour by business. We have an interest, therefore, in getting the best research - not simply grasping onto whatever sounds good whilst hoping no-one notices if it makes false claims as to what it actually proves.

We want to promote management by fact, not by fancy. Research that is lazy about the claims of proof that it makes is not helpful to the cause. Based on the evidence presented so far, that is the case with this latest research. The causation-not-just-correlation test is one to bear in mind to ask the tough questions of all the 'too-good-to-be-true' research findings still to come!