Studies show ESG works

There are plenty of papers suggesting ESG works. Here is a start on a partial review of some of the evidence from academic papers.

One of the hurdles to adopting "ESG" is partly the language. ESG is a poor acronym. Environment Social Governance conflates separate ideas and does not completely cover the whole area to which investors often speak about. It's also meaningless outside insider circles and even is poorly understood in many companies and in main stream finance.

Another barrier seems to be the standard of "proof" required compared to some other factors. In my view, it is unlikely we will ever find definitive "proof" partly due to a data problem  I briefly discuss here.  

There is often no "ESG fact" only opinion. There is an impression that an “ESG” rating is more factual than it is. This academic study shows a poor correlation between data provider ratings, around only 0.3 It is unlike a bond credit rating where agreement is fairly tight by the rating agencies. In my view, it is more similar to a brokers sell side report where different sell siders will disagree - an opinion. You can see this is if you look at an ESG score from a data provider. For instance, it is a fact that there is a combined chair/CEO at a company but the score XYZ rating agency then gives that item and the weight it has in the overall rating is an opinion, a judgment. This tick box approach can be misleading, in my view. Productive, engaged employees? Corporate culture? Human capital? Staff turnover? Those skeptical of ESG mostly agree these factors are important. An ESG person might call it the "S" in ESG. A CEO could say it's just running a good business. I call it common sense.

Back to the ESG studies...

Investors are still arguing if there's a value premium (is it P/E, P/B ?) still, whether there is a such a thing as smart beta or dumb beta, and whether factor investing works and which factors, if so! In investing... value, growth, quality, low volatility, blind factor 2 - none of this works all of the time - no rule of thumb does.

A poor fund manager with no skill will not be helped by an ESG tool box. Like a poor surgeon with the latest robotic cutting tools.... that surgeon will simply cut your artery more cleanly and swiftly ensuring a quicker death.

A skilled fund manager should be helped by the ESG tool box. A part of me is happy if other fund managers can't get to grips with it - that puts the odds more in my favour.

Still, for those looking for “proof” I think there are studies taken together that offer a strong level of evidence.

  1. “S” Works. Does the stock market fully value intangibles? Employee satisfaction and equity prices by Alex Edmans.

link here

“…These findings have three main implications. First, consistent with human capital-centered theories of the firm, employee satisfaction is positively correlated with shareholder returns and need not represent managerial slack. Second, the stock market does not fully value intangibles, even when independently verified by a highly public survey on large firms. Third, certain socially responsible investing (SRI) screens may improve investment returns…”

My take: while the study is correlational, it does control for industries, factor risk, and a broad set of observable characteristics. That makes it robust. And even if this factor is not the causal driver, it is likely to be correlational to another intangible which is the causal driver (eg good management processes).

It makes theoretical sense. Satisfied employees = more productive employees = better company = better stock returns.

This is confirmed by my conversations with CEOs and other business managers.

Employees are important. The culture they work in is in important. I doubt many people would argue employees are not important to a business.

One can  explain what might be happening.

Employee satisfaction is not easily tangible – it is an “intangible”. It is not reported in a company’s annual reports or the 10-K. It also is unlikely to impact near term earnings and cashflow, the near term tangible that many market participants are looking at.

However, most would agree in theory that it should impact medium to long term earnings.

So, what is happening is that this “intangible” (AKA extra-financial AKA non-financial AKA S in ESG AKA….. common sense! ) is not being “immediately priced” (market doesn’t know about it, is not assessing it, is not prepared to take the time horizon risk on it…) but this intangible is being priced when it subsequently manifests itself in cashflows and earnings announcements.

This is, in my view, robust evidence for part of the “S” component in ESG. It makes theoretical sense. Business managers agree. Academics agree. Investors, I speak to agree. It even makes sense to my mother.

2. Corporate Sustainability: First Evidence on Materiality by Mozaffar Khan, George Serafeim and Aaron Yoon.

"…firms with good ratings on material sustainability issues significantly outperform firms with poor ratings on these issues. In contrast, firms with good ratings on immaterial sustainability issues do not significantly outperform firms with poor ratings on the same issues. These results are confirmed when we analyze future changes in accounting performance….

Link here

These authors used a hand-mapping process, which you should read for yourself.

My take: So part of the problem is materiality of an item, its disclosure and then whether it is being used in practice in any case.

For example, a “reduce, refine, replace” policy. First, we would need to assess how material such a policy would be. For some companies, it will be more material than others, for some companies it may have limited materiality and not impact the company. Second, if there is a policy, is it disclosed? It might not be disclosed, but it might exist. Then regardless, of whether it exists or not. Is it being followed? Even if it does not exist, some company employees might do it anyway – especially if it was core to what the company might be doing – I could think of those working in an architect’s practice specialising in green buildings.

This type of matter is problematic on the large data sets that might fall under ESG.

However, it makes theoretical sense that performance on an intangible matter that does make an impact to a company would affect the performance of the company (both financial and presumably at a stock return level).

And thus disclosure and measurement of performance on such factors should reveal insights about the company and future returns.

I believe this study is evidence of that.

It is also evidence that non-material factors have limited impact. This again makes theoretical sense.

The impact of drug pricing regulation should not affect how much a clothing retailer sells, but it would impact a pharmaceutical company.

Similarly, the efficiency and robustness of a cotton supply chain is not going to impact a pharmaceuticals manufacturer but might well impact a clothing retailer.

Some intangible factors such as employees (noted above) may be more broadly universal across companies where as some are likely to be company or sector specific.

Even so, this study is good evidence on this idea. It uses Sustainability Accounting Standards Board (SASB) guidance on materiality and a robust orthogonalisation and regression process. The overall findings are also confirmed “using firm-level panel regressions that account for a host of additional firm characteristics such as analyst coverage, investments in R&D, advertising and capital expenditures, and board characteristics and firm or industry fixed effects.

OK. I will stop here for now.

Cross fertilise. Read about the autistic mind here and ideas on the arts here. On investing try a thought on stock valuations