The curse of BSG

4 Feb

4 February 2022

Continually over-claiming the case for ESG undermines the causes we’re trying to promote

Trigger warning: this is going to be a bit of a rant.

Barely a day goes by without a post appearing in my LinkedIn feed claiming magical properties of some ESG intervention or other on company performance or societal outcomes. Excitable claims are made for ESG investing, divestment, diversity, linking executive pay to ESG and so on.

To illustrate, here’s the latest example that popped into my feed (apologies to the authors – this is no worse than many other cases, it’s just the straw that broke the camel’s back).

This study compares diversity statistics, executive pay practices, and carbon action indicators. The study finds that higher levels of diversity and the presence of climate targets in executive pay are correlated with better climate action scores. They conclude that: “The data shows that there are several levers [to accelerate climate action] such as greater board gender and age diversity, more effective climate governance and climate-linked remuneration.”

This is very typical of the genre: what starts as correlation has, by the time the conclusions are reached, become causation. But the study has a number of flaws.

First, there are no controls for sector or geography. European firms have both greater gender diversity and tend to be taking more climate action, both in part because of more aggressive Government regulation and shareholder intervention. So it is likely the regulatory and shareholder environment that is driving both greater diversity and greater climate action rather than diversity driving climate action.

Second, the likelihood of reverse causality is ignored. Companies with a clear plan and targets for carbon reduction are much more likely to include those targets in executive pay than those without such targets. In other words, it is likely the existence of a plan that is leading to targets in executive pay rather than targets in executive pay leading to the existence of a plan.

On closer inspection, many such studies suffer from these same basic errors: lack of controls, confusing correlation with causality, missing the likelihood of reverse causality. These errors often completely undermine the claims made.

The spread of BSG

I call this phenomenon BSG. (See what I did there?) I define BSG as follows:

BSG is the practice of making claims for the causal real-world impacts (in terms of performance or societal outcomes) of a particular ESG activity, based on a study whose research design or execution is in fact incapable of supporting the claims made.

BSG often has confirmation bias at its core. Many BSG studies are produced by consultancies that have a commercial interest in the outcomes being studied. The two organisations that sponsored the study referred to above are a diversity and sustainability consultancy. I don’t for a moment question the motives or integrity of these organisations, but to adapt Upton Sinclair it is tempting for someone to believe something when their salary depends upon it being true.

Why does it matter? If the intent is right then who cares if we play a bit fast and loose with a few arcane concepts like controls and causality? It matters for four main reasons.

If we play fast and loose with the truth then we can’t complain when the other side does too

Here’s a study. Over the 20 years to September 2021, the proportion of women on FTSE-100 boards improved by 26 percentage points, compared to 15 for the S&P-500. But total shareholder return for the FTSE-100 has been only +136% over the same period versus +221% for the S&P 500. Commenting on the findings the lead researcher said: “This is the clearest evidence yet that having more women on boards destroyed shareholder value. Every additional 10% points of female representation costs 77% in cumulative shareholder returns. Attaining gender parity on boards would reduce the value of UK companies by a cumulative £2.3 trillion.”

The facts above are all true (although the quote is fictional). But it can immediately be seen that such a study would of course be entirely bogus and tell us nothing about the impact of diversity. It would certainly not validly support the claim that gender diversity is a bad thing. Call it reverse BSG. But sadly, many studies that point in the opposite direction are no more reliable. How can we call out studies that deny the real benefits of responsible business if we tolerate BSG studies that support our own position and preferences?

We risk focussing on the wrong stuff

If we think that more diversity automatically drives improved outcomes (it does not), we risk treating it as a numbers game and ignore the profound changes to working practices, culture, and the design of jobs that is required to make diverse teams effective. This is inclusion in the broadest sense. If we think this isn’t an issue, a recent set of case studies of diversity initiatives in financial services firms was almost exclusively focussed on changing numbers, rather than on changing the practices and job structures that would enable diverse teams to flourish. Avoiding the reality that greater diversity does not, at least on average, and so far, lead to better performance, we miss the much more important questions: Why not? And how do we change that?

Another example is ESG investment strategies, which are commonly claimed to outperform, despite the reality being much more nuanced. Divestment is a popular strategy amongst ESG index funds and certain categories of asset owner, especially university endowments. A recent paper was reported as showing that divestment caused reductions in carbon emissions. It showed nothing of the sort. The study did not demonstrate causation, and if it showed any results at all, it probably supported the more nuanced position of conditional divestment: for example holding oil companies, but then threatening divestment from the laggards within this sector who are making the least contribution to emissions reduction. Again this matters. If asset owners and retail investors pursue investment strategies that they think are making a difference but actually aren’t, then they are diverted from doing things that may actually be useful, and may meanwhile be incurring a cost in terms of higher risk or lower returns.

Knowingly doing stuff that doesn’t work is a breach of fiduciary duty

It’s become almost a truism that more ESG leads to better returns, whether we’re talking about companies or investment approaches. But this isn’t true. Focussed ESG activity can indeed add value, but indiscriminate ESG activity does not (and costs money). The lazy assumption that ESG = Value is resulting in investors and companies carrying out all sorts of activities that may, in fact, not be in their investors or clients’ financial interests. That’s fine if those clients or investors have given a clear mandate for that activity to be undertaken regardless of the consequences, but that’s rarely the case.

BSG studies create an environment where tough discussions about beneficiary preferences and fiduciary duty can be avoided. If every ESG issue is a win-win, leading to higher returns, then we don’t even need to discuss it. But the world is more complicated than this. Trade-offs are everywhere. A recent report by the Centre for Corporate Governance at LBS and The Investor Forum emphasised the need for absolute clarity from investors in how and why stakeholder issues are being prioritised, in order for this to be aligned with client mandates and fiduciary duty.

In the long-term BSG will back-fire and reduce rather than increase trust

There is lots of talk about the importance business increasing trust and the need for a reset of capitalism. I am myself a fully paid-up member of Alex Edmans’ Grow the Pie club: great companies deliver value through pursuing a purpose that benefits society. But not every ESG issue can be hitched to this wagon. As Alex himself reminds us, we need to be discriminating.

Indeed there’s already a suspicion in some quarters that the responsible business agenda is being hijacked by woke activists to bypass the political process. This may be leading to declining trust in certain segments of the population whose political priorities are not aligned with the zeitgeist. I am not a fan of disparaging use of the term “woke”, which I find a bit insulting to a younger generation that cares more about social justice. But it does highlight a danger. ESG issues frequently draw business into political territory, whether it’s climate change, diversity, inequality. Tolerance, or worse still eulogising, of studies that are shown to be BSG will simply increase suspicion of an elite stich-up.

Follow the evidence

Advocates of responsible business, in which I include myself, need to retain our critical faculties and resist the spread of BSG. We need to go where the evidence takes us, even if that requires us to address awkward questions or difficult challenges to our position. Rather than seeking out studies or massaging data to support our prior convictions, we need to retain our curiosity about the many nuances and contradictions within the literature on responsible business.

In this way, through being trustworthy and transparent, we can find our way towards a model of responsible business, informed by the evidence, that has a decent chance of producing the benefits for society that we instinctively, and through our experience, know to be there. And we may have more success in bringing sceptics with us on the journey.