One of the most common criticisms I hear about ESG investing is that the ratings do not agree. The same company can receive a high score from one provider and a mediocre score from another. For many, that is evidence that ESG ratings are unreliable.
I have started to wonder whether we are drawing the wrong conclusion.
In most areas of investing, disagreement is not a failure of the market - it is the reason markets exist. If everyone assessed risk in exactly the same way, there would be little opportunity to uncover value. Different views create price discovery. So why do we expect ESG ratings to converge when they are trying to evaluate something as complex as corporate behaviour, governance, and long-term resilience?
The first thing worth recognising is that ESG rating agencies are not necessarily measuring the same thing. They use different data, different methodologies, different weightings, and in some cases different definitions of what good looks like. Expecting identical outcomes is a little like expecting every economist to produce the same forecast. The variation should not surprise us.
What interests me more is where the disagreement is greatest. When one agency sees a leader and another sees a laggard, I do not immediately ask which one is right. I ask what they are seeing that the others are not. Large differences often point to something worth investigating - opaque governance, inconsistent disclosures, emerging controversies, or simply a business that is changing faster than traditional assessments can keep up. In other words, disagreement can be a signal that uncertainty is higher than average.
I have found this way of thinking much more useful than averaging the scores together. A consensus score may look reassuring, but it also smooths away the very information that deserves attention. Investors spend considerable effort identifying uncertainty in financial markets. We should be just as interested in identifying uncertainty in sustainability data.
None of this means ESG ratings are beyond criticism. Like any analytical tool, they have limitations and biases. But perhaps we have spent too much time asking whether the ratings are correct and not enough time asking why they differ.
If several independent experts cannot reach the same conclusion about a company, that tells me something important. Not necessarily about the quality of the company, but about the quality of the information available to judge it.
Instead of treating ESG rating divergence as noise to be averaged away, perhaps we should treat it as a risk indicator in its own right. Sometimes the most valuable insight is not the score itself - it is the fact that informed observers see the same company very differently.