Many investment organizations are currently revisiting their ESG policies, in response to the recent sharp increase in interest in this area. 

This is an important exercise. Developing an ESG policy should involve some serious thinking about objectives and beliefs. If the policy is to be successfully embedded into the fabric of the organization, the policy should fit with the existing culture and investment proposition. A good policy strengthens the investment proposition and creates internal alignment. It’s worth putting in the effort to get it right.

And once a policy has been created, you want to be confident that it rests on a solid foundation. You want to be sure there are no obvious gaps, and that the policy can be defended in the event of a challenge. This has become all the more important as the debate around fiduciary duty heats up, and as expectations evolve. Whatever your policy, there are likely to be those who accuse you of doing too little, and others who accuse you of going too far. 

A useful framework to this end is one that was first developed more than twenty years ago by John Ilkiw, Mike Clark and myself. This involves asking the following five questions:

1. What is the impact on the expected level of investment return?

2. What are the implications of the policy with regard to risk?

3. How will the policy be implemented?

4. Does the policy have broad stakeholder acceptance?

5. Is the policy properly documented?

Let’s briefly look at each of these in turn.

1. The impact on expected return.

You need to understand how the policy is likely to affect the long-term average expected return on your portfolio. There are three broad conclusions that might be reached: (a) the policy is expected to increase returns on average (b) the policy does not change return expectations or (c) the policy is expected to reduce returns on average.

The policy may aim to enhance returns, for example, through effective stewardship or through the identification of mispriced factors or themes. 

Other policies may accept a reduced financial return as a price to be paid for a positive social impact. These policies are generally not in the purely financial best interest of the beneficiaries, which means they will only be appropriate where the beneficiaries have explicitly chosen this approach.  

And a significant number – probably most – ESG policies neither increase nor decrease expected returns. That does not, however, mean that actual returns will be unaffected by the policy; indeed, there will almost certainly be some periods over which the policy results in a higher return than would otherwise have been achieved and other periods over which it results in a lower return. So this test is not about the impact ex post, it is about the expected effect based on the information and analysis available ex ante. 

2. The impact on risk.

While expected return is basically a one-dimensional measure, risk is anything but. 

Many ESG policies are explicitly designed to reduce various forms of risk: risk associated with the regulatory response to climate change or environmental degradation, for example; with cybersecurity; with ineffective board oversight and misaligned management incentives; and with a long list of other things. 

ESG policies can also accentuate some forms of risk, or introduce new ones. A policy that excludes certain sectors, for example, may reduce the degree of diversification that is possible within a portfolio. 

Thought should particularly be given to the risks that are difficult to quantify. Many of the issues that ESG deals with are only partially understood. We don’t necessarily know enough about them to measure how big they are. As I noted in a recent post, “ESG is where the action is because of – not despite – the fact that investors are dealing with incomplete data, with unknown and shifting relationships, with changing (and widely varying) client expectations, with difficult-to-interpret science.”

That these risks are hard to pin down does not make them any less real. Quite the opposite, in fact: it makes it all the more important to think through what they mean for the portfolio.

3. Implementation.

You should also consider what it takes to implement the policy in practice. It’s hard to generalize here: there is not much in common between the steps required to implement a policy based on hands-on stewardship or one based on divestment or a thematic policy based on identifying and participating in key trends. 

But whatever the policy, the time and resources required to implement it need to be understood, along with the possibility of unintended consequences. The importance of implementation should not be underestimated; as Heather Myers and I noted in a 2008 overview of these five tests, it is this area “that usually throws up the greatest challenges for investors seeking to pursue responsible investment policies”.

That remains largely true today. Progress has been made in the past few years on disclosure and on the development of analytical tools, for example, but there’s still a long way to go. 

One area that remains especially problematic is the question of how to measure results. Where ESG considerations have been truly embedded into the investment process, it can be impossible to disentangle the effects of the ESG policy from those of the other elements of the process. Risk monitoring is only as good as the model on which it is based. And the measurement of a portfolio’s social or environmental impact remains a very imprecise science at this point.

4. Stakeholder buy-in.

These tests were first developed for pension funds and other asset owners. For these organizations, giving thought to the perspective of all stakeholders is a useful protection for fiduciaries against the risk of allowing one’s own ideological preferences to inappropriately shape a policy. Similarly, for an asset management organization, the stakeholder test can help to ensure that objectives are properly framed.

One of the goals of our five tests is to ensure that the policy can be defended in the event of challenge. To this end, it is easier to get stakeholder understanding and acceptance of a policy at the outset than it is to do so later, if results prove disappointing. Stakeholder buy-in from the outset also reduces the likelihood that a well-designed policy is abandoned at the worst possible time.

5. Documentation.

And rounding out our five tests is documentation. Not the most interesting topic in the world, but absolutely necessary if the exercise is to have lasting value. Keep in mind that if you are legally challenged, the legal argument is more likely to come down to the process followed than to the outcome that resulted. It can be invaluable to have documentation of the rationale behind your process, as understood in the context in which it was created, as a shield against the arrows that hindsight might throw.

In conclusion, even though the ESG landscape has changed beyond recognition in the twenty-plus years since John, Mike and I first wrote down our five tests of an ESG policy, they remain a useful framework for any organization looking for reassurance that their policy is built on a solid foundation.

ENDNOTE (added October 2021): Not five tests, but six

It turns out something is missing from the five tests above. Because while the impact of the policy on the investment program is essential for trustees to understand, impact flows the other way too. So there’s a sixth question that must be considered, namely:

What is the expected environmental and/or social impact of the policy?

More on this additional point can be found here.