There are limits to how much of an ESG tilt you can introduce to your portfolio before it becomes more risky than a market index.
Employing a thoughtful environmental, social and corporate governance (ESG) policy has tangible benefits.
Not only will it contribute to better societal outcomes, but it will also reduce risk in a portfolio by avoiding companies that have the highest likelihood of collapse due to reputational damage or litigation, or simply having a less predictable risk profile.
But there are limits to how far you can tweak your portfolio before it starts to become more risky compared to a simple market capitalisation-weighted index. And when you go beyond those limits, chances are you are in breach of your fiduciary responsibilities.
You may argue that if you tilt on ESG, that you get a better factor footprint, but is it economically true that it is the same portfolio?
“You may argue that if you tilt on ESG, that you get a better factor footprint, but is it economically true that it is the same portfolio?” de Silva says in an interview with [i3] Insights.
“If it has a higher price/earnings (PE) ratio and a lower return on equity (ROE) than the cap-weighted index, then you are taking an active factor tilt in the interest of ESG that might not be something that retirees or superannuation investors want to take.”
De Silva and his team have built a risk model that tracks ESG factors and measures how each one introduces volatility to a portfolio. They use data from MSCI, but also from an alternative data provider, called OWL Analytics, which tracks more than 25,000 companies and updates their ratings monthly.
“Most risk models use fundamental factors, like PE or debt-to-equity ratios, but they don’t use ESG factors,” de Silva says.
“The question is then: ‘If I want to build a better portfolio with a focus on governance, then how much tracking error can I take in the portfolio before I have a fiduciary issue?’”
De Silva’s work was inspired by guidance from the United States Department of Labor, in which the department warned that while ESG policies could help reduce risk, fiduciaries should not “too readily treat ESG factors as economically relevant to the particular investment choices”.
“The Department of Labor sent out a letter that said you can’t give up your fiduciary responsibility by providing an ESG tilt,” de Silva says.
But if you can show that a portfolio has a similar exposure to economic factors, then you could make the case that you are not taking any undue risks in your ESG approach, he says.
“For example, if you take an equity portfolio, then what you need to do is agree on what the equity factors are, whether it is value, momentum, quality, small cap and so on. How much ESG tilt can you get, but yet have the same factor footprint?
“If your portfolio has the same PE ratio, price-to-book ratio and ROE, then you can claim fiduciarily that it is the same portfolio.
“You can get tracking errors under 3 per cent and still have approximately the same factor footprint. Once it is starting to get to 5 per cent then it starts to become impossible to build a portfolio that has better ESG and the same factor footprint.”
The (US) Department of Labor sent out a letter that said you can’t give up your fiduciary responsibility by providing an ESG tilt
This approach can also be applied when funds consider divesting from companies altogether.
In Australia, most of the large pension funds have divested their holdings in tobacco companies. But de Silva says any such measure should be accompanied with reweighting to companies with similar factor profiles as tobacco producers.
“When you buy a tobacco company, then yes you are buying tobacco, but you are also buying a consumer staples company with very low earnings variability,” he says.
“What you need to do is find other companies that have the same cash-flow variability, the same sales-per-share variability as tobacco companies.
“These companies then give you the same behaviour in different parts of the cycle as tobacco companies would.”
But he argues divesting from entire sectors would be difficult to justify.
“Tobacco is much easier than energy because it is a smaller group of companies. But energy is a good example, where if people say ‘look, I don’t want any exposure to oil producers’, then the tracking error becomes very large,” he says.
Not all ESG factors are created equal. Some factors introduce more volatility to a portfolio in terms of tracking error than others and it is possible to optimise an ESG approach for maximum ESG impact and minimum tracking error.
“Some ESG factors don’t introduce much tracking error and so you can get a lot of that factor in your portfolio, while others produce a lot of tracking errors. This way you can kind of balance off the different types of ESG factors,” de Silva says.
It is not simply a case of doing this exercise once and then loading up on the ESG factors that introduce the least volatility as the relationships are not always static. Investors should not only think about the volatility of the ESG factor itself, but also about its correlation to market cycles and how this correlation changes over time.
“Part of it is the covariance. There might be an ESG factor that has a lot of volatility, but if it is also very correlated to the market, then by taking it you don’t take a lot of tracking error risk,” de Silva says.
[i3] Insights is the official educational bulletin of the Investment Innovation Institute [i3]. It covers major trends and innovations in institutional investing, providing independent and thought-provoking content about pension funds, insurance companies and sovereign wealth funds across the globe.