ESG: focus on the right dimensions of ESG and avoid the “confirmation bias”
In a key note presentation, Alex Edmans, Professor of Finance at London Business School, voiced concerns that confirmation bias, i.e. accepting “evidence” only if it confirms what we would like to be true, is increasingly common in ESG.
“Inconvenient truths” concerning, for example, gender diversity, executive pay and environmental and social governance, should not be buried. “Evidence” put forward included choosing unpublished research over peer-reviewed published results, the use of results based on just two years of evidence, or lacking statistical significance, or where dividends of stocks had been ignored, or comparing results estimated using carefully chosen and inconsistent time periods.
Edmans asked the audience to dismiss common misgivings about ESG.
- For example, the claim that average ESG funds underperform. This is not relevant because stock selection for those funds also relies on traditional financial variables and includes trading costs, so it is difficult to separate all the effects.
- The argument that there is no unique definition of ESG should be ignored since there is no clear definition of virtually anything intangible: for example, there is no consensus in the analysts’ earnings forecasts and these are still used.
- Finally, the argument that ESG data is of low quality, self-reported and unaudited is overplayed. The same applies to non-ESG data: earnings are affected by accruals; R&D measures only input and not output; the corporate outlook and strategy are self-reported; and all this information is still used. ESG data is increasingly objective, independent and substantiated.
Edmans concluded that certain ESG strategies do outperform, for example, those centring on shareholder rights, management incentives, the responsible use of corporate jets, employee satisfaction, customer satisfaction, eco-efficiency or based on performance on material issues. However, investors need to focus on the right dimensions of ESG and need to be discerning about which data to use. The challenges in ESG implementation increase the value of good ESG strategies, he said.
Carbon: credit markets have bigger role to play in reduction of emissions
Ulf Erlandsson, chief investment officer at Strukturinvest, argued that credit is the most relevant asset class when it comes to attempts to reduce carbon emissions because of its larger size relative to other asset classes, and because ultimately companies will struggle to survive if credit is not made available to them. Measuring carbon emissions in credit portfolios is thus of key importance.
Erlandsson proposed a relatively simple model, ECO2BAR, which looks at CO2 emissions from an ordinal standpoint and takes a risk-based approach to measuring this in credit portfolios. The model is based on first dividing the universe into sectors, penalising most the biggest polluters, and then, in each sector, penalising the companies responsible for the largest carbon dioxide emissions.
He built the model while encompassing important credit alpha generation mechanisms such as shorting positions, taking leverage and using derivatives as well as accounting for explicit investments in green bonds. He showed that carbon reduction did not materially impact the alpha generation in a few credit strategies used as examples. He concluded that there is room for reducing CO2 in credit portfolios without necessarily sacrificing performance.
Engagement: successfully engaged companies tend to deliver higher stock returns
Tamas Barko of Mannheim Business School talked about ESG activism. He examined unique proprietary data on company activism from 2005 to 2014 involving more than 800 engagements by a large European asset manager.
Barko found that targeted companies typically have a higher market share and are covered by more analysts than a control group with untargeted peer companies. About 60% of the engagements were closed successfully by the asset manager and developing a good working relation was seen as the crucial factor.
Companies with low ESG ratings had a bigger chance on activism and on improved ratings thereafter. The reverse was observed at targets with high ESG ratings at the outset. Successfully engaged companies tended to deliver higher stock returns than the control group and unsuccessful involvements underperformed the control group. Lastly, activism did not impact any accounting and operational performance figures, except for increased sales growth after successful engagements.
To read more on our Environmental, social and governance theme, click here.