- September 19, 2014
AP4 already has US and emerging markets low-carbon mandates and plans to invest up to ... [more]
Research on environmental, social and corporate governance (ESG) and investments has advanced in rigour, coverage and volume, but data quality, and the problems of reverse causality are still concerns for academics looking for a meaningful relationship between ESG factors and investment performance.
A fundamental question about responsible investment is whether using ESG information enhances investment performance. Georgios Serafeim, assistant professor of business administration at Harvard Business School, says there have been many attempts to connect ESG and investment performance, but for him the quality of data remains the main barrier to any really conclusive outcome.
“Compared to the integrity of financial accounting with its data, auditing, mechanisms and measuring systems, ESG data is noisy, which means the probability of finding a significant relationship is less – it’s econometrics 101,” he says.
Jane Ambachtsheer, partner and global head of responsible investment at Mercer, agrees.
“In the past few years this has been a growing area of academic study and it has expanded in coverage across different asset classes. There is evidence to show there is not a performance penalty, but it is harder to make clear-cut the case in support of the positive investment case. There are still a lot of issues around quality of information and data,” she says.
Quality research counts
Ambachtsheer and Serafeim were speakers at a United Nations-backed Principles for Responsible Investment (UNPRI) academic-run webinar, which brought together academics and practitioners to discuss the developments in ESG investment studies and integration since the United Nations Environment Program Finance Initiative (UNEPFI) released its seminal 2007 report Demystifying Responsible Investment Performance.
A paper by Sweden’s Seventh National Pension Fund (AP7), which reviewed an additional 21 academic studies published after UNEPFI’s report, was also presented at the webinar. It focuses only on environment and social and omits governance studies.
The results of this review, The Performance of Socially Responsible, reinforce that there is nothing to suggest that responsibility for environmental and ethical issues in asset management in general either raises or lowers returns.
Two thirds of the studies in this report state that there is no obvious connection. And in the last third, five studies suggest a positive correlation while three point to a negative correlation.
With regard to AP7’s study, Ambachtsheer says funds labelled as Socially Responsible Investment (SRI) are a legitimate area of study, but it is difficult to compare across ESG as a screen of decision-making and ESG as an investment screen.
“From a fiduciary perspective the study provides comfort that you’re not destroying value. But it doesn’t answer whether ESG factors hold the key to better risk/return outcomes,” she says.
For Serafeim, it also highlights the problem that even when a relationship is documented it might be statistically significant but not economically so.
“There may be a certain effect and when you scale it by a standard error, it is a relatively big effect, but economically it’s not that significant.”
Serafeim also believes that when it comes to the academic study of ESG and investment performance there is a possible case of ‘reverse causality’.
“It’s a difficult one to solve. There could be a case of reverse causality, where financial performance is causing ESG, not the other way. This affects what you can take from the results.”
Patience will pay for performance
Serafeim presented at the webinar with his colleague Bob Eccles, professor of management practice at Harvard Business School.
They believe there needs to be more patience in the field and that material results will take a long time to appear.
“People want the answer before the experiment,” Eccles says. “Longer time frames are needed to measure the impact of ESG and performance.”
“It is hard to believe there will be a relationship between a rating and earnings of next year’s stock returns. It is hard theoretically to understand why there should be a relationship between them. It is not a fixed time but certainly not over one year, maybe five, seven or 10 years. A long-term perspective is needed – it is about long-term performance – and this leads back to why studies don’t find anything.”
Time will help heal the problems of reliable data too, says Eccles, pointing to the evolution of accounting standards over a 75-year time period.
However, as Ambachtsheer points out, perhaps the information asymmetry is also a period of opportunity, as information is at a premium. Research is already underway to supply this demand.
Recent work by Frank Figge and colleagues helps to assess data quality and studies on how investors use sustainability information by Anna Young at the University of Sydney Business School, DanielBeunza at the London School of Economics and Fabrizio Ferraro at IESE Business School are worthy examples. This type of work will help us to unravel the performance question and establish links realbetween ESG and investment.