Understanding the noise that comes with ESG investing

There's a big governance drive toward ESG. Interest accelerated when [former United Nations Secretary General] Kofi Annan put together the UNPRI [United Nations Principles for Responsible Investment] initiative, and it's become a hot topic in the boardroom. Demand for ESG implementation by managers has grown tremendously. There's a very broad consensus that these are desirable guidelines.

That's the easy part. It becomes much harder when you take a cold, hard look at ESG, and you start thinking about implementation and cost. The first stroke of ESG has been exclusionary policies — of stocks that have very poor scores in terms of ESG — which have some impact on portfolio construction. But it's not very meaningful, especially if these are long-only passive portfolios. If you go to the next step, which consists of saying, 'Let's fully integrate E, S and G in all of our processes,' in an equity market-neutral format and manage to constant volatility or risk, then you have much more meaningful impacts from deploying these strategies. When you simulate the cost of implementing them in an integrated fashion, you start coming up with some interesting data.

That's a problematic argument. Our research indicates that it does not enhance returns overall. The good news is that implementation of the governance factor G is naturally good and does enhance returns. However, the governance factor can be largely claimed by, and overlaps with, the quality factor. The bad news is the positive signal of G decays into noise after inclusion of E and S.

I have to caveat all this by the fact that this is not a very rich data environment. E and S factors are in large part produced by surveys that are sent to companies and answered roughly once a year. If you've been doing this for 20 years, you've got 20 data points. So in terms of data density, this is a less-than-ideal environment to apply statistics. Also, these surveys vary enormously among the firms that are producing them.

With that caveat, our finding is that you're essentially looking at E and S producing noise minus cost in statistical terms.

Our paper is about a fully integrated ESG strategy within the framework of an equity-market-neutral portfolio that's risk managed. It's quite specific. If you're constructing a passive portfolio that isn't risk managed, the implications are very different: It's much less costly.

We used our simulation capabilities that we use for all of our other strategies. We found that when regressing the ESG factors over stocks, most of the explanation for the performance of the stock could be found in the governance factor. Most of what you got from governance was, however, explained by quality as opposed to governance being a stand-alone factor.

We also found that large companies tend to score better on E than smaller companies. That is really a reflection of large companies being good at managing expectations regarding environmental impact, which means they have [investor relations] functions and a vested interest in reporting and scoring on environmental factors, whereas smaller companies do not.

It provides an insight into the cost of implementation. If you truly have an ESG policy and you want to apply it beyond simply having an exclusionary list, there's some serious costs and challenges. The higher the frequency at which you operate, the more implementation costs you're going to generate. If you're trading noise, that's noise minus trading costs.

The challenges are significant, and I would suggest that we take integration in bite sizes, and the most promising guideline is governance.

You need more data density, so you need disclosure to occur more often. We have accounting standards in Europe and the U.S., and everybody agrees as to what constitutes a sale or an accrual. This is rules-based, and it's not open to reinterpretation every year. We need to get to a point where we all agree as to what constitutes environmental impact for a given company or industry. It's going to take a concerted effort of companies, investors, regulators and political authorities in order to converge toward a set of environmental data that is dense and rules-based.

Social is very difficult. It's hard to see how you're going to get to a consensus regarding socially acceptable standards across G20 countries, let alone G30, or 40 or 50.

ESG is not something you can be anti ― being anti-good governance, anti-a better environment and anti-social equity. We are first and foremost scientists, and we look at ESG from an empirical basis. This paper provides a framework to think about how fast you can move on the spectrum from passive exclusion to a fully integrated ESG portfolio in active trading environments.

On the environmental side, I think there's reason for optimism that we can move to better standards of disclosures, but it's not going to happen quickly.

Through this research, we, as systematic quantitative managers, highlighted some of the inherent limitations of a wholesale integration of ESG information into a portfolio. We are nevertheless committed ― through various avenues, including our participation in the U.N. PRI Hedge Fund Working Group ― to contributing to the continuing developments in the industry. ■

Any description or information involving investment process or allocations is provide for illustrations purposes only.

Any statements regarding correlations or modes or other similar statements constitute only subjective views, are based upon expectations or beliefs, should not be relied on, are subject to change due to a variety of factors, including fluctuating market conditions, and involve inherent risks and uncertainties, both general and specific, many of which cannot be predicted or quantified and are beyond Capital Fund Management's control. Future evidence and actual results could differ materially from those set forth, contemplated by or underlying these statements. There can be no assurance that these statements are or will prove to be accurate or complete in any way. All figures are unaudited.