In this wide-ranging Q&A, Arnaud Sarfati, CEO and co-founder of Paris-based quantitative asset manager LFIS Capital, discusses how the firm is tackling ESG implementation in its quant strategies, touching on portfolio constraints, evolving regulation, auditing concerns, and “data exhaustivity”.
Sustainability and ESG (environmental, social and governance) factors are reshaping investment decisions at both ends of the manager-investor spectrum, and the way hedge funds and other investment managers select positions and build their trading books is coming under ever-closer scrutiny. The assortment of datasets and the risks of ‘greenwashing’ pose a particular challenge when it comes to consistently measuring and monitoring sustainability criteria across portfolios.
Paris-based quantitative asset manager LFIS Capital, which manages a range of alternative strategies spanning credit, managed futures, and risk premia, has been ramping up its ESG focus in recent years. Much of the firm’s work centres around how ESG and sustainability metrics can be better improved in order to help generate portfolio returns and drive alpha.
Hedge fund strategies that use computer-based trading algorithms to crunch numbers face a substantial hurdle in the face of inconsistent and infrequent ESG indicators, says LFIS CEO and co-founder Arnaud Sarfati.
In a wide-ranging Q&A, Sarfati – who previously spent 15 years at Société Générale Corporate & Investment Banking, latterly as global co-head of its Cross Asset Solutions Group, having earlier held a number of senior roles focused on financial engineering and equity derivatives – discusses how LFIS is tackling ESG implementation in its quant strategies, touching on portfolio constraints, evolving regulation, auditing concerns, and “data exhaustivity”.
What are the current hedge fund investor and manager approaches towards certain asset classes and instruments regarding ESG investing?
There is a significant amount of variety in what different firms are doing with regard to ESG investing, so it can be hard to generalise on specific approaches. What we can say is that, after an initial surge in interest in ESG investing, the market is entering into a more discerning second stage. Firms are increasingly challenged on the link between performance and ESG, specifically how they balance their fiduciary responsibility to deliver risk-adjusted performance with ESG constraints. There is also greater awareness and vigilance from investors concerning ESG data and potential ‘greenwashing’. To paraphrase French philosopher Denis Diderot: ‘It is not enough to do good. You need to do it well’.”
Can you describe the key challenges and hurdles that certain managers – either discretionary or quant – face when it comes to ESG?
A main challenge LFIS Capital and other quant firms face when it comes to ESG investing is the data. There are nearly 100 different ESG data and ratings providers and each has its own approach to sourcing, research and scoring. This calls into question the data itself. As an illustration, correlations between credit ratings from S&P, Moody’s and Fitch exceed 95 per cent, but the correlation between ESG scores from traditional data providers is much lower, between 11 per cent and 63 per cent.
Data exhaustivity is another issue, with companies reporting varying amounts of ESG data, which makes fair comparisons difficult. Many ESG providers have developed proprietary models to estimate unreported data, and investors are incorporating these judgment calls into their process without a lot of transparency.
Also, ESG data is not subjected to any official auditing process, raising questions of validity. For example, today’s ESG reports are where financial reports were dozens of years ago in terms of maturity. Moreover, ESG indicators are usually submitted annually or, at best, a few times a year. This very low frequency of data is insufficient for quantitative investors like LFIS which rely on reactive valuation and risk models.
Does the expansion of alternative data and the demand for data engineers and coders among hedge fund firms offer the chance for some managers to get an edge on competitors?
Like all quantitative asset managers, LFIS has always focused on data engineers, coders and IT profiles in our hiring. While it is true that we are now able to exploit a lot more data including textual data, processing and incorporating alternative data is not all that different from traditional financial data, so our hiring needs have not changed that much in this respect. When you decide to partner with an alternative data provider, it is important to understand the underlying methodology and be able to parse and integrate it efficiently. Here, quantitative hedge funds may certainly have an advantage.
What is the best way forward for quants and their approach to ESG to overcome data issues?
LFIS is a quant asset manager, so naturally when researching ESG, we started with the data. Our analysis uncovered certain drawbacks with traditional ESG data, including those already mentioned. To address these concerns, we launched a research axis to look at whether applying alternative intelligence, machine learning and NLP to textual data from news and blogs could provide data that was better suited to our needs, and developed alternative ESG data that is designed to be more reactive, predictive, transparent.
How is the changing regulatory backdrop on ESG and sustainable investing influencing where hedge funds and quants choose to invest?
Efforts to create a universal framework for ESG measurement that help to address the lack of consistency in ESG data and reporting is something we encourage. More reliable ESG data is, of course, in the best interests of asset managers and investors alike.
The changing regulatory backdrop, including the stipulations brought in by the EU’s Sustainable Finance Disclosure Regulation (SFDR), will help bring clarity to the ESG investing landscape, though more from the end investor perspective as opposed to the manager’s approach to where they invest.
However, with half of European funds expected to be classified as 8 (‘promote ESG characteristics’) or 9 (‘have ESG as an explicit objective’) by early next year, there is still the risk of greenwashing. In our view, there is no substitute for having our own methodology, having full transparency and prioritising agility in ESG investing. This approach is the foundation of our ESG programme.
