Synechron Aims to Normalize ESG Ratings

As sustainable and well-intentioned as environmental, social, and governance (ESG) investing is playing out to be, it can be a convoluted and messy space for investment firms to navigate. The quality of ESG data remains a challenge as disclosures are often inconsistent and not standardized. Also, ESG rating providers such as MSCI, Refinitiv, Sustainalytics, and Bloomberg—the latest to offer such ratings—differ in their methodologies and how they examine the same exact firms. 

This sector suffers from a lack of standards in the data, says Sandeep Kumar, head of Synechron’s buy-side accelerator program. 

“When, let’s say, Refinitiv or MSCI says, ‘Hey this particular stock has an ESG score of 70’, and Sustainalytics says ‘Hey, this [same stock] has a score of 35’, there is no way to match these scores,” he says. 

Investment firms are looking for standardization of ESG ratings, Kumar tells WatersTechnology, in the same way that credit ratings are now standardized, so firms understand what a “AAA rating” is, or what an “investment-grade bond” means. 

“It doesn’t matter whether you go to Moody’s or S&P, you can relate to these [credit] ratings. But there is no such grading of ESG ratings yet in the industry,” he says. 

To address this, some firms are building internal frameworks to better understand the impact of ESG factors on their portfolios. “Many firms, especially the big ones, have their own mechanisms of figuring out the ESG scores of companies—they don’t just rely on the ratings. Hedge funds like AQR do their own research, and have their own data engines and apply NLP on top of it to come up with their own scores,” he says. 

For this reason, Synechron will be launching a new accelerator—its term for minimum viable prototypes showcasing the use of new technologies applied to a general use case—by the end of October, which aims to solve at least two specific problems within the ESG space: normalization of ESG ratings, and predicting returns of ESG investments. 

The ESG workbench, which Synechron has been working on since May, will allow firms to bring in the ratings from the different rating providers and normalize the ratings looking at their past performance and how each provider calculates the “E”, “S”, and “G” scores. “By breaking this up, we are trying to combine and normalize these ratings and give a uniform view of these ratings,” Kumar says. 

It hasn’t been easy, he adds. 

“To give an example, Refinitiv has a positive rating, so, the higher the score, the higher the ESG rating, whereas Sustainalytics, they have it the other way around: If the ESG score is high, it means a lot of attention needs to be paid to this company’s environmental or social factors. So, a lower score means a higher ESG rating there,” he says. (For more on the technical components of the service, see “Technical Components” below.)

Another challenge is when there are new rating providers. Kumar says that more established players like MSCI, for example, have “perfected this and standardized how they have done it.” The newer entrants don’t have the historical ratings, “so there’s no way to test the ratings to show how firms have performed over the years. Just knowing the rating at a single point in time is not helpful,” he says. 

Synechron is working with RavenPack, which provides stock-specific and macro sentiment analysis and scores. It is considering how it can combine RavenPack’s analysis into its tool. 

Kumar says RavenPack has a “no human” approach in how it analyzes companies. Instead, it uses machine-learning algorithms and NLP. “The reason they do that is they don’t want to bring subjectivity in saying ‘Tesla is good’. They want to scan news articles, company filings, supply chains, and other factors and then come up with a rating,” he says.

ESG = Better Returns? 

Then comes the second problem Synechron looks to solve. Does investing in ESG result in better returns, and can ESG scores predict a portfolio’s performance? 

Early ESG investors may have pushed forward with their investments even though they were getting returns 1% lower than regular funds because they believe in the sustainability factors and “it’s the right way to do it”. 

“But as more investors have gone into ESG, those returns have gone away,” Kumar says. “There is research that shows that just because you are investing in [companies with good ESG scores], it will not give you better returns. It will take some time to figure out how to relate an investment performance with ESG.” 

In the Cloud

Separately, since partnering with Google Cloud in June, Synechron has been working to bring two new prototypes on the Google Cloud platform. The prototypes have been developed using open-source technology and machine learning, says Anand Chandra, senior director at Synechron.

The first is an extension of Liquify, a Treasury-as-a-Service API solution already developed by Synechron that gives banks’ customers access to FX cash management, hedging, liquidity management, daily cash management, and other cash investment products. ”We are working on two key prototypes, which build on cloud functionality. One of them is Liquify 2.0. We are going to extend the Liquify platform to include cash forecasting,” Chandra says. 

The Liquify model has an overnight algorithm called “cash sweeping” to look for end-of-day balances. This will be extended to include cash forecasting, tapping on Synechron’s AI platform—Neo—which comprises NLP, chatbots, robotic process automation (RPA), cognitive machine learning, data science, and robo-advisor accelerators. “The portal will leverage native AI and the analytics capabilities of the Google Cloud platform, with the end solution being a hybrid cloud product, built on foundations of open banking,” Chandra says.

The second prototype, a hybrid cloud-hosted framework, is an ESG trade enabler that integrates emerging asset classes on to a buy-side firm’s existing technology infrastructure.   

“One particular challenge for the buy side is that they might not have the technology to fully integrate new asset classes,” he says. 

Chandra says the two new products show how Synechron can work with Google Cloud capabilities. Synechron has around 150 engineers who have gone through the Google Cloud platform training and are working on how to leverage it for clients. “Our next focus is to invest in the capability build out across different locations,” Chandra says.