Podcast: Play in new window | Download | Embed
Subscribe: RSS
Topics Mentioned
- ESG Data Sources
- ESG Scoring Methodologies
Episode Transcript
Craig: I’m really excited to introduce our next guest. It is Ben Webster, co-founder and CEO of OWL ESG. Ben, hey man, welcome to the program.
Ben: Hey, thanks for having me.
Craig: I’m glad we could make it. We just saw each other in person back at the T3 conference. Good to see to see you.
Ben: It was great to meet you in person and that was an amazing event and I met a lot of amazing people.
Craig: I met a lot of people I know and people I didn’t know and it was great being around people and hanging out and getting that in person interaction, sort of serendipity walking around the trade floor and running into people that maybe you didn’t know, which you don’t get on Zoom so it was all good.
Ben: Yeah, it was my first time there. I really enjoyed that as well.
Craig: It’s always nice see people and see what they think. I think this is my eighth or ninth T3 conference, but yeah, it’s a big one for our industry. Where are you calling in from Ben?
Ben: Los Angeles, California.
Craig: Excellent, you sound great. I’m on the East Coast and everything is great over here, but let’s get kicked off. Can you please give us the 30 second elevator pitch for OWL ESG?
Ben: Yeah, for sure. OWL ESG is essentially an environmental, social, and governance ESG data and analytics firm. We work with a wide variety of institutional investor clients across the world, hedge funds, large asset management platforms, banks, insurance companies, financial technology platforms, advisors, wealth mangers, and so forth and so on and what we help them do is to make more impactful investment choices with the latest and greatest ESG data available on the market to help them serve their client mandates.
ESG Data Sources
Craig: One of the reasons why we’re talking and why we’ve been talking is there is no one ESG data source that everyone uses. There are lots of them, so can you talk about what the impetus was for your company and how you are solving some of these problems with all of the different providers of ESG data?
Ben: Yeah, happy to do that. Kind of the story goes back a decade approximately. My partner and I, Andrew Smith, we wanted to start an investment management company, primarily an ETF company, and we identified ESG, impact investing, sustainable investing as something that we’re both passionate about and decided that hey, we need to figure out how to execute this in an index that would power ETFs. We were looking at many different options, including creating our own data, but decided to first start by talking to the predominant ESG data analytics providers out on the market and without naming names, we started to get access to some data from a number of them and we quickly identified that ratings were very subjective, so a good example of what I mean by that is if you look at a bond rating, right, a credit rating, right, and you look at the ratings from S&P Global, Fitch, Moody’s, or others that exist out there and you compare them, any two of those bond rating firms, they’re typically going to be above a 95% correlation. That means if any given company has a good credit rating, like say an A or triple A with one credit rating agency, it’s going to be right around there with another.
Ben: That wasn’t true with ESG ratings. Any given company could have a great ESG score or a great E score, S score, or G score or other types of ESG scores and those similar scores could be anywhere from poor or dreadful with another provider, all the way up to great as well and so the correlation at that time was somewhere in the low 30% correlation, which we translate to mean that’s highly subjective and so, I come from a physics background, my partner came from a finance background, but with a heavy emphasis on data science and stats and math, what have you, and it just didn’t sit well with us. We were like, we can’t use these metrics because if we work with this one provider, this company would not be in our index. If we work with another provider, that company would be in our index and we just couldn’t sit with that, so what we decided to do is like well, you know what? Let’s see if we can work with multiple providers to create a consensus score that was ours that we could then power our ETFs and we spent a number of years trying to launch an ETF company while we had created our own consensus scores to solve that subjectivity problem and that’s kind of where the origination of our metrics or our core product, the OWL ESG consensus scores came from was we wanted to build investment products ourselves and the scores ended up having such resonance with anyone we spoke to about them that we just decided to become a data company.
Craig: That makes a lot of sense. You were scratching your own itch and decided it was something that you could then sell to other firms.
Ben: Exactly, so we created those scores to solve our own problem, but if you really think about, if we decide to create our own ESG ratings based off our own-and we can go into why the ratings are subjective in second, but creating our own ratings based on how we looked at the world, we would just be like any other provider, creating our own subjective view. By doing what we do and aggregating hundreds of sources and many of the largest ratings providers, we then can do something that takes it to the next level of objectivity and rate based on consensus, much like consensus earnings estimates are used to generate consensus from multiple subjective opinions of analysts on where the earnings are going to be for a company in the future and so that’s kind of what we decided to do and then I’m happy to get into why ratings are subjective if that’s of interest to you.
ESG Scoring Methodologies
Craig: Absolutely. I mean, I’d really like to know that. When I look at a lot of these data sources, I can’t figure out how they’re deciding which criteria to use. MSCI uses 35 parameters, how do they decide how to weight them, how they decide where this data is coming from? Is that data clean? If you’ve got 35 subjective data sources you are bringing into one score, how can anyone trust that score matches what any investor would think was a valid ESG number?
Ben: Yeah, so that’s a great question. Just to be clear, we’re big fans of what all the ratings providers do. It’s hard. Let me tell you why what they are doing is hard, what Morningstar is doing that’s hard, and all of the other providers out there and I do want to give them credit where credit is due. One of the main reasons why it is subjective is because ESG disclosure is not regulated. It’s not like when you have credit ratings, they’re all looking at regulated, mandatory financial disclosures that all companies have to disclose, so the credit rating agencies are working with a very consistent data set that for the most part-and we know there can be shenanigans that go into-that companies can do to make themselves look more solvent than they are and we all know that, but those are all known things that are pretty much-people doing credit ratings know about those and they have developed techniques on how to deal with it, but they’re all looking at a very standardized data set and so Moody’s and the S&P Global and Fitch and the others all look at the same thing.
Ben: The problem with ESG scores are because it’s not regulated. Any given company like the companies that ESG ratings and research firms would be evaluating, they can report-they don’t have to report anything. They can report on some things and not others. Two companies in the same industry can each be deciding to report on certain ESG metrics, but the metrics that one oil and gas company report may not be the same ESG metrics that another oil and gas company reports. When I say metrics, things like carbon data, water waste, whatever it may be, they’re not reporting on the same thing and they’re not reporting them in the same ways, so they could be reporting on carbon emissions, but maybe one is reporting on total carbon emissions and one is reporting on certain subsets of carbon emissions or they would be putting them into goals, not into past performance and it’s not even audited, so that’s difficult and so what you end up getting is if you’re someone at an ESG ratings firm, you’re trying to go okay, let’s figure out how we’re going to rate oil and gas companies or how we’re going to rate trucking companies and shipping companies and how we’re going to rate ship makers. You’re looking at any group of companies within a segment, let’s just stick with oil and gas, and the data is completely inconsistent, there are blanks everywhere, it’s not standardized. How do you create a standardized data set, standardized ratings whey the raw materials are so all over the place? By nature, they have to take subjective leaps. They have to go okay, how do I rate British Petroleum and how do I rate Exxon on an emissions score, triple A on emissions or whatever it may be, when they’re not scoring the same thing and they have to take subjective leaps to-they don’t have the same raw data, excuse me, so they have to take subjective leaps to do that. That’s where the issue arises.
Ben: Each of the different ratings firms make ultimately different subjective leaps and they all have their reasons for making those subjective leaps and frankly, they change it over time. They try to get better or in some cases, more marketable, or in some cases maybe they cut corners and in some cases they don’t. Without pointing at any given firm, it’s tough no matter how you slice and dice it and it’s a difficult issue and so what we’re trying to do is cut through all of that noise and go, source one, this is what they think about British Petroleum. Source two, this is what they think about British Petroleum and what are the differences in what they think and what are the similarities in what they think and when I say about what they think, what they think in regard to how British Petroleum is managing certain ESG risks and opportunities. How is British Petroleum managing their pollution risk? How are they managing their risk in the local markets they work in and how they may be destroying the environment or helping the environment and where they’re drilling, where they’re selling, all of those types of things. We are seeing on that topic and topic by topic, how source one and source two agree or disagree on how British Petroleum is doing and so we’re taking those advanced-we’re looking at this deeper and trying to isolate where the sources agree and making a standardized data set that’s useable, clean, consistent for clients.
Craig: That’s cool. There is a lot to unpack there. Can we talk about how many data sources are you bringing in?
Ben: Over 600 at this point.
Craig: I didn’t realize there were 600 ESG data providers. Are they equal or are they all the same?
Ben: There are probably like 50 or 60 ESG data providers in there, but around 13 or 14 major providers in there and then a whole bunch of specialists and then the remaining sources are other types of companies, NGOs, non-profits, and so forth and so on that they’re not ESG data providers, but in the process of doing what they do, they are collecting data points that are ESG data points. Like for example, a non-profit that may be working with the clothing manufacturing industry and are consulting with those clothing manufacturers on how to make their supply chain conflict free and free of labor violations and so on and so forth, so in the process, that non-profit or whatever type of entity that we’re working with, they’re gathering that data to do what that do and that exhaust gets given to us.
Craig: What is your tech stack? What did you build on the back end to make all of this work?
Ben: It’s very much cloud based. We run it on AWS. We’re using Python and a lot of the tools available in Python. Essentially, it’s a big process in data science, analyzing our different sources and running regression analysis and determining correlations to, again, identify when each of our sources look at a company, what is the ESG view of that source on that company and how does that compare to the view source by source on that company?
Determining Overlap Between Data Sources
Craig: In the regression you run, you are trying to determine the overlap between the different sources?
Ben: Yeah, exactly, determine hey, for this company, over time have they been looking at, for example, board diversity or water waste or C02 emissions and other emissions or gender pay gap or whatever the ESG issues that they rate? Over time has this source and that source continued to look at that issue as something that they’re scoring for that company? Sometimes there are issues where just because they call-like if you look at a certain source and they may call it emissions, but another source may call it carbon emissions and they’re rating those issues. Are they the same thing or are they different? We’re doing a lot of analysis of that type to identify how correlated those two are and that determines a degree of consensus on that issue, are they in agreement or not in agreement?
Craig: You’ve got your data science, your advanced statistics, you are running a regression, you’re detecting correlations to identify commonalities between sources and determine the overlap of the relevant ESG statistics, so what can companies-what do your clients do with this output?
Ben: There are a lot of different types of clients and I know that we have a lot of hedge funds, large asset manager shops, we have ETF issuers that use our data and what they’re doing essentially is constructing an investment product. A lot of that clientele is looking at our data and trying to find some sort of risk mitigation or some sort of alpha. They also use it sometimes to constrain their universe. For example, we only want companies that are in the top 50 percentile for their industry on our diversity in workplace rights score or I want to make sure that all companies in the bottom quartile on pollution prevention and climate change score are not in my portfolio, things like that. They’re doing to construct their universe and screen out or screen in companies into their universe as well, so that’s used and they’re doing it for reporting, so when they sit down with investors, we have a lot of-in the wealth space, in the wealth management space and the advisory space, people sit down with their clients and the advisors or wealth managers sit with their clients and go, hey, what’s important to you? Do you care about pollution prevention and climate change? Do you care about management ethics? I do, but not as much. I’m fine not looking at that metric or having companies that may score poor on that because I really care about pollution prevention and climate change. The long and the short of it is a lot of advisor work with a lot of female clients and it’s huge amongst female investors, millennial investors, and younger generation investors and a lot of RIAs and wealth managers are trying to future safe their firms, knowing that if you invest in serving those demographics and doing a great job on it and building your practice around what they care about and they very much care about ESG, that they’re going to have better client retention, a lot more asset management growth, and building a bigger and better practice and you need to engage with those clients and ask them, what do you care about? Let’s look at what we can measure with OWL’s data and let’s see of those ESG themes and issues, what do you care about and let’s build a portfolio and choose funds or stocks or bonds according to what matters to you.
Craig: What matters is that different investors care about different issues, so what’s the most important thing that you need in your data to be able to support those investors?
Ben: That’s a great question. What I would say is that the most important thing is giving them an easy means of having conversations with their investors and determining suitability and then reporting on it and ensuring that you engage regularly with them on whatever your updates are, whether they’re annual update meetings or quarterly update meetings. I think that the best means for doing that is through some sort of software as a service obligation. We are launching our own soon, but we work with a lot other financial technology providers that embed our data into their workflows and allowing-for example, we work FI360 and Broadridge, allowing the plan fiduciaries that are in that system to build their 401(k) plans based on choosing funds that are better than peers on different ESG metrics and so, it’s really that data that allows easy personalization and reporting that is the highest and best use, at least for the wealth advisory space.
Craig: I can imagine that having that ability to easily personalize your portfolio, no matter where it is being held, could be extremely important. What about having a wide securities coverage and a wide data coverage? Does that help in customization?
Ben: That’s really, really important and I’m glad you pointed it out. I think the biggest issues with the category of ESG data that’s called ESG ratings is the subjectivity we have already addressed. The other two are coverage and lack of having fresh, updated data. Starting with the coverage, typically, most providers cover 6000 to 8000 companies and they may cover some funds and what have you. OWL covers almost 30,000 companies, which is like 99% plus of the world’s market capitalization. We also cover approximately 60,000 mutual funds and ETFs and their various share classes and we cover, starting in a month and a half, we’re going to be covering about 200-I think it’s about 260,000 corporate bonds, so we can cover a wide variety of a portion of the portfolio, both in fixed income funds and equities, so that coverage is really important, but what’s also really great is that we update monthly because we have hundreds of sources powering our scores for any given company. There is always some source giving us some data to give us a fresh view on a monthly basis and update the score. The typical ESG rating vendor, they will score the average company once a year, so you would never make an investment when you’re analyzing a company where you have the latest information about them on earnings or sales growth or anything like that was 12 months lagged. That’s what is going on and so what we have is the freshest data in the industry, the most objective, with the widest coverage.
Craig: You don’t generate any data on your own and you don’t override a source, you just review the sources, aggregate them, and maybe provide some sort of credibility analysis?
Ben: Yeah, the credibility analysis is internal, but yeah, for the OWL ESG consensus scores, we are not overriding anything. We merely identify consensus and make sure that the data that’s powering us is quality, so if we’re detecting abnormalities in the underlying source’s data, our credibility analysis being internal, what we’re doing is we’re trying to identify bad data, not subjective data, bad, incorrect data that we can eliminate from optimization process and the aggregation process, but ultimately, what we’re providing to the client, it doesn’t have any of our subjective viewpoint on hey, is this company ESG or good for the world or bad for the world or what have you, it’s merely this is what the market of the world’s leading issue data providers, this is the market view on what this company-how this company is doing regarding to ESG. That said, you asked do we have any of our own data? We actually do have our own data, but they’re completely separate products that are parallel to our consensus scores, but we also provide what’s often called screening data. If you want to stay away from companies that are involved in animal testing or fracking, palm oil, weapons of mass destruction, whatever it may be, this is more of an ethical viewpoint of the world meaning hey, that activity may not be illegal, but I just do not want to be invested in companies that that’s what they’re peddling or that’s what they’re doing, so we call that our screening product, or ethical or principle based screening product. We also have a controversies product and those controversies are things like is the company embroiled in any litigation involving things like oil spills, environmental catastrophes? Are they involved in litigation about human rights litigation, labor litigation, corruption, accounting fraud, bribery, those types of things and last but not least, we have a whole bunch of carbon data, so if people want to directly look at hey, alignment with the Paris Accords or carbon emissions and lowering their carbon footprint, that is a separate data set that we do provide, but to be very clear, it’s a different data set that can be used in conjunction with the consensus scores, but it has a completely different process in its creation and maintenance.
Craig: Ben, you have said it all and we have run out of time. Can you tell the people listening where they can find out more about your company, OWL ESG?
Ben: Yeah, you can find us on obviously, the web at www.owlesg.com. You can find us on LinkedIn, Facebook, and other social platforms.
Craig: Great. Ben, thanks for being here. I really appreciate your time.
Ben: Thank you so much for having me, it was a blast.
Transcribed by Mobile Assistant.
