S&P Global Inc. (SPGI) Earnings Call Transcript & Summary

All right. Let's get started here, pretty impressive crowd for 4:30. You guys certainly want to hear what Doug has to talk about. So thank you, everyone, and thanks for making it this far in the conference. I'm very pleased to have S&P Global for our next session. We have Doug Peterson, President and CEO of S&P Global, here. Welcome back, Doug, to the conference, and thanks for always participating.

It's great to be here. I always like being back at this conference. It's a fantastic group of investors. And we've had a really, really great day. So thank you for organizing this.

Fantastic. As always, you could ask questions via Pigeonhole. It's pigeonhole.at and the passcode is SDC2023. So we'll get started. Maybe let's just step back here. There's been quite a bit of transformation in the company over the last several years. So when you think about S&P Global as a company, how would you describe it for sort of general audience in the crowd today?

Well, first of all, S&P Global is a company that's providing data, benchmarks, research, analytics for the financial markets and other markets, commodities, equities, real estate, fixed income, equities, et cetera, that allows people to make informed decisions. I know that sounds like a very generic answer. But we also anchor that off of the best brands in the financial services markets: S&P Global, S&P Global Ratings, S&P Dow Jones Indices, within the Commodity Insights business, Platts, IHS Markit products and services, Market Intelligence. And we also have a Mobility business with brands like Polk and CARFAX. So across the businesses, we're providing necessary, must-have data, so people can make informed decisions to make important investment decisions.

Right. To your point, you touch many end markets, transportation, commodities, financial end markets. As you look to those end markets, where are you seeing strength? Where are you seeing weaknesses? Just walk us through those end markets.

When I think about the immediate markets, there's been a lot of volatility in the last 1.5 years once inflation started kicking in. So we've seen the inflation and the volatility that's impacted some of our businesses. But let me take a step back and also give some perspectives on longer-term trends. One of the most important trends is capital market transformation taking place in the United States and around the world. In the capital markets, you see active to passive. You see private markets starting to come up. Securitization is changing at what it means. In Europe and Asia, you see markets moving from banking markets to debt markets and capital markets. Another big theme is what I'd call sustainability. It's ESG. It's ESG scores. It's everything that have to do with what you look at when you say climate and climate change and how you think about that space. And then there's also energy transition. These are another really large secular theme that's been positively impacting our business with tailwinds. And then the third is digitization and digital transformation. In the meetings I've been in today, I don't think there's one that we didn't talk about AI. And you look at all of these factors together, the three of them, transformation of capital markets, energy transition and digital transformation, we benefit from all of those. And those are giving us the tailwinds for our longer growth projections.

Okay. Since you brought up AI, I guess, we'll go to AI. I guess, the one thing we could say here is I think there's evidence that you've been investing in AI for a while, given the Kensho acquisition in 2018. But again, let's just step back here, talk through how you think about S&P's capabilities in AI and then we can talk about opportunities.

When we bought Kensho 5 years ago -- and for those of you who don't know Kensho, Kensho is a group of people based in Cambridge and Boston. We have about 120 specialists and mostly PhDs in computer science, physics, mathematicians, who have been providing us with services for the last 5 years. It's having an internal captive R&D specialty group. When we bought them 5 years ago, we had already been a B round investor in Kensho when it was a startup. And we were learning about what they were doing. And we had a vision that in the future, 5 years from then, which is now, and another 5 years, 10 years in the future, that all of us would be assisted in our decision-making by AI tools. Not that we were going to be replaced, but we would be assisted. We'd have a copilot that would help us make better, faster decisions. So over the last 5 years, we used Kensho as this internal R&D shop to help us develop products, services, linking data, speed of access to markets. And we've become really knowledgeable of how about -- how to use AI and how to use machine learning, both from an operational point of view as well as a product point of view and how it improves our delivery to our customers.

Okay. And how do you think about opportunities then in AI over the next 3 to 5 years?

When we look at what's happened with AI, generative AI is a new theme that's created a lot of excitement in the markets. And all companies are talking about what it means for them. Because we're starting from a point of strength that we understand what AI is, we've had Kensho now for 5 years in addition to other strong teams across S&P Global that provide AI and decision sciences services, we're starting from a viewpoint that we already know how to use learning data. We know how to protect our data. We know how to keep a firewall around it, so we can protect it inside of our own company. And we also have clean linked data, which is a great starting point. We can use it as learning data. I'll give you a quick example. When we first bought Kensho, we had 300,000 hours of earnings call transcripts. And then we had the actual earnings calls recorded and we had all the transcripts which were clean and absolutely completely purely reflected what was said. That meant we had digitized data. Kensho used that to build a product called Scribe, which we use today for earnings call transcripts and some other types of transcriptions. We have linking data opportunities. So we already know how to link data. Natural language processing gives us a base to, now to your question, it allows us to think forward about what are we going to do about processing in our data operations. What is it going to mean for learning and opportunities with areas where we use expertise? How can we enhance that expertise, do it faster, do it better? And then on the product side, how are we going to be able to use AI and generative AI to be able to let anybody here ask very complex questions and queries and give them answers that are precise, that are clear, they have an audit trail around them and you know it's the right answer?

In terms of commercializing some of those opportunities, what sort of timeline are you thinking about in terms of bringing that to market?

Today, we have five products in our marketplace, which are AI products that are generated by Kensho. And you can see those, like Scribe and Link and NERD. These are products that already give us that foundation of AI knowledge that we can use. We have a couple of prototypes we're building internally, a language model, which we call [ Fin LM ]. It's an internal model we're using right now. We're also going to be using multiple asset, multiple ChatGPT, working with people like AWS, working with Microsoft. We're not going to be able to do this alone. We don't have the capital to invest to build a really large language model. But we do have the data. We believe that we'll have a couple of products in the market sometime -- if it's later this year, early next year, I don't want to make a promise. We don't have a road map yet that specifically says when the products are going to be out. But you should expect, you're going to start seeing some piloting, some experimentation. We're already doing it internally. And we're going to want to get some relationships with clients to start piloting this with to see how we're going to commercialize these. But I'm really excited about what this is going to mean for us. And it's going to be one of those tailwinds for S&P Global.

Where do you think long term the value creation happens in AI? Do you think it's the data, where you are? Is it somewhere else? How do you think about value creation long term?

I think in valuation long term, one part of it is going to be in operations. We're already seeing the kinds of benefits we get from being able to link data, extract it, put it together in products and services very quickly in a way that you've got clear, high-quality data and services. But in the long run, we already know what it's going to mean for areas like medical services and law services, things like that, where you can extract large amounts of data, look at it using AI tools, generative AI that's going to allow you to have a completely different way of working. The same thing is going to happen in the financial services industry. So for analytics, for research, for forecasting, we think the tool is going to be -- allow you to move faster, to work faster. And we -- as I say from the very beginning, our view is these are tools that are copilots. It's not going to replace us. You still need to have judgment. You need to have experience to make decisions, investment decisions, investment management decisions, portfolio decisions, risk decisions. You still need judgment. So I think this becomes a tool that supplements us and allows us to make decisions faster and better informed. But it's not going to replace what we do.

Okay. Let's switch to the cloud.

Yes.

You announced a partnership with AWS. Can you just talk about the genesis of that partnership, the cloud journey at S&P Global and how we should think about again revenue and expense opportunities over time?

Yes. We had started our cloud journey about 7 years ago, when we still had a lot of data centers around the world that we either owned or we were leasing. And we decided that it really didn't make sense for us to be in the data center business. When you're a company like ours and you have 40-plus data centers around the world, you'd say, "Why do we have all these data centers?" So we started working with all of the cloud providers and generally speaking and moved a lot of it to AWS. We also work with Microsoft and Google. But we had a large bulk of our operations with AWS. And after the merger with IHS Markit, we started finding ways to consolidate even further. Our Ratings business, Indices and -- Ratings, Indices, Mobility and Commodity Insights are almost all the way in the cloud already. They're substantially in the cloud. There's very few data centers left that they're working on, if any at all. The largest final transformation we have is with Market Intelligence. We still have some servers there. We have 15 data centers overall in the company. And by the time we finish, we'll end up with about 6, having started with more than 40. And we believe at the end, we're going to need a few here and there, either for latency or for certain requirements, for regulatory requirements to have data in a specific location, on-site, onshore, et cetera. But for us, we found a lot of benefits from this. It means that we're not making these capital investments. We don't have to have those assets on our balance sheet. We're working with world-class organizations that can keep up to speed with all the latest technology. It also allows us to have really high-quality security and the best security. So the cloud strategy for us has been very beneficial for how we run the company. And then we've also been good at managing the expenses and having really strong partnerships that allows us to have a good expense management around this as well. So net-net, the cloud has been beneficial to us. And we now see there's also opportunities for innovation.

Okay. Great. Let's switch over to your Market Intelligence business, which is your biggest business, at least by revenue currently. It is the business where I think you've probably made the most investment, clearly with the IHS deal, but it also is your lowest margin business.

Yes.

What's -- maybe talk about the long-term vision for the business, particularly now with sort of the AI spin on that and how data becomes more important. And maybe what gives you conviction that this is where you should be investing most of your resources.

When we think about all the different capabilities and businesses across S&P Global, you still come back, no matter what they are, to data. And we think of Market Intelligence as a data factory, as a center of excellence that for S&P Global can provide services for the entire organization, that it is a center of excellence that allows us to manage the products of Market Intelligence itself but also become a center of excellence. As an example, Ratings360, which is a tool that we use to have an interface with CFOs and treasurers of all of our issuers, that's built off the platform of Market Intelligence. The Platts Dimensions Pro is built off of the platform of Market Intelligence. The data services that are provided help benefit the entire company. And at the same time, we have the different business lines. We have the Desktop. We have the risk services, which is the monetization of data from Ratings. We also have our advisory business and our delivery services. This has been one of our fastest-growing areas, where we're providing data, which is data feeds. And then finally, we also have our Enterprise Solutions. So we manage it in four different areas. But all of them are areas which are necessary for organizations like yours in different ways to have the data you need to make decisions. We've been investing in a combination of bringing in new datasets. The IHS Markit merger itself was a way to bring in much larger scale in the business. So we'd have the datasets to be able to provide the services that I talked about with capital markets and sustainability, et cetera. So this gives us a much larger scale and a global scale on the data. But each of the business lines there have really good growth prospects, user increase. We've seen a little bit of a dip in the results from the Enterprise Solutions' volume-driven businesses. A couple of businesses there are driven by volumes of IPOs and debt issuance and things like that, which have been down quite a lot. But then the flip side is that some of the other products in that have been able to offset that. But net-net, it's a set of services and business necessary for S&P Global to be a leader in data and analytics.

How are you thinking about the end markets to those businesses? I'd imagine banks and financials are up a good portion of that end market, given some of the stresses we've seen. Has that impact in...

Yes. So first of all, we're fortunate in that Market Intelligence is a diversified business. We're not reliant on any one single type of customer base. We have financial institutions with financial institution in buy side, sell side, insurance, pensions, banks, investment banks, brokers. We have governments, we have regulators, we have corporates. And with corporates, we have large users and especially large, global, sophisticated users. They have their strategic planning department, their treasurers. So we have a very well-diversified business. And that's one of the beauties of this business of Market Intelligence. So that's something that we start with, a diversified business. In terms of the product lines, we've seen user growth in the Desktop. We typically have our relationships with clients in an enterprise relationship as opposed to a per seat or per desk business model. We have an enterprise license, which means that it makes it very easy for new users to get onboard. And we see the usage going up, we see the users going up, which then goes -- figures into our contract negotiations at renewals, which are typically 2 to 3 years. You asked about the market in the financial services market. We have seen some elongation of the sales cycle as we've seen financial institutions have gone through some layoffs. There's a lot of volatility in the markets. There's a lot of risk in the markets. So we have seen some elongation of the sales process and the sales cycle. But it has -- we haven't seen any drop in terms of closing deals or any change in terms of our retention rates.

Okay. You mentioned the Desktop business being an area of relative strength. Some of the competitors, like Refinitiv, I've talked about, really aggressively investing in that area. So what are you seeing on the competitive front there? And sort of why do you think you're able to gain share?

Yes. So this is an area where, clearly, we have to start with the quality of our data and must-have data, like the Cap IQ data, the SNL data. Now with IHS Markit, we have certain reference pricing data, which is must-have for financial institutions. If you go on to the product controller or management control of any major investment bank globally, they have data. They have data they need that we have that allows them to manage their risk, to manage their positions. So for us, the most important thing is to have the must-have data for different type of these client bases that I've discussed. And then on top of that, we have to invest for quality delivery. We've invested in the actual look and feel. We just recently did a very small tuck-in acquisition, ChartIQ, which allows us to enhance and improve what is going to be the visualization tools, the graphing tools, the data linking tools, et cetera, in the Cap IQ platform. We've also been investing to upgrade the interface, these speed load times. Sometimes people can't believe I'm talking about page load speeds. But it's really important because if you're sitting around twiddling your thumbs, people get frustrated. You've got to be able to have the kind of quality of service so that people can get on, they can use it. It's easy to use. We have different personas and different types of user experiences, depending on what your role is, so we can automatically get you plugged into that. And so at the end of the day, to answer your question, the most important thing for being competitive is to deliver what our customers want in a way that makes it easy to use, it's high quality and it has the data that they need to make decisions quickly and run their businesses.

Okay. Let's talk about some of the opportunities in that business. One is private markets, you've talked about quite a bit. Maybe remind us progress so far in terms of capturing opportunity and where you see opportunities going forward.

Yes. The private markets business for us spans both Ratings and Market Intelligence, Ratings because we will be providing services to the private credit businesses as they start securitizing, syndicating and managing those credit portfolios actively. And then when it comes to private markets generally, we have a total of $100 million business last quarter. We see this growing at an annualized basis of about 12%, so greater than double digit. And what we provide is information for private credit products and private credit providers and private market providers themselves on their portfolios, portfolio analytics. But very importantly, we have a product called iLEVEL, which is an example of a product that allows investors in private market portfolios to have the data and access across all of their different providers. So if they have multiple relationships and multiple investments in different private equity and different private credit providers, they can aggregate their portfolio across all of them into a very simple, easy-to-use desktop that allows them to understand what are their concentrations, what are their risk positions. And we're seeing a lot of interest in this and a lot of demand especially now coming. As private credit has started growing, investors in private credit want those tools, so they can value their assets, they can look at concentrations. So this is an example of some of the things that we do for private credit and private markets. It's a very interesting area. We provide a lot of services and we see a lot of growth here.

Same thing, supply chain, that's also one all of your competitors are pushing hard into. So talk about how S&P is differentiated and where you see opportunity.

Yes. So for the supply chain, if you think about it, if you're a large supply chain manager in a large global organization, you have typically -- we've typically thought of what you need to do is how do we help you manage your counterparty risk. And we already have that product and service. We have the credit information, the credit analytics. We have reference pricing, et cetera, if you're a supply chain manager, so you can look at your hundreds of thousands of suppliers and understand what credit risk you have and how you are going to manage that. Well, now with IHS Markit coming together and rethinking the product, we can also add an information about shipping. We have some of the best shipping information in the world. We can put in prices on containers. We have Panjiva, which is a company we bought a few years ago, has information about what's coming in and out of ports and airports everywhere in the world. And you can now take what we have as a pilot of a supply chain console and not just bring the supply chain credit risk about your suppliers, we can also add in the other information you need about prices, shipping, transportation. If you want, we can add an information about products you might need to hedge and understand more about, whether it could be metals, it could be the price of oil or fuel oil or gas or jet fuel, et cetera. So we're finding this -- this is a growth area. Right now, I'd call it in the piloting stage. We have some really interested clients who are piloting it with us. And it's another one of those exciting areas that we're trying out with -- we feel like it's really important that we put a lot of toes in the water and try different areas for growth. And this is one of those that right now we've launched this supply chain console as a pilot. And we're quite excited about it because there's a big demand for this information.

Okay. Let's switch to the Ratings business. Obviously, the debt markets have been turbulent over the last few years. You look at the current state of the markets, how would you characterize growth, right, better or worse than expected? Maybe some color around any products or regions that are tracking better or worse.

Yes. Overall, as you know, I don't necessarily track credit on a monthly basis or a weekly basis because you need to see how it smooths out over a longer period, quarterly, annually, et cetera. But we saw a period in 2017, '18, interest rates were low. The markets were quite robust. We went into the pandemic and every corporate treasurer in the world raised a lot of debt. They wanted cash on their balance sheet. They needed a cushion. And at the same time, there's a lot of deal flow. So we saw a very large increase in the growth rates in the Ratings business in '20 and '21. And then in 2022, when inflation started kicking in, we saw more volatility. We saw very large drops in certain asset classes, especially asset classes like high-yield loans and high-yield debt. So we came out of last year, which was a very, very large drop, and we're seeing now some recovery. It's been lumpy. You can see a few weeks with really strong issuance and a few weeks with weak issuance. We see a lot of -- in this year so far, one of the big trends has been large global corporates, investment-grade corporates, they've been steadily coming back to the market. There hasn't been a lot of high yield. Structured credit has started coming back to the market. So you're starting to see again CLOs, that volume is coming back to the market. What we look at in the longer run is, first of all, the issuance itself is typically correlated most highly with economic growth. That's the #1 correlation is economic growth. Timing is impacted by interest rates. So as interest rates stabilize, which is one of the things we expect over time that we will get some level of stabilization of rates here and through the ECB in Europe, that will be positive for credit because people now have more ability to plan with stability in the rates. And then we look at some other factors, which, as an example, there's about $7 trillion of debt on balance sheets of financial institutions and corporations, which is going to be maturing over the next 4 years with about $2 trillion a year for the next 3 years. And we expect that, that will get refinanced and a lot of that will get pulled forward. I don't know when it will get pulled forward, partially because of that stability on rates and stability in the markets, et cetera. And then one last thing I'd mention is that there are some other really large capital investment programs that are going to be taking place globally. One of them is energy transition. When you look at energy transition and you look at what's going to be invested in grids, in new power plants, in renewable energy, in batteries, et cetera, and in the U.S., we have the Infrastructure Act, we have the CHIPS Act, we also have the IRA, we're starting to see that investment take place. And a lot of the investment is going to be driven by debt. So you have that kind of capital markets activity, which is going to accompany energy transition. And then there's also going to continue to be capital markets activity once the M&A markets come back again. So when I look out over a medium term, there's a lot of factors which will be bringing back a lot of volume to the market, which leads to our 6% to 9% medium-term target for growth rate for Ratings.

Okay, great. Let's switch to your Mobility business, which is currently actually your fastest-growing business, double-digit top line growth, so good stuff there. Can you just talk about the secular trends that are driving that sort of level of growth?

There's two very immediate factors which have been driving it. One of them was that last year, we saw that the dealers had incredible power in the markets because there was no inventory. So the dealers were kings. They were able to receive automobiles, used automobiles, new automobiles. The second they got on the car lot, they could sell them. We are now shifting back to a point where the OEMs and manufacturers now have inventory again. It's going back to the dealer. So there's a bit of a shift of where the revenue piles -- where the revenue opportunities are. We're very well hedged between back and forth between the dealers and the manufacturers. Right now, the manufacturers are also facing a second theme, which is really important for them, which is the transformation of the industry to electric vehicles. And that's changing their supply chains. It's changing how they're managing their businesses, how they think about inventory management. And so we have a very strong set of products which provide information about the entire supply chain of the automotive sector, parts, all the way to different kinds of spare parts that go into automotive, into the manufacturing. We're also looking at the forecasting for medium-term, long-term forecasting. We have products and services that allow the manufacturers and the dealers both to look at what are the expectations for who's going to buy what car, what kind of car they need, et cetera. And so these are the kinds of products and services. I would say that, on the one hand, the strength we're seeing in the manufacturing market, there's a little bit of an offset that now that the dealer market has shifted a little bit, we are seeing some weakness there. We're seeing a little bit longer sales cycles and we're seeing a little bit lower -- the retention rates have dropped a little bit. And so we mentioned that on our last earnings call that, that was going to hit us a little bit of weakness for the rest of the year.

And is that a cyclical trend? Or do you think that's sort of a new normal?

I think that we're going to have to watch very carefully what's the longer-term role of the dealers, especially with electric vehicles. But the dealers play a really critical role in the markets in terms of providing service, providing information, providing people ability to have test drives, et cetera. And then in the long run, they have a really important role to play. One last part I didn't talk about is CARFAX. CARFAX is a business that the second a car leaves a car lot -- it's a brand-new car. The second it leaves a car lot, it's a used car. And every used car, which is every single car that's on the road, because 100% of cars on the road are used cars, it doesn't matter if you just bought it 5 seconds ago, becomes a target for CARFAX. And CARFAX is growing really well. It has products and services that people actually input their own data. It links into other services. And it's a really attractive service that is also providing a lot of data, which is extracted for both dealers and for the manufacturers. And so that's a part of that ecosystem of the Mobility business.

Okay. Just curious, you mentioned electric vehicles impacting dealers. Curious, is there a way you look at that business and can segment the difference between growth rates between the ICE part of the business versus electric vehicles?

Yes, we do. I'm not the expert to answer the question in terms of the details. But we do very detailed analytics and forecasting across every market, down to regional MSAs -- at the global level down to each regional MSA based on ZIP codes and wealth levels and subsidies, what are the demands for different kind of cars. So we -- that's the kind of research and analytics that we provide that are used by all parts of the supply chain.

Okay. You bought Market Scan recently for that business. Maybe talk about what that business does, how it fits into the strategy.

Yes. So Market Scan is a business that provides detailed pricing information about automobiles. And what we had, when you looked at the services that we had, we were able to tell you what car somebody is going to buy, what model they were going to buy, when they were going to buy it, where they were going to buy it, but we didn't know how much they were going to pay. And we didn't have a tool that allowed you to take who, what, when, where and then what are they going to pay for it. And so Market Scan added into that portfolio of who, what, when, where and add in the what are they going to pay for it. So now we have a complete package that dealers and manufacturers can use to estimate what are the opportunities for them in individual cars, individual sales that help somebody manage not only the pricing and the sales process, it also helps them -- it added to our ability to provide more information about incentives, whether it's interest rate incentives, price incentives, what it's going to take to get somebody to buy this car. So it was a really powerful addition to our portfolio.

And typically, what's the monetization model for that business? Is it get better data, so the rest of Mobility...

That is monetized to the users. They pay a subscription fee for it.

Okay. So just [indiscernible] the added costs, it gets more revenues. Okay. Have a ton of questions here from the audience, so let me start firing away. First one is if rates are higher for longer, how do you see the impact on Ratings -- on the Ratings business?

First of all, I don't know what rates mean are higher. So let's assume that they kind of stay above the 4% to 5% range. I will go back, I'm going to date myself. But the very first CD I ever opened when I saved enough money after my first job in college was I got a CD at 16% at a savings and loan in San Francisco. And mortgages were 18% and 19% at that time. And the economy motored along, and we were used to rates. We just got really overly reliant on 0 interest rates for 12 years. The markets are going to reset at some point to a new level of rates, whether inflation is 2% or 2.5% or 3%. I know the target is 2%. But once we hit a rate and we hit a ceiling, people will have certainty. And what I believe people are looking for is certainty and some sort of reliability, so they can make decisions. There is a large back -- I already talked about the backfill of ratings, the $7 trillion on balance sheet. So at some point, that's going to have to get refinanced. And as I mentioned earlier, when we look at rates and ratings, it's not the determinant of the issuance level, it's more of a timing factor. So if rates stayed really high for a while and people hadn't gotten used to it yet, they didn't feel it stabilized, it could slow down issuance.

Okay. On ESG, how is your ESG solution different versus some of your peers, who've had first-mover advantage? And is there an M&A path to gaining more scale in ESG?

Well, first of all, when I think of ESG, I mentioned earlier that we think of it as a horizontal that covers all of S&P Global. And only one part of our ESG, or I'm going to call it sustainability offering, is a score. We have -- the scores that we provide have very high quality. They allow you to get an E score, an S score and a G score. And they give you the subcomponents. The methodology is very simple. It's very clear. We provide the underlying data. And so if you're an asset manager or a bank and you're using these scores, and you want to weight them differently, you can weight them differently. You don't have to use our weighting. You can take that score. Our underlying data is very high quality. We also have a product. We acquired a company from RobecoSAM a few years ago. It's the CSA, the Corporate Sustainability Assessment. At the time we bought it, there were 1,200 companies that were self-reporting. It's pretty voluminous reporting that goes into the system. We're now up to 3,000 companies. So one of our strategic advantages, having that base of data of 3,000 global corporations that are providing us with data from themselves, it's self-reported, it's really high quality. So we have a very high-quality ESG scoring process. And we use it in multiple ways across the company. But we think of ESG in a much broader way. And in my recent conversations with your organizations, with financial institutions, with corporates, the refrain I've been hearing is, "We're not relying on the scores anymore. We need to build our own ratings. We need to make our own models." And that is music to my ears because we have the best data. 7 years ago, we bought Trucost. And Trucost is the premier organization that provides information on over 15,000 companies on carbon, water, waste, chemicals. We also have physical data on millions of buildings, ships, ports, et cetera. So we have the data that you can use, the raw data that you can build in your models that you can use to develop your own models for ESG and ESG scoring. And so this is an area where we see a lot of growth. We see it's one of the most important trends we're seeing right now is a shift away from relying on scores. Maybe you're still using them. But people are saying, "I can't rely on the scores because when my regulators show up and when my asset owners show up, they want to know, 'Well, how did you come to that decision? How do you manage that portfolio?'" They don't want to just know that you used a score blindly. They want to know that you had your own methodology and your own approach to managing your portfolio and providing the information, whether it's E, S or G, it's climate, it's carbon, whatever the metrics you're going to use to provide that information back. They want to know that you made an active decision. And to make that active decision, we have the data and analytics that you can use to make that active decision. So I think that we're very -- we have a big differentiation compared to some of our peers because of the datasets that we have that we've built in the company.

Just to follow up on ESG, clearly a much more political topic today than it was maybe a year ago, as you speak to sort of the end users, is there any change in the level of demand, et cetera, for ESG-type products?

Yes. Despite some of the politicalization of ESG in the United States, we see that the demand is growing globally. And I'll give you an example. Last quarter, which was our Trucost data, I mentioned what Trucost is, the demand for that data grew 50% year-on-year over the first quarter. So the Trucost data demand, 50% increase. Last year, in the EU, 65% of the flows of new funds into asset management products went into some sort of an ESG strategy. So I don't see that shifting. I know that there are some very important aspects we need to think about when it comes to politics. But what we do is we provide data, research, analytics, benchmarks so that people can make informed decisions. We're not telling you what's good and bad. We're telling you what are the facts. And then you can make your own decision.

A question on AI. So longer term, to what extent can your 40,000 headcount be reduced as a result of human plus AI productivity tools?

I don't want to make a projection on that. But the question is how much would we either reduce roles or how many of those roles would still be around but would be used for a completely different approach? They could be used for sales or analytics, et cetera. I look at this as AI is going to be an opportunity for us to enhance the roles of people, make analytical roles more interesting, more exciting. And there probably will be productivity opportunities along the way. But I don't have a projection right now.

Okay. One on cloud. Any particular reason why you chose AWS versus other cloud providers?

Well, first of all, all the cloud providers are very professional. They do a really good job. We already had a strong relationship with AWS because we've been using them across both S&P Global and IHS Markit. So we started from a position of scale. We also see opportunities to innovate with them. They are very interested in some of the cogeneration of opportunities that we could have because of the data we have, because of the data they have. And they came with a really open mind and a lot of interest in building things together. And we see it as a strategic partnership. What they -- one of the things they find interesting in us is that because we're already considered a data company, they're working with a group that has clean data. It's consistent. It allows us to have a very strong partnership. So this really relates to a partnership approach and already having a lot of trust with a long-term relationship.

Just a follow-up on the cloud conversation. Do you worry about sort of capture and cost pressures, just to the extent that you have one cloud provider, essentially they can raise prices whenever they like on you?

Well, we have a 5-year contract. It's a $1 billion, 5-year contract. But our spend is more than that. So we are going to continue to be a multi-cloud company. We won't just have one provider. We feel like it's always important to diversify. And we'll probably continue to grow on our cloud expenditures. But the other thing you talked about, cloud capture, I think all of us recall over the last few years, as people start adopting the cloud, at some point, you realize you're probably spending more than you thought. But we've built a lot of really good controls in place in terms of spend, of use levels. We also have internal -- we have internal volume pricing as opposed to just kind of fixed pricing. So we're quite good about how we're going to control the expenses around this.

Okay. A question on Ratings and expenses. So how are you rightsizing the Ratings business or contingency planning in event of a lower-for-longer issuance environment?

Well, first of all, we're -- we have a business model in Ratings, which says that we're going to be global and we'll cover every asset class, which means that we are going to always look through the cycle. And it's unlikely we're going to be doing a lot of cutting. This isn't -- we don't think about Ratings as contingency because we need to serve the markets. And we have to keep that position of having a global multi-asset class, full asset class coverage. The other thing is that our Ratings analysts don't just issue ratings. They also surveil credit. And if you think about the market we've been in recently, this is a tough market. We've seen volatility. We've seen higher -- we're seeing a much larger increase in bankruptcies and defaults. We're seeing a lot more volatility when it comes to rates, when it comes to risk levels of different industries. So our analysts don't -- just because they're not doing an issuance of a rating doesn't mean they're not busy. They're very busy continuing to surveil the markets. And then the other thing they do is our analysts are incredibly busy answering phone calls for investors. And what time are you going to get more calls than ever from investors? It's when the markets are tough. And so our analysts are providing a window to the market from people that want to understand what's happening in the market, what are the conditions that could lead to a downgrade? What are the conditions that could lead to a default? And so our analysts are actually busier than ever. And we feel that it's worthwhile to keep the investment through the cycle and hold firm on how we're going to do. Clearly, we've been really, really good. We're an incredibly disciplined company when it comes to expense management. We are going to maintain that discipline. But there are some businesses like Ratings, where we found that we have a core level of capacity, we're definitely going to keep that in place.

Okay. Maybe to follow up on that, how do you think about the ceiling on margins, given very high incremental margins for most of your businesses? So how do you think about the ceiling?

Well, we start every year, when we begin our budgeting process, we always start it with a premise that we can continue to grow our revenues faster than we can grow our costs because this is -- people call it positive jaws. But we believe that there should be a positive gap between your revenue growth and your expense growth. And then we get into a really healthy debate about what are we going to do about that difference? And in many cases, we've said that, "Well, we need to invest some of that. We need to do some of them BAU. We need to continue to invest in growth." We have now what we call a Vitality Index, where we're tracking what growth level we're getting out of our new investments, so we can see new revenue being generated, new revenue growing. And we want to be compensated for that. We want you to understand how important it is for us to be developing new products. So that difference between the expense growth -- the revenue growth and expense growth is where we're going to take a step back and decide, "Let's now say how much of that do we need to reinvest, how much of that are we going to continue to grow the margins?" But we believe, based on our experience, that there's still room to grow margins.

Okay. Can we talk about capital allocation? How are you thinking about allocating excess capital? And also, if you could just tie into it, you recently sold the engineering business post the IHS deal. When you look at your portfolio, are there other places where you can still make optimization decisions?

Yes. When we think about capital allocation, those of you that know us well will know that before the merger with IHS Markit, we had a capital allocation model where we -- our target was to return 75% of our free cash flow to our shareholders through a combination of dividends and stock buybacks. And we're what's called a dividend aristocrat. We've been increasing our dividend for the last 50 years. So every single year for the last 50 years, our dividend has been higher than the year before. So we feel that, that is a really good track record. And we've been paying back that other -- to get the total of 75%. If we don't have anything better to do with the capital, we've some years paid 100% or more. After the merger with IHS Markit and looking at our capital investment needs, we decided that we could increase that from 75% to 85%. So our capital allocation model of free cash flow is to return that target of 85% to our shareholders. Now when it comes to capital allocation of our portfolio, we're quite disciplined at looking at each business, understanding how it fits in the portfolio. How does it work with our clients? Where are we going to invest? Where are we going to get the cross-value of something like Market Intelligence data expertise, technology expertise, Kensho expertise? How are we going to take advantage of that? And we're always looking at the portfolio. And we're going to continue to be very, very disciplined capital allocators. We made the decision on selling Engineering Solutions because at the core, they didn't own their IP. Their IP was, in a way, rented. We paid royalties for the IP for a business, a beautiful business. But we felt like all of our businesses, we own the IP. And there would probably be a better owner of that business than we were. And we found a very good buyer. It was a really excellent relationship with KKR. And we're very pleased with the relationship with them. But this was the kind of thinking we have around the portfolio. So we're going to continue to be disciplined. To the extent we might look at tuck-ins, we might look at tuck-outs, if that word exists. There might be little pieces here and there that don't fit. But overall, in the long run, consider us to continue to be really disciplined capital allocators in how we think about the portfolio.

Okay. Another question on AI. I guess, a big debate, winners and losers is sort of who owns proprietary data versus who's the data aggregator. I guess, where does S&P sit in that? And I'm guessing you're going to say proprietary. So is there a way for us to, I don't know, judge that, measure that from outside?

Yes, there's -- I'm going to be oversimplified. But there's two types of proprietary data. One type is things that we generate that we own with our IP and our brand that nobody else can do, unless infringing our brand or whatever. So we have this kind of -- that would be a benchmark. It's a price benchmark. It's a rating. It's an index. Those are proprietary data that are very hard to replicate and duplicate. There's other data that I would say the next level, which I would call must-have data. It's also proprietary. But it's built off of and based off of multiple sources of public data. So if you look at something like SNL or Cap IQ, we're literally taking information from 50 or 60 different sources. And we're bringing it together in a way that we're linking it, we're storing it, we're finding ways that we can turn it into a product. And you're saying, "Okay, well, that data is public data." But we have time series that go back, in some cases, 30, 40, 50 years like Compustat. And we've got that data in a way that it's already embedded in people's workflow. So do I call that proprietary data? Yes. I call it -- I don't think that's commoditized data. The commoditized data might be what you pull from EDGAR. So something we get from EDGAR, that's commoditized data. But once you take it from EDGAR and the financial database and the OCC and the Fed and this and that and then you get 60 sources of information and you linked it all together, so it's really easy to use, that becomes proprietary.

Do you think in an AI world where, I guess, an agent to go do all that work for you, that linking work and gathering work, does that lower the moat of that type of data?

The -- as long as we continue to add value to bring really high-quality search, delivery, visualization, et cetera, to the markets, this is going to continue to be used as decision-making, must-have data. That's the way we think about it. So one of -- really, what's most important to your question is we don't think we can sit still. We can't be complacent. We can't be arrogant. We have to look behind us, see who's coming up, nipping our heels. And it's really important that the most -- what we do is we add value and continually improve the way we serve our customers.

Okay. Another one from the audience. What's been the biggest surprise from the IHS field?

I think the biggest surprise has been some of the businesses that I didn't really understand as much during the due diligence and now bringing them in. There's some of the Market Intelligence financial services businesses, which are, for instance, the DVA, the valuation business, 13 million prices, which are developed and issued every single day that are used in the guts of every single financial institution globally. When you look at that, the original Markit business, you see that it was built in the quality. That was something that was really, really a big positive surprise, much more than I ever expected. And then I'd say the other one on the energy and natural resources business, what IHS Markit has brought for with Platts, when you put those together, that is much bigger and more powerful than I'd ever imagined.

Okay. Maybe the last one, given we have less than a minute here, I guess, Doug, you've, as a CEO, clearly very successful, given the stock price performance and the demand from investors to talk to you. As you look back at your tenure, what have you been most proud of? And then maybe looking forward, how are you thinking about succession planning? And is that even something that's a discussion?

Well, first of all, I'm so excited about this job. I love this job. And we have so much more to do. We just completed the merger and all the things we talked about with the tailwinds we have, the secular trends. This is what I love to do. I'm having a blast. I love this job. And for the last 10 years, just building a company like S&P Global, I'm probably most proud of the incredible people and the team we have. But our Board has a succession process that's outlined in our proxy. And what we think about is developing our people. At the core of the way we think about S&P Global, it's all about our people. And what's important for us is that any single person in the company is being developed and they have people that work with them that are being developed. And so that anywhere in the company, we don't think about succession planning as a CEO succession. We think about succession for everybody. We want to have a pipeline of talented, motivated, high-quality people that are willing and ready and able to do a lot of jobs in the companies. And that's the way we think about succession.

Fantastic. I think that's a good place to end it. Thank you very much, Doug.

Great. Thank you again.