MSCI Inc. (MSCI) May 5, 2026 Conference Transcript

All righty. Good afternoon, everybody. Thank you for being here. For those of you who don't know me, my name is Manav Patnaik. I cover business and information services for Barclays. We're very pleased to kick off our post-lunch sessions here with MSCI. And we have Alvise Munari, who's the chief product officer, also the chief head of client segments, I believe. Alvise, maybe just to start, if you could just maybe Maybe give us a little bit of a background. You're new to a lot of the people in the crowd and in the audience. Maybe just a quick bio of how long you've been in MSCI, what all you've done, and so forth.

I joined MSCI close to 11 years ago after spending quite a amount of time on the sell side working in the rivers. and I initially joined as the head of MIA then I got promoted to look after our clients globally that was about you know six years ago and about and then about two years ago the standard two years ago I was asked to become the first chief product officer in the history of MSCI and actually it's interesting that a company like MSCI had product heads for our different product areas but never had a single true product officer and then about a year ago I was asked to develop the the segment effort which is an important transformation how we think of the opportunity set in the good market at MSCI in that we want to focus on understanding in a lot more details what clients do in the day-to-day to essentially build more useful tools. And we realized that a dimension that is sufficiently detailed but not too detailed to become individual client-specific is to look at, basically, industry segments. Active asset managers, passive asset managers, pension funds, endowment foundations, world managers, hedge funds, so on and so forth.

And therefore, we do have now a fully-fledged segment strategy. Got it. Just going back to your point on you as the first chief product officer at MSCI two years ago, what was the catalyst for that? I guess there was, obviously, some deceleration in your business. What was the behind-the-scenes thoughts on that?

So as the chief of client, I had become very vocal with Henry about two challenges. One, that our solutions were not built on an integrated technology stack. And so as a result, they were not easily interoperable. And they were not easily interoperable for us to create further solutions and also for our clients to operate within their stack. So that was challenge number one. And, of course, it was clear to me that we were leaving value on the table because of that. And we were not helping our clients as much as we could because of that. And then the second issue is that I was convinced that we had lost focus on basic product creation discipline, and we did not have a clear function to connect the quantum of product needed to fulfill our growth model with a quantum of new revenues needed from new product quarter by quarter. So that was not done in a deliberate, disciplined, systematic manner. So for those, look, I guess two, three reasons, Henry said to me, okay, you've been complaining a lot. Go fix it.

Well, I mean, I guess credit to you that maybe it's not a coincidence, but I know on the last call, Henry said that MSCI has introduced more new products in 1Q26 than they did all of last year. And we've seen some of that momentum in the numbers as well. So what are some of the key initiatives that have led to that?

Look, first of all, I want to clarify one thing. So I think on earnings call, ARI might have either not been very specific or might have gotten slightly carried away. So let's say that either of the two, what is true is that in the first quarter of 2026, we introduced roughly as many new products as in the whole of 2024, okay? And then in 2025, we launched roughly twice as many products as in 2024. And so if you do the simple arithmetic, it would follow that in 2026, we will launch twice as many products as in 2025, which will be four times as many products in 2024. And then the reality is that we hope that by the end of the year, our exit velocity in terms of new product launches will bring us to roughly speaking 5x where we were in 2024. for that is our aspiration for exit velocity, new product creation for .

GARY GENSLER- Got it. And where are all these new products coming from? What are the key areas of focus categories?

GARY GENSLER- Look, very good question. So we are innovating across the board. Because at the end of the day, to grow, to generate new revenues, you need to basically bring to market solutions that basically add additional value. So everywhere, we need to innovate. Now, when I took over the challenge of being the first cheap product of Rameshia, I said, OK, well, look, where are we going to innovate first and most aggressively? In those areas where there's a combination of two things. One, we have done enough work to our technology stack to be able to innovate faster and two we understand the usability of the new product better and therefore the time to mind is going to be quickest and therefore we focused on our index business why because it's obviously the business that MSCI has been in the longest is the one which I would say we understand the best holistically and we had done already a lot of work to the tech stack leading to last year. And therefore, we could innovate faster. And so that's what we did. And so therefore, I think you've seen in the last quarterly numbers, our index revenue and run rate have shown an acceleration, to a large extent, because of that. So it's the contribution from new products, OK? Then we focused in areas where the market demand and was stronger than usual. And we also had favorable competitive positioning. So that was in the ecosystem centered and around private assets. So a lot of innovation there as well. Of course, we bought burges in October 2023. And we then made further smaller but very targeted and specific acquisitions more recently, which helped propel our ability to bring to market new solutions a lot faster, new powerful solutions a lot faster, and also did leverage a franchise that MSCI already had in terms of target clients and where we were quite reputationally established, so asset owners, and so in particular, you know, total portfolio solutions. So TPM, which we acquired through the Burgess Acquisition and then we enhanced through some of the more recent acquisitions, you know, proved to be something we could accelerate the market impact of quite quickly and we see that accelerating further. So look, it's a very simple number game. The Endowment Foundation and single-family office space globally has been very underserved in terms of fit-for-purpose total portfolio tools. What Endowment Foundation and single-family offices do is nothing other than total portfolio I mean, of course, they need to then understand some more specific aspects of what they do asset class by asset class, etc. But the most important aspect of what they do is the management of the total portfolio and the optimization of the total portfolio. So we, you know, we bring a fit-for-purpose solution which actually has everything they need in the right place with the right data because we had the private asset data thanks to the Burgess Acquisition. We power it up with some of the traditional MSCI strength like the factual analytics and all the all of a sudden you have literally a best-in-class winning If you add elements of sustainability and climate, which the asset owner community still is very focused on, all of a sudden you have something that is really unique.

Got it. To your earlier point on the technological limitations, I guess maybe a two-part question, was it AI that helped bridge that gap or some of these new technologies? And then you hired a new CTO, so what What does the new CTO bring to the table, I guess?

We hired a new CTO.

OK, sorry, yes.

We are looking for a new CTO. And so let me start with the first part of your question. So there was, on the one hand, work that we were already doing in-house, which we continued and accelerated, partly thanks to AI. We did some very successful targeted acquisitions, so Foxbury was in that domain, Vantager also in that domain, more recently Compass, which have literally given us bolt-on capabilities that have augmented the type and quantity of products that we can build and that we and service, and then also we focused on restarting the reverse switch engine. So you know those four things together are all necessary and I think that's what you will see continuing. Now from the point of view of the tech stack and data stack work, I think there was a recognition that MSCI is a financial tool and services company, but these financial tools and services literally consist of data, models, and technology. The models is what we traditionally have been excellent at, so it's research. The data and the technology had not been as clear and as great of a focus as it had to be. And so Henry took the decision to say, OK, we really need to have somebody focusing the data object and data asset, data architecture, and data strategy, and we need to have somebody really focusing on the product engineering platform and architecture side, and that's the person we're looking for. And so as soon as that happens, and I'm confident it will happen soon, you're going to have an MSCI where the management team is equally focused on all these three ingredients, as well, of course, as our clients.

Got it. Just touching on AI again, And Henry talked about how AI has been a godsend for MSCI. And so from your perspective, can you just talk about some of the different avenues of how AI is improving MSCI? I know it sounds like new product innovation, but maybe some other client adoption examples, et cetera.

So AI is, of course, a rich set of tools, a rich set of powerful transformative tools that allow to do knowledge work in a substantially faster, more ambitious, and more impactful manner at relatively limited cost as long as you manage it reasonably. For a knowledge company like ours, this is revolutionary. That's why Henry calls it a godsend. So first of all, the complexity and the scale of the research problems we can now tackle is literally orders of magnitude more ambitious than what he was three years ago. And that, I think, will, you know, as we get better at mastering new technologies and we also realize new and different ways that it can improve and transform what we're doing, I think this will accelerate. So as a researcher, you can now have 10,000 analysts working for you that can all process data extremely fast can all breathe on extremely fast or have access to a vast amount of information that they can absorb really quickly and keep the information current that's unprecedented okay so that's my view is literally for an ip company it's the most transformational event then of course there is the aspect of the efficiency speed, scale, and cost with which you can build all these new objects and then run them industrially after you actually conceptualize them, right? And of course you know various type of AIs are of tremendous help there. So you know the quality, scale, and size of QA you can do using machine learning is unprecedented. Now, this didn't need LLMs, but you can use LLM to run machine learning even more efficiently to do even better QA. So you combine the two, all of a sudden, you can have incredible results. So you've got better research ideas. You can build them more efficiently. And then the third substantial transformation is that AI can help make it a lot easier for for your clients to consume the tool, the content, and the services that you produce for them. And that, I think, we are just at the beginning of discovering. Because it's not that this is just like a flick of a switch and everybody is ready to go. Both sides need to get ready. They need to make certain changes. They need to adopt certain technological protocols. And then it will become a lot easier for this to happen. But, you know, in some of the more recent solutions that we have launched, of course, the AI empowerment is inbuilt in the solution that we built, and then the client just sees the transformation. So, you know, we launched this new service called MSCI Index AI Insights. So, at the first level, this is just a way to basically, you know, quickly and precisely retrieve information about MSCI vast index IP array, okay? So instead of having to, you know, call somebody who then calls somebody who then calls somebody and gets you information, you access the tool, which is available as a standalone application to what we call MSCI 1, but it's also available to MCPs and APIs in both Cloud and ChatGPT. My assumption is that it will be available in many other ways as well, right? And you get a lot of answers very quickly, right? So figuring out what's going on your index, figuring out why the value is this, not that, start figuring out what happens with this stock or that stock very easy. However, we've done more than that. So there is a very highly trained engine that basically we spent quite a while honing that you can use to do more sophisticated reasoning. And so you can start asking the engine to do a lot of things that beforehand you would have had literally a team of people helping you do, then calling MSCI, then giving you back the answer, and iterate, and then maybe after two weeks, get you what you needed.

Now you can actually do it in a matter of minutes. Got it. You know, one of the buzzwords being used is MCP. And a lot of companies are trying to provide stats on their partnerships, et cetera. How do you look at your current relationship with the LLMs? Like, what kind of revenue model, business model are you now? And where do you think it eventually evolves into?

Yeah, look, I think we're at the very beginning of the journey, right? So as it's always the case at the beginning of a transformation, you don't know exactly how the new world will be structured. You don't know exactly what patterns or protocols will prevail at scale versus which one will not. But I think it's a fair bet to assume that at least for the largest users or content and services, they will want a programmatic way to interact with it. They already do. So by the way, we've been using APIs, like many of our peers, competitors, and larger clients already for 10 plus years. The APIs were not architectured in the way you need to have it now to make everything easily AI fungible, but the concept is not new. Now APIs plus MCP is different, because basically you're saying, OK, I'm going to provide a universal a Lonely Planet guide that tells you exactly how to use and where to find my stuff in a way that basically means that you don't have to actually spend a whole lot of time figuring out how to go by the APIs. So that's a transformation. And then if both sides develop agentic capabilities that are permeable through the APIs, I I think that's where the real transformation begins. And I think that's what we are at the early stages of. To be honest, right now, we're seeing more direct, tangible client impact on the capabilities that we have identified and make available to them by our applications than by an MCP and API. So for instance, MSCI, Index Insights, we have many, many, many more users that that come on to MSCI1 than ones that go through the cloud MCPs. We do have a few that go through the cloud MCPs. I mean, a few means, probably a few hundred. But we do have many, many more through MSCI1. Same thing with many of our more complex tools and services. For the time being, the client will tell us, look, if you go through your own thing, we are more comfortable. We are more sure that we're going to get the right answer. We're going to get the right reasoning. And so for the time being, we prefer to use that. Is that going to change? It depends what you're using the engine for. If it's just retrieval, and then you use your own logic, then I think you're going to go through generic APIs and MCPs. If you're a very large scale player, all you're going to focus on is the retrieval bit. And I think you're going to go through that. If you're not such a scale player, and so you want to benefit from the specialized purpose-built I think you're going to go to our channels. Now, your point about pricing models and the evolution of the economic curve. Look, any way this is going to shape up, it's clear that it will get people to call more data, more services, more functionality, more tests, more what-if scenarios, because there is a possibility to do it. And a lot of these things are useful, right? So people didn't use to do it because it was very complicated. So you just really ask what you couldn't figure out yourself. But as it becomes easier, you're going to say, sure, why do I want an answer which is directionally correct? I want the exact answer. I'm going to ask the question. I'm going to do it programmatically. And I can ask 1,000 questions, 100,000 questions, a million questions, 100 million questions, right? So the way I think this is going to evolve is that we're going to have a layered pricing models whereby you're going to have access to a certain set of modules of data and functionalities depending on what you decided to pay for. There will be then a consumption layer, where there'll be a basic consumption pattern included in the basic fees. You ask 10 questions a day, we're not going to charge you more money. You ask 1,000 questions a day, maybe we start thinking about it. You ask 50,000 questions a day, we're sure we're charging more money, right? Because we like to support all the traffic, right? And so there will be charges for that. Then there will be the question of whether you actually ask the engines to do bespoke analysis for you. Simulate a portfolio, do a stress test, try out a different score of this type or that type. build a custom index, build a custom model, OK? And so we'll charge you by objects. When you decide to build a new custom index, we'll charge you some money. You decide to basically run money off it, and there is a clear relationship on how you're going to run the money. We'll charge you ABFs like we do today if you are somebody who's licensing and indexed to an ETF. So I think that's going to be the evolutionary curve.

You mentioned custom indices, so let's touch on that. But I think historically, people interchange custom indices with self-indexing, and they thought that would be the death of all the index providers. That never happened. You guys have grown your custom indices business in the low double-digit rate for a long time. And last quarter, it was into the 20s. So can you just tell us what's been driving the good growth? What drove the incremental growth in first quarter?

OK, so there is, I think, a long-term trend, and there is specific events. So the significant growth and growth that you saw last quarter is driven by both. So on the one hand, we do have a strong long-term trend, which I can talk about in a second. And then we also had a very substantial licensing event with a big investment house that basically decided to substantially increase the amount of IP they're licensed from us in that space. And that helped the sales. Now, that did not happen in a vacuum. So we see that literally as the logical manifestation of the long-term trend. What is the long-term trend? The long-term trend is that the financial industry wants more and more portfolio recipes. Why does it want more and more portfolio recipes? Because no two human beings, no two institutions have the same portfolio objectives. They have similar KPIs, potentially. They have similar challenges to tackle, but no two of them want exactly the same portfolio object. And of course, because it really was not possible to build a truly customized and supportable individualized portfolio solution, we didn't do it. We found all sorts of proxies. People used to have, in a few off-the-shelf model portfolio options to offer to their clients, few versions of the same basic building blocks. and, you know, various combinations of this. Now, same with indices, okay? MSCI built its reputation building benchmarks. And the philosophy was a benchmark for all, okay? It's important, a benchmark for all. So basically everybody who has essentially the same KPIs should be happy with the same benchmark, right? Because you're starting to say, okay, what's the right way of doing X, Y, Z? Well, here it is. Here's the benchmark that tells you exactly what is the right way to do X, Y, Z. And so if you want to understand whether you're getting the right service, not the right service, whether you should go a bit more left, a bit more right, you look at the benchmark, and you see how much and why you're deviating from the benchmark, OK? You say, OK, great, so we have all figured out the ideal ways to do certain things. But let's face it, none of us is that ideal person. None of us need the stylized benchmark portfolio. We need different things. And so you say, OK, well, can index technology help me to literally formalize the expression of what I need specifically for my personalized portfolio? And the answer is, of course it can. In fact, it's the best technology available in the world to do that, right? Because it can literally encode all the KPIs needed by the individual investor, institutional or retail, into a programmatic set of steps that you can then feed to a computer and, of course, propagate forever, unless the KPIs need to change, in which case you can adapt And be sure that as long as the KPIs are still the one the clients need, the investment recipe will do the job the right way, OK? And so all of us say, wow, why shouldn't we do that? And that's exactly what has been happening. So the technology has progressed first slowly, then much more rapidly to support this. People have understood first slowly and then more rapidly that this can be applied in all sorts of contexts, by the way, including in supporting active asset management. And therefore, that's what you are seeing. You're seeing that in all context of investment and financial life, people are starting to look at this. So whether it's in the context of building model portfolio for individual world management clients, in the context of building multi-asset class benchmark for asset owners and insurance companies, in the context of building bespoke benchmark for investment institutions that want to make sure that their PMs are doing the job really as best as they could, whether it is for the more traditional applications, ETFs, whether it is to build structural products, Whether it's to build, customize, systematic investment recipes for pension funds or hedge funds, you can do it by an index.

Got it. Maybe just a quick follow-up. In the indices business, but especially on the custom side, I think Foxberry, you mentioned, was an acquisition you made. Recently acquired Compass as well. What did those two bring to the table? Because the impression is you guys are an index powerhouse.

So as I said, MSCI built its reputation and its business on the benchmark for all era. Now, many components of the infrastructure that you need to support indexation services at large were natively built during the benchmark for all era. You know, best-in-class corporate action engine, best-in-class way to calculate including factors that have to do with free float, market structure, and so on and so forth. And a whole host of other things similar to this, right? When it came to, you know, enabling very specialized custom index portfolio construction recipe, the engine had not been built for that. Foxberry had been built exactly for that. So that's why we bought it. So we say, okay, we have the best-in-class central engine that does all those things that are needed at scale for a standardized massive index business. We then have the calculation engine that is very robust and reliable. We need to improve the portfolio construction engine. So we bought Foxberry, we spliced it in, and that's how you you actually get a massive acceleration in the custom equity index business. Now, again, MSCI was built to do benchmark for all, especially in equities, especially in equities, right? So we entered the fixed income business five or six years ago. We're now growing there. It's a very attractive proposition. But let's face it, we did not have the same capabilities. And so we started working with Compass to satisfy needs from, for instance, the structural product market, where people say, gee, guys, you're so good at equities, especially after you bought Foxbury, but we see demand for multi-asset class stuff. We see demand for nonlinear payoffs that embed option component, futures component. And, you know, we were doing that, but we could do it only small scale. So we started working with Compass. We saw that they were truly excellent at it. They had built a really impressive engine that could do this at scale with a high degree of flexibility. So all asset classes, all type of payoffs, so long, long, short, option-based, future-based, option-and-future-based. They also handle commodities and cryptocurrencies.

And so we said, okay, maybe we should buy them. Okay, that's super helpful to appreciate that. Maybe let's just touch on the two other deals you did recently as well, Vantager and PM Insights. I believe they both fall into your private assets category, where Burgess is obviously your main asset. So what are those two assets? How do they fit into the plans there?

Yeah, so Vantager we bought because of the private asset business, but also because the capability they develop we think will be really helpful for us to have and then further develop across the board so what manager had developed is a highly capable highly specialized ai-powered engine for you know data extraction to a model in the context of of private assets documentation. You know, basically take a whole bunch of PDF, a card load, and basically tell the engine to fit a specific data model you've pre-created in a very dependable and competent way. So you can actually turn millions of PDF pages into very highly conceptualized structured data that you can then use in the investment process. That's that's why it's important, right? This is, of course, immediately super helpful in our private asset business, but we think it's going to be helpful across the board. Well, because we think that the future of the investment industry, on the one hand, will keep on being about analyzing time series. But on the other hand, it will be about building very specific knowledge about individual assets. Some of it is done by the traditional financial analyst process, which does a fantastic job. a lot of it will need to be enriched by looking for all sorts of data sources in all sorts of unstructured forms with this type of capabilities we will be critical to build okay and so PMI look we got to know PMI because remember we still are the benchmark guys right so we build all the possible benchmark that one could ever think of in the equity space we now well we're now doing any in fixed income, although in fixed income there was already other people, so we need to find ways to differentiate ourselves, which we're doing. And then we say, OK, we're going to do it cross-public asset classes, thanks to the acquisition of Compass. And then we say, what about the private assets? And in private assets, we've indeed started to build systematically fit-for-purpose benchmarks for all asset classes, sub-asset classes, is different type of cuts, different type of regions, and so on and so forth. However, those objects, they have the liquidity limitation that the underlying asset has. So if you want to start bridging the gap between public asset benchmarks and private asset benchmarks, you need to try and find something in between. So PMI had developed very powerful technology based on having all sorts of agreements with industry players in the secondary market to be able to essentially estimate evaluated prices. And so we built with them a ticking venture index. That worked out quite well. So we said, OK, well, we'd like to do this more broadly. And so we said, OK, why don't we buy them, right? And so indeed, this will accelerate our ability to build really usable, in this case, not venture, really, indices. for the private asset industry, first on the equity side. On the credit side, we're doing it, by the way, also. We're doing it a different way, so there'll be something hitting the news soon, hopefully, there as well. But as we heard, I don't know how many of you heard the launch presentation for Maris, which I thought actually was really interesting. At the end of the day, what our investors are asking them is, how do you get your marks? How do you get them? And look, in private credit, I believe that probably the answer that someone like Aries would give their investors may not be perfect, but would be reasonably solid. Probably we can do better, because that's what we do for a living, but it is reasonably solid. In the private equity space, I think there's a lot of work to be done. And so we aim to bring transparency to the world, make it easier to start seeing and comparing the returns across all ways of holding equities, and therefore help the investor community be more effective in how they allocate capital.

Okay, we've got five minutes left. So let me focus on analytics. So when it comes to indices, I think consensus is it's a benchmark, probably not disruptable by AI, private assets is evolving. But on the analytics side, can you help us appreciate what the moat there is versus all these AI solutions that could potentially threaten that?

So look, the analytics business is a vast business. So there certainly are aspects of what people do, including ourselves in analytics, that are substantially more prone to AI disruption and that are other parts of what people refer to typically as analytics that are substantially harder to disrupt. OK? So let's talk about what is easier to disrupt. So anything that has to do with workflow is to disrupt. And even easier if the workflow is not particularly proprietary, not particularly unique, not particularly insightful into what a specific client or type of clients are doing. OK? And so somebody who's providing relatively standard risk of performance attribution on liquid public assets, That's going to get disrupted, OK? At the other end of the spectrum is where you are building proprietary risk and performance models based on 40 years plus of very granular data that has been cleansed in an extremely disciplined and pedantic but fit for purpose manner across a multitude of markets that you can therefore use to understand and the structure of your portfolio exposure across the board in a basically single coherent manner. That is not so easy to disrupt. Now, if you then also say, OK, let's think about the index benchmarking analogy. So we built the benchmark. The benchmark essentially has standards. The standards are useful because not only they tell you what the theoretical performance of a certain object should be, but they also to help you crystallize norms that are going to be useful, even if you do totally freestyle portfolio construction. Same with models. So enter new technologies. And we can now build new models and calibrate models in a way that it customized to clients in a matter of hours. It used to take us months. All of a sudden, we can innovate a lot more in the model space as well, OK? So my view is, therefore, there's going to be the opportunity to actually extend the mode thanks to AI, as opposed to seeing the mode get eroded because of AI, OK? So I think it's a question of where you are in this spectrum. So the model capabilities are the hardest to replicate. The very generic risk and performance analysis are the easiest to disrupt. But the total portfolio services may get easier to disrupt, but there is a lot of ingredients you need to understand how to put together in a coherent, systematic, and dependable manner. So I think they are a lot closer to the far end of the model side than the other end. So I guess MSCS makes it more to the far end. Is that what you're saying? Well, what I'm saying is that I think that there is a way to do all of this in more intelligent and more value-additive ways of using AI if you understand the finance first and foremost and you understand the client. But in particular, within the model and total portfolio solution space, I think the advantage we have is so substantial that it would be very difficult for a very smart engine to actually replace us anytime soon.

In the interest of time, one last question, 30-second answer. So, you know, everyone talks about the slow decline of the asset managers, and that requires its own fireside chart probably, but you've been growing a ton on the trading and hedge fund side. How early are we in that segment penetration process? Could that keep growing to help offset the asset?

Listen, I always tell people, look, if you look at the large scale trading operations, and whether it's the trading operations of a broker dealer, so it's officially called of a trading operation, or whether it is the trading operation of a proprietary market maker, so it's called the market making operations, or it is a trading operation of a large platform hedge fund, so it's called the investment desk or the pods, the point is they all do equities and fixed income. Within equities, they do idiosyncratic linear, systematic linear, non-linear, so typically often involving one or more world-based strategies. and they do much more complex relative value, OK? Right now, we play a significant role in equity, linear, systematic. So we got plenty of adjacent areas to capture. Now, what do I think is going to happen, OK? The world of investment and therefore of trading and therefore all of finance is converging decisively to a portfolio-centric architecture, OK? The unit of conceptual exchange will be not the individual idiosyncratic risk. It will be the portfolio. I mean, of course, some people need to slice and dice idiosyncratic. Of course, you do this for a living, right? But the reality is the majority of people will understand portfolio. They won't even want to touch the individual exposure, right? In fact, it's probably best for everybody that the vast majority of people never touch it. In that journey, you need somebody that basically offers the standard rules that everybody can agree around and therefore communicate with. That's what MSCI aims to do across all asset classes, all form of portfolio trading, linear, non-linear, convex, non-convex, around the whole globe.

We're out of time there. Thank you so much for the user. Thank you, everyone.

Thank you.