Thomson Reuters Corporation ($TRI)

Good afternoon, everyone. Thanks for joining us. Curt Nagle, I'm the new Senior Business and Information analyst here at BofA. Session today, during lunch, is going to be Thomson Reuters. Really, really pleased to have CEO, Steve Hasker; and CFO, Mike Eastwood with us. We're going to structure this as a fireside. Any questions, if we have time, we'll field the audience. But Steve and Mike, again, thank you for coming. I really appreciate it.

Thank you.

And why don't we kick off? Yes. So as you can probably imagine, obviously, very topical. We're starting these conversations on AI. In terms of the first one I ask, Steve, you've spoken really, really directly, and I think really -- really directly about the criticality, right, of professional as you put it, fiduciary grade A models, right, versus general purpose models when we're thinking about all these potential risk and opportunities, right? And how much of that matters specifically for your legal end markets and of course, for tax accounting and audit as well, high cost of failure, right? So in terms of data assets, domain expertise, right, and increasing the use, right, not integration, but use of models. How are you differentiating yourself from both established competition and entrants?

Yes. Thanks, Curt. So we've also got Rawia Ashraf, who manages the CoCounsel product suite here. And of course, Gary Bisbee who's a prime mover in our finance and operations at TR. So thanks, Curt, for having us, and thanks for the question. So as we sort of enter this agentic era of AI, we see it as an opportunity for growth. We see our TAMs expanding and -- in legal and in tax and we see at least 4 assets that pretty clearly differentiate us as it pertains to serving fiduciaries. And when I say fiduciaries, I mean lawyers, tax preparers, auditors and the like. So these are people who have to be right, right? They can't use a probabilistic solution, they need a deterministic solution. They need to be right, and they need to be able to verify, constantly verify and validate their arguments and their results. And so the 4 assets that we bring to bear, that equip us to provide fiduciary grade AI versus sort of broader knowledge work, AI or even professional grade AI for salespeople or coders or whomever are as follows: The first is our vast repositories of unique and proprietary data. So if you don't have it, you can't ground and train your output and you don't have a source of truth, you just don't have it. And so we have it in 1 or 2 other players, and we've got more of it than anyone else. The second is thousands and thousands, at last count, 4,500 highly trained domain experts that we have repurposed -- and Rawia has been a leader in this, repurposed them to train our agents and so we have, for example, CoCounsel bench, which basically will train CoCounsel to perform a particular set of tasks. So it will plan, it will reason and it will execute. And before it gets released into the product, that bench of experts need to validate the results and they just sign off it. And once they sign off it, Rawia pushes it into the product. And so first is data, second is experts. The third is our data privacy and protection. So we provide an ironclad guarantee to our customers, which is their input will not become part of our AI output. And we suggest that they ask any other provider that question because to the extent you've got a start-up with a great product or a great UX or whatever they might have, but that it's learning from their workflows and from their IP and embedded in that product, that is a third rail issue for fiduciaries, right? That is a third rail issue for fiduciaries. The idea that you're going to train a tech player to perform the tasks just in the long term, we don't think is durable or sustainable. We think that's going to be a real sort of moment of correction, if you like. And then the last, which is -- so we don't talk about much, but if you're a subscriber to Westlaw or Practical Law and you've been a subscriber a period of time, and now you're a subscriber amongst the million users of CoCounsel, we have an extensive support network. So you can actually pick up the phone in the middle of the night and say, I am working my way through a particular piece of litigation research. And Westlaw has produced the following answer and I don't understand it. And I'm in front of the judge tomorrow morning. Help me, explain to me where my logic is wrong. And by the way, we have the same for our tax calculation engines. I'm at line 1079 of the state return. And I don't understand how to apply the depreciation schedule of my client into the deduction that I can or can't claim. And so I think when you talk about the sort of -- some of the start-ups, some of the foundation models, ask them whether they're going to have that level of support for fiduciaries because fiduciaries are doing 2 things. They're generating arguments, and then they're constantly validating and checking it. And AI cannot be trusted to check AI. So the data, the experts, the data privacy and protection and the governance and then the support, we think, positions us uniquely to serve fiduciaries and 1 million CoCounsel users and the uptick in the growth of our Legal business as well as the successful deployment of Westlaw Advantage, which is the first deep research agentic product that truly unleashes the benefit of those things for the -- in the conduct of the agent. That gives us great confidence that the upcoming launch of the next version of CoCounsel will be similarly successful and differentiated.

Got it. Maybe sticking on that point, Steve. Yes, just in terms of the importance of -- maybe we'll just zero in on CoCounsel Legal, I think new one's coming out in maybe a few months.

Yes. So Rawia has got an alpha at the moment, with a couple of hundred customers. Goes into beta on April 20. And then as soon as we start getting the right signal, we'll put it into general release in the summer.

So yes, in terms of what to me, sounds like a much more integrated tool, right, across kind of the whole corpus of...

It's really the first time we've taken content and the expertise and embedded it into the agents.

Right. So in terms of -- a lot of ways you could look at it, whether it's stickiness, whether it's wallet share, whether it's upgrades, how should we think about that?

I think it's a bit of all of those. If you think about TR's Legal business, we've traditionally sort of historically been very focused on litigation and very focused on the law firm, potentially sort of less focused on transaction and less focused on Corporates on the General Counsel. And so we think CoCounsel both strengthens our position within the litigation space and the law firm space, but it enables us to extend in a really meaningful way into the transactions activity within our legal customers. And also build on practical law, build on legal tracker and the sort of early forays that we've made into the General Counsel's office. And certainly, the announcement that we made at the last earnings call that Microsoft General Counsel group are adopting CoCounsel top to bottom, we think is sort of the start of some interesting growth around corporate and corporate legal.

Interesting client, too. Yes. So maybe talk a little bit more generally about what you're seeing in terms of across the different segments, renewal cycles, upgrade cycles, and you've got a lot of new product coming in. How much of that is driving the 9% organic, at least in legal? And just over the next, let's call it, I don't know, 24 months, what you envision for an upgrade cycle because of all these new products coming on.

Mike, do you want to take that?

Yes, happy to. We had a similar question in our last session before lunch today. And the question relates to in regards to what's driving the 9% organic growth for Legal Professionals. And the way I frame it, it's the entire portfolio that's lifting it. I think there's some natural reaction thinking that it's our CoCounsel given the significant growth that we have with it and thanks to Rawia for it. So the CoCounsel is certainly a key component of the Legal Professionals growth of 9%, but also Westlaw. Many of you will remember, we launched Westlaw Advantage in August of 2025 at ILTACON. It is performing incredibly well, as we mentioned on the last earnings call, and we're very pleased with the Q1 sales of Westlaw Advantage. Also within Legal Professionals, we have practical law, which is how Rawia joined us 13 years ago. That business continues to perform incredibly well, $650 million of revenue, about 60% of that is in Legal Professionals, 40% is in corporate. So Curtis, really the entire portfolio that Legal Professionals that has -- is performing really well and driving the 9%. I'll just put a plug in with the May earnings call, we will start providing you with information on how Legal Professionals, excluding government is performing. That's something that I think will be very important because we did have some cancellations that we discussed during the February earnings call. I think you'll be very pleased with that trajectory of Legal Professionals, excluding government.

Great. We always love more disclosure. Steve, going back to you, just to quickly touch on the Anthropic relationship. You gained a lot of exposure, up on stage with them a couple of weeks ago. I guess 2 fairly simple questions. Just one, just define the relationship you have with them, I think there's a little bit of confusion in terms of scope, depth or however you want to put it. But that would be the first one.

Yes. So Anthropic are a vendor to us today. I was actually there in San Francisco with them on Monday. So we've built our agentic products, so Westlaw Advantage, the next version of CoCounsel to be model-agnostic. So in other words, we can and we do change to the best foundation model. So we've used ChatGPT over time. We've used Gemini over time, and we're currently doing a lot of the work on top of the latest Claude models. And we think that at the moment, those are the highest performing enterprise models. And so they are a vendor to us, and we do it constant -- as you expect us to constant evals on whose model is performing best for our products and our use cases. We have, in addition, developed our own large language model specific for legal. It's called Thomson and we will talk about it more and more over the coming months. And the reason we'll talk about it is that Jonathan Schwartz, who's a former Google DeepMind scientist. He has developed a model, which is outperforming the very best foundation models for some of the specific legal tasks that we are going out after. And so if you think about that from Mike and Gary's perspective, it gives us a degree of optionality in terms of both the cost and the speed and the accessibility to that and the exclusivity of that. And if you're thinking about it from Rawia's point of view, it's a great means in which we keep the foundation model as honest in terms of accessing the very best model and getting favorable terms. So that's really the relationship with Anthropic. And we think that CoCounsel and Cowork will fit together quite nicely. If Cowork is well suited to provide AI tools to knowledge workers more broadly and in some cases, professionals, were at the fiduciary grade and Cowork is not designed for the fiduciary graph because it doesn't have the content. It's not supported by thousands of experts, different data privacy and governance rules and not the customer support.

Okay. And then just one more quick question on Anthropic just with the -- and I think this has changed today, the supply risk designation of the U.S. government. Presumably, that may preclude maybe even yourselves or clients from using it, I guess, being agnostic, is that not an issue?

Yes. I mean it's not an issue because we are model agnostic. I mean, obviously, Rawia's teams want to spend time developing the product and just driving further and further into more sophisticated use cases rather than chopping a model out and so it's not completely costless to us in terms of time. But being model-agnostic means that should that supply chain risk designation hold, we can move away from that model. We hope that that's not the case, but we're not sort of overly exposed to that.

Okay. Fair enough. Yes. So just sticking with AI, another, I think, maybe developing or a large theme, right, is how pricing models evolve over the next few years. The ones that have the mission critical, the fiduciary, proprietary data, however you want to call it, and potentially moving to more consumption models, value-based models or based on outcomes for your customers. I guess to what extent is Thomson Reuters considering that? I guess just given the potential for changes in the legal workforce. It seems like it's pretty pertinent. So how are you looking at?

Well I think -- I mean I think the first sort of misunderstanding that a lot of the sort of ups and downs, mainly downs that we've seen on in terms of market caps and valuation. So for the first one is I think there's sort of not an understanding of what it takes to be a fiduciary and what it takes to serve fiduciaries. I think -- and sort of undervalues that set of aspects. Sorry, what was the other part of your question?

Well, on pricing, just, it'll look...

At the moment, we -- the other, I think, misunderstanding was that we might have seat-based pricing.

Well, it's not so much the seat-based pricing, right? But with -- against all the points you made about the value and I mean, fiduciary, right?

Yes. So typically, we've got enterprise-based pricing, and it is sort of -- it's in some way, shape or form tiered. Where we will move, particularly as the industry goes from the billable hour to value-based pricing and more and more of the work is performed by agents and ideally our agents, we will look to shift to value-based pricing. So we'll be able to sort of price differentially for more sophisticated tasks. We'll be able to price differentially for much more volume in usage. I think it's too early today to sort of prescribe exactly how that's going to progress and what benefits that's going to provide to our shareholders. But certainly, we're keeping a key eye on the overall usage of the product in terms of user numbers, the degree to which it's been used for highly sophisticated tasks versus more sort of repetitive operational routine tasks and ultimately, the impact on our equity partner profits, and the impact on the sort of efficiency of a general counsel.

Yes. So at this point, I mean, you basically preparing to, I guess, you call it a valuation framework for that, not something that you're having active discussions with customers.

No, we certainly are having active discussions with customers. But you can understand the tools at the moment, I had at least one general managing partner of a law firm here in New York said to me. Look, the tools are 18 months ahead of the change management in terms of the rewiring of the underlying legal work that's performed within his teams and so as that change management starts to pick up steam and the impact starts to flow through, through his organization and ultimately to his bottom line. That's when we'll be able to kick the value-based pricing.

Sure which is I think other thing you commented on, right? So tools are ahead of the usage.

Which is sort of no different than any other significant tech development. It takes a while for the sort of processes to be rewired to take full advantage.

And the rate of change.

No, exactly.

Yes. Okay. Maybe Michael, I'll throw this one to you, just changing up a bit if we can talk about the audit tax and accounting side of your business. I think fairly long-standing structural issues in terms of being able to source labor. You've got a tax engine, right? You've had that for a while. It's the core of the business, a couple of new acquisitions, right, to supplement that. It kind of sounds like you're building more of an end-to-end ecosystem, right? I think right, integrating AI.

I would say we have built that. Sorry to interrupt your -- I'd say we have built it. If you go back over the last 3 years to January of 2023, we've had 4 key acquisitions with SurePrep, with SafeSend, Materia and Additive. Those acquisitions, coupled with the existing assets have allowed us to really fill the gap that existed 3 years ago. So we go to market today with the full end-to-end solution, which gives us the confidence when we talk about 2026, we've committed to 11% to 13% organic growth for our Tax & Accounting Professional business, which we now call it Tax Accounting Audit Professionals there. Our confidence remains there, just given the -- we have transactional revenue within the TAAP business. So there could be some variations quarter-to-quarter. But the confidence overall is significant there. You touched on the determination engine, which is critically important for us. Steve talked about the talent of Rawia, but we have Kevin Merlini, who joined us from an acquisition about 2 years ago. He's just doing a fantastic job for us our ready to review, ready to advise products. So if you think about our confidence level in that business, we have a full suite of offerings given the recent acquisitions, the organic product developments that we have there. So we're really, really pleased with the progression.

Okay. Very good. And then maybe just sort of a who knows question, but in terms of kind of meeting those needs for taking up, right? The work of your clients, is that something that's happening as we speak? Are you getting more requests to basically digitize, I suppose.

Certainly within the Tax and Accounting Professional market, I think most people realize in regards to the level of professionals there. We've had significant retirements. I think is what you're getting at which is advantageous for us. We know the tax laws are complex. The U.S. is very complex. Brazil, when we think about our Dominio business in Brazil, that's grown 20% CAGR over the last 12 years since we acquired it there. So the inherent complexity but also the shortage of workers within Tax and Accounting Audit Professionals, that weighs into our favor in regards to us helping our clients every day.

Okay. Fair enough. Michael, I'll stick with you for just a couple more. Margins, right? So 100 bps this year and thereafter the next 2. One of the strongest outlooks, I think, at least in terms of what's guided at least in infill services, maybe just break down kind of some of the biggest contributing factors and obviously, the fixed cost leverage, maybe mix in terms of the operating segments. And then they reemerge on how we work initiative? How much is that kicking in. Another kind of big theme in terms of, I guess, labor and asset utilization, which could be pretty meaningful. So just break that down and kind of level of confidence in the 100 bps?

Yes. Just to level set, we were slightly over 39% EBITDA margin for calendar year 2025. We've committed to 100 basis points improvement in '26, '27 and '28. What gives us the confidence? We have very significant operating leverage, given 60%, 65% of our cost base is fixed in nature. That's been consistent. I oftentimes get the question, in the last 3 years is more of the AI and now agentic AI. Do I see any significant change in that operating leverage? I do not. So that's intact. Second, you mentioned the reimagine our workforce which more simply stated, driving productivity. That is a key contributor in regards to our confidence. So those 2 are kind of the pluses, slightly offsetting those items. We continue to reinvest in 2026. We will in '27 and '28. And in areas where we think we'll have significant returns to help us accelerate organic growth, our guidance for this year is 7.5% to 8% for total TR, approximately 9.5% for the Big 3. So those reinvestments will help us there. And then also, we are absorbing some dilution from acquisitions in 2025, and we've closed 1 acquisition in 2026. So given the operating leverage and our line of sight in regards to the progress and what we know -- we think we will achieve on the productivity initiatives. Those are the factors that give us confidence in our EBITDA margin expansion for '26, '27, '28.

Okay. And maybe I'll just kick it back to you, Steve, in terms of reinvestment priorities, where are you thinking you can perhaps accelerate growth? I mean what are you most excited for and where you're dedicating the most capital in the business.

I mean, it's fairly well balanced across the Big 3. So within Legal, we think that, as I mentioned before, we think that CoCounsel is an engine for growth and TAM expansion in the transactional space and in the general -- in the General Counsel's office. As Mike said, within Tax and Accounting and Audit Professionals, we have these anchor properties around tax calculation engine, which are very, very deeply entrenched within the corporation's tax group. They're lightning fast. They're very accurate and they're cheap to run. And so the opportunity is to use agents to link into those and to perform more and more of the tasks to alleviate the talent shortage that Mike talked about. And as we think about our Corporates business, we think more and more Fortune 1000 companies have to sort of come out of -- to some extent, Excel based calculations of tax and come into ONESOURCE. We think that indirect tax and e-invoicing is comfortably a double-digit or better growth proposition for the time horizon, particularly as the invoicing mandates roll out across the world. We think that the General Counsel is going to undertake the effort as with every function and every corporation to use AI to automate that plays to CoCounsel. And then, of course, we have a really interesting proposition that's been -- was designed mostly for the U.S. federal government and security classified work in the form of the CLEAR and TRSS assets that we think has a growing role to play in AML and KYC in particular in that sort of broader risk landscape. So as we look across the Big 3, we see a nice balanced portfolio of growth opportunities that ought to see us get into the sort of double digits in most cases for those franchises.

Okay. I've got a big audience. So maybe I'll just pause for a second if anyone wants to ask questions. Happy to field them. Okay. No problem. So a couple more. Mike, kind of a perspective question. So transitioning to a new role, right? Maybe the chair of the Thomson Foundation, I think, come April 30 -- April, 30-year career at Thomson. I think it'd be valuable just in terms of kind of into your 30 years biggest takeaways, how the org has changed? And what you're most excited for in the coming years?

Yes. I'll remain CFO through May 8. Our next earnings call is May 5, so we'll participate in the May 5 and then May 8. Just to clarify, that point. So your question on regards to perspective or reflections. The most important word is gratitude, nearly 3 decades. Thomson Reuters afforded me more opportunities than I deserved. I had the opportunity to work in nearly every business, including many that we have divested over 3 decades and also had the opportunity to work in many countries throughout the U.S., Argentina, Brazil, Switzerland and Canada, which was very rewarding personally and professionally. I think it also helped me with the business, hopefully. The second item at the gratitude is what the team has accomplished. Steve and I have been in our chairs 6 years, March 15 will be the sixth year anniversary. And if you benchmark and Gary can check me here, 2019 versus 2026, kind of a 2x factor comes into play in 2019. Our total TR organic growth was 4%, our guidance for 2026 is 7.5% to 8%. If you look at our Big 3 organic growth in 2019, we were at 5%. Our guidance for 2026 is approximately 9.5%. If you look at free cash flow, 2019, we were $1.1 billion, free cash flow in 2026, $2.1 billion. So that's what I say, the 2x factor. And over that time period, the team has also significantly increased our EBITDA margin. We were 31.5% back in 2019, we'll be over 40% in 2026. And the way I view it, that's not about Steve Hasker and Mike Eastwood, that's about the team. And that's kind of my last point reflecting on your question is optimism. I'm -- I can't say this in front of her, a little bit pissed that my wife is encouraging me to retire. I hit 60 years old in 2 weeks, and the level of optimism I have for the business, given the talent that we have here, the $10 billion worth of capital capacity, if you go out to 2028, we're very low leverage now, roughly 0.5 turns. Take that up, we can do a hell of a lot for our customers and our shareholders in '26 through '28 with that. So those are just some initial thoughts in regards to...

Can I add 2 things? I think firstly, I've witnessed firsthand the tension between Mike who wants to stay for another couple of years and sort of fight our way through this moment of skepticism from some of the investors versus Mike, the man of his word, and the promises we made to his wife in terms of retiring at -- in or around 60. I think the other -- the only thing I'd add to Mike's point is the rate of innovation. So when we started our roles, TR used to put out a new version of Westlaw every 3 or 4 years and was good at that. so not here to denigrate our predecessors or anything like that, but we didn't do much else. There wasn't a lot of innovation across the portfolio whereas now we're putting out new versions of Westlaw from sort of start to finish in deep research and agentic in 4 months from initial ideation to general availability in 4 months. Rawia and her colleagues have completely rebuilt and will relaunch CoCounsel starting now in alpha, beta April 20th. And then as I said, released through the summer, CLEAR Investigate, which is an Agentic AI offering. New additions to and integrations around Pagero and ONESOURCE. And you go across the portfolio, and we're now I think a truly content an AI-driven tech company that's able to go toe to toe with the most innovative players in and around B2B software in the world. And so our sort of commitment is to just continue to improve that velocity and impact of innovation through the Big 3 and we have the luxury been very, very focused on those fiduciary professions, and we're not going to get distracted from that. We don't see a need to sort of branch out into different areas or try different things. We think there's lots of growth, as I said, within and across those Big 3 segments.

Yes, Curt, if I could just supplement there. The pace of innovation, which we discussed a lot in the sessions this morning, especially the last 3 years, with AI, agentic AI, just within our company, the pace of movement has just been incredible.

Got it. And with a very healthy balance sheet.

Very healthy.

So yes, maybe just ending on that and then I'd love to do just a quick lightning round world association. So yes, I mean, kind of your point capital allocation or deployment is a bigger theme. The leverage, you just did a $1.2 billion buyback and capital return. I think in terms of, again, very specific focus on your core markets, right, but there is a desire to maybe do a big bolt-on, right? So in terms of just thinking about, let's just call it, the next 24 months, the balance between the buybacks and where the valuation is versus potentially acting at the right valuation on a larger bolt-on?

Yes. I mean I think I'll let Mike sort of talk about the capital allocation. I think the sort of beating we've taken over the last few months and more broadly over the last 6 months, in a sense, is energizing because it makes us the underdog, and we sort of love that. I personally love that. And so we've really taken this opportunity to get our folks fired up about competing and outrunning everybody. And so that's the first thing. I think the second thing is we look across the landscape, we've got $10 billion, $11 billion in dry powder between cash on hand, leverage what is 0.3, 0.4x at the moment and a business that's now producing north of $2 billion in free cash flow. And we're somewhat optimistic that some of the sort of disruption is going to create some interesting opportunities for us. You're not going to see us do anything that is a head scratcher. We might do something bigger if we really saw the value for our shareholders and we felt that it was an asset that's around which we're an advantaged owner, we might do bigger sort of returns of capital and share buybacks and so forth. But our mindset and stance is to take advantage of this moment in every possible way we can. And we think we've got the franchises, the talent and the balance sheet to do so.

And the energy?

And the energy.

Yes. I would say we'll remain disciplined in regards to our balanced capital allocation approach and strategy there. This year was our fifth year of increasing dividends, 10%. We recognize dividends are more important to certain geographies than others. But I would anticipate a continued increase of 10% per year. Obviously, that has to be approved every year by our board, but that's an important aspect. Steve talked about strategic M&A. We talk internally in 3 tiers. Tier 1 for us are bolt-ons, think about with valuations of $500 million to $1 billion, then Tier 2 larger then Tier 3. Your point there, Curt, we're certainly willing to consider larger than bolt-ons, which are to Tier 1. Will we consider a Tier 2? Yes. But certainly, the level of confidence, the thresholds has got to meet some pretty high expectations there. And that still leaves us, to your point, Curt, for capital returns. We did the $1.2 billion that we announced on April 25. Questions about why we did. I think it was [ Bentley ] this morning asked about the $600 million return of capital under Canadian law and paid up capital there's a provision. We received proceeds for monetizing our stake in the London Stock Exchange Group. And the tax benefit expires May 7. That's why we did the return of capital, if there are questions there. So I think just given the low leverage that we have, the capacity that I mentioned, $10 billion plus it gives us flexibility as we move forward, whether it's dividends, strategic M&A or more capital returns. And I think if you think about strategic M&A, I think Steve touched on it. But right now, we have about $400 million within the indirect tax e-invoicing space, opportunity to scale, $400 million in revenue in that risk, fraud and compliance area that Steve mentioned. So opportunities to scale. So -- and then certainly, continued very focused expansion in areas internationally like Brazil, Southeast Asia.

Okay. Board Association? lightning round.

Rawia is smiling because she's way better at this than me so maybe we'll swap.

Legal workflows?

CoCounsel.

Tax engine?

ONESOURCE.

Westlaw?

Preeminent.

Preeminent. Reuters?

The source.

The source. General-purpose AI models?

Wide space growth opportunity for TR.

I like that. Okay. Very good. Really appreciate the conversation.

Thank you. Curt. Thanks very much. Cheers.

Thank you.

Thank you.