ICON Public Limited Company (ICLR) Earnings Call Transcript & Summary

All right. Hi, everybody. Thanks so much for joining us this afternoon. I'm Elizabeth Anderson. I'm the healthcare technology and distribution analyst here at Evercore. Very pleased to be joined by ICON, Brendan Brennan, CFO as many of you know. Kate Haven, IR, as many of you probably also know as well.

Okay. So maybe just to kick it off with sort of the question of the hour for the last many hours. Investors continue to be worried about a slowdown in pharma R&D spend. So how -- can you kind of give us a lay of the landscape right now? And how do you think about tracking it as we've gone through the last couple of months?

Yes, it's been an interesting period. We've obviously seen probably 2 divergent elements, particularly from our point of view, over the last while. I think CROs to start off with have been somewhat insulated from other reductions that we've seen in R&D because, obviously, people are still keen to get through with their Phase II and III developments. It's the nearest thing to commercialization. They need to focus on that now. So I think it's clear as well, but they're still in that part and looking for ways to make more efficient their R&D spends on a year-over-year basis. That's certainly the case. But I do think, as I said at the beginning of this, there's 2 divergent pieces. There's the biotech industry, and there's the large pharma. And those 2 pieces are not firing on the same kind of -- at the same point. So the biotechs we've seen, obviously, had a torrid time of it from '22 into '23, funding levels have been down, IPO market has been closed. It's been a difficult period of time. I think what we saw in the beginning of mid-ish of this year, actually, June, July time frame was actually really kind of better traction for them. And I wouldn't say it's -- I don't think anybody's doing the victory dance, far from it. But I do think what we've seen is stabilization in their marketplace from a funding environment perspective. And that's been positive over the last month or 2 months, continues to be from the January period -- or sorry, July period onwards. And we've seen that in our business wins in Q3, and it still looks to be positive as we come into Q4. There's no great precipice there that should make that particularly worse. In fact, we've come to an interest environment where people are generally more positive about the future on interest rates. And so actually, I think they're gaining a bit of traction and some of the long-term healthcare investors, who have been sitting on the sideline are going back into that space. So I think that is a good news story. That certainly helped us from a business wins perspective in Q3. We expect it to continue to help us in Q4 to get back into the kind of mid-1.2 ranges. Pharma is a different animal at the moment. It's got that kind of shadow hanging out there in terms of the Inflation Reduction Act that's thinking about how it shapes its R&D spend over the next number of years to, as I said, get as many drugs into Phase II and III and really push as much productivity as possible. I think as part of that, they've been pulling back in other sectors, pulling back on discretionary spends, maybe more pulling back in discretionary commercialization spends, but also in any phase clinical development and preclinical discovery a little bit and focusing more on that Phase II and III area, which obviously we're -- this is the core of our organization. What we've also seen, I suppose, the other trend that we've referenced on a couple of our earnings calls as well, is that kind of blending of what they're looking for from a Phase II treatment provider. So traditionally, obviously, and still to this day, the weight of it is still full at service outsourcing. So we take the projects, which we often involve in the design, and we run it from sites effectively. I think the blending of the idea of that between that and FSP, which is more that they hire or effectively, we hire staff to them. So their clinical monitoring force globally might be an ICON employees who are embedded in the pharma company. So they're looking at how do I get something in between those 2 points. Can I have a situation where I have some clinical monitoring, some full-service outsourcing and a blend of 2? Do people have to work in my SOPs, can they work on the CROs' SOPs? How much of the management of that workforce can I outsource as well? Obviously, we lead the field in FSP, and we're one of the biggest organizations in the world in full-service outsourcing as well. So I think from that perspective, it feels like a good time for us to be in the marketplace, and I think that's playing really well from the interactions we're having with our partners at the moment as well.

Got it. That makes sense. Is the right precedent in sort of this pharma spending to think about like prior large pharma M&A from like the, I don't know, early 2000s, where you have just had kind of like initially, these organizations were combined, they would cut spending, but then like, I don't know, some period or months later, maybe 6 whatever months later, there's actually more of an opportunity for you guys as you're like saying, hey, I understand you're reducing your budget, hey, we're 20% or 30% cheaper, like here's a great way to keep the pipeline while also hitting the cost targets.

No, I think that's on. We've seen that not only in that period, but like directly after the great financial crisis, where you really saw pharma companies looking to show that they're being effective with their own cost control on their own cost base. And really, that's obviously that was the advent of the big strategic relationships. So we've seen that, and we continue to see that. And that's certainly something that's weighing on their minds at the moment. How do I -- and it's particularly the C-suites of these organizations who are looking at how to fundamentally we do our research and development, how do we get a better bang for our buck in terms of this. I think there's always an opportunity for large-scale CROs, such as ourselves. And it's particularly, especially with the flavor of this one that is more runs to FSP, I think that for ICON particularly given our experience at developing those types of relationships, given our experience of running that type of business, it suits us particularly well in this particular environment. But I do think it's always an opportunity. We're out there with existing customers, we're asking them what are we not doing for you today that we have that service that we could be doing, how do we bulk up in the existing relationship. And for new customers, we're exploring what kind of delivery service they want, how much FSP, how much full service and how we're best placed to do that blend. So it's been a very active period in terms of conversations around that these last couple of months into the end of this year has been really active from that perspective with those large pharma customers.

Got it. No, that makes sense. And how -- is there anything to call out in terms of like therapeutic areas switches [ recently ]?

No. I mean it's -- we're kind of through the kind of that very heady time of the vaccines piece, obviously, through COVID. I would say that particularly the CROs are well down the other side of that hill at this point. So that's not an incremental headwind from '23 into '24. Obviously, oncology, rare disease, CNS are still very, very prominent in terms of numbers of drugs in development. People obviously are talking about metabolic. It's an area that people are focused on. It's still a relatively small part of the overall landscape. It's probably maybe just in or around or just under sub-10%. So it's in and around that kind of number. Obviously, it's probably going to be a high growth area, and we're all going to look to that as well. And I think, we have a lot of experience in that space. So we'll be looking to take any advantage we can there. But the traditional weighting towards oncology, CNS and other indications very much back where it was prior to COVID.

Got it. Okay. That makes sense. And then any sort of changes in RFP flow as we're thinking about 4Q?

No, at this stage, it remains solid. And we talked about being up nicely year-over-year on a trailing 12-month basis. That persists into Q4. I think the mix of it is pretty decent as well. So it's not -- we've seen obviously -- as I mentioned, we saw kind of good recovery in biotech and emerging biotech, small biotech, particularly in Q3. That remains stable, is the best way to put it as we go into Q4. So that's good. And we see a fair amount of interest from the mid and the large. The large guys, it's not -- it's kind of a different animal. It's not technical business wins. That's not what you're kind of having more long-term conversations with them around their sourcing model. And that sometimes shows up in an RFP flow and sometimes it doesn't, because that can be -- it's more binary in terms of how that's won. But generally speaking, I'd be pretty happy with where things are, and certainly have been at least as good as they were in Q3.

Okay. That's certainly good to hear. Are you seeing any sort of pullback in terms of like you obviously are very much focused on Phase III and like on the Phase IV sort of like real-world evidence part of the business. Is that somewhere where you're seeing any kind of like pullback in spend from the pharma side?

No, not to this point, fortunately. It's been a really good growth business for us. A lot of those studies are often mandated now. They're not optional. They used to be more optional in the past, it's often mandated at this point. So that still remains an active space. That's been a really good growth sector for us. As part of our organization, we've seen that grow nicely into the double digits over the last number of years. So -- and the business wins trend has been very solid there as well. So it's not -- actually, the expectation certainly is that in 2024, it has another good, solid year of pretty solid growth. One of the other things that we're talking about and thinking about and I've been discussing over the last year or so is the quantum of, is this -- what phase of the cycle are we at with large pharma in terms of how much business they put out there? Are they going to be eager now to put actually more products out into Phase II/III work in the anticipation that they need to offset this IRA or offset their own internal patent clips? That's been something that's been an interesting dynamic. I think from my read of it, they're at a point at the moment where they're figuring out what method they want to do their outsourcing in and with how many providers they want to that outsourcing. I don't think the rubber has quite met the road yet in terms of any glut of additional trial work that they want to do as they go into 2024, but it will be very interesting to see. I mean I would agree with the thesis that they do need to put more work into that phase to get more of a commercial bang for the buck. But to this point, it's still more in the form of working out the structure. We'll hopefully see more drugs come into that and into more steady starts as we go into '24.

Got it. And if we think about some of your sort of data and analytics portion of your business, that should be sort of relatively more insulated from some of these pressures to in terms of just like number of seats and that's sort of a steadier business overall. I think about like Symphony or...

Yes, it is. It is. It's -- I would say there's a little bit of a discretional spend there as well for us. So it is a little more impacted, I would say, that part of the organization. It's a relatively small, as you know, Elizabeth. It's single digits in terms of the revenue contribution to ICON. So they're having, I would say '23 has been a tougher year for them than '22 and I think we're still seeing that in the back half of the year. So they've got probably tougher kind of hill to climb as we go into '24 off the back of the business wins. So again, you can see that discretional decision-making around where money is being spent happening even in the ICON organization.

Okay. No, that makes sense. Have you guys ever talked about the percentage of your revenue that comes from the FSP business?

I mean we've talked about it kind of from a notional perspective. It represents circa 20% of our total revenues, but it's -- it's a very solid business. On the combination of our 2 legacy companies, ICON and PRA, obviously, we brought together the largest FSP players in the world. But even as stand-alone, ICON had a lot of experience dealing with FSP in that type of market. I think people get a little concerned and maybe appropriately about the margin profiles in that business versus the full service part of the world and how the mix shift will impact on overall margin profiles. I think what I would say is that we've got a lot of experience running that type of organization. And obviously, with the bigger scale of the organization now is in the combined organization, it represents an important part of our contribution margin and our EBITDA indeed. And I think we're also known as an industry -- or in the industry as a company that is really very focused on cost control. And I think if you've got that discipline as an organization, if you're experienced in how to deploy the recruitment solutions that you need to develop good FSP-type relationships, you understand what the customer's expectation is of the employees from an experience perspective, you can really make that business work. I mean we've got -- we're a business that does 21% EBITDA profiles even with a higher proportional part of our business being FSP than the rest of our peer set. So I think it's a really, really strong business. And it can have -- when managed well, it can produce very, very solid EBITDAs.

Got it. And if we think about sort of the flow of that, like I'm just trying to think of like how to frame, it's roughly 20% now like -- and it's maybe towards the higher end of what we've seen recently. Like what is the typical range? Like does it sort of range in like maybe 15% to 25%, that's kind of the right range to think it is? Is it like 18% to 20%? I'm just trying to think of like if we're talking about that sort of broader shift towards it, like I appreciate what you were saying in terms of the contribution, like how big of an impact are we really talking about in terms of that shift? I'm just trying to think through...

In revenue percentage terms?

Either revenue percentage terms or how you think -- yes, I was thinking about it in those kinds of terms, but...

Yes, yes. So I mean I still...

Just like a broad...

Yes, I think it's probably -- in the short term, this is more around evolving conversations that we're having with customers. So -- and again, the way this business is awarded is that you kind of have to sign a new relationship and then we'll take a year of it into backlog at a time. So I do think that 80% of our business for next year or there or thereabouts is already decided by the time we get to the end of Q3. So for '24, specifically, I don't know if it moves it all that much. Maybe it's a couple of percentage...

It's not going to go from like 20% to 50%. That's I'm trying to like...

That certainly wouldn't be my expectation. And certainly, we're seeing a lot of demand in this space. And I think maybe if we're sitting here at this time next year, we can have that conversation again. But certainly, at this stage, we would see it moving maybe in the 20% to 25% range in -- over that period of time. But it's -- I don't think it's -- no, I don't think it's going to be 50% of...

Right. And it also is like it would depend on the mix of biotech too, right?

Of course. This is peculiar to those top really 20, 25 companies in the world.

Yes. No, that makes sense. Okay. That's very helpful. So I think one area where ICON probably gets less credit than average is sort of on your suite of tech assets. So can you talk about the tech investments you've made sort of since the PRA acquisition and how that plays into the business?

Yes, sure. We have the -- I mean, obviously, it's one of the areas where we're really very focused on and we kind of we break it out into numerous compartments. So it's not just -- it's -- certainly, we do have software solutions that we offer and certainly are very fundamental to our business, but we also think about it from the perspective of automation and robotics. And probably the third leg of the stool, if you like, is AI and how we deploy AI in our organization and how we're further deploying it into new tools that we're using in the marketplace. So I think if you look at the kind of the fundamental part of the software systems that we have, there's some really well-established systems like FIRECREST that's really informative and useful for the doctors at the actual patient sites and really helps them to get a quick understanding of where the trial is at, where this patient is at, what the actual visit -- what's going to occur at that actual visit. And that gives them really kind of again, takes the pain out of the process for them. We're integrating that now. We're one of the big pieces that came across on the PRA acquisition, which was our DCT, our decentralized platform. And that's going to allow the doctor then to have the ability to actually run the DCT trial in the same kind of format. And we have a DCT platform that's very scalable, very quick to scale. Obviously, any time you're building a DCT, it has to be bespoke. It has to be for that particular trial. So I think one of the advantage of ours is it's very, very quick to scale. It's quick to build. We can get that up at industry-best time frames. And again, it's interoperable with the FIRECREST tool, which is effectively the site enablement tool for doctors. So I think that's -- they are some of the bigger pieces that we've certainly focused on postacquisition from a technology perspective in terms of what really moves the dial. I mean the other bigger one to mention is, and this is somewhere where we have been augmenting this particular tool with AI is, OneSearch. OneSearch is our data aggregator that allows us to design clinical research. And it's obviously then we're now layered an AI level on it. So it's also looking at a much broader suite of information, being able to interrogate a much broader suite of information than it was in the past. And what it allows us to do, again, is it produces, from an AI perspective, a first design of the clinical trial perspective. Then, of course, we have people who come in and review that in the same way we do often work with AI. It's not just AI in isolation, it's AI and people overlaying their own experience. But that's giving us the ability to actually go to customers saying, although this might sound -- maybe won't sound counterintuitive, but certainly, when we look at the vast majority of the data we're seeing, it's telling us that these are those particular sites that are going to be faster for recruiting your patients for your particular indication on your trial. And that can have a meaningful impact in terms of speed to patient recruitment, which is the big metric in our industry. So both of those systems have both FIRECREST and OneSearch won an awful lot of awards in terms of leading platforms in the space. And then to maybe get to more of the -- one more system that we just recently launched from the AI perspective, which is Cassandra. And that's kind of a system, again, that looks at your trial, and it will give you a 90% to 95% accuracy in terms of telling you whether you'll be mandated by the FDA or the EMA to do a postapprovals trial. And of course, if you're collecting the information, if you know that when you're doing your Phase II or III trials, you can collect more information and have much more of the work done, so that it's not as hard a task when you get to that point. So again, it's just another really good example where we kind of feel like we're leading the field from an AI, and we have a center of excellence around AI. And as I said, the Cassandra award has been -- or has been awarded numerous awards in the industry for the use of AI in the space. So that's something we're very excited about. It doesn't stop there. It goes into the organization, though, when we look at, as you guys know, a lot of what we do around automation, robotics is very important to our cost base and driving the efficiency of our cost base, and that's somewhere where we're going to continue to look at that. So I see this as being really core critical to our strategy as we move forward. Our customers and whether it be mix of FSP or just customer expectation around cost delivery and efficiency, we're going to have to be more efficient. We're going to have to be more automated in how we do our services. And so this push towards the use of technology, the use of automation and what we do on a day-to-day basis and the use of smart AI is going to be a big strategic part of how we move forward.

And how -- in terms of that, like how much of that is sort of deployed and sort of in use with trials now versus some of this, it still sounds like it's more in expansion mode, right, like that. I just sort of -- we think about the opportunity now versus a couple of years?

Cassandra is a live product. You can buy that today and deploy it on your studies if you want. Likewise OneSearch is something that we can deploy today if people -- when we're working with our customers, and it's broad-based. It's not one particular customer set. They would often use OneSearch to be able to get better visibility. I would say that we're still -- we're doing -- some of the work that we're doing around FIRECREST and the DCT platform and merging that is still ongoing. So that's something that we're working on at the moment. But as I think of the future and you think about how things like digital twins or biosimulation might have an impact upon early clinical development, be it Phase I or Phase II, about how much of that could be done in silico and get us to decision points faster. They are the areas that we're really excited about how do we develop and how -- as a clinical research organization, are we on the front foot of the development of that. I think one of the things that we're very conscious of is that we want to be the people who are really expert in this. We don't want to leave it to our customers to try to figure their way out and do it on a piecemeal by piecemeal basis. If we can bring a kind of a global solution to how this is -- how you can deploy this kind of technology. You can apply it to a biotech. It doesn't have the same resources as a large pharma, and they can use that advantage that we can bring them. So it's very exciting from our perspective.

Yes. No, that makes sense. Have you been able to sort of quantify the impact on patient recruitment time lines? Like is that something you said like over the -- as we think about it since, I don't know, PRA, we've improved it. I don't know, whatever percent.

Yes. No, in terms of our OneSearch tool, we have seen where it can actually be 20% more efficient than the normal process around patient recruitment. That's certainly something in a metric that we continue to track. And we're looking to always improve upon that as well. Now it's not -- all of this is pretty therapeutics dependent. So obviously, it's going to be different in different therapies, but that's a bit of a broad brush measure, but that's certainly the kind of quantum we've been seeing.

I mean the other piece we track, too, is reducing non-enrolling sites, for instance, as they are still a big problem in the industry. And it's not just around patient recruitment, but it's around site start-up and making sure we're selecting the right sites right out of the gate, and that's obviously improving overall time lines.

Yes. No, that makes sense. And then in terms of like do clients are sort of specifically seeking out these sort of like solutions or they're just kind of like, please run my trial, I'm not sure how you get there, but I want to get there. Like how does -- maybe the answer is both. You get sort of all of that. But like how specific are they in their, I guess, in their request versus you're just like oh, it's part of our wraparound like that we...

For -- in most instances, the client -- they have to be introduced to the solution. Yes. So -- and that's fair unless they have that experience or they've used it before, they're not going to have that experience. So -- but it's something that we kind of go with the full suite of what we can offer in the first place on it in any environment. And then they -- it's a bit of a smorgasbord. They can kind of pick and choose as they feel it's going to really benefit them in terms of their clinical research process.

They also want to know what's deployable today, right? I mean they're less interested, I guess, and obviously, we have a number of pilots that we run across the organization in AI and in tools generally, but they want to understand where we've had experience in bringing those tools to them. And obviously, the more the better in terms of what we've actually done in being able to show that to them.

Right, so that they're not the guinea pig on whatever that deployment is. Okay. No, that certainly makes sense. Okay. So maybe if we sort of go from sort of that -- sort of broader level down to sort of 2024. If we sort of think about 2024 conceptually, how do we think about the puts and takes on the gross margin line? I think you have mentioned FSP as a mix. I think you've talked about historically that you sort of see that cap being sort of 29.5%, 30%, something like that. But if we think about next year, like are there -- what other points are we -- what am I missing on that?

No, I think it's not a bad summary in front of us. We think about -- obviously, in the third quarter there, we were at 29.8%, which was -- we're very, very happy with. We talked 1.5 years ago about when we're kind of setting some of our medium-term targets to be in the range of 29% to 30% from a gross margin perspective. I think as we go into '24, yes, I think FSP mix is something that we need to consider. And maybe that has a little bit of a headwind at gross margin, still very much comfortable in that 29% to 30% range. So it's not moving back that much, if at all, then of course, we will continue to do the hard work to try to be as efficient as possible and kind of keep it towards the top end of that range. But yes, there could be with business mix shift a little bit there, a little more pressure on gross margin. I think the corollary to that maybe or the components of that, that you need to consider is then we've also -- at the same time, we talked about 29% to 30%, we talked about 8% to 9% SG&A.

Yes, that was going to be my next question.

There you go. And that's somewhere where we've been really successful where we're at 8.8% in Q3. And we've said 8%. That's kind of where we want to get in the medium term. So we do feel that we want to see about 30 to 50 bps of margin expansion next year as we go into 2024, for the full year '23 versus '24 EBITDA. And that's going to be a combination of both of those things for the leverage on SG&A, maybe a little bit of a headwind on gross margin. But all in all, we feel comfortable that we can still perform that.

Got it. And then in terms of that and maybe this is sort of a part for both of those components. Like the pricing roll through that was sort of caught all the CROs sort of as inflation took off and was a little bit of a headwind that should be -- you should be getting to the point if inflation is coming down, where you're getting -- and you're getting better pricing presumably on like a bigger percentage of your contracts like that should almost start to become, at some point, a little bit of a tailwind or not. That's not the right way to think about that?

I don't know if I think about like given that it's a keenly priced industry and you do have large pharma really kind of trying to belt that at the lower end. I'm not sure I'm putting too much show. I'm putting too much weight into that. I think what the nice part of the whole inflationary environment is of course, is on our salary and our cost base and the salary increase costs. We've seen really, really good levels of turnover in terms of it dropping substantially from where we were a couple of years ago. We're now below where we were at historically from a turnover perspective in our headcount. So it's a really stable headcount base. And I think the inflationary environment on that side of things will help us a little bit more this year. So I think when I think about it from a pricing and cost perspective, I think the cost perspective is more of the benefit than the pricing at this point.

Got it. Okay. And then maybe turning to what you were just saying about sort of the 30 to 50 basis points of annual margin expansion mostly on the SG&A side. How do we think about the opportunities within the global business services model that you said that maybe that also plays into some of those wage things that you just sort of mentioned?

Yes, and there's a -- I suppose that probably is a mix of headcount and where we locate folks globally. I think we've done a really good job in terms of the ICON model has always been a standardized process on large systems with cohorts of folks in centralized offices and oftentimes they're in lower cost locations like Mexico and India. And that's the way we've actually rolled it out across now the combined organization as well. So we're I would say 75% of the way to kind of achieving where we wanted to be from day 1 in terms of that. So we have the systems in place, we have the standardization in place. And we have a lot of the teams actually shifted to those locations. So we've seen a huge benefit from that in terms of that cost base, in terms of that SG&A structure. And that continues to be a big part of how we play out to that 8% target we're talking about as we go forward. I think the other part that maybe we haven't explored as much yet, is about the continuation of automation and the continuation of using efficient process and how we really -- and this is not as much of about -- people often say, is this about crunching the headcount, it's not necessarily with that. But it does allow us to maintain headcount and continue to grow the top line and kind of break that connection between necessarily having to increase headcount with revenue growth. So it's both in our gross margin and in our SG&A and it's about how much automation can you bring to bear on the lives of people who are doing stuff that really they don't find exciting, interesting, how many kind of rote tasks can you take out of the lives of people. It doesn't matter whether they're a frontline CRA or whether they're someone who's working in AP, making payments to our suppliers. It really is about how much of that can we be focused on. And that's where we have that center of excellence in our organization around automation, around robotics and robotic process automation. And that's -- as I said, we keep that separate from AI. We keep it separate from the frontline systems because we realize that there is real hard work to be done in automating an awful lot of what we do. But that really, again, helps us drive our ability to be efficient, to be margin focused as an organization, but also to make sure that we're being extremely cost efficient and back to our customers again and giving that benefit back to them also.

That makes sense. And I think in your prior statement, you were just talking you're sort of 75% of the way there on the global business services model. Is that kind of like you think you'll be all the way there, or maybe you're never all the way there, right? Maybe that's the wrong way of thinking about it, but the next sort of closing that rest of the 25%, is that gap, is that kind of like a through 2025 kind of event, and that's the way we should think about it? Or is it just think about it as a longer tail than that?

I mean, I think certainly, it's medium term. So it's maybe over the next couple of years, whether that's '24, '25 and '26, but I wouldn't necessarily all has be done by the end of '25. But certainly, the way I view it at the moment is that certainly we want to be moving through that over the next couple of years.

Okay. And have you ever quantified sort of the impact of some of these like robotics and automation things that you've talked about?

I mean not in dollar terms. We talk and think about it in terms of numbers of hours that are done through automation in the organization. And we've talked about having certain internal targets. We're looking to -- we kind of started off with saying something like could we get to a target of 1 million hours of automation being done by process automation or robotics and then our kind of our next target was to double that. So these are significant numbers in terms of, I suppose when you think about that on an annual basis and the amount of work that people would have been doing to drive that same level of delivery, they are significant numbers. And that's why, I mean, it's a well-established process in the organization. We started this about -- myself and Tom O'Leary, the Chief Information Officer, started this process of OPI and really looking into it about 6 or 7 years ago. And it's been something that we've continued to invest in, continue to develop our own internal skill sets around, because I think we know that as the further we go, the bigger we get as an organization, the more important it's going to become that we're extremely efficient.

Makes a ton of sense. Okay. Maybe moving away from the margin line. I'm talking about some of the nonoperating items. Can you walk us through the $100 million of interest expense improvement and potential timing there?

I think on interest, particularly, we have -- that's kind of our default position. And what do I mean by that? I mean that if we follow the same course we've been doing over the last 2.5 years, in terms of our levels of debt paydown on our term loan B, particularly, we should be able to deliver on the $100 million. So it's much around, yes, we put -- we are now 60% fixed versus floating interest rates with hedges put in place. But it also means that if we continue that aggressive level of paydown, we should be able to decrease our total interest bill for next year by circa $100 million. I suppose the opportunity that isn't spoken to and that is obviously, we've just been made investment grade by S&P. We're hoping that Moody's will follow suit before the end of the year. That gives us the opportunity then to consider kind of refi-ing that entire term loan B, moving to an investment-grade bond-type structure where we'll be circa 2x debt to EBITDA by the end of the year. So we're kind of solidly in that rating territory. And then, of course, if we do that, a, we can still get the $100 million of savings or should be able to get $100 million of savings depending on interest rates in the first quarter, of course. But we can also then, obviously, have a situation where from that point onwards, we don't need to make that quantum of debt paydown on a quarterly basis because obviously, you've got fixed rates and you've got fixed bonds and they're going to be coming up over a period of time, and we're happy with that level of debt in the organization. That gives us a lot more room for maneuver on M&A and from even a stock buyback perspective. So I mean that's where we'd like to be. Our default position at the moment is, assuming nothing happens, we can still get $100 million down, but we'd like to be able to be in a position to refi and have that cash and availability to use it to continue to build out the organization.

Okay. And both -- you need both ratings agencies to put you into invest grade.

Yes, absolutely.

That makes sense. And then how do you sort of think about that sort of fixed versus variable rate exposure? Well, I guess 2 questions. One, the $100 million assumes current interest rates going forward?

Yes, -- yes, the current curve of expectation.

The current curve of expectation. Okay. And then two, how do you think about the mix of sort of variable versus fixed rate debt at this point?

We're comfortable with that actually. There would have been a time, I suppose, and if you look at our peers, they're probably more like 80% fixed than we are. We're about 60% fixed, which I think right up until maybe in the last couple of months, people would have said, maybe higher is better. As people think now we're topping out on the interest curve, people are actually saying, "Oh, hang on, now is not the time to fix more." So funny enough, I was in New York just 2 days ago talking with some of our banking colleagues about what the mindset in the marketplace is around that. And they were saying, actually, 60-40 is a pretty good place to be right now because you can't take the opportunity as those interest rates starting to fall. That's the big question, though, right? That's the one everyone's scratching their head about. Are the Fed going to move in March, April? Or is it going to be higher for longer, like they've been saying for up to this point? So that's the piece, I think, that we need to scratch our heads on a bit and think about it. But right now, I'm pretty comfortable with the 60-40, particularly in the context of being able to possibly actually flip the entire term loan B and move to a more fixed environment as we go forward.

And we've been able to obviously aggressively pay -- I mean, the structure of the term loan B, obviously, we've been aggressively paying that down and we still have the opportunity to do that on the floating piece of that -- of the term loan B.

No, that makes sense. And how do we think -- okay, so maybe that's obviously improved -- an improvement in terms of cash outlays, if you're not paying this interest expense and et cetera. How do you think about sort of on the cap deployment front, like what's most attractive to you in this environment, considering sort of valuations and the billable assets?

I mean, certainly, the preference is for additional M&A. I mean it's sort of a return to the pre-PRA acquisition strategy, which is more of a string of pearls sort of bolt-on in nature. And it's really focused around sort of the ancillary areas of the portfolio and our service offering where we're not quite at the scale that we'd like to be. So it's areas like laboratory services. And even that late phase area we were talking about earlier, Phase IV in general. Site and patient solutions, technology. I mean there's a number of areas where we feel like we have -- it would be a nice opportunity to scale that in a faster manner than organically growing those businesses as we have been doing. So the challenge, of course, is that valuations in the space are still quite high.

Yes, that's what I was going to ask.

We have been trying a lot of private equity interest. That's always the push and the pull in terms of the attractiveness of opportunities from a strategic perspective and how we weigh that against some of the financial valuation metrics. So certainly, we'd like to see that step up in '24, I think that's our expectation. And then alongside share buyback and certainly could be more aggressive on the buyback front if we obviously don't deploy as much in M&A.

Okay. And you're talking about a step up in M&A, you're talking about year-over-year basis, that's kind of what you are thinking about.

Obviously, we did it very small.

Yes, that's okay. I just didn't know if that was like a pre-PRA commentary step-up for that recently. Okay. That certainly makes sense. If we think about also maybe on the sort of cash flow front, how are you -- how is the collections process going? Like how are we -- how is that process been unrolling as we get through the end of the year?

Yes. It's -- it remains hard work. It remains a very focused attention to each of your debtors and making sure that you're having open and active communication with them around the timing of their invoices, the timing of our billing, all of those pieces. I think we've done a good job on getting back to a better position to this point. We're at 49 days of DSO at the end of Q3. Good for me, given our credit terms across our book of business, is a bit -- is in the mid-40s. So that's -- so optimal is mid-40s. So we're not 1 million miles off what optimal looks like. In order to get there, I think we're very focused on doing exactly what I said, which is following up with customers, making sure they're happy with their invoices, making sure there's no specific issues or reasons why they can't make payments on that invoice. And as we look at the pressure on where it's coming from, it really is the large pharmas that are creating the pressure in the DSO. It's not -- we don't have really significant DSO exposure in our biotech or midsized customers. And you've got large pharmas who have credit terms of circa 90 days on average. So it is -- that's -- we -- obviously we're exposed to that as part of our -- obviously, our mix shift of FSP is obviously, that's much more large pharma as well, as you can imagine. So that's been the continuing trend during the course of the year. And so it's all about that. I think we have the ability to maybe get down those 2 or 3 days in the last quarter of the year. That's certainly what we're focusing on and a lot of hard work going into making that happen. But I don't feel at this point like there's anything that we're doing that's structurally incorrect or we're not getting after billing points or any of it. All of that is kind of done and dusted. It's really just trying to get cash out of folks at this point in time with the interest rates today where they are at the moment. It's pretty tough. So it takes a lot of the work. But we're very focused on, as I said, getting that 2 or 3 days down. And I wouldn't say I'll be declaring victory at that point, but certainly, I mean mid-40s is in and around the right level for where we should be.

Yes. No, that makes sense. In terms of the free cash flow improvement, any other sort of key areas besides that, that are sort of something you guys are thinking about?

From a cash flow perspective, our balance sheet is pretty simple. It's not unlike our P&L account to that extent, if you can keep your DSOs stable, there's a bit more front-loading in the cost base with taxes and insurance costs in the first half of the year and some data cost licenses that we have that usually get renewed in the first half of the year. So it's a little heavier on cash usage. So cash usage in the second half is always a bit better. I think we saw that particularly even with the improvement in DSO in Q3. We still saw a very significant or fairly significant part of our cost base being just our total amounts of payments we need to make in Q3 versus Q2. So we expect that to be the case in Q4 as well. But it's not -- I mean the major lever to pull is still DSO.

Got that. And are you subject to any of the Pillar 2 impact?

Of course. Yes. Yes, like most of our non-U.S. companies or companies that have presences outside the U.S., yes, absolutely. The global minimum tax rate is not a new thing. That 15% global tax rate is coming in from '24 and we will be certainly subject to that. We're about 15.5% of effective tax rate at the moment. We were about 16.5% last year. In the kind of medium term, I've been saying that, well, we can expect -- I suppose if you have a minimum of 15%, I notice most people have been saying that year-over-year impact from '23 to '24 could be quite significant, like 1% or 2%. I think it's probably for us, maybe it's in that percentage range between 15.5% and 16.5%. And again, in terms of giving you maybe kind of a little more color on that, I would say at this point, from what I'm seeing at the moment from my team internally is probably at the higher end of that range. So I would say that 16.5% is probably the way to think about that. The good news, if that's the way you want to color it, is that, that should stabilize there afterwards. There is no -- at least at this point, there's no big other OECD implementations that we can think of that mean that we should move significantly out of that 16% to 16.5% range as we go forward into '25, '26 and onwards. So it feels like we're kind of back to a more post-transaction and all of the other pieces and much more even stable tax rate as we go forward.

Got it. Okay. That makes sense. And then I think just the last side on the capital deployment, any changes in sort of like the ongoing business needs as you think about some of these like initiatives that you've been talking about, like should we think about sort of like CapEx and things like that as being fairly consistent going forward?

Yes. I mean it's interesting. Most of our CapEx is not anything other than people's time there. It's a lot of what we spend on CapEx and we spend. I mean we will probably spend -- we spent about $150 million last year. It will be in that, maybe a little less this year, but certainly not ballpark. We'd like to -- I mean, I think as we go into every year, we want to spend a bit more. So we'd like to see a bit more than that next year. But it really comes down to the internal costs of people's time because software as we all know is a service now, so you pay for it on your P&L account. So CapEx really has shifted over time from being the acquisition of licenses and hardware to really people's time. So we feel like we are doing a good job. As I said, we have a lot of internal folks who are focused on automation, and on AI and system development that are going to move it out of the company. And we're able to able to service that at that kind of $150-ish million on an annual basis. So not expecting any dramatic changes at the moment on that one.

Got it. And as we have just a little bit more than a minute left. What do you think at this point is that is most misunderstood about ICON by investors?

I think things are certainly -- I mean, there was -- the level of -- it's been an interesting year. This has been, I think, a better year at actually telling the story, I think for people getting their heads around the risk profile of the CROs of getting past the point of being overly concerned about integration with PRA. We've obviously seen the margin profile come through really strongly as well. I think if I was here even earlier in this year, I would have said that I think the most missed part is how successful we've been at margin development in the context of bringing 2 20,000-people service organizations together. That has been a phenomenal story and have been hard fought, and right across our organization, everybody is -- takes a part of the share of the glory of that one. In fairness, I think the market has a better understanding and a better appreciation for that now. And I think that's something that has been pleasing to see over the last while. I think it's -- there are still a lot of nervousness out there. There's still a lot of other pieces at play. How the biotech is going to impact, how large pharma is going to impact. But I do feel pretty good about where we are in the marketplace at the moment. I think we've got a very competitive offering. I think our mix, as I said, at the beginning of FSP and full service, they will be able to blend that is probably unmatched at our scale in the industry at the moment. So I think if there was anything that I would say that is -- that maybe that's a -- which is we feel like we're in a really good place in a solid marketplace where we can take advantage given our unique business mix at the moment.

Sounds great. Thank you so much, Brendan. Thank you, Kate.

Thanks all.