GE HealthCare Technologies Inc. (GEHC) Earnings Call Transcript & Summary

Thank you, everyone. It's Patrick on the U.S. MedTech team. Disclaimers, none of you are going to read them, none of you care, but they're here, so I'll put them back there. And much more importantly, thank you so much to both Pete and Jay as CEO and CFO of GE HealthCare for coming and agreeing to do this.

Thanks for having us.

Yes, and meeting everybody. It should be a great discussion. And I think it'd be a lot of fun. It's definitely a subsector that is near and dear to my heart, and so it's always fun to chat. I think with that, why don't we kind of open with, Pete. It's been around 2 years GE HealthCare being listed as a public company. What would you say that has been some of the key takeaways and surprises for you, and then also for you, Jay, over that time period?

Yes. I mean, look, it's hard to believe. It will be 2 years January, first week of January that we spun out of GE after obviously being part of the company for over 100 years and integrating all these different global markets. But I was a GE guy, Patrick, you may not know 15 years. I left for 17 and came back. And so the challenge with that is there're certain things you think you know and there's certain stuff that you still know, but you got to make sure you understand which is which. But the interesting dynamic, and I'll just start broader for the whole market is just how much more surprise-wise imaging, in particular, is becoming indispensable for all the great devices of other companies are implanting and driving. And then even on the drug side, that many drugs because of how expensive they are, the right diagnosis and precision decision point becomes super critical. And so that's a huge underlying driver of our growth. In this world, in the past, where atop cost to driving health care was too many MRIs. Now doing more MRIs could actually reduce the cost of a drug or a device because you're making the right diagnosis. So that's one category. And I'd say the other big category is how fast AI inside devices in radiology has grown. I mean we have over 70 approved FDA devices that are integrated in. There's a lot more on the [ come ], other folks in our space, same kind of thing. And I think it's exciting because it has probably one of the biggest potentials to change kind of the care delivery. So I don't know, Jay, your thoughts?

Yes. I think some of the innovation opportunities are incredibly exciting. Our pharmaceutical diagnostic business, some of the AI advancements that we're pursuing, all of those have been sort of delightful surprises for me entering the company a little over a year ago. But I would also say the execution focus of the team. If you think about this idea of having a lean culture in place, there are so many different elements that contribute to that. I've been really impressed with our ability to execute on that. I knew there would be normal margin opportunities related to a new spin-off, but some of the focus in terms of transforming how we go to market, how we operate our manufacturing facilities and the team's execution ability with respect to that has been really impressive.

And I'd just say the last piece, I'd say, is that, as a company coming out, the opportunity to continue to find ways to accelerate growth with focus on different markets, but at the same time, be able to improve margin is a rather rare scenario, but it's one of the advantages, I think, of what we have from the cost and the -- both fixed as well as gross margin improvement at the same time being able to continue to grow with exciting areas. I mean the area Jay just mentioned, radiopharmaceuticals is one we're quite bullish on, and it has a lot of interesting disease state expansion areas as well.

I definitely want to hit on that and margins. But I might start with a slightly nerdy imaging question because, yes, why not. That installed base globally, not just for you, but the industry as a whole has been shifting over the last, call it, 10, 20 years, x-ray arguably a bit less, CT, PET/CT more, MR, like there's been a lot of changes. How are you thinking about the different modalities, their growth rates, and what it might mean for your service book longer term?

Yes. No, it's a good question. I think one of the underappreciated kind of assets is our service business, the brake-fix business. If you talk to our customers and where we have high net promoter scores, they love their service team. Because, in many cases, it's running the factory within the hospital system. If you're imaging equipment, if your capabilities aren't running, you can't run your system. And so how well you do that and the intimacy that's created is a big deal. Look, I mean, in general, the more sophisticated the equipment, the more the contract attachment rate goes up. And I think that applies to us as well as other folks. So a sophisticated MRI, we may have a 75%, 80% attachment rate or CT scanner, an 8- or 10-year old product, there's many other people that might be able to work on it. And so this trend of actually faster growth in MRIs, obviously, PET/CT, advanced multi-head NucMed cameras, all of those things are evolving actually bode well for service attachment. Then you layer on AI inside, and you also then layer on potentially other services with that. We're -- we definitely see that the opportunity to continue to advance what service is from just break-fix to a much larger play is there.

Is it a fair assumption to make that the attach rate of service is higher in the U.S. for kind of everybody than in, say, EMEA? There's more third-party providers in EMEA than in the U.S. on the service side. Is it fair assumption?

I think it's fair to say that in kind of the mid- to lower tier. But when you get into the higher-tier items, pretty much around the world, you see a better attachment rate to those higher-end sophisticated products. I think what will be interesting to play out, as you do have more advanced software components in the products, just how many people want to take the risk, so to speak, to go to a different service play, at least within the first 3 or 4 years.

How are you seeing within the modalities like MRI, are you seeing any change in mix, like number of 3 Tesla systems versus 1.5 or anything like that over time? And have you seen any shift within the modalities of the spec [indiscernible]?

Yes. I think it's an interesting play. I would say, in the MR world, you clearly see more of the rise of 3T being a multipurpose system. It hits an interesting sweet point, although, at the same time, a 1.5 system or others with embedded artificial intelligence you could argue you can have more capabilities with a little bit better [ siding ] costs. I think one of the factors that is playing out in certain developed markets is, you can buy our product or anybody else's product at a given price and the installation cost of doing your build-out could be more expensive than the actual product. And so thinking about upgrades or certain products that give you high-quality output, but at the same time, may actually reduce your overall costs are also playing into it.

I mean maybe if we can like flip to AI. It's one of those like buzzwords that for a lot of subsectors gets thrown around a lot, but actually, I don't know, RSNA in like 2015, everyone -- it's real, if you know what I mean, the radiology, but it also means a lot of different things from automated image diagnostics to image recomposition to like scan speed times. Could you break down for like how you think about how AI interfaces for you at GE HealthCare, and what it means for the business going forward into these buckets?

So we think about it in 3 large buckets. I mean it's kind of AI inside, AI at the departmental, AI at the enterprise level. And the majority, to your point, Patrick, today is about AI inside the products. And if you break that down even further, you have a component of automated capabilities that a human used to do that the machine does faster. So you're imaging a mother. You have to get the baby's circumference of the head, other metrics. You used to have to draw that all out. Now with the right AI models, in seconds, it has a much more accurate done. So that exam just got compressed by 10, 15 minutes. So that's one example of productivity and workflow. There's other sides here about actually how the device works. So like in MRI, forever, certain amount of positive charges of water in your body, right, creates an image. Much of that signal is actually doesn't create an image. An AI algorithm can predict which vectors are going to be used to recon and not, and you can reduce that out, you can actually have a better image and have a faster reconstruction. That's changing how the device itself works. That's in place. And the third is actually in diagnosis or directive diagnosis where you take a look, and give you an example in like mobile X-ray, this was a big deal in COVID, is having a nurse not a tech go into a room where someone may have -- they are coughing, do they have pneumonia, getting the X-ray done and actually having -- give you a -- basically a green, there's no COVID pneumonia, or a red, they have it, or in certain cases, a yellow, a certain follow-up. And so the diagnostic is part of it. And I think you're going to continue to see that part grow. And then this next area, which is departmental solutions are when you start having multiple devices work together, data flow between devices, data flow between different areas. And then I think the third, when you get into enterprise, there's a combination of how the hospital is running an AI to be able to give real-time prognostic help on what to do. And the other side of that one then is multimodal diagnostics. So we all know if you go through an oncology case, there might be 5 different imaging modalities. There might be biopsies. There might be omics, genomics and such, and 6 or 7 different people look at those and it comes together in a report. It's the perfect opportunity for machine learning or GenAI down the road to what did you find on all of those? And what does that determine on what's the next step? So that's the broader sequence of how we're looking at things. But clearly, the next 2 to 3 years, the big value is this optimization of inside the devices.

How the conversations with the customers around this are going? How do we not get to a place to take an analogy and thinking back to the slice wars on CT and that side of things, where the capability gets built into the system, but the innovators don't really get paid for it. How do we prevent that happening and making sure that the customer is really going to be aligned their interests or aligned with you guys on getting the value out of that software?

Yes. I think the difference is -- in this case is the more -- at least our perspective, is the more that you start focusing on disease states or areas so that from end to end, just a faster exam doesn't do anything unless you actually have more throughput. And I think one of the things that we've been really focused on is how do you actually take a look at how the money flows? So can the customer make a reasonable return on this? How does the outcome change for the patient? And how does the productivity for the users, how does that change? If you can click all 3 of those, you can create more value, which typically translates more value for us. If you can't, you run into certain challenges. And I think one of our industry issues in the past has been just more technology for technology's sake. Doesn't necessarily translate for a better outcome, and it doesn't translate to more margin or value in the industry. And I do think we're in an interesting window, Patrick, where both of those things can happen. The other side is, most customers now are moving to the cloud for their own capabilities. And we, as companies, are really spending a lot more time on SaaS capabilities. And if you really want to implement these AI models, they're not going to all be resident on a machine. They're going to be too expensive. They're going to be in the cloud, right? And you're going to leverage that capability. That will then bring this opportunity for customers to have more of an OpEx model to be involved, and I think that's going to create some stickiness, back to the services question, new services lines that are just not capital in nature.

Maybe to pivot a bit and, Jay, have a topical one for you. China obviously is coming up a lot in discussion across all subsectors. What -- from you guys like what's the latest that you're hearing on the ground about how the market is developing, thoughts, how that moves forward? And it's a changing topic, but how are you feeling about the stimulus? And if that does or doesn't have an effect and when?

Yes. China has definitely been a source of much discussion for us. We don't really have a substantive update versus what we shared at our earnings call. During the earnings call, we lowered our guidance for the full year to reflect a fairly dramatic slowdown in China that we're seeing. And a lot of that comes down to reluctance of hospitals to buy in advance of clarity around the stimulus package. We are -- our view on China, we've been manufacturing there for 30 years. We sell so much -- we sell 70% of what we sell in China in our 2 big businesses is made in China for China. So it's a market we're committed to, and we have a very optimistic long-term view about the market. Just look at the trajectory over the last 10 years and kind of the desire and need to grow health care spending to support the broad population that they have. So we feel very good about it. But it's been one of these things where there's been a lot of volatility. And to reduce the volatility, we just took out the number as it relates to any kind of normalization of the China market in our guidance this year. We've seen some activity in terms of things entering a pre-tender bucket, so some discussions happening amongst provinces, hospitals and so on. Our teams on the ground are very close to that. So we're seeing some of that perk up. But as far as actual tenders as a result of stimulus, we haven't seen it yet. And our expectation is we'll see some at some point this year, but from a sales standpoint, we're not really counting on that.

Can only be depressed for so long. I mean, to keep the competitive environment there, APAC as a region as a whole, Japan has some unique competitors, but China also has some unique competitors that are a little more isolated to China. I would argue they haven't had a lot of success outside. But how are you seeing the competitive environment in China overall and some of the domestic competitors there?

Sure. Look, there are formidable local competitors in China. The 2 most prominent ones, United and Mindray, have made significant investments in R&D and gained ground in China and outside of China over the last several years. What I would say is our focus is on continuing to differentiate through innovation, continuing to differentiate through close customer relationships and really with a strong service offering. With those 3 ingredients in place, we've been able to protect our share in China and other markets around the world against everybody and have had some success against those marks, but we're not underestimating those competitors for a second. They're quite good.

And I think to the point, just Jay hit on this, is that the more -- anybody at any given time can maybe have similar features in a given product and maybe a lower price or a lower cost at any given time. The more that those work across multiple products, the data flows synchronously across that way. And there's clear outcomes for how you solve a given disease state area or productivity area, which is, again, the direction we're moving. The more it gives you a more sticky relationship with that customer, whether it be here or China or any other market. And so that's a critical part of when you think about why GE HealthCare, which makes us different. And we didn't hit radiopharma yet, but this idea of a radiopharmaceutical, which we have expertise on how to make drugs, a PET/CT system, which we have deep engineering expertise how to make there, and we also have integrated data systems on how that data will flow for AI, the more we connect those, that's how we build more of a capability that is differentiated in the marketplace.

The -- I mean it's a -- I guess it's a good time to pivot to the radiopharma. And the PDx business overall had a little bit more time since the proposal from CMS. Is there any sense -- maybe worth initially level setting the audience for those who might be in the weeds, what's going on and what the implication could be for you guys?

Yes. Maybe I'll start and then, Jay, feel free to jump in. So for many, many years in the nuclear med space, primarily driven by in the PET world in the United States for CMS, so our 65 and older population, the pharmaceutical diagnostic was treated like a supply. So it was reimbursed at maybe a couple of hundred dollars. So even if the customer had to pay $1,000 or $2,000 for the agent, they only got $200 roughly reimbursement. And again, this is just kind of example ballpark-wise. So then the growth of using -- doing a study with a new agent, it had to be really compelling clinically because the customer was going to maybe breakeven or lose money. What has just recently changed is in the preliminary outpatient rule, which will be solidified in January, maybe some tweaks to it, would be that those would not be reimbursed as a supply anymore, but actually as a drug. And so a product that could be $1,500 or $2,000 would have a reasonable reimbursement close to what the customer pays for it. So as you can imagine, that changes the economic profile for a customer to use PET and radiopharmaceutical more routinely. Why is that important? Well, maybe the last 10 years, there hasn't been that many new drugs. FDG has been around forever. There hasn't been a lot of other things in the NucMed side. There's now a rise of new molecules. So Flurpiridaz, which is a product to use for cardiac perfusion, the Vizamyl product tied to amyloid beta plaque, our DaTscan product, which helps determine Parkinson's and actually being used in [indiscernible], Cerianna for breast cancer, so there's quite an interesting portfolio. It helps really change how someone will look how they'll use the product. And I think, Patrick, one of the things that I would lay out there, say, for the Alzheimer's products that are out there, the therapeutic ones, part of the diagnosis of amyloid beta plaque, you could do some type of a CSF stick, a lumbar, and see if someone actually has a quantification of it, but you're still not going to have the precision you would have in PET. But if you're losing money on one and you breakeven or make money on this one, you may decide to go for the lower one. Now this actually gives patients a better tool to give more precision on their diagnosis. So we think this ultimately is a big deal over the long run. It's not something that changes trajectories overnight because you need equipment and pharmaceuticals together. But this is very promising, and we think this is a space that there's going to be a lot of interesting breakthroughs on the therapy side, which will drive the diagnostic component, which we're a critical player in.

I think the Street has something like mid-singles in the PDx over the long term. It sounds like there's a possibility you end up running ahead of that given the innovation pipeline. Is that a fair statement?

Yes. I mean, we would assume that PDx, into the future, with -- if these rules play out the way we discuss, will be one of the key drivers for us.

Being a key drivers, maybe margins as well, I think, is kind of another component of the GE HealthCare story. I guess, firstly, big picture, maybe, Jay, what do you think is -- not to front run your Investor Day, what would you say is the key, most critical factor that's enabling you guys to hit the margin targets?

Yes. I would say that, first of all, to your point, we're very pleased with the margin progress since the spin and certainly in the first half of this year. If you look at, it in the first half of the year, and as you know, sales growth is extremely beneficial to margin. But in the face of essentially flat sales, we expanded our gross margin 110 basis points. We actually declined SG&A. And so we grew EBIT 50 or 60 basis points in the face of R&D that was growing in the teens. So really nice margin equation for the first half of the year. And then as we looked at the balance of the year, we were able to absorb essentially a $500 million reduction in sales and deliver the EPS that we originally hoped to deliver during the year. That's our current expectation. So overall, really good progress on the margin. And I would say a few things are enabling it. We have this intense focus on price, and that has clearly been an element that we've benefited from. This idea of a lean culture, a lean culture has a lot of things to it, focus on safety, quality, delivery, but also cost. And so that has played out in what we call our variable cost productivity initiatives. We had performance where those initiatives have outpaced inflation. So we were net to the good on our cost programs allowing price to just kind of flow through. And then, in addition to that, we're intensely focused on G&A leverage. And a lot of that comes as we come off the transition services. During the call, we pointed out a couple of things that our IT team was able to do this year, reducing IT spend in this year by $60 million. One was moving some of our servers to the cloud, and the second related to consolidating application support. Those 2 things were enabled by coming off the TSAs, but those are examples of things that we'll continue to do into next year in IT and other functions like finance that will allow us to draw down and improve G&A leverage. So as I think about the critical ingredients, it's all of those things, plus the impact of innovation. Because remember, when we launch new products, typically, they come at a higher price and a lower cost. So you get a really nice margin benefit. So you put those ingredients in place, and collectively, they've given us a high level of confidence in our ability to drive margin along the lines of the medium-term targets that we've shared.

And I would just add, I think Jay covered it really well. The R&D investments have a double benefit if done the right way, which is value that you can capture more in the market, but also we're rationalizing platforms. And I've talked about a lot of this in the past in other forums. I think some of our competition has done a better job than us over the years. But we're really now getting these platform leverage. And so if you offer 8 things to the market, and behind the scenes, it's really 3 components, to the suppliers, you can over double the amount of business you give them, which gives you lower input costs. You can do different things on concentration of how you manufacture. And we're well down the road on that. That's one of the ways to drive our imaging margins over time. And the other part that will come is more SaaS offerings. So we've invested pretty heavily to get a SaaS backbone in place at GE HealthCare that many of these software opportunities that we have, that we have as a capital transaction. For a customer, instead of having to step up to pay that money upfront, having that actually as a monthly stream, creating stickiness, but also opening up to many more customers to come in because we're going to have more and more of these AI tools and stuff coming out.

I did have a company that, in the end, gobbled by another one, they were going to IPO. And they commented that they had a record low safety issues since they installed fire detection systems and sprinklers. And I said, why didn't you have them before? And they were like, we just didn't. I said, what prompted you to put them in? It's because the local fire station burned down, which I find beautifully ironic. I don't know it was very -- I like that one a lot. Comparing divisional margins is always incredibly dangerous because cost allocation -- I mean there's 1 billion reasons. But actually, what you think that was kind of interesting for anybody who hasn't seen a CT scan or MR being built, it's incredibly involved, right? It's a very involved process. Do you think that's been a big part of the spread that -- SKU is the wrong word, but the offer and breadth and the complexity of that has been a big part of that margin differential between you and one of your peers?

It is, but I would say it comes down to this. So I would say, in the eyes of the customer, if you were to see our products a couple of years ago versus our major competitor's, you would see the similar offerings that are out there. But behind the scenes, we might have had 8 or 9 lines of code to build each of the systems discrete, which means there are 8 or 9 software teams, that's more expensive. If you wanted to move one application to another system, you actually had to port it over. Now we're going to have completely one line of code, which actually means speed, agility and seamlessness. But just simple examples. If you had a CT line 8 different tables that you were buying, and you could say, let's get it down to 3, which we have, short, long and a heavier weight table for bariatrics, and they can fit on the 5 scanners that you have, you can imagine to the supplier that's making this how much more volume you could give them, the focus on quality systems, all of those things together. So how does that translate? That translates into lower cost and actually speed to market. Both of those help us on the margin front.

Jay, maybe -- I get this question a lot, and I think it's typically for people who haven't maybe looked at the industry very often. But I often get the question of like, if order books are running at x, call it, low single digits, surely, that means the next 12 months organic sales growth will be -- I mean I'm bringing it up in order to give a platform to sort of help people understand the backlog componentry and the growth. So maybe help people understand the interplay between order books and actually what the revenue growth ends up being like?

Yes. I think it is a good question, it's one we get as well. There are really 3 elements that we look at when we consider the health of the business. One is order growth for sure. And I think it's a fair number to look at. And our order growth ex China in the second quarter was 6%, inclusive of China was 3%. So fairly robust in the face of that volatile market. The second thing that we look at is this idea of book-to-bill, which is essentially a measure of how much in excess of sales are your orders. Now it's important to note that these are not comparable versus competitors in the space because we treat things differently. We include service at a 1:1. We include our PDx business at 1:1. So revenue equals orders. So it does kind of lower the [indiscernible] versus how others might look at this. But in our case, in the second quarter, very robust, 1.06x. And then the third thing that we look at is the actual backlog that we have in place. Now there's a couple of different flavors of backlog. The one that we call out on the earnings call, in my view, is probably the most relevant one, which is the total backlog. And that sits at record highs of $19 billion. It was up sequentially. I believe it was up year-over-year. So very robust backlog. So you have to look at those 3 things together. And the reality is, we had outsize order performance a couple of years ago in a very intense manner as a result of COVID, and in particular, in our PCS business. And so that definitely impacted kind of the cadence of orders for years to come, but it's represented in this robust backlog that we have in place. And so as we look at it, those 3 ingredients together point to a healthy backdrop. I was very encouraged by what we saw in the second quarter, in particular in the U.S. market. Very, very robust orders, a nice capital environment. And if you look at all 3 of these things together, we kind of say it's supportive of the mid-term dynamic that we're hoping to see.

Yes. And I would just add, I think, Patrick, it's a great question because if you look at the rest of MedTech, us and a few players have this dynamic that's very different. So it's not a natural with the flow products where someone places the orders, you can [ send ] those catheters or implants that week and that's how it shows up. And so you could have a year, 1.5 years, 2 years of low single digits and still have mid-single-digit sales like vice versa. Just because you had mid-single or high single-digit orders doesn't necessarily you're going to see that sales translate within the following couple of quarters because there's timing, right? And some of these are site dependent where there might be 18 months to have the site ready to put it in. I think the other aspect, and this is actually a positive over the long run for us is, when you hear us talk about enterprise deals or big customer relationship deals that are 5, 7, 10 years, once they've committed to us, we may only put the orders in for the first year, 18 months, but my guys aren't having to sell and convince them for the following 7 to 8 years. And so whatever they buy, which we had a fleet plan for them, we know what it's going to be. And barring something catastrophic in their financials, they're going to buy that. And that is fundamentally reoccurring revenue that we don't really log at that point. Some of our competitors log it all upfront. We believe we only take the first 18, 24 months where we have hard POs. But the reality of it is, once they're committed contractually, you're going to see continued business. And as that business grows, that's going to give us even more stability, I would say, in reoccurring revenue.

Those large enterprise deals, if you have to characterize even just qualitatively, I mean, they've been growing as a proportion, but roughly today, of business, how much do you think they represent?

You want to take a stab at it?

Yes. I mean I don't know that we have quantified that...

I had it put to me as 1/3, not for you, for somebody else put it to me as 1/3 for their business.

I think it's in that area or slightly below will be my guess. But I would say, in the United States, it's one of the growing trends. The more that there is aggregation of IDNs, the more they want to leverage their spend. I think you're seeing a little bit more in Europe and some of the pan, multi-country players that actually now are working with centralized health care delivery networks. But it becomes an important aspect. And again, I think if you then layer on and say SaaS opportunities to help them run their system, that's what becomes the most important attribute of the relationship. And plugging in hardware to actually supplement that becomes secondary at some point down the road, which again, why it's so important that a disease state, a smart device as well as a digital strategy, both cloud and AI together, is critical.

And so for the classic conference question for both of you. What are you surprise you get asked that more, or what are you surprise is too much focus on either way?

China maybe or no?

Yes. I think China this year will be around 11.5% of our business, 12%, something along those lines. probably getting 25% to 30% of our questions on that. I think one of the important points of emphasis of our team is driving through to free cash flow. And so it's not only all the revenue growth and operating income growth that we're driving, but from a lean standpoint, this intense focus on being efficient with your working capital balances and CapEx to drive through to free cash flow. Because, in our view, it really does change the complexion of the balance sheet and what we can do long term. I am surprised we don't get more of that in our discussions, but that's okay because ultimately, we do view it as a real value driver, and we're going to continue to emphasize that. I don't know what you would say, Pete.

I mean it's changing now. The other one would just be, you touched on that, is radiopharmaceuticals. Again, it's not that natural with a MedTech group of investors and customers. And so it's not that natural for someone to pull out a pharmaceutical question. But again, in this world where if you're going to do theranostics, this idea of a radioligand that goes to a cancer cell, unloads it's payload, just kills the cancer cells and out. And you can't do that without actually doing the assessment with one device following -- follow-up in another and then computing it together, you have to be able to talk about all of those. And so that's now starting to rise based on folks like yourself in a lot of questions that are happening. And then I would think the other one is the service business always isn't the most sexiest business. But it's a very, very important customer satisfaction, sticky business and will become more than just break-fix in the future as we bring in more software capabilities.

Perfect timing. Pete, Jey, thank you so much.

Thank you.