UnitedHealth Group Incorporated ($UNH)

Great. Welcome back to the Barclays Global Healthcare Conference. My name is Andrew Mok. I'm the facilities and managed care analyst here at Barclays, and we're pleased to welcome back UnitedHealthcare to the conference. Welcome.

Thank you.

With me on stage, I have Wayne DeVeydt, CFO; Patrick Conway, CEO of Optum; Bobby Hunter, CEO of Government Programs; and Julia Murphy, Vice President of Investor Relations. Julia, why don't I kick it to you for opening comments?

Great. Thank you. [indiscernible].

Great. To kick things off, Wayne, as you look at the first few months of the year, how is performance tracking relative to expectations across your major business lines?

Yes. So maybe to start off and just kind of anchor everybody in the audience. As a reminder, we guided for greater than 8.5% growth for the year. I think it's fair to say our guidance was both prudent and hopefully conservative. We took a view that we would grow across all of our lines of business this year. We're 2 months in. It's early, as you know, and 2 months isn't a trend necessarily. But I would say that we're encouraged with our strategy. We're encouraged with our pricing and product design. And I would say things are lining up very nicely with the expectations we laid out in our January call around full year outlook at this stage. Again, early in the year, but very encouraged across the board.

Great. Well, let's dig into the Medicare segment. I think medical cost trend ran around 7.5% in 2025 and your 2026 bids and guidance assume an acceleration to roughly 10%. So can you help us understand what's driving that increase from 7.5% to 10%?

Yes. Do you want me to jump in?

Yes.

Yes. Thanks for the question, Andrew. So maybe just to ground on just one of the words you used, you talked about an acceleration of trends. So maybe I'll just kind of break it into a couple of components for you. We completed 2025, and we talked about more in that kind of mid-7% range. And then there's kind of a roughly 250 basis point expansion of that trend to get to what we have projected for 2026. Importantly, inside of that number for '26, the core utilization, what we call residual trend, we're assuming constant. So we're not assuming acceleration, not assuming deceleration, a constant assumption there. And then when you think about the bridge to where we landed for 2026, it's really kind of simplistically about 2/3 known rate elements. So I think the physician fee schedule is a great example of it, right? The return of the doc fix for '26 after not having it in '25 creates a bit of a meaningful kind of trend turnaround as you step into the year, maybe inpatient rates with where those ultimately came in is another good example. And then the other 1/3 we've talked about is more of an accommodation for some of the unknown risk elements. And obviously, those unknown risk elements come to fruition in different ways. One of them that we had in our mind certainly was around tariffs. More to play out there. We're 2 months into the year. So I'm not going to predict what's going to happen on that front, but I feel good about the accommodation we have there on that unknown piece. So I think to maybe kind of just play off what Wayne said, we took a very prudently conservative approach when we thought about pricing for '26, when we thought about the assumptions for '26, really early in the year, 2 months in, but I feel good about kind of where things sit right now and remain appropriately cautious as we look at what's happening, what's emerging, and we're using all of our tools to make sure we got our eye on everything.

Great. And on the 2027 Medicare advanced notice, there's been a lot of discussion on the rebasing of the risk model, which places greater coefficient weights on skin subs at the expense of more chronic conditions. That seems like a fairly obvious flaw in methodology. Is that argument resonating with the current administration? And how would you -- how likely do you think it is that CMS revisits the calibration here, potentially rerunning that regression after normalizing for those factors?

Yes. Yes. So the good thing is we submitted our comment letter. That comment letter is out there, and we really kind of stepped through all the key elements that we focused on in our overall response. The one that you're talking about, the rebaselining was certainly one of them. And I don't necessarily disagree with the premise. I think that one of the challenges when you use 1 year of data to calibrate the model is there's going to be anomalies in any given year, right? And in 2024, in particular, that anomaly, most pronounced was around wound care and some of the skin substitute elements. So what you would see in our comment letter is we actually gave a nice little graphic of kind of how that distribution plays out, and you can kind of see how we are viewing that. It does show obviously skin sub is a bit of an outlier. So what we wrote to because I think there are challenges, this is a complicated -- recalibrating the model is not an easy task. So I don't want to make it sound like it's something easy for CMS to do. It's a challenging process. So whether or not they could actually kind of fully recalibrate and do something different in this time frame, I don't know, I'd kind of leave that to them, but that's a tall order. So what we wrote to is, do you consider a deferral so that you can rerun it and take the time to maybe take a little bit of a modified approach. Or do you phase it in, right? And similar to other risk model changes in the past, do you do more of a phase-in to kind of mute the impact and maybe give a little bit of time to then do some of that recalibration work. And I think any of those solutions we've kind of proposed to them are as things that we're supportive of. And ultimately, we just -- we want to get it right. Right? We want to get it right. They want to get it right and move the industry forward in the right spot.

Have they been receptive? Is the dialogue constructive, do you think?

So during the comment period, right, they're somewhat limited on what they can say. So I'm certainly not looking to put words in their mouth because there are very clear rules of the road of what you can do in that engagement. But we have had opportunity to meet with them, right? We've had opportunity to meet with them. And just like we do it kind of in all comment periods, really ground our perspective in data and just bring the data forward and kind of let the data speak -- and we've had the opportunity to do it, which I am incredibly grateful for to just bring our data to the table and engage in that dialogue. What they do with it, obviously, is their discretion. But we'll continue to bring that forward as we identify new things, learn new things and be as supportive as we can in that dialogue that we're having.

Great. Another point you raised in your comment letter was that AI technology and revenue cycle tools are structurally increasing productivity and revenue yield across the health care system. These are relatively new dynamics that CMS may not have had to consider in the past. So from your perspective, what specifically needs to change in how the administration builds its forward forecast so that technology-driven effects are reflected accurately rather than mistaken for temporary noise or coding behavior?

Yes. Yes. So I mean -- and we wrote to it that way because that is something we've seen, and we've talked about it, right? We talked about on our earnings call, some of the emergence of that trend acceleration in 2025, obviously, that had an impact on us that we then had to consider when we thought about pricing and the design for '26 and beyond. And I think what we've really grounded for CMS and the folks that we've talked to when we've had these discussions is about the importance of, again, really grounding the utilization, the unit cost and the forward-looking assumptions in just the reality of what we're seeing in health care costs today, right? And so we're -- I think specifically, we're not asking for an accommodation in '26 and '27 assumed trends in growth rate of acceleration, we're not, but we also don't believe that a significant deceleration is probably the right spot to be either and certainly not something that we're seeing in the provider behavior, the utilization of services, the intensity of services. So for us, it's really about when we look at what's happening in physician utilization or inpatient utilization, we're looking more for consistency, right? Consistency from what you've seen in '24, '25 and how that's projected forward to '26 and '27. And the same thing really applies around the rate element, more about consistency than assuming an acceleration or deceleration. And I think that would get the industry and the overall funding level to a spot where the rate does much kind of more appropriately reflect the actual cost of health care today and the importance of benefit sustainability and other things for the consumers that rely on MA.

Great. Let's move on to some of the other business lines. In Medicaid, I think you reiterated expectations for margins to be down about 100 to 170 basis points. Can you first give us a sense for how Medicaid medical cost trend developed in the second half of '25? And what level of trend you're assuming for 2026? And then related to that, can you provide an update on any conversations around 4/1 and 7/1 rate negotiations?

Maybe let me first frame up kind of our outlook and some of the key dependencies. So -- and then I'll have Bobby comment a little bit more on trend and what we're seeing. But the one thing to keep in mind is that we know that, generally speaking, states are a bit slower in responding to trend in the beginning periods in time. And so part of our goal is to recognize that some of these rate increases are coming in on July 1, right? So relative to our expectations of 6% to 7% kind of rate increases, we're optimistic based on the early negotiations and early feedback. But as you know, until you get to July 1, you don't really know where this plane is going to land. And so I think that's an important part. And with that in mind, it was important for us as we finished out '25 to take a conservative posture on trend because again, until we have more clarity around the rate environment, it would be inappropriate and obviously not prudent to lean in. And so I think it's fair to say that our assumptions at year-end were conservative, and we're seeing some of that play out early. But nonetheless, this is really about rate and where we're going to go. But Bobby, anything you want to add?

Yes. No, super well said. I would just say that from a pure trend perspective, I mean, we're assuming pretty good consistency between '25 and '26. And in '25 for us, trend was above where the rates came in, and we're assuming the same thing is going to exist in '26. We're doing everything we can to address affordability to partner with states. We've got some kind of known hot items, I think, for Medicaid broadly around behavioral and specialty and some of the home care services that we're partnering with states to try to find solutions and act on those as quickly as possible, but there does still continue to be a disconnect between the actual trend and the rates that we're pursuing.

Great. And turning to the individual ACA exchanges. I know it's a small business for you, but there is a dynamic situation. So I did want to touch on it. You previously indicated exchange membership could be down more than 500,000 for the year. As we approach mid-March, can you share how effectuations are tracking so far? And how much of that expected decline you anticipate to see in the first quarter?

Yes. So 500,000 is still kind of the right zone for us. So we're not moving off of that number. I would say as you pace through the first couple of months, obviously, what's happening is some of the grace period wear off and disenrollment that comes from that. And I would say with what we've seen the first couple of months. I think that we're still tracking towards that realization of the 500,000. Think about roughly half of that likely as a result of some of that kind of grace period time frame. So I think kind of end of March, beginning of April time frame. And then the other half really plays out throughout SEP as just kind of more natural wear off occurs of the enrollment and then with some of the SEP rule changes, less of a kind of pipeline to fill that gap.

Great. Turning to OptumHealth. I wanted to clarify some of the fourth quarter items and run rate. OptumHealth underperformed guidance by, I think, roughly $600 million in the fourth quarter. You attributed that primarily to onetime items. Can you walk us through the key drivers of that shortfall and help us understand how much of the $600 million is truly onetime as we think about the right run rate looking ahead?

Yes. So about 70% is onetime items. The other 30% investments, think value-based care investments, clinical investments, infrastructure investments to support value-based care, and those continue to run into this year, to be clear, as we make those investments for the long term. We're confident in OptumHealth in the guidance we put out for this year and the growth in that guidance, as Wayne alluded to. We're also confident in the long-term trajectory of OptumHealth, both on the value-based care platform and the integrated nature of that platform. And as a reminder, it's a diverse set of businesses. So we also have services, ambulatory surgical centers, home health, hospice, so a diverse platform that we're confident in the long-term growth.

Right. And you also booked a $620 million PDR for expected 2026 contract losses. Were those related to external third-party contracts or internal UHC contracts? And mechanically, should we consider that a tailwind for 2027?

Yes. So I'll start, Wayne, feel free to add in. So external and do think of it, those contracts, we will either exit them or get the rate update we need for 2027. So that does provide that tailwind for 2027.

Yes. The way to think about it is with the lost contract, in theory, there's an accounting benefit that comes through in 2026 because you're amortizing the PDR. We are carving that out for our non-GAAP so that you can see we're not taking that benefit when we provide our guidance. But conceptually, that benefit becomes real and tangible in '27. And the way to look at it is, I think a question we get is, but are you really going to be able to get rates where you need them on that big of a chunk of a loss reserve contract. But these contracts range from losing a negative 2% margin up to, say, a 10% margin, 11% margin. And so there is a very large chunk that's very close on the J curve to where the profitability needs to be for us. And so we do think we'll recapture a substantial portion of that, meaning we're either going to get priced right or we're going to exit it. And so ultimately, it does become a tailwind for 2027.

Great. And sticking with Optum, after stripping out Optum Financial, which was running around, I think, 44% margins in 2025, the stand-alone OptumHealth pro forma margins finished at around 1.5%. Could you outline the key building blocks of the margin bridge from current levels back to the 6% to 8% long-term target? And what underpins your confidence in that margin target without the higher-margin Optum Financial business?

Yes. So we are still confident in the long-term margin target. So let's break down the business. You've got integrated value-based care, where a couple of things that are positive attributes to call out. One, you're at or above that target margin for the more mature cohorts of patients. You're also at or above that target margin at about 30% of geographies. So that shows you we can get the performance to that level in integrated value-based care. And we've rightsized the risk platform, both in terms of geographies, arrangements where we pulled back from PPOs. And as Wayne and I just talked about, also contractual changes, we were able to accomplish much of that in '26 and then we'll move forward in '27 and are in those negotiations right now with our various payer partners. So that's sort of the integrated value-based care. The other attribute I'd call out within OptumHealth and these businesses service that integrated value-based care platform. You've got things like home health and hospice, double-digit margins growing, performing well. Ambulatory surgical platform, around 20% margins growing, performing well. So within OptumHealth, you've got a diverse set of businesses, including ones in a double-digit margin territory. You've also got payer and employer services, as we've said, double-digit margins growing, performing well. So it's a diverse business that there's multiple pathways to get back to that target margin range.

Andrew, the one thing I would help people anchor on is if you think about this year, we said we would improve margins by at least 30 basis points. That's our floor. We would obviously hopefully do better than that as we progress this year. But that's carving out the loss contract reserve we talked about, which obviously next year is a dollar-for-dollar improvement in margins. And the one thing we've highlighted is a lot of these on the J curve are right at the point of an inflection where you'll start to see the positive. So I think as Patrick said on our year-end call, we're not looking at basis points for 2027. We're looking at points of improvement in the margin in '27. And this year is kind of that final year of lapping the investments, getting through the loss contract window, and then you'll really see the ramp-up happen in '27 and '28.

Yes. And the 2 big -- as Wayne said, the 2 -- the rate notice and where it lands is a big factor for '27, which you heard from Bobby and we agree. It's not reflective of the underlying cost trends. So we'll watch that closely and our payer negotiations. So we're confident in the progression over time, as Wayne said.

Great. There's been a lot of moving pieces in OptumInsight over the last few years. Taking a step back, can you help us understand the composition of that business today, maybe bucket the earnings into core categories like SaaS, consulting, managed services and financial. And then talk about how you would rank order the investment priorities across those groups going forward.

Yes. So you've got a payer-oriented business where we work with about 8 out of 10 payers across the country, by the way, through things like authorizations, clinical programs, really supporting their back office through technology. That business is performing well. And as you put AI behind that business, it creates even more opportunity. You've got a provider-oriented business, I think RCM, et cetera. Once again, a business with a strong base. We also work with about 8 out of 10 hospitals and health systems across the country in some way. And once again, you put AI behind that, tangible example. There's a product called Crimson, which I -- when I used to work in health systems, I used 20 years ago, still exists, still a big customer base. You put AI behind it. It's now Crimson AI. Actually, the savings, and we put out a release on this, the savings, the impact, et cetera, with our customers is greater, and you're seeing greater sales. So a tangible example of powering product by AI. The other one that we've talked about publicly is OptumReal, real-time settlement of claims really across payer and provider, partnering with UHC first and now partnering with other payers and providers to settle that claim in real time. So the payer knows what they're paying, the provider knows what they're paying. And by the way, the consumer knows their co-pay. Now bringing Optum Financial, which you may ask more about. Now Optum Financial can help settle that transaction. So that -- and so we're increasingly moving to AI products, software-driven products that we're seeing a lot of momentum in the marketplace. Now I will say this will be a journey. If you're selling products now, that's hitting '27, '28 and beyond. We're still developing a whole portfolio of products, investing about $1.5 billion into AI across enterprise this year. So very bullish about the long-term trajectory of OptumInsight and Optum Financial. I also want to say, if you think about this year, it's an investment year really building up to that long-term growth.

Great. Speaking of investments, I think you pointed to $1 billion of mostly AI-enabled cost efficiencies in 2026. Can you give us a sense for how much runway is left on these efforts and how we should think about the impact beyond 2026 and specifically looking at what impact we should expect on the G&A ratio?

So we're in the very early innings. And I would literally put the AI journey is the bottom of the first inning at best. And while we've taken out almost $1 billion in costs going into '26 related to AI, we think that number actually incrementally grows going into '27 and incrementally grows as we move into '28. Ultimately, the impact on the OCR is one to be determined. And the question is about journey and pace. What I can tell you is our $1.5 billion we're investing this year is still not enough for where we want to take this company in the next 2 years. And so we've already started building the further pipeline of how can we accelerate that even more aggressively in '26 and then going into '27. Ultimately, the impact on the OCR should be significant. Again, we should be talking points, not basis points. And I think this is an industry, as you know, that generally has moved in the basis points because of a top line revenue. We're talking about moving at points because of real cost takeout and real efficiencies. So I don't want to commit to a number, but I would say ultimately that there are functions where 70%, 80%, 90% should be fully automated from start to finish. Just a point of reference though, you might find interesting. We had probably our best January enrollment and call center experience that we've had in the company's history. We obviously used AI as a major component of it. In one single day, we answered over 3 million calls. We had over 45,000 of those happen simultaneously. The average speed of answer was 18 seconds and first call resolution was north of 90. It's it's really unbelievable when you look at our history of what we've done and where we've been at. And candidly, the agents are smarter by the end of the day than they were at the beginning of the day, and they're not sick. And there's lessons we can learn though, and we're taking those lessons for kind of round 2 as we go into '27 and doubling down on those investments. So I think you're going to really start to see real value creation in terms of G&A takeout. And I think you're going to see it really start to compound aggressively in '27 and beyond.

Yes. Two just super brief additional examples in RCM, as you use AI and technology, you bring more people out of the process, increases the margin profile. In the Rx space, and Wayne alluded to this, we onboarded a record number of clients, incredibly satisfied and actually now using it, the vast, vast majority of your transactions in Rx can be solved AI plus digital. And actually, it's a better consumer experience and it's less costly.

Great. Maybe the last few minutes here. Another item I wanted to touch on today was within OptumRx. There has been several PBM reform items in the last few months, including legislation that enforces rebate pass-through in the commercial market. How do you expect these changes to impact the business over the next few years?

Yes. We were ahead of these in OptumRx. So it will not impact the business. So we had already announced 100% commercial rebate pass-through rolling that across our contract base. We've already done cost-based reimbursement to remind people for all drugs, all pharmacies, not a niche number of drugs. We've done already things in prior authorization and reauthorization changes, just removing friction. So our OptumRx team has a ton of momentum. They've had record sales years for a number of years now. And what's resonating in the marketplace is the transparency we provide for clients, the clinical programs that drive affordability and an integrated solution, including with UnitedHealthcare and other payers that we serve, where we integrate medical and pharmacy and do that well on behalf of they serve. So we're winning in the payer and the employer market in OptumRx and doing very well.

Great. And with greater clarity on sort of the legislative environment, it sounds like you're well into your investment cycle on the AI side. As you return to a more normalized capital deployment in the second half of the year, how should we think about the priorities across share repurchase, dividends and M&A? And within that framework, are there any key strategic priorities you see within M&A?

So maybe to bring it together, we said we anticipate generating at least $18 billion of free cash flow this year. We will be maintaining and growing our dividend. Historical practice has been the case, and we've already got approval from our Board to continue that practice. So you'll see that occur this year. And that leaves us a decent amount of powder for debt paydown, buybacks and M&A. Relative to debt paydown, though, we have committed that we would get our debt to cap closer to 40%. We fully expect to be there in the back half of this year with just regular terms coming due in Q1 and Q2, we're just paying it down. We will be actively in the buyback program again this year. And then while we have a decent amount of powder for M&A, I mean, candidly, based on the intrinsic value that we are currently trading at on a discount basis, there's probably no better acquisition than ourselves right now. So you'll see us probably get more aggressive in that space if the market dislocation continues to -- mislocation continues to exist. But last thing I would say is, look, we think that OptumInsight and OptumHealth is a big part of the future of this company. And we believe in the VBC model with high conviction. And we know we have to prove it to everybody in this room that it's there and the margins are coming back, and we're confident we'll do that. And I would say fintech on the OptumInsight, we're going to go all in on AI and fintech, and we think we can become an important part of the health care technology ecosystem over time.

Great. Well, with that, we're out of time. So thank you so much for joining, and please enjoy the rest of the conference.

Thank you.

Thank you.

Thank you.