agilon health inc
NYSE:AGL
Earnings Call Analysis
Q3-2023 Analysis
agilon health inc
We begin with a warm welcome on Agilon Health's third quarter earnings call of 2023, led by Vice President Matthew Gillmor.
Agilon's journey of empowering primary care doctors in transforming health care is ongoing, with more than $250 million reinvested into local primary care this year due to effective care by partner groups. The company has strategically sold its Hawaii business to hone in on its core markets, which differ in operational structure from Hawaii and are central to Agilon's growth strategy, such as non-delegated multi-payer relationships and joint ventures with large physician groups. This sale bolsters Agilon's focus on integrated senior care across Medicare Advantage and traditional Medicare populations, critical to its future success.
Financially, Agilon is performing robustly, with total member count reaching 508,000, a 43% surge, and revenues climbing 75% to $1.2 billion, topping guidance. Excluding Hawaii, partner market growth was even more impressive, with a 49% increase in membership and revenue growth at 85%. Despite rapid expansion, adjusted EBITDA shows a reduced loss, resonating with the company's model's capacity to thrive with growth. The raised guidance for 2023 reflects steady growth and strong financial resilience.
Agilon is forward-looking, with high visibility into 2024's driver's thanks to a predictably growing partner class for 2024 and prudent reserve actions in Medicare Advantage. The quarter's positive outcomes, along with the sale of the Hawaii business and conservative reserving, underpin the increased and narrowed adjusted EBITDA outlook for 2023 to a range of $6 million to $18 million. Membership and revenue forecasts are also revised upwards, excluding Hawaii. However, the medical margin outlook has been reduced modestly to reflect a conservative approach to reserving and utilization trends, setting a diligent foundation for future performance.
Closing remarks by Steve, Agilon's representative, amplify confidence in the company's strategic focus post the Hawaii sale, leadership in full risk model adoption, and improvement in clinical and financial performance. The utilization of better data and resources is leading to exemplary outcomes for stakeholders. Agilon's innovative approach is redefining health care utilization and ensuring continued success into 2024 and beyond.
You don't have any saved screeners yet
Hello, everyone, and welcome to the Agilon Health Third Quarter 2023 Earnings Conference Call. My name is Carla, and I will be your operator for today's call. [Operator Instructions] I will now hand you over to your host, Matthew Gillmor, Vice President of Investor Relations. Matthew, please go ahead.
Thanks, operator. Good afternoon, and welcome to the call. With me is our CEO, Steve Sell; and our CFO, Tim Bensley. Before we begin, we'd like to remind you that our remarks and responses to questions may include forward-looking statements. Actual results may differ materially from those stated or implied by forward-looking statements due to risks and uncertainties associated with our business. These risks and uncertainties are discussed in our SEC filings. Please note that we assume no obligation to update any forward-looking statements. Additionally, certain financial measures we will discuss on this call are non-GAAP measures. We believe that providing these measures helps investors gain a better and more complete understanding of our financial results and is consistent with how management views our financial results. A reconciliation of these non-GAAP financial measures to the most comparable GAAP measure is available in the earnings press release and Form 8-K. Following prepared remarks from Steve and Tim, we'll conduct a Q&A session. During the Q&A session, we'd ask everyone to please limit themselves to one question so we can get through as many questions as possible. With that, I'll turn things over to Steve.
Thanks, Matt. Good evening, and thank you for joining us. Momentum across our business remains strong, and we are making rapid progress against our vision to transform health care in 100-plus communities by empowering primary care doctors. As an indicator of our success, this year we are on track to reinvest more than $250 million into local primary care based on the high-quality cost-effective care being delivered by our partners. These results are enabling our partner groups to expand access to preventative services, improve quality outcomes and drive value for the communities they serve. I want to thank my Agilon colleagues and our partners for their trust in each other and their belief and support in a network that is shaping a better health care system for our country. Before discussing our performance for the quarter, I wanted to take a few moments to highlight our decision to sell our Hawaii business. As we have discussed with you, Hawaii operates very differently compared to our core partner markets in the Continental U.S. In our core partner markets, we leverage a common operating structure. This operating structure is centered around a long-term joint venture with a scaled physician group and non-delegated multi-payer relationships with health plans and CMS. The key differences in Hawaii's operations, namely the lack of a joint venture partnership with a large physician group, a much smaller senior patient physician panel size and Hawaii's delegated MSO infrastructure made our Hawaii business increasingly less strategic to Agilon and created a drag on our financial results. We are pleased to transition MDX Hawaii to a new owner that is better positioned to invest into the business and optimize its delegated infrastructure, which should benefit patients and the community at large. The sale of Hawaii will enhance our ability to focus even more specifically on our partner markets at a point in time when we are driving increasing success across a PCP's entire senior panel. This quarter and going forward, you will hear us consistently highlight the power and importance of our integrated senior business across Medicare Advantage and the traditional Medicare populations. Our ability to generate successful outcomes for patients, PCPs and the system in a multi-payer full risk model is increasingly unique. Our focus on this opportunity will pay dividends in the years to come. Turning now to the quarter. Partner market performance was again extremely strong across both MA and ACO reach. All of our key financial metrics were generally in line or above our guidance ranges, especially on an underlying basis. Our results continue to demonstrate the unique power of our model to inflect profitability while driving significant growth. During the quarter, our total membership on the platform increased 43% to 508,000 members, and revenues increased 75% to $1.2 billion. This was above our guidance and was supported by the successful onboarding of new PCPs and faster pull-through of members, particularly in new markets. Adjusted for the sale of MDX Hawaii, our partner market growth was even stronger with total membership up 49% and revenue up 85%. Even with our faster membership growth, adjusted EBITDA continues to inflect higher, increasing more than $20 million year-over-year to a loss of $6 million for the quarter. This was in line with our outlook and supported by strong medical margin gains in our partner markets across MA and reach. Adjusted for Hawaii, our partner market EBITDA was positive $6 million for the quarter, well above our outlook. And it was even stronger on an underlying basis as our results included some net negative development from 2022. Our combined medical margin across MA partner markets and reach was strong in the quarter, with MA generating $111 million and REACH generating $55 million. These results demonstrate the power of a PCP focusing on the most complex patients across their entire senior panel with differentiated information and care team resources. As an example, our year 2 plus partner markets in MA and REACH both generated over $130 per member per month in medical margins year-to-date. Think about the value delivery to our PCP partners and the health system when we generate this magnitude of savings across an average panel size of 400 to 500 senior patients. From a guidance perspective, we have raised our membership revenue and adjusted EBITDA outlook for 2023. This was supported by the growth and margin progression in our MA partner markets, including REACH and the sale of MDX Hawaii. We also plan to maintain a more conservative reserving approach as we exit 2023. And this is intentionally reflected in our medical margin outlook for MA and will support our future performance in 2024. Our ability to execute against our adjusted EBITDA targets during 2023 and enhance our visibility to 2024, continues to reflect the strength and durability of our model. As I have discussed with you previously, the key differences in our model are driving differentiated clinical and financial performance, unlike the traditional fee-for-service system, which predominates across health care. All patients in our model have a fully aligned or attributed relationship with their primary care doctor, and through Agilon's platform, our PCP partners have the data and resources to proactively impact patient care. This translates into specific differences in the way health care is utilized and managed, and this shows up in our business in very tangible ways. For example, we continue to have outstanding results in the standardized Star ratings measures. For 2024 Stars, the percentage of our membership in 4+ Star plans will increase modestly to approximately 84%. However, as most of you know, plan-level star ratings also include the performance of non-Agilon providers. For our year 2 plus partners Agilon's specific performance is 4.3 stars and increased nicely year-over-year. This was despite more stringent cut points and relatively flat star ratings industry-wide. We continue to excel in areas like preventative cancer screens, diabetes control and medication adherence. Our quality results will drive meaningful value to our patients and the health care system in the years ahead. And because of our high member retention, Agilon and our physician partners will share in this value over the long term. Additionally, our ability to drive differentiated cost performance was clearly evident in the recently released ACO REACH results for 2022. This data allows Agilon to compare our performance against the unmanaged fee-for-service system as well as other value-based care models. During 2022, our Reach ACOs drove nearly 10% savings relative to the Medicare benchmark. This was more than 2.5x better than the program average and our results included more than 90,000 beneficiaries in diverse markets. Agilon also returned nearly $30 million in guaranteed savings to the Medicare Trust Fund last year. As you can see from our REACH results this quarter, our differentiated cost performance has carried into 2023. Looking forward, we believe our leadership position as the platform and network moving physicians to full risk has grown considerably. This is a function of the magnitude of savings we are generating across the entire senior panel of a primary care doctor. Our timing is also important. As primary care physicians and the broader system increasingly need a scalable solution for multi-payer full risk. This dynamic is driving the ongoing inflection we are seeing from a demand perspective among both physician groups and health systems. For the class of 2024 new partners, we now expect 25,000 new ACO REACH members, and we are increasing our expectation from 100,000 to 110,000 new MA members. At this point in the year, we are nearly complete with our payer contracting cycle, which gives us better visibility into the membership pull-through. We also now have a clear line of sight to a very strong class of 2025 with multiple new partners signed, including independent groups and health systems. Implementation for this class has already begun, which should bode well for future performance. Before turning the call over to Tim, I wanted to offer a few comments on 2024. We remain highly confident in the trajectory of our adjusted EBITDA inflection and expect to share an initial view in early January. As we have discussed previously, we operate in a very forward-looking model, and our visibility into the key drivers for next year's performance are quite high. First, we have a large and growing class of 2024 new partners with an attractive margin profile that should be meaningfully accretive to adjusted EBITDA. This is a function of the longer implementation cycle for this class and targeted investments we have made around technology and centralizing key processes. Next, the combined strength of our 2023 run rate margins across our integrated senior business will have key positive implications for 2024. First, our REACH performance will carry forward, driven by our ability to maintain the cost differential compared to the benchmark. And second, the reserve actions we have taken in Medicare Advantage should reduce the risk of negative claims development next year. Finally, we remain very confident in our ability to manage the new risk adjustment model starting in 2024. The impact [ at ] Agilon from the V-28 model change is relatively modest. And given the limited maturity of our partner markets and senior membership, our ability to mitigate this impact is very high. With that, let me turn things over to Tim.
Thanks, Steve, and good evening, everyone. I'll now review highlights from our third quarter results and our updated outlook for 2023. Starting with our membership for the third quarter, total members live on the Agilon platform increased to approximately 508,000 including both Medicare Advantage and ACO reach. Our consolidated Medicare Advantage membership increased 58% to 420,000, driven by the addition of new partner geographies and 9% growth with our same geographies. Adjusted for Hawaii, MA membership in our core partner markets grew 69% to $384,000 and our same geography growth was 12%. Reported revenues increased 75% on a year-over-year basis to $1.2 billion during the third quarter. Year-to-date, revenues increased 73% to $3.5 billion. Revenue growth was primarily driven by membership gains in new and existing geographies. On a per member per month basis or PMPM, third quarter revenue increased 11%. This was primarily driven by benchmark updates and membership mix, including higher benchmarks in several new markets. Adjusted for Hawaii, revenues increased 85% to $1.1 billion and revenue PMPM increased 10% during the third quarter. For our MA business, medical margin on a reported basis increased 42% year-over-year to $108 million during the third quarter. Year-to-date, medical margin increased 67% to $408 million. While this was below our outlook for the quarter, the difference was primarily driven by performance in Hawaii. Adjusted for Hawaii, medical margins increased 46% year-over-year to $111 million for the third quarter and was in line with our expectation, even with approximately $6 million of net negative development. On a per member per month basis, medical margins across our core partner markets increased by 4% year-to-date to $119, driven by the maturation of markets and member cohorts. For our year 2 plus partner markets, medical margin PMPM increased 17% to $134 on a year-to-date basis. MA medical margins for the quarter included a net $8 million in negative development, including $9 million in prior year claims, offset by $1 million in prior year revenue, $2 million of the net development was attributed to Hawaii and the remaining $6 million was attributed to our core partner markets. The negative claims development this quarter was almost entirely isolated to system issues with a single payer related to supplemental benefit costs. As we discussed last quarter, we are making focused investments to improve our visibility into data gaps with payers. Excluding the year-to-date development, MA Medical margins would have been approximately $125 PMPM in our partner markets and $143 PMPM in our year 2 plus partner markets. We think this is an important metric as it better reflects the year-to-date run rate performance of our MA business in light of the sale of MDX Hawaii and the actions we have taken to minimize the risk of negative development in 2024. For our ACO reach business, we continue to be very encouraged with the performance, which again outperformed our guidance this quarter. REACH generated $55 million of medical margin in the quarter and $117 million year-to-date, which is a twofold increase from last year. Additionally, on a per member per month basis, REACH profitability this year is roughly comparable to the year 2 plus MA partner markets. This level of performance is encouraging and underscores the power of our multi-payer full risk platform. While REACH is not consolidated in our financial results, we do think it is relevant to think about our medical margin performance on a combined basis across our MA partner markets and reach. This is because our combined MA and reach performance is what drives our key profitability metric, adjusted EBITDA. For 2023, our combined medical margins for MA partners and REACH are consistent with our original guidance expectations. As Steve mentioned, this sets a strong foundation for performance in 2024. Our adjusted EBITDA on a reported basis was negative $6 million in the quarter compared to negative $26 million last year. On a year-to-date basis, adjusted EBITDA was positive $28 million compared to negative $20 million last year. As a reminder, adjusted EBITDA includes geography entry costs primarily associated with new partners that would generate revenue in 2024. The year-over-year gain in adjusted EBITDA reflects the increase to our medical margins across both MA and reach as well as platform support leverage, partially offset by performance in Hawaii and the net negative development. Excluding Hawaii, our adjusted EBITDA would have been positive $6 million for the quarter and $42 million on a year-to-date basis. From a utilization standpoint, composite utilization across MA and reach was generally in line with expectations. For MA, we did observe an increase in utilization during May and early June, which resulted in some in-period development that we recognized during the third quarter. This was contemplated in our guidance and trends moderated back towards normalized levels during the third quarter. For REACH, utilization during the first half of 2023 developed favorably relative to our expectation. As a reminder, our results and reach better reflect our real-time performance against the unmanaged fee-for-service system because of how the benchmark mechanics work. Turning now to our outlook for full year 2023. Please note that our updated guidance excludes MDX Hawaii for the full year. We have raised and narrowed our adjusted EBITDA outlook to a range of $6 million to $18 million from a prior range of $0 to $23 million. We have also updated our membership and revenue outlook, which are both higher on an underlying basis, excluding MDX Hawaii. From a medical margin perspective, our revised outlook is $455 million to $470 million and is approximately $50 million lower than our prior range. This reflects two factors: First, the removal of Hawaii represents about $20 million of this change; Second, one of our primary goals is to exit 2023 with appropriately conservative reserving posture in MA. In light of this, we continue to proactively refine our model to account for utilization trends as well as any potential blind spots with health plans. We have assumed utilization will not moderate any further from recent levels, which accounts for $30 million of the change to our MA medical margins. We think this is a prudent approach and is [ informed ] by our decision to maintain a more conservative reserving posture. We are pleased to make these decisions, which provide a strong foundation for future performance while still modestly raising our adjusted EBITDA guidance. Full details on our fourth quarter and full year guidance can be found in the earnings press release. With that, let me turn the call back to Steve for some brief closing comments.
Yes. Thanks, Tim. Before opening up the lines, I want to emphasize three key points. First, the sale of our Hawaii business enhances our ability to focus even more specifically on our differentiated core partner business and positions us for continued success in '24 and beyond. Second, our leadership position as the platform and network, moving physicians to full risk has grown considerably, and you can see it in the clinical and financial results we are driving and in the accelerating demand from physicians in independent groups and health systems. And third, with access to better data, resources and incentives, our primary care doctors are actively managing the way health care is utilized across their entire senior panel, which is yielding meaningfully better outcomes for doctors, patients and add on. With that, we are now ready to take your questions.
[Operator Instructions] We will now take our first question, which comes from Lisa Gill from JPMorgan.
Just wanted to just follow up on the medical cost side. So Tim, you talked about system issue with a single payer, data gaps with that payer. If I remember last quarter, you also talked about increasing reserves to try to account for some of this going into the back half of the year and then we saw that the prior period development here in this quarter. Can you just talk about where you are on that data gap, the systems issue? Do you feel like that's one fully behind you? And then secondly, when talking about utilization, you said in line, you saw a little bit of a bump in May and June. What we saw with some of the Medicare Advantage players in the most recent quarter is that in the quarter, they actually saw higher costs than we're projecting even beyond May and June. And one of the largest players talked about some COVID hospitalizations that they saw towards the end of the September quarter. So just want to really understand what you saw, what your expectations were and if you have anything on the COVID side or if you've seen anything on the COVID side?
So Lisa, I'll start, and Tim can chime in. I mean, I think I'll start with we raised EBITDA guide on the year because of the strong overall performance across MA and REACH across nearly 500,000 senior patients, and that really sets a strong foundation for '24. From a utilization perspective, composite utilization was in line with our overall expectations. As Tim kind of outlined, we did see a step-up in Q2 utilization in MA and REACH. The MA was within our guide, REACH actually developed favorably relative to sort of what our expectation was around that intra-period. In Q3, we've seen a deceleration. And our guide makes an assumption on utilization that will be flat through the end of the year, and that's reflected in the reserve pass that Tim talked about in terms of an extra [ $3 million ]. But Tim, you want to talk about?
Yes. Lisa, just a couple of things specifically to your question. First of all, on the prior period development, I think we have made a lot of progress this year and partnering with the payers out there to try to close some of these gaps. Essentially, some of these information issues are caused by just the complexity of our model, and that's also -- it's also a feature of the model, right? We've got over 30 payers, over 100 payer contract combinations in our markets. But having said that, I think we made good progress. The issue in this quarter was a very specific issue with just 1 payer. Hopefully, that was a lingering issue that we've now figured out where the gap was and should be okay with that going forward. In terms of the utilization trend, remember last quarter when we talked about this, we said, "Hey, we haven't seen the spike up yet. We didn't have enough information from May or June to really see what some of the payers were referring to. We did want to make sure that we had covered the possibility that there would be some higher utilization in our guidance going forward. As it turned out, as we just reported, we did see some increased utilization of mail that was greatly started to moderate in June. And as we've seen so far, started to continue to moderate in early Q3 as well. I'm not completely surprised by that compared to some of the comments that I've made from some of the big payers. I mean, I think our model and even some of them have said should be performing better than the average out there. So the fact that we saw a spike up in some moderation down is probably just a factor also of the strength of our model. On COVID specifically, we're actually seeing -- if you go back to last year, which is -- you might want to think about that as the first really sort of post big COVID year. We did actually see some seasonality with COVID starting to spike up last year in kind of the August and September time frame in terms of utilization. We did see some of that again this year, although actually lower spike up in COVID this year than even what we saw last year. And the overall magnitude of hospitalization from COVID in terms of a spike up versus the baseline is not really material and not causing any material change in our medical margin performance or our medical margin outlook.
We will now take our next question from Justin Lake from Wolfe Research.
I wanted to start off, I've got a couple of questions. I want to start off on the REACH performance. So you said it was $55 million of medical margin. So give or take, half your medical margin in the quarter, what were you expecting it to be in the original guide?
Yes. Justin, just real quick before we can obviously talk a lot about the performance of REACH just continue to be really positive in the quarter as it had kind of year-to-date so far, but the $55 million of medical margin that we talked about for REACH is not part of our consolidated results. So the $108 million of medical margin that we reported, which does include Hawaii. In addition to that, if you look at the footnote showing the unconsolidated ACO REACH entities, we had $55 million there. and about $18 million of EBITDA offloaded [indiscernible]
I apologize [indiscernible].
Yes, if I can just add to that, Justin. I mean I think we're driving really strong medical margin performance across the entire physician panel across both REACH and MA. And so the $55 million that Tim talked about is driven by beating that national benchmark by more than 300 basis points. In my remarks, I talked about the '22 results. You're seeing that in '23. But the same things that are driving success in REACH are driving success in MA, and our partner market medical margins came in at $111 million in MA which is sort of in our guide. And when you adjust for that PPD, it's actually at the high end of our guide. And I think it's just a function of this model that we've got around high touch and our ability to drive better access cost and quality outcomes. So that's what I'd highlight.
Yes. I mean the ACO REACH population has been more traditionally an unmanaged population. So there is a big opportunity for us there. We tended to be even as we move through this year. We have very good current data from CMS on our members and we tended to be a little bit conservative. So we actually saw even positive development as we moved into Q3. We had guided about a $12 million EBITDA flow-through for ACO reach, we ended up with some positive development flowing through from the previous quarter -- $18 million overall, so far during the year, the way you make money in ACO REACH is the way you perform well [ when ] AC REACH is essentially you just outperformed the Medicare fee-for-service reference population on cost performance, and we're outperforming that on a year-to-date basis by over 300 basis points. So the model really is starting to prove to be very valuable on the ACO REACH side.
And then Justin, what was your real question? Sorry.
Well, the other question I was going to ask on ACO REACH, and again, I could be -- it's been a long couple of days, I might be off again. But my recollection of the Investor Day was that you guys actually kind of titrated lower your expectations for ACO REACH. I think I was looking at it and you would expect that, I think, in 2026 to be, am I recalling the number right, $35 million of profitability?
That's right.
Okay. So it sounded like you had kind of gotten more conservative or some thought process there on like, hey, let's take that number down a bit. But 6 months later, you're running at that run rate already looks like. Is that a number that you think we should be looking at differently than from the Investor Day? Are you thinking about that business differently?
Yes, absolutely. I mean our emphasizing the power across the entire senior panel is hopefully one of the big takeaways that you're going to have. The way the mechanics and the ACO REACH model work is you carry forward your performance, and you need to beat that underlying benchmark, which we would expect we're going to do again next year. And so I think we feel like we're going to have much stronger performance going forward. We've already seen it this year-to-date. And it's -- we have much more valuable and readily available and consistent information, which is giving us high confidence in that.
We will now take our next question from Stephen Baxter from Wells Fargo.
I just wanted to try to clarify the discussion a little bit on the revisions to the medical margin guidance. So I think I'm following you that you're saying $20 million of the lower $50 million to $55 million revision on medical margins related to pull in Hawaii out. So I guess that leaves, call it, the [ $30 million ] for the continuing partner markets. I'm just trying to understand kind of the [ $30 million ] in the context of the quarter you just had. It does look like including, I guess, wanted to strip out the Hawaii results out of the quarter. it doesn't really look like you were kind of off your medical margin guidance that you provided for the third quarter. So just trying to understand why you have $30 million lower medical margin on a core basis if the quarter was relatively in line [ ex ] Hawaii, it kind of implies that it's primarily related to the fourth quarter. Just trying to correct my understanding of that, hopefully, if that made sense.
I mean I think a key goal for us was to raise and be adjusted EBITDA in the year. We were able to do that based on the strong performance across REACH and across MA, to your point, in those partner markets, they were very strong. The second goal was to really put ourselves in a position from a reserving standpoint in which we could significantly mitigate the chance of any negative development into 2024. And so I think that, coupled with this utilization outlook, which we think is prudent that the deceleration we're seeing in Q3 will not necessarily continue. That's what puts that $30 million on top of the $20 million that you take out for Hawaii. So that's really the logic around it.
Yes, I think that's right. I mean we -- as we close Q3, if you put the prior period development aside and you take the Hawaii, which was -- did perform below our expectation for the quarter out of it, our overall medical margin PMPM did perform well within the range that we had guided to. So we were very happy with that. On the other hand, that was a combination of higher revenue as we synced up our final -- got all the mid-years in and synced up our final or the next round of our [ WAF ] estimates for the year as well as some higher medical expenses that flowed through from that May and early June spike that we saw. So we just looked at the rest of the year and said, "Hey, if we just assume that there's not really any further moderation, and we want to continue to make sure that we minimize the possibility of any prior period development bleeding over into 2024. That's an additional $30 million of just medical expense versus what we had in our previous guidance. And we were able to do all that in addition with the strength of the ACO REACH business and the other actions that we just identified, we're on top of that in a position to essentially slightly increase our overall adjusted EBITDA guidance for the year.
We will now take our next question from Gary Taylor from Cowen.
A couple of questions. The first one, just on Hawaii. Why was that -- I think I'm talking at the EBITDA line, if I wrote it down correctly. But why was that an $11 million drag in the third quarter when I think in the first half, it was only a $2.5 million drag? Like in all those sort of metrics on Hawaii got quite a bit worse in the third quarter.
Yes. I mean, generally speaking, as we move through the year, seasonally, our medical margin does drop down and in Hawaii and particularly, that drives some pretty big deleverage because it's a delegated model, we have a higher cost basis there. And so in the third quarter, without any acceleration of medical margin, you would see a bigger and bigger drag on EBITDA as we move through. And your math is correct, the absolute drag of Hawaii in the quarter was about $12 million of the EBITDA line.
Got it. So there's normal seasonality. The second one was, I guess, I'm kind of probably not the only one struggling a little bit with the negative PYD that keeps showing up, but the concept also that you've been building reserves. But net-net, how do we think about carrying? End of the day, the PYD says the prior year's cohort medical margin wasn't as good as you originally thought the boosting of reserves says this year, medical margin won't be as high. How do we think about carrying that in over the next couple of years? Like how does this development reserve building impact how you're thinking about modeling the cohort medical margin development?
I can start, Tim, and you can jump in. I mean, Gary, I think the thought process is we've been building this year to remove the issue for next year. And so this year, we have the impact of the negative PPD in our results. We also have the impact of the build, which is there to try and put us in a position where we don't have it going forward next year. So I think you step off with a very strong run rate across reach and MA. I think we have a plan to not have PPD recurring, which we've had in the last couple of years. So that should really help as we do that. And then we have a really strong class coming in for '24 that just told you is larger than we had previously told you. And the performance of that as we expect to be at the very high end of our typical margin range. And so all of those things, I think as you look towards '24, are real positives for us. And that's been our goal, our goal has been to deliver on the EBITDA for this year and to try and take this issue around negative PPD off the table. And we've had the ability to do that because of the strong performance.
We will now take our next question from Jailendra Singh from Truist Securities.
One quick clarification if you're willing to share, who was the buyer for Hawaii asset? Was it a related party? And then my main question is on the topic of utilization trends. Like one of your fair partners, Humana, talked about recently about having higher MLR from their PPO book of business, which I believe is like almost more than half of your membership, curious what you are seeing in terms of utilization trends among PPO members versus HMO members. And one more kind of same topic, like some payers have talked about offering Flex cars and OTC benefits as they view this as an advantage next year, how have been your discussions with them about getting reimbursed for this pressure as you are effectively taking on for them?
So that's 3 questions in one. I'll go down -- so the buyer is an experienced California-based organization that's really strong in delegated model services, so really strong claims, very strong in customer service. And we think they're a great match with MDX Hawaii. So that's the first point. In terms of our PPO experience, we've talked about this before. We are probably the largest risk-based player in terms of PPO in the country. Our PPO business is just over 50% of our membership, it's also the fastest-growing component. And our PPO business is -- performance is in line with our HMO business. And I think the reason for that is the differences in our model. a large payer with a broad network versus our high-touch PCP patient model, which has the ability to guide that, that patient on where they're going to go for specialty care. At our Investor Day, we shared over 90% of specialty referrals come through that primary care physician even in a PPO model that is allowing us to really deliver cost-effective care. And so our PPO experience is very strong. And then just on the -- I think your question is on the Flex card side, our conversations with folks is that on an aggregate basis, we're probably seeing a reduction in those year-over-year in terms of the total dollars across our population that will be out on those.
Our next question comes from George Hill from Deutsche Bank.
Question one is actually pretty similar to Jailendra that I want to ask on a higher level than I know that kind of going into the back half of the year, we were looking at a lot of the Medicare Advantage ACOs to ratchet back on benefit design, given the kind of the poor rate environment for '24, the changes to the risk model, but it looks like most of them preserved benefits despite kind of the tough rate environment. I was wondering if you could -- and you kind of communicated that the negative rate environment would flow through to their risk-bearing care providers. So I guess maybe just walk us through at the high level how you guys are thinking about what looks like a preservation of benefits and what looks like a tougher revenue environment? And just kind of how you're thinking about that with respect to medical margin and EBITDA margin?
Yes. So I mean, George, I think our outlook on '24 is strong for the reasons I talked about strong run rate, strong class coming in for '24 being able to manage that new risk adjustment model very well. I think our view on aggregate supplemental benefits is that there will be a net pullback on those for the markets that we're in. It's a market by market basis, we don't have a lot of [ D-SNP ] in our markets. And so that's kind of the view. And so that would be neutral to a net tailwind, meaning the supplemental benefit experience from '23 to '24. But the big drivers are really the step-off in the run rate, the strong class of '24 and then just our ability to drive kind of medical margin maturation year-over-year.
And George, I'd add one thing. I think when you look at our geographic exposures relative to some of the companies you were referring to, we've got a very different geographic exposure and a very different patient population that we serve and it -- that's what really drives our view that the changes that you talked about are very manageable for Agilon.
Our next question comes from Whit Mayo from Leerink Partners.
Yes, I've got just two really quick ones. But Steve, you've talked about the investments you guys are making the 60s blind spots on trend. Is there any update or anything you can share that gives you confidence that you're seeing some of these gaps begin to be improved here. And I just want to be clear, I'm a little confused here on the guidance here. But I think it was last quarter, you guided to a $30 million increase in the reserves within the third quarter range. And so are you saying there's another $30 million now on top of the previously contemplated $30 million? Sorry, I'm just -- I think some of us are a little confused.
Yes. So maybe we can start with the guidance. I think last time we talked about a total of $60 million in Q3. And what we are saying is there's another $30 million in Q4. So that is the combination of those two. I think it's -- and we're able to do that because of the strong overall performance across. In terms of the investments that we're making, I do think we're making progress with our payer partners on the data submission, I think it's really in terms of getting that information from there where we're getting it in a faster way. But the other thing I would tell you is that our REACH real-time data tells us that we're tracking pretty well with that. and that we're able to manage things within kind of the ranges that we expected. Inpatient continues to be down, outpatient is up, but we will take that trade kind of all day long. And that's true across MA and across reach. But those investments we made in the new data [ offset ] that we brought in is making us better for that. And we are building reserves so that we make sure we're adequately reserved if development does come through.
So can I ask just one -- I'm sorry, I'm confused on this point. So does the guidance -- is it 60 total? Or is it 90% now?
Yes. So when we came through the second quarter and said, "Hey, we're going to basically increase our outlook for reserves because we didn't feel like we had clear visibility to some of the increases in utilization that the big payers were putting out there. So now we've actually seen that. We did see that spike up in May, moderated a lot in June came back down, continued to come down in July. So we've now accounted for that, and that was part of our increase on a rate basis of $60 million of medical expense that we had forecast or that we had guided through coming out of the second quarter, and we've accounted for that all now in the third quarter results that we just published. As we look forward now to the fourth quarter, we've assumed that even though the spike in May clearly has moderated down some that we're not going to forecast that it's going to moderate any further. Our guidance as [ than ] anticipated, it's going to moderate any further, and that would essentially represent an additional $30 million that we've put into our medical margin or medical expense forecast for the second half and is the second reason why our medical margin forecast for MA has come down. So it's an additional $30 million makes this assumption that we don't see any further moderation in claims and puts us, we believe, in a strong position to minimize the possibility that they'll be negative development bleeding through into 2023. And we're able to do that at the same time maintaining and actually slightly raising the midpoint of our overall EBITDA guidance for the year.
We will now take our next question from Adam Ron from Bank of America.
I have a quick one on cash flow. It seems like adjusted EBITDA is up $50 million year-to-date year-over-year, but cash flow from operations is actually down. It looks like working capital used to like doubled year-over-year. Just wondering what's driving that if it's a timing thing, if we should expect a step-up in Q4 and just generally, how we should think about EBITDA flow through into cash flow?
Yes. Adam, thanks for the question. It's always a timing thing with cash flow. Our cash flow when you think about what our final payments and our big payments come in from payers when years are settled up and our surplus has settled up. We typically do have a lag of when EBITDA is generated versus when we actually see the cash for that. So the fact that we have a large number of new payers and a large number of new markets this year, I think something like 34% of our members are actually from this year or on the MA side are actually from our new markets, that just exacerbate that delay essentially. So we always say that the EBITDA that we're generating this year, we would primarily see transitioning into improved cash flow in the next year. So a little bit in the third quarter, we also have just with the sheer number of new payers and new markets we have, a little bit of a timing issue even when we would normally get some of that settled up. And so we've got maybe a little bit less cash payments that came in, in Q3 that will come in through the rest of the year. But yes, we always do have a timing disconnect between the EBITDA we're generating and the cash that we receive for that performance.
Our next question comes from Jamie Perse from Goldman Sachs.
First one, quick clarification on the $60 million medical margin reduction year-to-date. How much of that is from changes in utilization patterns that you've seen and how much of that is from insulation from future prior period development? And then my second question, just on new membership for next year, the 25,000 from ACO REACH, 110,000 from MA. You guys have previously said a higher starting point on medical margin per member per month next year. I think you've given above 60 for that on the MA side. How should we think about that in light of what you're seeing in utilization and the new reserving policies? And a similar question on ACO REACH. Just do you think that 300 basis point delta versus the benchmark can hold next year?
So maybe I'll take the second one first. So I think that we think we're going to be really strong with this class of '24 next year and above that $60 PMPM range. And our implementation has gone extremely well around that. It's part of our investments in technology, it's part of a faster sales cycle. So there's a longer implementation period. So that's on the MA side. On the REACH side, we believe our performance is going to carry forward for us. it's always in relation to sort of the net or the macro utilization across the entire Medicare book overall. But every year, we've been positive relative to that. Anywhere from 100 to 300 plus basis points. So I think we would expect for all the reasons we talked about earlier that, that would continue for us on a go-forward basis.
Yes. And just to follow on then back to the beginning of the question, just a follow-on the REACH question. I mean one of the things that we've talked about is, of course, REACH, the revenue benchmark, what we're going to get paid basically moderate within the year based on utilization. So as long as we continue to do a good job, which we have been doing of outperforming the Medicare fee-for-service benchmark performance, then we'll continue to drive really good performance on REACH. And as I said, this year, we're outperforming that fee-for-service population with our population by over 300 basis points. So our model is really having a very positive impact. You really see the power of the model on that. It was previously largely unmanaged population. In terms of the guidance that we put out for medical expense in the fourth quarter, we basically -- just to reiterate what I said before, looked at it and said, "Hey, let's kind of take a viewpoint that the spike up in May that moderated down somewhat in June or July is not going to moderate down further. And based on that, we're putting a full another $30 million into our medical expense forecast and therefore, medical margin forecast for the fourth quarter. We think that, that puts us in a good position to meet one of our key objectives, which is let's be appropriately reserved to not have any prior period development bleed over into 2024. So I don't really want to divide it into categories as much as to say, we've looked at the expense trend and tried to be appropriately conservative to try to make sure that we're preventing or at least minimizing the possibility of that happening again next year.
We will now take our next question from Elizabeth Anderson from Evercore ISI.
One, I was hoping you could speak sort of qualitatively on -- I know you've talked about how you're doing like additional reserving and et cetera, to help sort of manage the utilization as we go into 2024. I think you also mentioned in your prepared remarks that you're also making some qualitative changes on that. So just wondering how you're sort of thinking about that in terms of the claim management. And then secondarily, I was just wondering if you could comment a little bit further on the change in trajectory of geographic entry costs and how that has sort of trended versus your initial expectations given the large size of the class.
So I think Elizabeth on your first one, Tim walked through the math on the reserves I think maybe what you're speaking to is we have added an SVP of Data Solutions and we are working much more closely with payer partners to make sure that we're getting data in a consistent and more ready fashion around that. So we're making progress on that. Do you want to talk about [ Geo entry cost ]?
Yes. And I was just going to say to add to that, I wasn't quite -- Sorry, I wasn't quite sure what the promise of the first question, but if it's along those lines. The other thing I would say is on prior period development, is, obviously, we did have some key partner market prior period development in this quarter that we just reported. But the good news is I think we have made a lot of progress across our broad payer platform with improving the quality and the timeless of information we're getting from them. And if there's a silver lining in this quarter, it was really a very specific limited issue with one payer or so. Hopefully, the investments we're making there, the investments and more expertise that we brought in is going to really pay some dividends going forward. So we're feeling better about that. On geographic entry costs, we were a little bit better in the quarter than what we had forecast. We're trying to make sure that we're -- since that's a new way that we're essentially reporting our adjusted EBITDA this year that were a little bit conservative in that number, but we were a little bit better. But I think overall, going forward, our all-in geographic entry cost, we still expect to be in the $400 to $600 range per new member that we're implementing for the next year. So I think we're going to see a lot of movement and we'll probably continue to stay within that range.
Our next question comes from Sean Dodge from RBC Capital Markets.
This is [ Thomas Keller ] on for Sean. Just one on the higher acuity clinical programs you'll have. [ What are the ] capabilities already fully implemented across your -- the older cohorts 2018, 2019, I guess, during '22. And so I guess what I'm looking to understand is kind of your latest thoughts on the upside from the rollout of these capabilities for those cohorts that are already generating medical margin PMPMs kind of north of [ 200 ].
Yes. Thanks for the question. I mean I think one of the distinctive parts of our model is that we've got these programs around renal around [ Palitive ], around high risk management, that are part of the care team around that PCP, we get much better enrollment rates and much better impact. That's how we're able to drive things like negative inpatient trends. To date, we're about 60% implemented across all of our markets. Our earliest markets do have those rolled out. By the end of '24, we will have them across all markets, including the '24 markets. So I think we do believe that there's upside from these clinical programs as we get them more fully rolled out.
Our next question is from Jack Wallis from [ Gargenheim ].
Just to follow up to the cash flow question from earlier. And just given the moving pieces this year as to help Hawaii larger than normal new class for next year on the incremental reserves. You do think that it's reasonable to expect the free cash flow positive next year? Is that maybe more realistic for '25 time line?
Yes, we're still -- and one of the things I said before is we had a very big class coming this year. Actually, that class is performing really well. I think that 2020, cut to 2024 is going to be even a stronger performing class, but we're really pleased with the large class that we brought in this year and where they're performing as all of the incremental EBITDA that we're generating this year kind of flows into next year, that's going to be a big supporter of our ability to generate positive free cash flow this year. So for now, we're still on track to do that. Obviously, we'll keep you updated as things move through next year. But we're -- at this point, we're still feeling pretty good about next year being kind of a transition year into positive free cash flow.
We have no time for further questions. So with that, I will hand back to Steve Sell, CEO for final remarks.
Thank you, everyone. We appreciate your interest in our company, and we look forward to updating you on future calls. So we will talk to you soon.
This concludes today's call. Thank you all for your participation. You may now disconnect your lines.