Bottom line. Humana’s 2Q26 headline “beat” looks more “in-line” against the proper context. Healthy reserves serve to support the EPS floor, not an impending earnings inflection. Overall, the print provided greater confidence that 2026 pricing and reserving assumptions are holding, but It did not provide the same validation for 2027, where the story becomes tougher for investors to underwrite. In 2027, HUM will need to capture better year-two economics from its 2026 cohort of new individual Medicare Advantage members, with full-year individual MA membership still expected to grow ~25%. At the same time, the company intends to exit plans affecting approximately 600,000 members and reduce benefits elsewhere. Management has shown that the new 2026 cohort is not creating an obvious early cost or engagement problem; however, it has not yet shown how much of that cohort it can retain, what the retained mix will look like, or how those moving pieces translate into durable Insurance segment margin.
That said, the lackluster stock reaction to the quarter was consistent with the large hurdle created by the run-up in the stock since the 1Q26 report. From the April 28 pre-1Q close through July 28, Humana shares increased 69.2%, while Bloomberg 2027 adjusted EPS consensus increased only 7.1%. The stock declined 6.0% on the report, a reaction consistent with a higher expectations bar rather than a weak quarter. By August 7, it had recovered to $385, only 1.0% below the pre-print close, while consensus was another 1.6% higher. In other words, the market largely retained the recovery narrative after the initial reset.
The headline “beat” looks more “in-line” with context
Adjusted EPS of $7.61 beat FactSet consensus by 5%, and full-year EPS guidance of “at least” $9.00 was reaffirmed. Overall, this was a decent enough quarter to keep the recovery story on track for now, but not enough to assume any near-term upside to management’s targets. The underlying drivers of the quarter’s 5% EPS beat were also modestly less impressive than the headline suggests. HUM’s 91.1% consolidated MLR was directly in-line with consensus, while the 91.2% Insurance segment MLR missed consensus by 16 bps. It is important to note that the Insurance segment MLR was in line with management’s qualitative guidance of “slightly above 91%.” The $7.61 2Q26 EPS was also in line with management’s guidance that 80-85% of full-year EPS would fall in the quarter.
The profile of the 5% EPS beat was more consistent with modest overall revenue upside than with a true margin-driven operating beat—the latter would have been more supportive of the stock here. Similar to EPS, adjusted pretax income beat consensus by approximately 5%. Revenue and medical membership were 0.7% and 0.9% above consensus, respectively. However, the adjusted operating cost ratio was 20 bps worse than consensus. That context around the underlying drivers and reaffirmed EPS guidance moves the takeaway for the quarter to more “in-line” than the headline suggested.
The read on reserve quality also appears to support the “in-line” read of 2Q26—i.e., reserves appear healthy, but there is nothing to suggest they are incredibly conservative or leaving them a substantial cushion for future quarters. Favorable 2Q26 prior-period development declined to $53M from $161M a year ago. IBNR remained at approximately $8.9B at quarter-end, up from $7.0B a year ago and roughly flat sequentially after the substantial 1Q26 build. The quarter’s results were not driven by an unusually large reserve release, suggesting that underlying operations/trends were likely as they appear in the results. Operating cash flow also benefitted from the IBNR build and the timing of a $1.05B Medicaid state-directed payment that settled shortly after the quarter ended. All in, the quarter reduced near-term downside risk, but it did not yet establish a broader earnings inflection.
Again, the unchanged “at least” $9.00 2026 EPS guidance reinforces that distinction. Bloomberg 2026 adjusted EPS consensus was $9.10 as of August 8, essentially unchanged from $9.11 on July 28. Year-to-date adjusted EPS is already $17.91. Management expects an adjusted loss of approximately $1.00 per share in 3Q26. Bloomberg 3Q26 adjusted EPS consensus was a $1.02 loss as of August 8, closely matching management’s approximately $1.00 loss expectation. If full year adjusted EPS were to finish exactly at $9.00, then the implied 2H26 result is a loss of $8.91 per share. That highlights how little the 2Q26 print generally can serve as a conventional earnings run rate. The print therefore did more to validate the 2026 floor, than it did to raise the 2026 ceiling—which, to be clear, should be good enough to sustain the stock at this point in their recovery.
Sources: Humana; StreetAccount; Bloomberg; Graver Research analysis. Consensus providers and contributor counts vary by metric.
The growth-to-margin handoff is now the core HUM debate
Individual MA membership reached 6.454M at June 30, an increase of approximately 1.204M, or 23%, from year-end 2025 and still tracking toward roughly 25% full-year growth. Through the limited claims window available, management said cost trends for both new and existing members were in line with the planned 7%-8% all-in medical and pharmacy range, with overall performance toward the better end. New-member engagement was in line with, and on some measures better than, renewing-member engagement; dual-eligible performance was also consistent with the rest of the book. That is meaningful given the size and speed of the enrollment increase. However, the favorability assessment was based on only about four months of completed claims, and Humana did not disclose a cohort-level benefit ratio, margin, or reserve position.
The year-two margin logic itself is reasonable. Retained members should benefit from better diagnosis and payment alignment, deeper care-management engagement, and lower acquisition and onboarding costs. Management said the all-in marketing, acquisition, co-op and onboarding expense for a first-year member is approximately twice the year-two amount. The entering cohort also had characteristics that should help: in March, management said approximately 70% of new members were switchers, approximately 70% were enrolled in four-star-or-better contracts, and roughly 75% came through more attractive channels. Still, these are mechanisms and mix attributes, not a quantified earnings bridge. Humana has not disclosed the retained-cohort denominator, per-member cost reduction, risk-score improvement, or basis-point margin benefit.
The 2027 bid cycle is where that bridge becomes more difficult. Humana is exiting plans that affect approximately 600,000 members and expects to recapture just over 40%, similar to its experience in 2025. At exactly 40%, 360,000 affected members would not be recaptured; management’s “just over 40%” expectation implies a modestly lower residual. Humana also did not disclose how much the exit pool overlaps with the 2026 growth cohort. The central tension is that the benefit changes and exits intended to repair margin could interfere with retaining the members expected to produce the year-two lift. Management expects meaningful 2027 margin progress, but repeatedly declined to quantify it because final membership size and mix remain uncertain.
Sources: Humana 2Q26 earnings release and earnings call; Graver Research analysis.
Expectations already require an Insurance-led recovery
Post-print consensus as of August 8 leaves little ambiguity about where the recovery must come from. Bloomberg adjusted EPS consensus increases from $9.10 in 2026 to $16.34 in 2027 and $27.27 in 2028. The nearly 80% increase expected in 2027 comes with only 3.0% revenue growth. Over the same period, Insurance operating income rises from approximately $170M in 2026 to $1.46B in 2027 and $3.03B in 2028, while CenterWell increases much more modestly from $1.61B to $1.73B and $1.86B. The direction is clear: top-line growth and CenterWell cannot carry the modeled earnings recovery. Insurance execution has to do the work.
At the August 7 close of $385, Humana traded at approximately 23.6x 2027 Bloomberg consensus EPS and 14.1x 2028. That relationship reflects the scale of the expected earnings recovery rather than, by itself, proving the stock is expensive. It also creates a demanding proof requirement. The 2027 EPS estimate range remains unusually wide at $10.39-$20.75, and the stock has already recovered nearly all of its initial post-print decline. With the 2026 floor better supported, incremental equity value increasingly depends on evidence that Humana can reach its at-least-3% sustainable 2028 pretax-margin objective. The 2Q result improved the probability of that outcome at the margin, but it did not establish it.
Source: Bloomberg; Graver Research analysis.
What matters from here
The next tests are more specific than another quarterly EPS beat. Medical-cost trend and reserve development must remain stable through the seasonally difficult second half; the final 2027 benefits and exits must preserve an economically attractive mix; and AEP must show that Humana can retain enough of the 2026 growth cohort and recapture enough of the exit-affected membership to realize the expected year-two benefits. Final CMS Stars data and Humana’s December 10 investor update should also provide better evidence on benefit-year 2028 economics, preliminary 2027 membership and the path toward Insurance margin recovery.
For now, 2Q26 supports greater confidence in 2026 delivery without transferring that confidence into 2027-28. Stable trend, stronger-than-feared retention and a quantified Insurance margin bridge would strengthen the recovery case. Worse exit-related attrition, weaker mix, adverse reserve development or a disappointing Stars outcome would weaken it. The quarter indicated that the current growth cohort has not broken the near-term model. Whether Humana can keep enough of that cohort, and earn an adequate margin on it, remains the thesis-defining question.