UnitedHealth Group (UNH): A Deep-Dive Value Investing Study

Hunting ten-year five-baggers among great growth stocks — pressing the upside question: "Can it get much bigger?"

• QHow high is its market ceiling? Is it enlarging an existing pie, or creating an entirely new market?5/10

  A

  Conclusion: The ceiling is still enormous in absolute dollar terms, but UNH has long been the biggest bite of this pie. What it does is "make an already mature, and increasingly regulation-squeezed, existing pie bigger and deeper," rather than create an entirely new market. Measured by Baillie Gifford's LTGG yardstick, this is precisely the root reason it struggles to deliver "5x in ten years": when you are already the whale in the pool, the remaining incremental space is limited, and every extra bite is more likely to draw the regulator's harpoon.

  First, how big is the pie itself and how much of it does UNH hold. The report notes that 2025 consolidated revenue reached $447.567 billion (official basis $447.6 billion, up 12% year over year), which by itself is close to one-tenth of total U.S. national health expenditure. It is not a startup staking out claims in a trillion-dollar blank-slate market, but a super-platform that has already swallowed a large slice of the U.S. healthcare payment and services system. On Medicare Advantage (MA), its most important growth runway, KFF data show that in 2025 it already covered 34.1 million people, or 54% of eligible Medicare beneficiaries (just 19% in 2007); in other words, the biggest penetration dividend of "moving more people from traditional Medicare to MA" has already run more than halfway.

  Is there still incremental room? Yes, but it belongs to "enlarging the existing pie" rather than "creating a new market":

  • Natural expansion from an aging population. The CBO projects that MA's share of all Medicare beneficiaries will rise to 64% by 2034, and with baby boomers continuing to turn 65, the pie is still growing; but this is a curve whose slope is flattening, as MA enrollment growth from 2024 to 2025 already slowed from 7% the prior year to about 4%.
  • Deeper penetration into "healthcare services" (Optum). The real new space lies in Optum keeping more of payers' money circulating within its own care, pharmacy, and technology systems, rather than on the insurance side. The report discloses that Optum already supports more than 122 million consumers and that Optum Rx processed about 383 million adjusted scripts in the quarter; this brings "more links of the existing pie" in-house rather than opening up a new pie.

  More importantly, Baillie Gifford would ask whether "the ceiling is truly reachable." Two hard constraints lower the effective ceiling here. First, the power to distribute the pie's profit is in the government's hands: the report notes that 44% of 2025 revenue came directly from CMS, and although CMS's final MA/Part D payment policy for 2027 brings an industry payment increase of about 2.48%, more than $13 billion (a short-term tailwind), payment rules, star ratings, and risk adjustment can be tightened in the other direction at any time. Second, sheer size has itself become the enemy of growth: the company has explicitly projected that UnitedHealthcare will lose 1.3–1.4 million MA members in 2026 (some measures even put it at 2.8 million), actively exiting unprofitable counties; this shows that in the current regulatory and cost environment, the marginal return of "getting even bigger" is negative.

  In one line for Baillie Gifford: UNH's absolute market ceiling is very high, but it is "the largest holder of an existing pie," relying on rising penetration plus deeper moves into the service side rather than creating an incremental market; when the penetration dividend is past halfway and greater growth increasingly bumps against the regulatory profit ceiling, this kind of ceiling gives weak support to "5x in ten years."

  Jun 11, 2026
• QCan its revenue at least double over the next five years? Is growth driven mainly by volume, price, or new businesses?3/10

  A

  Conclusion: Almost impossible. Doubling revenue over the next five years (which requires a compound growth rate of about 15%) is far beyond UNH's realistic growth center; growth will be led by "volume," with "price" capped by regulation and "new businesses" too small to carry a doubling. This is the item where it falls furthest short in Baillie Gifford's ten growth questions.

  First, lay out the hard numbers a doubling requires. The report discloses 2025 revenue of $447.567 billion (official $447.6 billion). Doubling in five years means reaching about $895 billion in 2030, a compound rate of about 14.9%. Yet UNH's own 2026 revenue guidance is only "greater than $439 billion"; note that this figure is below actual 2025 revenue, because the company is actively exiting loss-making MA counties and expects to lose 1.3–1.4 million MA members in 2026. For a company whose next-year revenue guidance is "flat or even slightly lower" to suddenly run 15%+ annually over the following four years lacks a realistic basis.

  Next, look at the historical growth center. The report's main financial table shows that from 2021 to 2025 revenue grew from about $287.6 billion (i.e., $287.6B on a billions-of-dollars basis) to $447.6B, a CAGR of about 11%–12%, and this curve is decelerating: growth of 14.6% in 2023, 7.7% in 2024, 11.8% in 2025, and a sharp drop to 1.9% in Q1 2026. Even maintaining the 11%–12% center of the past five years would put revenue at about $750–790 billion in five years, still a clear gap from doubling; and the current deceleration signals show that even holding the old center is not easy.

  Breaking down the three engines of "volume, price, new businesses," one by one, here is why they cannot support a doubling:

  • Volume (enrollment/prescriptions/service visits) — the main engine, but slowing and even reversing. MA penetration has reached 54% per KFF, with growth slowing from 7% to 4%; UNH itself is still shrinking its MA membership. Optum's number of people served (over 122 million) and Optum Rx's prescription volume (about 383 million adjusted scripts per quarter) remain the most tangible sources of incremental growth, but they are mid-single-digit to high-single-digit organic growth, not doubling-level momentum.
  • Price (ability to raise prices) — capped by regulation. The report accurately notes that UNH's pricing power is "present, but limited," bounded by contract cycles and regulatory limits, rather than Coca-Cola-style direct end-price increases. Premium rates depend heavily on CMS payment policy, and it was precisely because "price (payment) failed to keep up with medical costs" that 2025 operating profit fell about 41% year over year (2024 $32.3B → 2025 $19.0B). The path of doubling driven by price increases is essentially blocked.
  • New businesses (Optum deepening, value-based care, technology/AI) — right direction, but too small and just stumbled. Optum is the real deepening engine, but in 2025 Optum Health turned to an operating loss (about −$278 million on the official reported basis, and about −$1.115 billion as the report records on the company's restated Q1 2026 basis), showing that value-based care expansion is currently in "repair" rather than "re-acceleration." It can contribute increments, but it cannot recreate another UNH of the same magnitude within five years.

  The honest conclusion for Baillie Gifford: doubling UNH's revenue in five years is unrealistic, and the reasonable expectation is mid-single-digit to low-double-digit organic growth (and 2026 may be flat because of the active contraction). The base color of growth is "volume-led, price-limited, new businesses passing the baton but insufficient in scale." This is a steady compounding machine, but not the "double in five years" high-growth target that Baillie Gifford is looking for.

  Jun 11, 2026
• QFive years from now, what will take over as the next growth engine? Does this "second curve" exist today?5/10

  A

  Conclusion: The second curve does exist today, and it is called Optum, especially Optum Health (value-based care plus an owned physician/clinic network) and Optum Insight (healthcare data and technology/AI). But it is currently a successor "stepping up while ailing": Optum Health just turned to an operating loss in 2025, and it must first prove it can earn money steadily before it can "take over as the next growth engine."

  Why say the second curve already exists rather than being wishful thinking? Because it already contributes a real revenue base, not a slideware story. The report discloses that, on the restated Q1 2026 basis, 2025 revenue for Optum's three businesses was: Optum Rx about $154.726 billion, Optum Health about $100.051 billion, and Optum Insight about $21.034 billion. Optum as a whole supports more than 122 million consumers. In other words, relative to the "first curve" UnitedHealthcare (insurance, 2025 segment revenue of about $344.903 billion), Optum's curve of "extending from payer toward healthcare services/pharmacy/data" already matches or even exceeds the core insurance business in scale. This is exactly the structural difference that sets UNH apart from pure insurers (such as Elevance and Humana).

  But what Baillie Gifford would press on is "whether this second curve can carry future growth, and whether it is healthy." Three layers must be pointed out honestly here:

  • Optum Health (value-based care, which should be the sexiest second curve) just took a fall. In 2025 it went from huge profit to an operating loss (about −$278 million on the official reported basis; the report records about −$1.115 billion on the company's restated Q1 2026 basis), due to loss-contract reserves, rising medical cost trends, and structural adjustment in the value-based care business. The report accurately flags from this: "building a bigger platform does not equal building stronger returns." A curve that is still losing money and needs to rebuild its margin can only be counted as a "potential engine," not an "already-ignited engine." The good news is that in Q1 2026 Optum Health's operating profit already returned to about $1.1 billion, and the recovery is happening, but it needs more quarters to prove it is not a "toothpaste-squeezing rebound."
  • Optum Insight (data + technology + AI) is the highest-quality but smallest curve. 2025 revenue was about $21.034 billion and operating profit about $3.461 billion, the healthiest margin of the group (about a 16% operating margin). Massive claims, prescription, and clinical data are UNH's true long-term assets, and there is plenty of room to use AI for actuarial work, coding, and operational efficiency. But at $21 billion in scale, it cannot become the main engine driving the entire group's growth in the near term, and is more like a "profit-quality ballast plus a long-dated option."
  • Optum Rx (PBM) is a cash cow rather than a growth curve, and it sits in the regulatory crosshairs. 2025 revenue was about $154.726 billion and operating profit about $7.193 billion, the largest in scale, but it is a mature middleman business and sits in the "bull's-eye" of the FTC investigation: the FTC found that the three largest PBMs control about 80% of the prescription market, and alleged that their markups on specialty generic drugs generated more than $7.3 billion in excess revenue over six years. Its regulatory beta is too high; it is not a growth pole you can extrapolate with confidence.

  So "what takes over five years from now"? The most likely answer is: Optum Health completes its margin repair and again becomes a dual growth-plus-profit engine, while Optum Insight uses data/AI to keep improving overall profit quality. This second curve physically already exists today and its revenue already has scale, which is UNH's real trump card; but its current state is a "successor still in repair" rather than a "successor who has firmly taken the baton." For Baillie Gifford, this means the existence of the second curve can score high, but the certainty of its "sustainable high-quality growth" must be discounted, which is also one of the core reasons the report assigns "Watch" rather than "Buy."

  Jun 11, 2026
• QWhat is its core competitive advantage? Will this moat widen or narrow over the next three to five years?6/10

  A

  Conclusion: UNH's core competitive advantage is the trinity moat of scale, data, and licensing that comes from deeply vertically integrating "payer + healthcare services + PBM + data technology"; this moat is still very wide and very deep overall, but its direction over the next three to five years is "broadly stable, locally narrowing," as regulation is systematically chipping away at this moat.

  First, where the moat is strong, a part Baillie Gifford would score high and one backed by hard data:

  • Scale advantage (very strong). The report discloses that UnitedHealthcare served about 49.1 million consumers in the quarter, and Optum supports more than 122 million consumers. Enormous scale translates directly into stronger actuarial capability, bargaining power with hospitals and drugmakers, network density, and expense-amortization capability; in a business whose core is controlling medical costs rather than selling healthcare, scale is itself pricing power.
  • Data advantage (very strong). Massive claims, prescription, utilization, and clinical data form an ammunition depot for pricing, risk adjustment, and cost management, hard for competitors to replicate in the short term.
  • Licensing and regulatory barriers (very strong). Insurance, PBM, managed care, banking, and data compliance create extremely high entry barriers; to take on UNH simultaneously across insurance, PBM, physician networks, and government programs, only a few leading companies in the U.S. can do it.
  • Market position (leading). In the MA market, per KFF research, in 2024 UnitedHealthcare and Humana held the highest market share in about two-thirds of counties, together accounting for about 59% of MA enrollees. UNH is a "super-platform player across multiple profit pools," rather than a single-niche champion.

  But what Baillie Gifford truly cares about is whether the moat widens or narrows over the next three to five years, and here one must be honest: the direction is "locally narrowing," with three specific pieces of evidence, all pointing to regulation rather than technology:

  • The PBM-side moat is being cut by regulation. The FTC found that the three largest PBMs control about 80% of the prescription market, and alleged that their specialty-generic-drug markups generated more than $7.3 billion in excess revenue over six years. Optum Rx stands right in the bull's-eye, and the compression of PBM economics by regulation in the future is a high-probability event, which would directly thin the most profitable segment of the moat.
  • MA-side "excess profit" is being watched by the judiciary and Congress. The company has confirmed that the DOJ is conducting both criminal and civil investigations, focused precisely on MA risk-adjustment coding (whether "upcoding" to inflate reimbursements); the Senate has also publicly questioned its coding practices. The report accurately classifies this as "regulatory beta far higher than surface defensiveness." Any tightening of CMS risk adjustment, star ratings, or audit standards would rewrite the share of profit UNH captures.
  • The "synergy myth" of vertical integration was partly disproven in 2025. The market used to assume "the bigger the platform, the more synergy, the more automatic the quality and efficiency gains," but Optum Health's operating loss in 2025 and UnitedHealthcare's operating profit falling about 40% year over year (down to $9.4 billion) show that the moat's "automatic widening" did not happen unconditionally.

  What will not disrupt it? Technology. The report's judgment is accurate: health management needs technology and AI, but technology mostly changes efficiency and cost structure and is unlikely to make a large insurance/managed-care platform "lose its position instantly." The real disruptive force is regulation, and regulation's method is to "thin out its excess profit pool," rather than "make UNH disappear."

  The conclusion for Baillie Gifford: the moat's "width/depth" is still industry-leading and irreplaceable in the short term (this scores high); but its "future trend" switches from "clearly widening" to "broadly stable, locally (PBM/MA excess profit) narrowing." Baillie Gifford prefers companies whose moat is widening; UNH satisfies "wide" but not "becoming wider," which is the key gap between it and an ideal LTGG target.

  Jun 11, 2026
• QIf its core business is disrupted, does it have the DNA to reinvent itself? How does it treat mistakes and bad news?5/10

  A

  Conclusion: UNH's "self-reinvention DNA" is medium. It has a successful track record of remaking itself from a single-line insurer into a "payment + services + PBM + data" platform (Optum is the evidence), but the way it handled the bad news of 2025 leaned toward "repairing scale and reporting basis," rather than candidly admitting to shareholders that the old earnings model may be structurally re-rated, which discounts its "sincerity about reinvention."

  First, fill in the implicit premise of this chained Baillie Gifford question: when the core business is disrupted, does the company have the DNA to reinvent itself? How does it treat mistakes and bad news? These two questions must be viewed separately.

  The first question, reinvention DNA: yes, with historical evidence. UNH's most powerful reinvention evidence is that over the past decade-plus it remade itself from "the largest health insurer in the U.S." into the Optum healthcare-services/pharmacy/data empire. The report discloses that by 2025 Optum's three businesses together already matched the core insurance business in scale (Optum Rx about $154.726 billion, Optum Health about $100.051 billion, Optum Insight about $21.034 billion). This shows it is not a company that would collapse from a single-point disruption; it has the organizational capability and capital strength to "migrate deeper into the value chain and recreate growth poles." And its true moat (data + scale + licensing) gives it room to maneuver even when the core insurance business is under pressure, rather than being cleared out overnight by technology.

  But Baillie Gifford also looks at "whether the reinvention is evolving toward higher quality." There is a warning here: the 2025 operating loss in value-based care (Optum Health) occurred precisely in the deepest area where it was most aggressively "reinventing," showing that its reinvention can also hit potholes and that not every expansion creates returns. The reinvention DNA is real, but the "success rate" of that reinvention has declined in recent years.

  The second question, how it treats mistakes and bad news: this is UNH's current soft spot most worth watching. There are several pieces of evidence:

  • Governance moves during the crisis lean toward "remediation" rather than "admitting fault." The report discloses that the board held 22 meetings in 2025 (heightened oversight, a positive signal); but in December 2025 the company additionally granted "Performance Improvement Awards," which the report accurately characterizes as "crisis-repair incentives, rather than high-quality pay design that arises naturally in an up-cycle." In other words, after performance ran into trouble, the company's first thought was "how to stabilize management," which is not the most reassuring posture toward bad news for shareholders.
  • The response to regulatory allegations is "firm denial plus attribution to external attacks." Facing the DOJ's confirmed criminal and civil investigations, the company called media reports a "sustained attack" based on "incomplete data" and "misunderstanding," and stressed that CMS's independent audits affirm its coding accuracy and that a special master had found no evidence of wrongdoing. This response may in fact hold up, but its posture is defensive; Baillie Gifford prefers management that "candidly acknowledges problems and reviews them openly." One point to its credit is that the company also committed to bringing in third-party oversight of risk-adjustment coding, which counts as a pragmatic corrective move.
  • The truly candid side shows in "proactive contraction." The company is willing to admit that part of the MA business is unprofitable and actively exits, expecting to lose 1.3–1.4 million MA members in 2026; this is the rational discipline of "bowing to bad news and cutting loss-making businesses," a plus. It shows that at the business level, management does not shy away from confronting problems.

  The conclusion for Baillie Gifford: UNH's self-reinvention DNA is real (Optum is textbook-level evidence), and it is willing to subtract when facing business-level mistakes (proactively shrinking MA); but on "governance- and regulatory-level bad news" it leans more toward defensive narrative and repair-type incentives rather than a thoroughly transparent review. On balance it is "moderately positive, but with trust noticeably lower than before 2023," consistent with the report's judgment. Baillie Gifford would agree that it "cannot die and can turn around," but would keep reservations about "whether it can candidly re-rate itself to where it should be."

  Jun 11, 2026
• QDoes management (especially the founder) have a long-term horizon and interests deeply tied to the company? Are they willing to sacrifice current profit for five to ten years out?4/10

  A

  Conclusion: In "institutional design," management aligns with shareholders (high ownership requirements, clawback provisions, long-term incentives), and its strategic language is also long-term oriented; but it is not founder-led, its absolute ownership share is very low (management and directors together about 0.19%), and post-crisis pay moves in 2025 show it is "willing to sacrifice the short term to repair scale, but not necessarily to sacrifice the short term for intrinsic value per share." The depth of alignment is medium, not the kind of "all-in net worth, founder long-distance run" structure that Baillie Gifford most loves.

  The core of this Baillie Gifford question is three things: long-term horizon, depth of interest alignment, and "whether they are willing to sacrifice the present for five to ten years out." Let me look at each.

  Long-term horizon: present, but governance concentration is on the high side. The company is currently led by Stephen Hemsley as both Chairman and CEO. The report accurately notes this double-edged sword: shouldering both roles during a crisis can raise decision-making efficiency, but it also elevates key-person and governance-concentration risk. The board held 22 meetings in 2025, indicating heightened oversight during the crisis, a signal of long-term responsibility. But UNH is not a founder company; it lacks the natural long-termism of "a founder betting the rest of their life on the company," and is more like a professional-manager-led mega-institution. The "founder-style long-term horizon" Baillie Gifford prefers can only count here as "institutionalized long-termism," and its quality must be discounted.

  Interest alignment: the institutions are in place, but absolute ownership is very thin. The report discloses two sets of key facts: first, the 2026 proxy statement shows that management and directors together hold about 1.7081 million shares, only 0.19% of shares outstanding (at the current roughly 908 million shares and a price of about $404, the absolute amount is meaningful, but its share of the company is tiny); second, the company's updated ownership requirements call for the CEO to hold stock worth 10 times base salary and other executives 2–3 times, with all named executives meeting the requirement, plus a Dodd-Frank-compliant clawback policy.

  This is the tension the report exposes: institutionally there is an effort to align with shareholders (10x-salary holdings, clawback, long-term incentive framework), but by absolute ownership share, it is not the kind of structure where "management has most of their net worth tied up in the company." For Baillie Gifford, a combined 0.19% share means management is "in the same boat" as outside shareholders, but their own plank on the boat is small; interest alignment is "adequate," not "deep binding."

  Willing to sacrifice the present for the long term? A clear warning signal appears here. Three pieces of evidence:

  • Pay was voted down by shareholders with their feet. The report discloses that the 2025 Say-on-Pay vote won only 60% support, and the company issued a dedicated response in its 2026 proxy; this in itself shows shareholders are unhappy with how its pay matches performance.
  • Repair-type incentives granted after the crisis. In December 2025 the company additionally granted Performance Improvement Awards, which the report accurately characterizes as "crisis-repair incentives, rather than high-quality pay generated naturally in an up-cycle." This implies that when performance faltered, management prioritized protecting its own incentives rather than showing shareholders a "sharing the pain with you" posture.
  • Capital allocation leans toward "repairing scale" rather than "repairing value per share." This is the most concrete piece. The report notes that the company's average buyback price was about $501.67 in 2022, about $529.85 in 2024, and about $454.82 in 2025, while the current price is only about $404; that is, the large buybacks of recent years were generally bought at prices more expensive than today, lacking the contrarian discipline of "only buying back at meaningful undervaluation." A management team truly minding intrinsic value per share would be more aggressive near the $234 52-week low, rather than buying back continuously at high prices.

  To be balanced about the positives: in 2025 the company generated about $16 billion of free cash flow, still maintained dividends (an annualized dividend of about $8.84 per share) and buybacks, and preserved continuity in capital allocation; proactively exiting the loss-making MA business also reflects some long-term rationality. So this is not a wasteful or untrustworthy management team.

  The conclusion for Baillie Gifford: management is honest, rational, and has an institutionalized long-term horizon, with interests "directionally aligned" with shareholders; but it is not the deeply founder-bound type, absolute ownership is thin (0.19%), and its 2025 pay and buyback behavior shows it is more willing to make trade-offs for "repairing scale / stabilizing the team" rather than "repairing intrinsic value per share." Both alignment and long-termism can only be rated "moderately positive"; this is exactly why the report rates "management and capital allocation" 3/5 and marks "is management trustworthy" and "is capital allocation rational" as "uncertain / uncertain, leaning weak" in the checklist.

  Jun 11, 2026
• QIf it disappeared tomorrow, how much would customers miss it? Is its way of growth sustainable, not reliant on harming society and regulation?5/10

  A

  Conclusion: If UNH disappeared tomorrow, the system would "miss it very much": it is the actual carrier of health insurance, pharmacy, and medical services for tens of millions of people, irreplaceable in the short term; but the second half of this question (whether the way of growing is sustainable and not reliant on harming society and regulation) is the crux, and UNH flashes a red light precisely here: a substantial part of its past excess profit came from links that regulators and the judiciary deem "potentially harmful to the public interest" (MA risk-adjustment coding, PBM markups), which is the biggest hidden danger to its growth sustainability.

  This Baillie Gifford question has a double implicit premise, both of which must be answered: indispensability, and social/regulatory sustainability.

  First half, indispensability: very high. The report discloses that UnitedHealthcare served about 49.1 million consumers in the quarter, Optum supports more than 122 million consumers, and Optum Rx processed about 383 million adjusted scripts in the quarter. This means it is deeply embedded in America's "utility-grade" healthcare infrastructure: the insurance claims, prescription dispensing, and healthcare-service fulfillment of tens of millions of people all run on its systems. The report accurately notes that switching costs are "moderate-to-strong": switching employers, government programs, pharmacy networks, and technology interfaces is not easy, and the licensing barrier is "very strong." If it disappeared overnight, no single player could immediately take over this payment + services + data network, and the entire medical settlement system would fall into chaos. So "would customers miss it": yes, and it is a must-have level of missing. The report rates demand stability and moat strength as pass/4 respectively, and this half holds up.

  But Baillie Gifford would press a sharper version: are customers "unable to do without it" or "fond of it"? These two differ. On NPS/reputation, the U.S. public and physician community have long been highly dissatisfied with large MA insurers' and PBMs' "claim denials, prior authorization, and opaque drug pricing"; this "can't do without it but don't like it" relationship is exactly the breeding ground of regulatory risk. Its indispensability is closer to an "entrenched must-have" than a "beloved brand."

  Second half, whether the way of growth is sustainable and does not harm society and regulation: this is UNH's structural soft spot. Three pieces of hard evidence:

  • PBM growth is accused of "harming the public." The FTC report found that the three largest PBMs (including Optum Rx) control about 80% of the prescription market, and stated that they marked up various specialty generic drugs by hundreds to thousands of percent, generating more than $7.3 billion in excess revenue over six years, on drugs used to treat cancer, HIV, and other serious conditions. Regulators directly label this kind of profit as "at patients' expense"; growth obtained this way is inherently unsustainable, because what it invites is precisely regulatory backlash.
  • MA growth is accused of "upcoding to defraud insurance." The company has confirmed that the DOJ is conducting both criminal and civil investigations, focused on whether MA risk-adjustment coding was manipulated to inflate reimbursements; the Senate has also publicly questioned it. The report accurately notes that 44% of 2025 revenue came directly from CMS; that is, a considerable part of the profit comes from taxpayers, and the way it is obtained is being examined by the judicial system for whether it "harmed public finances." The company firmly denies it, says CMS audits affirm its accuracy, and promises third-party oversight of coding; these are genuine defenses, but the very fact of "being under criminal investigation" shows that the social sustainability of its growth method is in doubt.
  • Public perception and political pressure. Large health insurers' claim-denial behavior has become a focus of U.S. public opinion in recent years, and politically "fixing the insurance/PBM middlemen" is almost a bipartisan consensus. In this environment, the bigger UNH is, the more it becomes a target.

  To be balanced: UNH's core business (giving tens of millions of people access to health insurance and healthcare) has legitimate social value in itself, and it is not an antisocial business; the problem lies in the source of its excess profit, namely MA coding and PBM markups, which happen to step into the gray zone of regulation and ethics. Among Baillie Gifford's ideal "growth that does not rely on harming society" companies, UNH satisfies only "the core business is legitimate" but not "the source of excess returns is clean."

  The conclusion for Baillie Gifford: indispensability scores high (if it disappeared tomorrow, the system would be paralyzed and customers would miss it); but social/regulatory sustainability scores low: several of its most profitable links are being simultaneously challenged by the FTC, the DOJ, and Congress as "at the expense of patients and taxpayers." On this double question, UNH is half excellent and half dangerous, and overall can only be rated "a shade below passing," and this is precisely one of the fundamental reasons its valuation multiple has been suppressed over the long term.

  Jun 11, 2026
• QWhat is this business's unit economics (gross margin, incremental returns)? Does it get better or worse as scale grows? Where is the money it earns spent?5/10

  A

  Conclusion: UNH is a classic "high-turnover, low-margin, enormous-scale" business: unit economics (gross/net margin) are inherently thin, and 2025 just proved that this thin profit is quickly eroded by cost trends; greater scale brings "stronger cost control" rather than "a step-change upward in margins." The money it earns is spent mainly on buybacks, dividends, and M&A, but buyback timing in recent years has been on the expensive side, and the incremental return on capital reinvestment can only be rated above-average, not impressive.

  This Baillie Gifford question breaks into three parts: unit economics, returns to scale, and where the money goes.

  Unit economics: thin, and fragile. UNH's business model is essentially "managing other people's money and earning a thin spread." The report discloses that the largest cost item in 2025 was medical costs of $313.995 billion, against revenue of $447.567 billion; that is, for every $100 of premium collected, about $89 is paid out directly in claims (the medical cost ratio, MCR). The MCR history the report gives: 83.2% in 2023, 85.5% in 2024, 89.1% in 2025, falling back to 83.9% in Q1 2026. This line shows the vital point of its unit economics is "controlling costs": every few points the MCR rises, a large chunk of the already-thin profit is swallowed. The direct consequence is that 2025 net margin was only 2.7%, operating margin plunged from the 8%+ of 2022–2024 to 4.2%, and operating profit fell about 41% year over year ($32.3B → $19.0B).

  This is completely different from the high-gross-margin, strong-incremental-return software/platform businesses Baillie Gifford favors: UNH is "thin margin × enormous scale," not "thick margin × compounding leverage."

  As scale grows, do unit economics get better or worse? The answer: the cost side gets better, but margins do not structurally rise because of scale. Scale indeed brings stronger actuarial, bargaining, network, and expense-amortization capability (the report rates "cost advantage" as strong and "scale advantage" as very strong). But 2025 cruelly proved that scale cannot fend off exogenous medical cost trends and regulatory payment pressure: the company serves nearly 50 million MA/commercial members and Optum covers more than 122 million people, yet its profit was still halved in 2025. Unit economics also vary widely across businesses: Optum Insight's operating margin of about 16% ($21.034 billion revenue / $3.461 billion profit) is the highest quality, Optum Rx is about 4.6%, and Optum Health was even negative in 2025. So "bigger scale → better unit economics" holds for UNH only on the "cost control" dimension, not on the "margin level" dimension: it is a business where scale determines survival but the ceiling is nailed down by costs and regulation.

  Where does the money go? Cash generation is still strong, but reinvestment quality has a question mark. The report discloses 2025 operating cash flow of $19.697 billion, capex of $3.622 billion, and free cash flow of about $16 billion (above net income to shareholders of $12.056 billion). Where this money goes:

  • Buybacks of $5.545 billion + dividends of $7.916 billion + $4.509 billion of cash for M&A and other. Cash-generation ability is beyond doubt (from 2021 to 2025 FCF exceeded net income every year).
  • But buyback timing is on the expensive side: the average price was about $501.67 in 2022, about $529.85 in 2024, and about $454.82 in 2025, all above the current roughly $404 and far above the 52-week low of $234.60. The report accurately notes that the problem is "whether to buy back at meaningful undervaluation, not whether to buy back at all."
  • M&A incremental returns are in doubt: Optum Health's M&A expansion in recent years has come with rising integration complexity and tail risk (2025 also involved an increase of about $799 million in the Change Healthcare-related loan-recovery reserve), and "building a bigger platform" does not necessarily mean "building stronger returns."

  A data point supporting "incremental returns above-average rather than top-tier": the report notes that the company's cumulative buyback amount over these years was enormous, yet common shares outstanding fell only from 941 million at the end of 2021 to 906 million at the end of 2025; the extent of share reduction is not striking, showing that large amounts of cash were consumed in "high-priced buybacks plus offsetting equity-compensation dilution," and the cumulative efficiency of per-share value can only be rated above-average.

  The conclusion for Baillie Gifford: unit economics are thin and fragile (a rise in MCR eats into profit), scale only strengthens cost control rather than raising the margin ceiling, and the reinvestment (buybacks/M&A) of earned money is above-average but not impressive. This is a real cash machine, but not the Baillie-style high-quality growth exemplar of "bigger scale, sweeter unit economics, fiercer compounding"; the report characterizes its financial quality as "still a high-quality cash-flow machine, but 2025 has proven it is not a perpetual-motion machine exempt from re-rating," consistent with this.

  Jun 11, 2026
• QFor it to rise five-fold in ten years, what conditions must hold simultaneously? Are these conditions realistic? What expectations does today's share price imply?2/10

  A

  Conclusion: For UNH to rise five-fold in ten years (about 17.5% annualized), three things must hold simultaneously: "a sharp margin recovery + sustained double-digit revenue growth + valuation multiples re-expanding from a suppressed state"; yet at the current price of about $404 with a TTM P/E of about 30x, the market already implies the optimistic expectation that "margins will recover successfully and regulation will not deal a substantial blow." In other words, the good outcome has already been partly priced in, a five-fold in ten years lacks the odds, and a five-fold is more like the "ceiling of the optimistic scenario" than a base-case expectation.

  This Baillie Gifford question needs two implicit premises filled in: what conditions must hold simultaneously for a five-fold in ten years? And what expectations does today's share price imply? Let me take them one by one.

  First, what today's price implies. A unified price anchor: as of early June 2026, UNH traded at about $404, with a market cap of about $370–375 billion (i.e., about $370–375 billion), a TTM P/E of about 30.6x, and a dividend yield of about 2.25% (the report's snapshot was $388.47 / about $352.8 billion on May 22). The report estimates: on the company's 2026 guidance of "adjusted EPS greater than $18.25", the forward adjusted P/E is still about just over 21x. A key comparison: UNH's current TTM is about 30x, while peers Cigna is about 12.3x and Elevance about 17.8x. This premium shows that the market has already priced UNH as a winner whose "margins will recover to normal levels," rather than pricing it on the bleak reality after the 2025 halving. The report accurately says: "the current price reflects recovery imagination more than deep pessimism." So buying today buys a reasonable business return rather than the odds of a clear undervaluation.

  Next, which conditions must hold simultaneously for a five-fold in ten years, and honestly assess whether they are realistic. Five-fold in ten years = about 17.5% annualized. Break it into "earnings growth × valuation change":

  • Condition one: margins must recover sharply and persistently. In 2025 the operating margin fell to 4.2% and Optum Health turned to a loss. To achieve a five-fold, it must at least return to and stabilize at the historical 8%+ level. In Q1 2026 the MCR fell back to 83.9% and operating profit returned to $9 billion, and the recovery has indeed begun; but the report stresses "more quarters are needed to prove it is not a one-off toothpaste-squeezing rebound." Realism: medium-to-low (direction is right, but it must persist for years without costs biting back).
  • Condition two: revenue must grow at double digits continuously. But the company's 2026 revenue guidance is only "greater than $439 billion" (below actual 2025), and it is still actively losing 1.3–1.4 million MA members; MA penetration has reached 54% per KFF, with growth down to 4%. Realism: low (see the revenue-doubling question; it can barely double in five years, let alone support a five-fold in ten).
  • Condition three: valuation multiples must re-expand from a suppressed state. But the current 30x TTM is already a high level held up by "recovery imagination," not a low one; for a five-fold, the market would have to award a higher multiple than today amid regulatory clouds. Realism: low (regulatory beta is high, and the multiple is more likely to be suppressed than to expand).
  • Implicit veto condition: regulation must not deal a substantial blow. DOJ criminal + civil investigations, FTC's scrutiny of the PBM, and 44% of revenue depending on CMS: any substantive tightening at any of these points would cause the three conditions above to collapse simultaneously. This is the sword of Damocles hanging over the five-fold narrative.

  Stack the three together to see the odds. The report's owner-earnings discount gives three tiers: conservative intrinsic value $230–300, fair $320–430, optimistic $460–560. The current roughly $404 already falls in the "upper part of the fair range." In other words, the market has already pushed the price to a "neutral-to-optimistic" position: a five-fold (corresponding to a share price of about $2,000) is far beyond the ceiling of the optimistic scenario ($560), and it would require "margin recovery + double-digit growth + multiple expansion + regulation cutting it a break," four positives delivered continuously for ten years, which probabilistically is blue-sky within blue-sky.

  The honest conclusion for Baillie Gifford: the three conditions required for a five-fold in ten years (sustained margin recovery, double-digit revenue growth, multiple re-expansion) are all of low realism and hang together under the sword of regulation; and today's price of about $404 has already prepaid the optimism of "successful recovery," not a discounted purchase. Baillie Gifford wants targets whose "upside imagination far exceeds already-priced expectations"; UNH is exactly the opposite: upside is limited, and the downside (the report estimates an extreme scenario falling toward $170–225, about 40%–55% drawdown) is real. This is the mathematical root of its suppressed score in the ten questions and the report's "Watch" rating.

  Jun 11, 2026
• QWhy hasn't the market realized all this yet? Is it that they can't understand it, look down on it, or can't see far enough? What will become the "narrative turning point"?3/10

  A

  Conclusion: The market has in fact already "realized" it: in 2025 UNH plunged all the way from a high of about $600 to a 52-week low of $234.60, falling more than 42% for the year, which was exactly the market voting with its feet to re-price the old earnings model. Today's rebound to about $404 reflects "the market betting it recovers successfully." So the real question is that the market has not made up its mind between "structural re-rating" and "cyclical over-punishment," rather than that the market "can't understand it"; the narrative turning point depends on which way this bet lands.

  This Baillie Gifford question asks "is the market unable to understand it, looking down on it, or unable to see far," and requires filling in "what will become the narrative turning point." First correct one premise: UNH is not an "overlooked, unpopular stock"; it is a global top-40 giant by market cap, extremely well-covered, with no information gap of "no one researching it." Its pricing disagreement is "the same smart money giving diametrically opposite readings of the same set of facts."

  Answering "can't understand / look down on / can't see far" one by one:

  • Not "can't understand it." This company is researched extremely thoroughly by Wall Street. The market fully understands its business.
  • The core is a tug-of-war between "looking down on it" and "misjudging it." What happened in 2025 was a violent "bear move": operating profit halved about 41% year over year, Optum Health turned to a loss, DOJ criminal + civil investigations came to light, and S&P and AM Best downgraded the outlook to negative in June 2025: the market beat it down from a "textbook compounding machine" to a "problem platform under high regulatory pressure." Today's price of about $404 (TTM P/E of about 30x, clearly above Cigna's about 12x and Elevance's about 18x) is the market starting to "go long on recovery": betting that Q1 2026 MCR falling back to 83.9% and guidance raised to >$18.25 means the turning point has arrived.

  So what is the real focus of the market's disagreement? The report nails it: whether "the 2025 profit break is ultimately cyclical noise or structural re-pricing":

  • The bulls (implicit in the $404) believe: the cost trend is temporary, UNH's scale, data, and licensing moat is intact, the Q1 recovery is the evidence, the old 8%+ operating margin will return, and so a 30x P/E is reasonable.
  • The bears (whom the report calls "a bear case worthy of respect") believe: rising MA costs, Optum Health capital returns below expectations, PBM political risk, and M&A negative externalities may be structural re-pricing rather than a cycle; if so, UNH's "normalized EPS" may be lower than $20 rather than above it, and then today's $404 is not cheap.

  This is why the market "has not fully realized it": the answer to this binary question has not yet been revealed, rather than out of ignorance; it needs time and more quarters of data to disprove one side.

  What will become the narrative turning point? This is where the question lands, in two sets of observable signals, upward and downward:

  Upward turning points (confirming "it is a cycle, the recovery is real," with the narrative returning to growth):

  • For 2–4 consecutive quarters, Optum Health steadily returns to positive and its margin recovers meaningfully (Q1 has already returned to about $1.1 billion of operating profit, but it must continue).
  • MCR stays stable in the 83%–84% range, with GAAP operating profit improving (rather than relying only on the "adjusted basis").
  • The DOJ/FTC investigations end or settle with "no material earnings hit," and the regulatory clouds clear: this is the strongest catalyst for multiple expansion.

  Downward turning points (confirming "it is structural, the re-rating is real," with the narrative shifting decisively toward a value trap):

  • DOJ/FTC/CMS events escalate from "noise" to a "material earnings hit" (fines, business restrictions, tightening of risk-adjustment rules).
  • MCR rises again and cannot be absorbed through pricing and cost management; Optum Health loses money again.
  • The company keeps buying back at high prices or doing highly complex M&A, turning "earnings recovery" into "stagnant value per share."

  The honest conclusion for Baillie Gifford: the market has already gone through a violent "bearish re-rating" (falling to $234.60) and is now swinging toward "bullish recovery" (rebounding to $404), rather than being "unable to understand" UNH; the essence of the disagreement is the still-unrevealed binary question of "cycle vs. structure." The narrative turning point will come from the landing of two variables, Optum Health's margin and the DOJ/FTC regulatory outcome, rather than from "the market suddenly understanding." For investors like Baillie Gifford who look for "great growth the market has not yet realized," UNH's awkwardness is that its story is "whether it can recover to its past good" rather than "whether it can grow into a greater future"; this is inherently not an LTGG-style narrative, and is more like a "key watch" target waiting for price and facts to offer better odds, consistent with the report's "Watch" rating.

  Jun 11, 2026