Samsung Electronics: A Long-Term Owner's Perspective

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 a slice of an existing pie, or creating an entirely new market?5/10

  A

  The absolute value of the ceiling is extremely high, but Samsung is almost entirely "enlarging and carving up one giant existing pie after another," rather than conjuring an entirely new market out of nothing—and that is precisely the key reason it falls short of the kind of target Baillie Gifford loves most: one that "redefines a category and monopolizes a new ten-year track."

  First, how big the pie is. Samsung spans four mature-to-semi-mature markets of very different character: memory (DRAM/NAND/HBM), foundry, display panels, and end-consumer electronics such as smartphones, TVs, and home appliances. Their combined addressable market is on the order of trillions of US dollars; in 2025 the company's full-year revenue reached a record high of KRW 333.6 trillion, and in Q1 2026 it posted a record single-quarter revenue of KRW 133.9 trillion. Space has never been the problem.

  The problem is that the "market structure" of these tracks is basically already set; Samsung is fighting for share within the existing base, rather than pioneering an incremental category:

  • Smartphones are already a mature, competitive market. The report notes the global smartphone market grew only about 2% in 2025, with Apple leading at a 20% share and Samsung close behind at 19%; this is a classic case of "grabbing a slice of an existing pie," not creating a new market.
  • Memory is a cyclical carving of the pie. On the strength of technology and ASP, Samsung returned to No. 1 in DRAM in Q1 2026, with about 38.6% share, but the total size of this pie is set by the AI-compute capex cycle, and Samsung is contending for share and pricing timing within the existing supply-demand structure.
  • TVs are already a slow-growth market. Citing Omdia, the report notes Samsung TVs have been No. 1 globally for 19 consecutive years—a solid lead, but "No. 1 for 19 straight years" itself shows this is a highly mature market with almost zero category growth.

  The only thing that barely carries a "creating a new market" flavor is HBM (high-bandwidth memory), a new product form of the AI era, along with adjacent new directions such as automotive electronics (Harman) and robotics/data-center cooling. But even HBM looks more like the old DRAM pie being re-enlarged and thickened under AI demand, and the "right to define" this new incremental curve currently rests mainly with SK hynix—whose HBM share is about 57%; Samsung is a chaser, not a pioneer.

  By Baillie Gifford's yardstick: Samsung's ceiling is high enough that "it will not stop growing because it hits the top," but its growth narrative is "maintaining and modestly expanding share across multiple trillion-scale mature markets," rather than "I created an entirely new market that did not exist before and belongs to me alone for the next decade." The former can produce a great cyclical platform company, but it can hardly deliver the kind of nonlinear "5x-in-ten-years" imaginative upside that Baillie Gifford seeks.

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

  A

  Doubling revenue within five years (from about KRW 333.6 trillion in 2025 to about KRW 667 trillion) is extremely hard and should not be taken as a baseline expectation; even where it looks possible in the short term, it would rely almost entirely on "price" (the memory price upcycle)—the least sustainable driver—rather than repeatable volume growth or structural new businesses, and that is the core reason it does not clear Baillie Gifford's "double revenue in five years" bar.

  Doubling within five years requires a compound annual revenue growth rate of about 15%. Laying out Samsung's real trajectory over the past six years shows how demanding that is: 2020 revenue KRW 236.8 trillion, 2021 KRW 279.6 trillion, 2022 KRW 302.2 trillion, 2023 pulling back to KRW 258.9 trillion, 2024 KRW 300.9 trillion, and 2025 a record KRW 333.6 trillion. The five-year (2020→2025) CAGR is only about 7%, and it even went through negative growth during the 2023 memory downturn. For a giant already exceeding US$300 billion in revenue, doubling its past-five-year 7% growth to 15% has no historical precedent to support it.

  Breaking down the three drivers one by one, none can carry a "doubling":

  • Volume (shipments): all three end markets are mature. The report notes the global smartphone market grew only about 2% in 2025 (Apple 20%, Samsung 19%), and TVs have been No. 1 for 19 straight years but the market has zero growth. Selling more phones, TVs, and panels cannot support a doubling.
  • Price (ASP / rising memory prices): this is the real engine behind the current revenue surge. In Q1 2026, DS semiconductor revenue soared year over year, and the memory business, on the back of higher ASPs, set record single-quarter revenue and operating profit, with semiconductor operating profit up about 48x year over year. But rising memory prices are cyclical—the report repeatedly stresses that full-year 2023 operating profit once fell to just KRW 6.57 trillion, proving the direction of prices can reverse at any time. Relying on price hikes to prop up revenue can work for a year or two, but not a sustainable doubling over five years.
  • New businesses: Harman (automotive electronics), robotics, and data-center cooling (the 2025 acquisition of Germany's FläktGroup) are all adjacent expansions, far too small to contribute the increment a doubling requires; Harman's 2025 revenue was only KRW 15.8 trillion, less than 5% of total revenue.

  The honest conclusion: over the next five years, whether Samsung's revenue "can set another record" is probably yes (as long as the AI-memory upcycle does not collapse); but "at least doubling" would require the memory super-cycle to run for several years without receding, plus volume, price, and new businesses all firing at once—not a reasonable assumption to use as an investment baseline. The report's neutral DCF assumes Samsung's ten-year revenue growth is only 3%–5%, which itself denies the realism of doubling in five years. What Baillie Gifford wants is "a doubling achieved through repeatable volume growth and new categories," whereas if Samsung's doubling happens, it would be a one-off pulse of cyclical dividend, far lower in quality than the Baillie Gifford paradigm.

  Jun 10, 2026
• QFive years out, what will take over as the next growth engine? Does this "second curve" exist today?4/10

  A

  Samsung's "second curve" does exist today, and it is exactly the two hottest areas right now—HBM (AI high-bandwidth memory) and advanced foundry—but both curves share a fatal problem: they are not new engines Samsung monopolizes; they are tracks it is chasing, with the lead held by its rivals. This makes it hard for Samsung to meet Baillie Gifford's requirement of "a clear, leading, self-owned new growth pole five years out."

  Laying out the candidate successor engines one by one:

  First candidate: HBM / AI memory. This is the largest increment right now. In Q1 2026, DS-segment revenue was KRW 81.7 trillion and operating profit KRW 53.7 trillion, with semiconductor operating profit up about 48x year over year, and Samsung began HBM4 volume production in February 2026. The problem is that the frontrunner on this curve is SK hynix—whose HBM share is about 57%—and HBM's margins are several times those of ordinary memory. Samsung is a chaser on this sexiest of curves, not a pioneer; moreover, HBM is essentially an extension of the old cyclical DRAM business, not a new species independent of the cycle.

  Second candidate: advanced foundry. In theory, foundry is Samsung's most imaginative "non-memory second curve," but the reality gap is enormous. The report notes Samsung's pure-play foundry share in 2025 was only about 7.1%, second place but with a vast gap to the leader; external data show TSMC at about 70% vs. Samsung at about 7.2%, a gap that has widened to 62.7 percentage points. Samsung's 2nm GAA process just topped a 60% yield in Q1, with a target of turning foundry profitable and pushing for 20% share by 2027—meaning this curve is still losing money and still in the "prove itself" stage, far from the point of taking over as a profit contributor.

  Third candidate: adjacent new businesses. Robotics (an increased stake in Rainbow Robotics), data-center cooling and HVAC (the 2025 acquisition of Germany's FläktGroup), and automotive electronics (Harman). The report is restrained here: the directions have logic, but "proving high returns will still take time," and they are too small (Harman's 2025 revenue was only KRW 15.8 trillion) to become a company-level successor engine.

  Baillie Gifford-style judgment: Samsung does not lack "second-curve candidates"; its trouble is the opposite—the one most sure to ramp (HBM) has its lead held by a rival and is still governed by the cycle; the one with the most structural imagination (advanced foundry) is still chasing and losing money today. So the successor engine five years out "exists today," but whether it can truly become a growth pole that is "Samsung's own, ahead of others, and cycle-transcending" is an unproven open question, not an established fact you can bet on. The report's "Watch" rating is, to a large extent, waiting for these two curves to convert chasing into leading.

  Jun 10, 2026
• QWhat is its core competitive advantage? Will this moat widen or narrow over the next three to five years?5/10

  A

  Samsung's core competitive advantage is "a combination of multiple moderate moats"—brand, enormous scale, vertically integrated manufacturing capability, capital intensity, and whole-supply-chain resource orchestration—rather than any single wide moat; looking three to five years out, this moat is broadly stable overall but narrowing in pockets, and it does not fit the "continuously and clearly widening moat" feature Baillie Gifford prefers.

  First, acknowledge where it is "strong," and view it business by business, because the biggest feature of Samsung's moat is that it is "strong unevenly":

  • Memory scale and manufacturing: this is the hardest piece. On the strength of technology and ASP, Samsung reclaimed No. 1 in DRAM in Q1 2026, with about 38.6% share; together with its NAND leadership, the unit-cost advantage from scale, yield, and vertical integration is real.
  • Brand and channel: citing Omdia, the report notes TVs have been No. 1 globally for 19 straight years; Galaxy, home appliances, and panels enjoy global recognition and broad carrier/OEM/B2B channel relationships.
  • Display manufacturing and whole-chain orchestration: synergy across multiple technology stacks such as OLED, advanced packaging, and image sensors makes it one of the few companies able to run such a complex manufacturing system.

  But whether the moat "widens or narrows" comes down to the two most critical fronts—and these two are precisely narrowing or being suppressed:

  • The HBM front is being widened by rivals, and Samsung is chasing. The sexiest memory increment of the AI era is HBM, and the frontrunner is SK hynix—whose HBM share is about 57%—with a real first-mover advantage. Samsung only began HBM4 volume production in February 2026, as a latecomer. Whether this moat can be widened over the next three to five years depends on HBM4/HBM4E qualification and ramp, which is an unproven item for now.
  • The advanced-foundry front's moat is far weaker than TSMC's. The report bluntly states foundry is clearly weaker than TSMC; external data show TSMC at about 70% share vs. Samsung at about 7.2%, a gap widened to 62.7 percentage points, and TSMC's 2025 wafer revenue was about 74%–77% from 7nm and more advanced nodes, a runaway lead in commercializing advanced nodes. Samsung's 2nm GAA just passed a 60% yield in Q1, with profitability targeted only for 2027; this moat is currently "narrow and being widened at a loss."

  The moat in end-consumer businesses is on the shallow side: the report is clear that phones and TVs are more easily caught up with; the global smartphone market grew only about 2% in 2025, and Samsung's 19% share slightly trails Apple's 20%—the brand is sticky, but the moat in end hardware was never deep.

  The honest conclusion: Samsung's moat is "present, but not pure enough." What is truly widening is the cyclical barrier of memory scale; what is narrowing or being suppressed is the two most critical fronts of the AI era—HBM leadership and advanced-foundry quality. The report therefore gives a "moat strength of 3/5." In Baillie Gifford's terms, this is a moat that is "stable overall, narrow in pockets," rather than the compounding kind that "keeps widening while rivals only fall further behind"; this is also one of the core reasons it stays at "Watch" rather than "Buy."

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

  A

  Samsung has the hardware DNA to "reinvent itself after being disrupted"—historically, on several occasions when a product line was disrupted, it survived and overtook rivals thanks to a vertically integrated manufacturing system and continuous heavy reinvestment; but the way it deals with mistakes and bad news is closer to "head-down catching up, low-key error correction," lacking the transparent culture Baillie Gifford values most—"open post-mortems, telling the full story of failure, and rapid iteration at the organizational level"—which is a discount factor for a long-term owner.

  First, the DNA to reinvent itself—here Samsung is actually not weak:

  • Its moat is itself "endless reinvestment". The report discloses that in 2025, cash outflow for PP&E alone reached KRW 47.5 trillion and purchases of intangible assets KRW 4.63 trillion; this ability to keep plowing huge amounts of cash back into fabs, advanced nodes, packaging, and display means that when a technology path is disrupted, Samsung has the financial firepower and manufacturing system to place a new bet.
  • Multi-business diversification provides a buffer for reinvention. The report notes Samsung spans memory, foundry, display, end devices, and automotive electronics, so a misstep in a single arena will not be fatal—precisely the structural backbone of "self-reinvention." When memory fell deeply in 2023 and full-year operating profit dropped to KRW 6.57 trillion, the company was not seriously wounded, held up by its other businesses and a strong balance sheet.
  • An extremely strong net-cash cushion. The report estimates Q1 2026 net cash of about KRW 119.2 trillion, the most direct guarantee that "even if it errs, it will not die, and it has the means to start over."

  But on the soft dimension of "how it deals with mistakes and bad news," Samsung is clearly weaker:

  • It has indeed erred, and admits it is catching up. The most typical case is falling behind in HBM and advanced foundry—the report bluntly says the HBM first-mover advantage is not on Samsung's side (SK hynix HBM share about 57%), and the foundry gap with TSMC has widened to about 62.7 percentage points. To its credit, Samsung has not covered this up but has set clear remedial targets—foundry profitability by 2027, a push for 20% share, and 2nm yield progress—a pragmatic error-correcting posture.
  • But the error-correction style is "grinding away quietly," not "open post-mortems". Samsung lacks a tradition of candidly dissecting strategic mistakes in letters to shareholders; combined with the "chaebol governance discount" the report stresses and the series of Lee Jae-yong legal disputes (in July 2025 the Supreme Court upheld his acquittal in the 2015 merger case, but he had previously served time and was pardoned in 2022), external minority shareholders find it hard to get a transparent window into "how management views its own mistakes."
  • Its handling of bad news on the labor front is also on the passive side. In May 2026, under government mediation, it reached a tentative wage agreement with the union to avert a strike, but the report notes the deal's design drew some shareholder groups' questions about legality/value transfer—a case of "holding the problem down" rather than "fully resolving it."

  Baillie Gifford-style judgment: Samsung "has the DNA to reinvent itself"—that is affirmed; its capital strength, manufacturing base, and business diversification let it place a new bet even if the core business is disrupted. But the way it "deals with mistakes and bad news" is pragmatic correction and low-key catch-up, not open, transparent organizational learning. The former ensures it will not die; the latter is what Baillie Gifford uses to judge whether a company can "keep evolving gracefully"—Samsung passes on the first and is on the weak side on the second.

  Jun 10, 2026
• QDoes management (especially the founder) have a long-term vision, with interests deeply aligned with the company? Are they willing to sacrifice current profit for five to ten years out?5/10

  A

  Samsung's management undoubtedly has a long-term vision and is willing to sacrifice current profit for years down the road (its high, counter-cyclical CapEx is proof); but on the point Baillie Gifford values most—"the interests of the founder/controlling owner being deeply and cleanly aligned with all shareholders"—Samsung loses points: it is a classic Korean chaebol control structure, with the controlling family steering through complex group governance and a long-standing "governance discount" versus minority shareholders, so interests are not fully aligned.

  First, "long-term vision" and "willingness to sacrifice the present for the future"—on these two points, Samsung is actually quite outstanding:

  • Counter-cyclical heavy reinvestment is the hardest evidence. Even in the 2023 trough year of a memory downturn with full-year operating profit falling to just KRW 6.57 trillion, Samsung kept up huge capital expenditure and R&D, without cutting investment for the sake of short-term profit. In 2025, cash outflow for PP&E still ran as high as KRW 47.5 trillion. This is classic "betting for five to ten years out" behavior.
  • Clear multi-year strategic targets. Foundry has set a 2027 profitability turnaround, a push for 20% share, and continuous 2nm GAA yield progress—a path of short-term losses endured for long-term positioning.
  • Institutionalized shareholder returns. The report and company disclose a commitment for 2024–2026 to total shareholder returns of 50% of free cash flow over the period, with regular dividends of KRW 9.8 trillion per year, retaining the flexibility to return capital early and cancel treasury shares—showing management does not pursue expansion at shareholders' expense.

  But on "interests deeply aligned with the company," Samsung differs fundamentally from Baillie Gifford's ideal founder paradigm:

  • This is not a structure where a founder holds a large stake and sits in the same boat as minority shareholders. Samsung is a giant enterprise controlled by the Lee family through chaebol cross-shareholdings. The report explicitly points out the "typical Korean chaebol-style governance discount," noting that its root is not only the presence of a controlling shareholder but also the long-term effect on minority-shareholder sentiment of the series of Lee Jae-yong legal and governance disputes (the 2015 merger case saw his acquittal upheld by the Supreme Court in July 2025, but he had previously served time in another case and was pardoned in 2022). "Governance risk cannot be treated as zero" is the report's own phrasing.
  • Management's personal direct holdings are very small. The report notes that the equity incentives the company grants to employees and executives are small relative to total share count, with limited dilution—meaning the professional-manager layer's direct interest alignment with the share price is not strong, and there is a mismatch between control rights and cash-flow rights.
  • A new bargaining party has emerged in profit distribution. In May 2026, under government mediation, the company reached a tentative wage agreement with the union to avert a strike, but the report notes the deal drew some shareholders' questions about legality/value transfer—showing that how profit is distributed among shareholders, the controlling family, and employees is itself a not-fully-aligned point of tension.

  Baillie Gifford-style judgment: Samsung's management scores full marks on "long-term vision," and passes—even excels—on "willingness to sacrifice the present for the future" (counter-cyclical reinvestment is real skill); but on Baillie Gifford's core proposition of "the founder's interests being deeply and cleanly aligned with the company and all shareholders," it is a chaebol control structure rather than a founder-led one, with a governance discount and interest mismatch. This is exactly why the report lists "can management be trusted" as "uncertain" and keeps the company at "Watch"—strong operating ability, but the governance discount has never been removed.

  Jun 10, 2026
• QIf it vanished tomorrow, how much would customers miss it? Is its growth sustainable and not reliant on harming society or regulation?6/10

  A

  If Samsung vanished tomorrow, the global tech industry would immediately fall into a severe supply shock—it is a key supplier of DRAM, NAND, HBM, OLED panels, and premium phones/TVs, making it highly indispensable; at the same time, its growth model is broadly sustainable and does not rely on harming society, but along two lines—geopolitics/export controls and governance/labor—there are "regulatory and social-sustainability" risks that need long-term watching. On this question, Samsung scores high on "indispensability" and above average on "social and regulatory sustainability."

  First, "how much customers would miss it"—on the indispensability half, Samsung is very strong:

  • It is a key upstream supplier for AI compute. In Q1 2026, on the strength of technology and ASP, Samsung reclaimed No. 1 in DRAM with about 38.6% share, is the NAND leader, and has already volume-produced HBM4. With AI servers extremely hungry for memory and a memory shortage emerging across the whole industry, if Samsung disappeared, cloud providers and OEMs would be nearly unable to make up this capacity in the short term—the hardest kind of indispensability.
  • It is equally critical in display and end devices. OLED panels (supplying major customers such as Apple), TVs that have been No. 1 globally for 19 straight years, and the world's second-largest smartphone brand (19% share in 2025, second only to Apple's 20%).
  • But "indispensable" does not equal "irreplaceable". The report honestly notes that in HBM SK hynix leads (about 57% share), and in advanced foundry TSMC holds about 70% while Samsung is only about 7.2%. In other words, in the two most cutting-edge segments, customers have "backups" if Samsung disappears; what is truly hard to replace is its "enormous capacity" in memory and display, not every one of its "most advanced technologies."

  Now, "whether the growth model is sustainable and does not harm society or regulation"—this half is mixed:

  • The fundamentals are healthy and it does not profit by harming society. Samsung sells chips, panels, and end hardware; there is no data-monopoly, addiction-based, or regulatory-arbitrage business model (the report rates its "data advantage as moderate to weak, not monopolizing profit through user data"). This stands in sharp contrast to many platform companies under regulatory scrutiny, and is a plus.
  • But geopolitics/export controls are a structural long-term variable. The report is clear that Samsung's China capacity and equipment supply remain subject to the US annual licensing regime—"the company faces fresh regulatory uncertainty every year." This is not "harming society," but it does make its growth sustainability subject to external regulation beyond its own control.
  • Governance and labor are the soft spots of social sustainability. The chaebol governance discount the report points out, the series of Lee Jae-yong legal disputes, and the May 2026 tentative wage agreement reached under government mediation but drawing shareholder questions about its legality all show that Samsung still has friction to work through in "its relationships with various parts of society (shareholders, employees, regulators)."

  Baillie Gifford-style judgment (dual standard): First, "indispensability"—Samsung is highly indispensable in its enormous supply of memory and display, but has substitutes in the most cutting-edge technologies, so it scores high overall but not full marks. Second, "social and regulatory sustainability"—the business model itself is clean and does not rely on harming society, but export controls and the governance/labor lines constitute ongoing regulatory-social risk, so above average. This company "would be sorely missed by the whole world if it vanished tomorrow," but whether its long-term growth can "avoid being repeatedly constrained by regulation and governance" remains an open question requiring long-term tracking.

  Jun 10, 2026
• QWhat are this business's unit economics (gross margin, incremental returns)? Do they get better or worse as it scales? Where does the money it earns go?5/10

  A

  Samsung's unit economics look beautiful in boom years, but they are essentially "asset-heavy + cycle-driven," rather than the asset-light compounding model Baillie Gifford loves—"the bigger the scale, the higher the incremental returns, the more the cash is saved as it grows." Its gross margin and incremental returns swing violently with the memory cycle, and the large amounts of cash it earns must be continuously plowed back into fabs and advanced processes, so free cash flow is far less smooth than headline profit.

  First, look at the current unit economics—at the cyclical peak they are indeed very strong:

  • Gross margin and operating margin are rising. Citing the audited figures, the report notes Samsung's gross margin rose from about 38.0% in 2024 to about 39.4% in 2025, operating margin from about 10.9% to about 13.1%, and net margin from about 11.4% to about 13.6%.
  • The memory business's incremental returns are extremely high during upcycles. In Q1 2026, DS semiconductor revenue was KRW 81.7 trillion and operating profit KRW 53.7 trillion, with semiconductor operating profit up about 48x year over year—memory is a classic case of high operating leverage, and when volume and price rise together, the increment falls almost directly into operating profit.
  • Profit does convert into cash. In 2025, operating cash flow of KRW 85.3 trillion was clearly higher than net profit of KRW 45.2 trillion, showing that earnings quality during the upcycle is genuine.

  But "does it get better or worse as it scales"—Samsung's answer is "better but not cash-saving," which is exactly its fundamental difference from Baillie Gifford's ideal model:

  • It is a "the more it grows, the more it must reinvest" model, not a cash-saving one. The report puts it bluntly: this company is by no means a "the more it grows, the more it saves" business. In 2025, cash outflow for PP&E alone was KRW 47.5 trillion and purchases of intangible assets KRW 4.63 trillion, over KRW 52 trillion combined, with segment depreciation and amortization totaling about KRW 46.9 trillion. To hold its memory process, advanced foundry nodes, and display technology, it must keep plowing huge amounts of cash back in.
  • Free cash flow is not smooth and is doubly suppressed by the cycle and CapEx. The report estimates 2025 rough free cash flow at about KRW 33.2 trillion, far below net profit; even more striking, total free cash flow for 2021–2023 was only about KRW 18.8 trillion, while shareholder returns over the same period reached KRW 29.4 trillion—returns even exceeded free cash flow, showing that cash freedom came under severe pressure in trough years.
  • Incremental returns collapse when the cycle reverses. During the 2023 memory downturn, full-year operating profit was just KRW 6.57 trillion—with the same capacity and the same scale, incremental returns can crash from extremely high to extremely low in an instant. These are cyclical unit economics, not structural ones.

  "Where the money it earns goes"—the report is very clear on this:

  • The bulk flows back into reinvestment: fabs, advanced nodes, packaging, display, and new businesses (robotics, data-center cooling, automotive electronics).
  • The rest is institutionally returned to shareholders: a 2024–2026 commitment to return 50% of free cash flow, with KRW 9.8 trillion in regular annual dividends, plus treasury-share cancellations. Capital allocation is "above average, but short of excellent."

  Baillie Gifford-style judgment: Samsung's unit economics in boom years (gross margin near 40%, extremely high memory incremental returns) are indeed impressive, and scale does bring real procurement/manufacturing/R&D synergy; but its essence is an "asset-heavy model of high fixed costs, high CapEx, and a strong cycle"—the money it earns must be plowed back in large amounts to hold its position, free cash flow swings violently with the cycle, and incremental returns cannot compound steadily. This is far removed from the asset-light paradigm Baillie Gifford prefers—"the bigger the scale, the more cash is saved, and the higher the incremental capital returns"—and it is the basis for the report judging "stable free cash flow" and "excellent returns on capital" on the checklist as "Fail/Uncertain," respectively.

  Jun 10, 2026
• QFor it to rise fivefold in ten years, what conditions must hold simultaneously? Are those conditions realistic? What expectations does today's share price imply?3/10

  A

  For Samsung to rise fivefold in ten years from its current price (closing at about KRW 302,500 on June 10, 2026, with a market cap of about KRW 1,922 trillion), its market cap would have to surge toward about KRW 9,600 trillion—which requires a whole string of highly difficult conditions to hold at once, and at today's price the market has already made "the AI-memory upcycle continuing for the long run" its baseline expectation, leaving almost zero margin of safety for a "5x in ten years." Frankly, this is not a blue-sky target Baillie Gifford would bet on at the current price.

  Conditions that must all hold for a 5x in ten years (each indispensable):

  1. The AI-memory upcycle must be an "8–10 year new normal" rather than a "2–3 year spike". This is the most critical and most fragile assumption. Current profit relies heavily on rising memory prices—in Q1 2026, semiconductor operating profit rose about 48x year over year. But the report repeatedly warns that full-year 2023 operating profit once fell to just KRW 6.57 trillion, proving this is a cycle. A 5x requires this upcycle to run long enough to "make high profits permanent."
  2. HBM must go from chasing to leading. Samsung has to regain dominance in HBM4/HBM4E and reverse SK hynix's roughly 57%-share lead—otherwise the fattest slice of profit keeps going to its rival.
  3. Advanced foundry must deliver a qualitative leap. Samsung needs to materially narrow its foundry share of only about 7.2% (versus TSMC's about 70%, a 62.7-percentage-point gap), and must hit the targets of lifting 2nm yield from the low-60s% in Q1 to volume-production levels, turning foundry profitable by 2027, and pushing for 20% share.
  4. High CapEx must buy a real lift in share/profit center, not merely sustain scale—the report warns that if investment buys no improvement, shareholders will only get "ever-larger scale with mediocre per-share value growth."
  5. The governance discount narrows, and AI demand forms a higher profit center—rather than a single short-cycle pulse.

  Multiplying these five conditions together—each uncertain—the joint probability of all holding at once is very low. That is why the report's neutral DCF gives a fair value of only KRW 170,000–240,000, with a ten-year growth assumption of just 3%–5%—far short of any 5x imagination.

  What expectations does today's share price imply?

  • Surface cheapness is an illusion. The current forward P/E of only about 5.5x looks extremely low, but that precisely treats 2026's windfall as the future base; the trailing P/E of about 24x shows that on realized current earnings, it is not cheap.
  • Almost no risk compensation. The report estimates the trailing earnings yield (about 4.2%) is basically flat with the 10-year Korean government bond yield (about 4.195% in May 2026)—meaning you bear Samsung's high volatility, technology-execution, and CapEx risk yet get almost no extra return premium.
  • The price already sits above the upper edge of the optimistic scenario. The report's optimistic intrinsic-value range is KRW 240,000–320,000, and the current roughly KRW 302,500 is already in the upper part of the optimistic range, above the "clearly-overvalued watch zone" the report defines (>KRW 270,000). In other words, the market has already priced in most of the optimistic script of "upcycle continues + catch-up succeeds."

  Baillie Gifford-style judgment: a 5x in ten years is not mathematically impossible, but it requires a string of low-probability events to happen at once—a permanent upcycle, an HBM comeback, a foundry qualitative leap, high returns on reinvestment, and governance improvement. And today's price of about KRW 302,500 not only leaves no discount for these uncertainties but has already priced in fairly optimistic expectations. The report's "Watch" rating, with the more attractive buy range set at KRW 140,000–190,000, essentially says: buying anew at the current price means paying for a very strong but not necessarily sustainable earnings peak, rather than buying a 5x-in-ten-years compounding machine at a discount.

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

  A

  The premise of this question needs correcting first: the market has not "failed to realize" Samsung's value—on the contrary, it has already fully, even somewhat optimistically, priced the AI-memory upcycle into the shares. On June 10, 2026, the share price was about KRW 302,500 with a market cap of about KRW 1,922 trillion, and the stock had just set a record high of KRW 370,000 on June 2. So the more accurate way to ask is not "why hasn't the market realized it" but "where exactly the market's disagreement lies, and whether any direction remains unpriced."

  First, to be clear: today's Samsung is a "fully priced name at a cyclical peak," not a "forgotten cheap leader." The report's strongest counterargument makes it plain: Samsung's current "sense of cheapness" (a forward P/E of only about 5.5x) is an illusion, because it comes from abnormally high current profit depressing the denominator; on a trailing P/E of about 24x it is not cheap, and the trailing earnings yield is basically flat with the 10-year Korean government bond yield (about 4.195%), so risk compensation is inadequate. The market has not "failed to understand it"; it has already priced in the optimistic script.

  So the real disagreement centers on three things—an inability to judge accurately, rather than any failure to understand or disdain:

  1. Is the upcycle a "new normal" or a "spike"—this is the biggest bull-bear divide. Bulls believe the AI-memory shortage will last for years and make high profits permanent; bears think this is just another peak of the memory cycle that will ultimately revert to the mean (after all, 2023 operating profit once fell to KRW 6.57 trillion). The market has not "failed to notice" this; it "cannot be sure"—and that uncertainty is itself the core tension in the current valuation.
  2. Whether Samsung can convert "chasing" into "leading". The market clearly knows SK hynix leads HBM at about 57% and TSMC's foundry is about 70% versus Samsung's about 7.2%—all already priced in as a "Samsung discount." The disagreement is over how large that discount should be and whether the catch-up will succeed.
  3. How large the chaebol governance discount should be. The governance issues the report points out, the Lee Jae-yong legal disputes, and labor-management bargaining have long suppressed the valuation center—also a known factor, already partly priced in.

  What would become the "narrative inflection point"? Here is the truly valuable part of this question—the key events that could reprice the market (up or down):

  • Upward inflection points: (1) HBM4/HBM4E winning large-scale qualification from top customers such as Nvidia, with share materially overtaking SK hynix; (2) advanced foundry 2nm yield breakthroughs, the 2027 profitability target being met ahead of schedule, and landing heavyweight external customers—this would rewrite the "cyclical discount" into a "structural quality premium," the strongest upward narrative inflection point. (3) Governance reform (enhancing minority-shareholder rights, stepping up treasury-share cancellations) narrowing the discount.
  • Downward inflection points (equally to be watched): (1) memory prices peaking and falling back, customers destocking, or AI capex slowing, proving this is just an ordinary cyclical spike—the report warns "profit may fall faster than revenue"; (2) HBM/DRAM share being lost again; (3) foundry 2nm customer ramp and profitability improvement failing.

  Baillie Gifford-style judgment: Samsung's problem is that "the market already sees it very fully, even a bit optimistically," rather than "the market can't see far and there is a bargain to grab." Its greatest current danger is the "high-profit illusion" the report repeatedly stresses—mistaking abnormal profit at a cyclical peak for something permanent—rather than undervaluation. The inflection points that could truly change the long-term narrative are the realization of two things: an HBM comeback and an advanced-foundry qualitative leap; until they are proven, today's price of about KRW 302,500 looks more like paying for optimistic expectations. This is also the fundamental logic behind the report's "Watch" rather than "Buy" rating, with the attractive buy range set at KRW 140,000–190,000—what it waits for is "the price returning to a level with room for error" or "the catch-up being genuinely proven," rather than "the market waking up."

  Jun 10, 2026