Samsung Electronics (005930) 🔎 In-depth
KOSPI · Price as of 2026-11-12 · Metrics 2026-11-13 · In-depth research 2026-11-07
Samsung Electronics is a diversified electronics company built around memory chips such as DRAM and NAND flash, alongside smartphones, TVs and home appliances, contract chip manufacturing (foundry) and displays. Today, memory, including high-bandwidth memory (HBM) for AI servers, generates most of its profit. Preliminary Q3 2026 results showed revenue of ₩195.25 trillion and operating profit of ₩107.44 trillion, extending its run of record quarters to four in a row; the semiconductor division contributed 99% of operating profit, and net cash stood at ₩197.6 trillion at the end of the quarter. The key point now is two-sided. While the company expects the memory supply shortage to deepen in 2027, its earnings power, reflected in a forward P/E of around 4.13x, and its remaining shareholder-return capacity are clear strengths. If memory price gains fade or the won keeps strengthening, however, profits could shrink through the very same channel.
This page organises public market and filing data for information purposes. It is not a recommendation to buy or sell, nor investment advice. Metrics labelled “forward” are our own unverified estimates. Please verify with the original DART filings and decide at your own responsibility.
Data sources
Prices & market cap from the FSC securities price API (data.go.kr) · filings & financials from OpenDART (FSS) · listings from the Korea Exchange.
Prices as of 2026-11-12. Methodology is public on the methodology page.
30-second brief
Auto-extracted from the computed values.
What do the SourceComputedEstimateReading tags mean?
Source filings and exchange data · Computed calculated from public data · Estimate our own unverified projection · Reading drafted automatically, then rule-checked. See AI & Automation.
Core valuation metric
This stock's effective sub-sector is “Memory Chips” (Semiconductors & IT Components · Semiconductors), a type typically read first through P/B.
Memory chipmakers ride the DRAM and NAND price cycle, so profits balloon in upturns and can swing to losses in downturns. That makes earnings-based multiples unreliable, so price-to-book (P/B) — the share price against the company's asset value — is the first lens.
Price against assets alone says little about where the cycle stands. Reading it together with price against this year's expected earnings shows how far profits have recovered.
This note explains how each sector is typically valued and is educational context only, not investment advice.
At-a-glance assessment financial health · growth · profitability · valuation
- Debt ratio, current ratio and interest burden all look healthy.
- Revenue rose 10.9% year over year, and the pace is slowing (3-year trend: rising).
- Most recent quarter (Q1 2026) revenue was 69.2% higher than a year earlier.
- ROE is 17.6% (controlling-interest basis). It is above the sector average.
- Operating margin is 24.2%.
- A forward P/E of 4.26x puts the price low relative to earnings.
Ownership & governance As of 2022-12-31
Largest shareholder Samsung Life Insurance 8.51% (corporate)
Controlling bloc incl. related parties 20.89%
With the controlling bloc holding 21%, control is maintained but the free float is relatively large.
🔎 In-depth analysis Reading
Samsung Electronics has four main businesses: the DS division, which handles semiconductors; the DX division, which makes smartphones, TVs and home appliances; its subsidiary Samsung Display; and Harman, its automotive electronics and audio subsidiary. Right now, nearly all of the profit comes from DS. Of total operating profit of ₩107.44 trillion in Q3 2026, ₩106.5 trillion came from DS, a 99% share. Memory accounted for ₩140.7 trillion of DS revenue of ₩147.1 trillion, and the DS operating margin (operating profit as a share of revenue) was 72%. Memory comes in two main types: DRAM, which reads and writes data quickly, and NAND flash, storage memory that keeps data even when the power is off. The most closely watched high-value product is high-bandwidth memory (HBM), which stacks DRAM in multiple layers and sits right next to an AI accelerator. The company began mass-production shipments of HBM4 in February 2026 and said it sharply increased HBM4 shipments in Q3. In May it shipped samples of the next generation, 12-high HBM4E, and it plans mass production in 2027. Alongside HBM4, it is also growing shipments of high-performance server DRAM (DDR5) and data-center storage drives (SSDs). The DS division also houses contract manufacturing (foundry) and System LSI, which designs chips for smartphones. Starting with HBM4, the bottom-layer chip (the base die) is made on Samsung's own 4-nanometer process, and the foundry business cited higher base-die revenue as a reason for its improved Q3 results. Having memory and contract manufacturing under one roof thus gives the company a way to meet demand for customer-specific HBM. The DX division, by contrast, posted Q3 revenue of ₩52.1 trillion and an operating loss of ₩0.8 trillion, its second straight quarterly loss after Q2. The reason is that when memory prices rise, so do the costs of the smartphones and TVs that use that memory. Samsung Display posted Q3 revenue of ₩8.0 trillion and operating profit of ₩1.1 trillion, while Harman posted revenue of ₩4.3 trillion and operating profit of ₩0.6 trillion; Harman's profit includes one-off items such as U.S. tariff refunds. Samsung also holds stakes in Samsung Biologics (31.2%), Samsung Electro-Mechanics (23.7%) and Samsung SDI (19.4%), among others. However, its proportional share of those companies' earnings (equity-method income) was only about ₩0.4 trillion in Q3, a small part of overall earnings.
The latest close is ₩208,000 and the market capitalization is ₩1,216.0 trillion. The price sits below its 20-day moving average (₩214,600) and below its 60-day moving average (₩245,642). It is under both its short- and medium-term moving averages, so the trend looks subdued. The RSI (a supplementary indicator that gauges the strength of gains versus losses over the past 14 days on a 0-100 scale) is 42.9, a neutral level. The one-month change is -10.3%, the three-month change is -9.6%, and the position relative to the 52-week high is -42.6%. Relative strength versus the KOSPI is 71 (on a 1-99 scale, converted from returns against the index over the past year with more weight on recent performance; higher means stronger than the market). It is stronger than roughly 71% of all stocks. Over the past three months it outpaced the index by 2.5%. Chart interpretation is best done alongside trading volume and the dates on which disclosures occur.
To be accurate, the financials need to be re-examined on a Q3-end basis. The multiples on the on-screen cards are calculated from 2025 full-year earnings or Q1-end equity, so they miss the profits that built up in Q2 and Q3. The first metric to check for a memory company is P/B (how many times net assets the share price is). Earnings swing sharply with memory prices, but equity moves more slowly, which makes it a steadier anchor. The 2.49x P/B shown on the card is based on Q1-end equity of ₩473.96 trillion. Recalculated using confirmed end-June equity (the controlling shareholders' portion, ₩565.06 trillion), it is 2.27x; using total equity of ₩614.11 trillion at the end of Q3, it is 2.09x. ROE (return on equity), which shows how much a company earns on its equity in a year, is 41.6% on a trailing 12-month basis. For a company earning more than 40% a year on its equity, a share price of just over twice net assets is not high. Fellow memory maker SK Hynix trades at about 3.3x its Q3-end equity. For earnings multiples, the reference period matters a great deal. The P/E (how many times one year's earnings the share price is) of 26.62x is calculated from 2025 full-year net profit attributable to controlling shareholders of ₩44.26 trillion. This year, that net profit already reached ₩193.58 trillion in Q1-Q3 alone, so the 26.62x figure says little about the company today. Recalculated on trailing 12-month earnings (Q4 2025 through Q3 2026), the P/E is 27.47x. The forward P/E for this year, which adds a Q4 estimate to the three reported quarters, is 4.13x. The balance sheet has strengthened further. The debt ratio (liabilities relative to equity) was 29% at the end of Q3, and the current ratio, which compares assets convertible to cash within a year with debts due within a year, was 310%. Subtracting borrowings of ₩17.19 trillion from ₩214.82 trillion in cash and short-term financial assets at the end of Q3 leaves net cash (cash remaining after repaying all debt) of ₩197.6 trillion. That is up roughly ₩30 trillion in three months from ₩167.54 trillion at the end of June, and equals about 15% of market cap at the November 3 close. Enterprise value (EV, market cap plus net debt), the price of acquiring the whole company, also needs to be reassessed on this basis. Because of the large net cash position, EV at the end of Q3 was ₩1,085.63 trillion, less than the market cap. EV/EBIT (EV divided by trailing 12-month operating profit) is 3.96x, EV/EBITDA (EV divided by operating profit before depreciation) is 3.37x, and EV/Sales is 1.83x. In other words, the business itself, excluding cash, is valued at less than four times one year's operating profit. Free cash flow, meaning cash from operations minus capital spending, was about ₩200.1 trillion over the last 12 months, roughly 15.6% of market cap. This ratio is called the free cash flow yield, and it means the money that funds dividends and treasury shares accumulated to nearly one-sixth of market cap within a year. In Q3 alone, operating activities brought in ₩98.6 trillion, while ₩26.9 trillion went into property, plant and equipment. Note, however, that operating cash flow can include items such as advance payments under long-term supply agreements, so it is defined differently from the free cash flow the company uses to calculate shareholder returns. The card's enterprise-value metrics and dividend yield are based on end-2025 net cash, results through Q1 and 2025 dividends, so they differ widely from current figures. Profitability has also risen sharply. The Q3 operating margin was 55.0%, and 46.1% on a trailing 12-month basis. It is lower than SK Hynix's (about 74.5% over the last 12 months) because the loss-making DX division and the display business pull the average down. Dividends have grown sharply as well. Dividends per share for Q1-Q3 were ₩372, ₩374 and ₩4,601, totaling ₩5,347, for a dividend yield of about 0.8% at the November 3 close. The Q3-end equity and cash figures are summary values the company disclosed in its earnings materials ahead of its quarterly filing, so they may differ slightly once that filing is out.
Over five years, revenue grew modestly while profits swung widely. Revenue was ₩279.6 trillion in 2021, ₩302.2 trillion in 2022, ₩258.9 trillion in 2023, ₩300.9 trillion in 2024 and ₩333.6 trillion in 2025, an average annual increase of 4.5%. Operating profit over the same years was ₩51.6 trillion, ₩43.4 trillion, ₩6.6 trillion, ₩32.7 trillion and ₩43.6 trillion. The pattern shows profits nearly vanishing in 2023, when the memory market was weak, and then recovering. 2026 is on a different scale altogether. Operating profit rose every quarter: ₩57.23 trillion in Q1, ₩89.49 trillion in Q2 and ₩107.44 trillion in Q3 (preliminary). Cumulative Q1-Q3 operating profit of ₩254.17 trillion is about 5.8 times the 2025 full-year figure of ₩43.60 trillion, and cumulative revenue for the period is ₩500.63 trillion. Compared with the same period a year earlier (revenue of ₩239.77 trillion and operating profit of ₩23.53 trillion), revenue more than doubled and operating profit grew more than tenfold. For quarterly comparisons, the default yardstick is the same quarter a year earlier (YoY), because memory combines seasonality with a price cycle, and comparing only with the previous quarter can easily distort the picture. Q3 revenue rose 126.9% and operating profit 783.1% from a year earlier. Because Q3 operating profit a year ago was low at ₩12.17 trillion, these growth rates look heavily inflated. Even against the previous quarter, revenue rose 13.9% and operating profit 20.1%, so the upswing carried on into Q3. The driving force is memory. Q3 memory revenue was ₩140.7 trillion, more than five times the ₩26.7 trillion of a year earlier. The company said strong server demand lifted memory prices across every application, and that it focused shipments on AI products while capacity was tight. Net profit did not grow as fast as operating profit. Q3 net profit of ₩75.14 trillion rose 4.9% from the prior quarter, below the growth in operating profit, because financial income swung from +₩4.7 trillion in Q2 to −₩9.5 trillion in Q3. The won/dollar exchange rate fell sharply during the quarter, but the company has not yet broken down this loss, so it is a point to check in the quarterly filing. This year's forward P/E of 4.26x is built quarter by quarter: reported Q1 and Q2 results and preliminary Q3 results, plus a Q4 estimate, rather than a single quarter multiplied by four. The Q4 estimate reflects five conditions. First, the company said strong semiconductor results will continue in Q4 and that company-wide results will grow. Second, contract prices that manufacturers set with large customers point higher for DRAM in Q4, but the increases are narrowing, and NAND showed signs of a pause in price gains in October. Third, the company said it will expand shipments of HBM4, high-performance DDR5, server SSDs and high-capacity QLC SSDs; with output from new fabs not yet available, Q4 growth must come from raising the mix of higher-priced products within existing capacity. Fourth, the average won/dollar rate in October was in the ₩1,330s, with the won more than 5% stronger than the Q3 average (in the ₩1,410s), so a currency headwind similar in size to Q3's was built in. Fifth, the U.S. tariff refund in Harman's Q3 profit was treated as non-recurring and excluded, and the estimate also reflects higher depreciation from expanded capital spending and the top corporate tax rate, which rose to 25% this year. On that basis, Q4 profit is estimated to rise from Q3, consistent with the company-wide growth the company has indicated. The figure could change depending on whether DRAM price gains hold throughout Q4 and whether the won strengthens further.
Recent disclosures fall into three areas: earnings, shareholder returns and investment. First, earnings. On October 29, the company released preliminary Q3 results: revenue of ₩195.25 trillion, operating profit of ₩107.44 trillion and net profit of ₩75.14 trillion. The initial figures announced on October 8 (revenue of ₩195.00 trillion and operating profit of ₩107.40 trillion) were revised slightly upward in a corrective filing on October 29. The company said it posted record quarterly revenue and operating profit for the fourth straight quarter. These are preliminary figures that have not yet been reviewed by the external auditor, however, and they may change in the quarterly filing due in November. Next, shareholder returns. On the same day, the company declared a Q3 dividend of ₩4,601 per share, ₩30.0 trillion in total (payable November 27). That is about 12 times the previous quarter's ₩374 and confirms the roughly ₩30 trillion Q3 cash dividend flagged in the August 21 filing. In that August 21 filing, the company put its remaining 2026 return pool at about ₩90-110 trillion under its policy of returning 50% of free cash flow for 2024-2026. In its October 29 earnings presentation, it said that after deducting the ₩35 trillion already paid or committed, it plans to return ₩55-75 trillion through cash dividends and a substantial treasury share repurchase and cancellation. The split and timing will be finalized at a board meeting in late January 2027. A June 24 media report of a large treasury share repurchase, which the company said in a July 23 follow-up filing it was reviewing for stock-compensation purposes but had not decided, was effectively settled by the August 21 treasury share acquisition decision and shareholder-return filing. Under that August 21 decision, the company acquired 56,856,305 common shares for about ₩15 trillion (an average of ₩263,823) between August 24 and October 6. The purpose is employee stock compensation, not cancellation, and the company said it will exclude this outlay when calculating its shareholder-return pool. Finally, investment. In an October 29 fair-disclosure filing, the company set its 2026 capital spending plan at about ₩84.4 trillion (₩76.8 trillion for semiconductors and ₩3.9 trillion for displays). That is about 60% more than the ₩52.7 trillion spent in 2025; memory spending goes to infrastructure such as new fabs, while foundry spending centers on its Taylor fab in the U.S. On June 29, the company disclosed a long-term vision to invest about ₩2,450 trillion in Korea over 2026-2040 (about ₩2,100 trillion in semiconductors), with the caveat that it is a guideline that may change with market conditions. There were also stake transactions within the group. On October 30, the company decided to acquire from its subsidiary Samsung Display the stake in China's TCL CSOT (10.16% after the deal) for about ₩1.27 trillion, with payment due in December after regulatory approvals. It also decided to subscribe for its full allotment of ₩666.2 billion in a rights offering by its affiliate Samsung Biologics (payment November 17). On October 12, 7,667,100 common shares held by Hong Ra-hee were transferred to Chairman Lee Jae-yong; this was a transfer within the family, and the largest shareholder group's combined stake remained at 19.69%.
Start with the facts to watch. Operating profit of ₩254.17 trillion in Q1-Q3 this year is about 5.8 times the 2025 full-year figure, and net cash at the end of Q3 is ₩197.6 trillion. The P/E is 27.47x on trailing 12-month earnings and 4.13x on a forward basis that includes the Q4 estimate. That is the same low single-digit range as SK Hynix (about 5.4x on trailing 12-month earnings), which rides the same memory cycle, meaning the shares are priced low relative to the size of the company's earnings and cash. On the demand side, investment by large data-center operators provides support. Q3 capital spending (on property and equipment) was $51.36 billion at Alphabet, $77.26 billion at Amazon and $39.37 billion at Microsoft, each roughly double the year-earlier level. Amazon kept its 2026 capital spending outlook of about $220 billion, and Meta narrowed its 2026 outlook to $135-145 billion. Nvidia, the largest AI accelerator maker, said in its August earnings materials that its supply commitments rose from $119 billion in the previous quarter to $279 billion, mostly for memory procurement. If these companies keep raising spending in their 2027 investment plans (usually presented with earnings in January-February), the floor under server memory volumes should hold. Some, however, are already spending more than they generate from operations. Alphabet's Q3 free cash flow was negative, and if funding conditions worsen and investment growth slows, memory suppliers' pricing power would be the first thing to weaken. On the supply side, the key question is when new capacity arrives. Samsung lifted its 2026 capital spending to about ₩84.4 trillion, and SK Hynix said its first Yongin fab will begin full operation in 2027. Micron's annual filing explicitly names new fab construction across the industry and investment by Chinese government-backed CXMT (DRAM) and YMTC (NAND) as oversupply risks. Fabs take time to go from groundbreaking to mass production, so there is almost no effect on Q4 supply. If output from several companies' new fabs arrives at the same time from the second half of 2027, price gains could slow. Price is the key variable driving current earnings. The company said strong server demand lifted memory prices across every application in Q3. Micron also guided September-November revenue ($61.5 billion ± $1.5 billion) above the prior quarter's actual ($54.23 billion). Industry contract prices show DRAM still rising in Q4 but by smaller amounts, while NAND showed signs of a pause in gains in October. The way prices are set is also changing. SK Hynix built a mechanism into its long-term agreements with 14 key customers that reflects market conditions within a set price range, and Samsung also said it will build a stable sales base through long-term agreements. The more such contracts spread, the less prices rise in upswings, but the firmer the floor in downturns. As long as contract prices keep rising, Q4 profit should grow along with them; if NAND prices stay flat for long, NAND's profit contribution is the first to slow. On the technology side, changes that reduce memory use are reaching real products. Nvidia said in its August earnings release that Groq 3 LPX, inference hardware that uses SRAM (ultra-fast memory built into the chip) and DDR5 instead of HBM, has entered full mass production. In March, Google Research unveiled a compression technique that shrinks the memory AI uses to hold conversation context (the KV cache) to at least one-sixth. If such technologies spread widely, HBM use per AI task would grow more slowly. That said, LPX carries large amounts of standard DRAM, so part of the demand is shifting toward standard DRAM rather than disappearing, and there is no official sign yet that these changes have shown up as lower memory prices. Among policy and macro variables, the exchange rate affects earnings most directly. The company said the stronger won reduced Q3 operating profit by about ₩10 trillion versus the prior quarter. The won was more than 5% stronger against the dollar on average in October than in Q3, so unless that reverses in November-December, Q4 will face a headwind of similar size. Conversely, a weaker won would become a tailwind. Memory was left out of the first phase of the 25% tariff (under Section 232 of the Trade Expansion Act) that the U.S. imposed on some advanced AI computing chips in January 2026, but whether it will be included in a second phase has not been decided. With the U.S. having ended its blanket authorization (VEU) for fabs in China at the end of 2025, whether equipment import approvals for the Xi'an NAND fab in China continue is another variable to watch at year-end. There are weak spots within the company as well. The DX division posted operating losses of ₩0.8 trillion in each of the last two quarters because of higher memory costs. If memory prices rise further, smartphones and PCs carry less memory and ship in smaller numbers, which in turn feeds back into memory demand. Nvidia also said consumer PC sales slowed because of higher memory and system prices, and Samsung Display expects weak demand to persist into 2027. Inventory also bears watching. Samsung's inventories were ₩71.99 trillion at the end of Q3, similar to the end of June (₩71.39 trillion). Micron warned that if volumes customers took on beyond their needs to meet long-term contract obligations pile up as inventory, the result could be oversupply and falling prices. The ₩15 trillion in treasury shares is earmarked for employee compensation, so it cannot be counted as reducing the number of shares in circulation, and the split of the remaining ₩55-75 trillion in shareholder returns will be decided in late January 2027. To sum up: as long as the memory shortage deepens in 2027 beyond this year's level, as the company has said, and data-center investment continues, the low forward multiple, ample net cash and remaining shareholder-return capacity work together as strengths. Conversely, if contract prices turn lower or large customers slow their 2027 investment plans, profits would shrink through the same channel, and a prolonged strong won would deepen the decline.
🔎 Valuation vs peers Undervalued
In Korea, SK Hynix is effectively the only direct memory-chip peer (99% of Samsung's Q3 operating profit came from DS, and 96% of DS revenue from memory); DX (appliances, TVs and smartphones), displays and Harman made up only about 2% of Q1-Q3 operating profit, and foundry profit inside DS is not disclosed separately, so LG Electronics (appliances and TVs), LG Display (displays) and DB HiTek (foundry) serve only as references for individual business lines. The table's P/E, P/B and ROE are site-calculated values (current price ÷ 2025 annual earnings and Q1-end equity), an outdated basis for two memory makers whose earnings have changed sharply; recalculated at Q3-end, SK Hynix trades at a P/B of about 3.30x (market cap ₩1,021.23 trillion ÷ total equity ₩309.09 trillion), a trailing 12-month P/E of about 5.41x (net profit attributable to controlling shareholders of ₩188.90 trillion) and a trailing 12-month operating margin of about 74.5%, versus Samsung at a P/B of about 2.09-2.14x, a trailing 12-month P/E of 27.47x and an operating margin of 46.1%.
| Peer | P/E | P/B | ROE |
|---|---|---|---|
| SK Hynix | 20.82x | 5.58x | 45.75% |
| LG Electronics | 23.26x | 0.87x | 4.26% |
| LG Display | 15.71x | 0.55x | 3.43% |
| DB HiTek | 13.98x | 1.57x | 13.04% |
The assessment is Undervalued, for three reasons. First, multiples relative to earnings and equity are low. The forward P/E (the share price relative to this year's expected earnings), built from three reported quarters plus a Q4 estimate, is 4.13x. On a like-for-like trailing 12-month basis, Samsung is at 6.0x and SK Hynix, which rides the same memory cycle, at about 5.4x, both in the low single digits. That is far below LG Electronics, LG Display and DB HiTek (16-25x), whose businesses resemble Samsung's other units. P/B (the share price relative to net assets), the first metric to check for memory makers, is about 2.1x on Q3-end equity, lower than SK Hynix (about 3.3x). Given that the company earns about 42% a year on its equity (trailing 12-month ROE), the price relative to equity is not high either. Second, there are supporting and offsetting factors. Net cash of ₩197.6 trillion, a structure that spans memory through contract manufacturing and packaging, and ₩55-75 trillion in remaining shareholder returns are supports. The loss-making DX division and the weak display business dragging down the average margin, along with heavy sensitivity to exchange rates, are offsets. Third, the card's P/E of 27.47x is hard to use as a basis for judgment. It is calculated from 2025 net profit attributable to controlling shareholders of ₩44.26 trillion, while the company earned 4.4 times that, ₩193.58 trillion, in Q1-Q3 this year alone. Including the preferred shares, which trade below the common shares, in market cap lifts the forward P/E to about 5.0x, but the assessment does not change. This view rests on the company's outlook for a deeper supply shortage in 2027 holding true. If memory contract prices turn lower or large data-center customers slow their 2027 investment plans, the case weakens.
Price history Close · MA20 · MA60
The latest close is ₩208,000 and the market capitalization is ₩1,216.0 trillion. The price sits below its 20-day moving average (₩214,600) and below its 60-day moving average (₩245,642). It is under both its short- and medium-term moving averages, so the trend looks subdued. The RSI (a supplementary indicator that gauges the strength of gains versus losses over the past 14 days on a 0-100 scale) is 42.9, a neutral level. The one-month change is -10.3%, the three-month change is -9.6%, and the position relative to the 52-week high is -42.6%. Relative strength versus the KOSPI is 71 (on a 1-99 scale, converted from returns against the index over the past year with more weight on recent performance; higher means stronger than the market). It is stronger than roughly 71% of all stocks. Over the past three months it outpaced the index by 2.5%. Chart interpretation is best done alongside trading volume and the dates on which disclosures occur.
Relative performance stock vs index · start = 100
Excess return vs index · 3M +2.52% / 6M +6.00% / 12M +57.06%
Key metrics Computed vs whole-market median
Valuation
The P/E of 27.47x is above the whole-market median (12.03x). The P/B of 2.57x is above the whole-market median (0.77x). That said, this P/E is based on last year's (trailing) results. With recent quarterly earnings up sharply, the trailing P/E can look higher than it really is, so a precise read is best done on this year's expected (forward) earnings.
Enterprise value (EV)
EV = market cap + net debt. It reflects cash and debt, so it captures the real cost of the whole business that market cap alone misses; lower multiples are cheaper relative to earnings or sales.
Intrinsic value (DCF estimate) Estimate
Model output — not a price target. This is what the formula returns under the assumptions below, not a view on where the share price should go.
DCF (discounted cash flow) estimate — discount rate 11.0%, initial growth 10.0%→terminal 2.0%, 10-yr forecast, free-cash-flow basis, forward earnings power normalized 3x. A reference range that shifts materially with assumptions.
Profitability & financials
Return on equity (ROE) is 17.6%, above the whole-market average (3.0%). The operating margin is 24.2%. The debt ratio is 30.9%, so the financial structure is stable.
Growth FY2025 · annual report (consolidated)
| Item | 2023 | 2024 | 2025 | YoY |
|---|---|---|---|---|
| Revenue | $194.1B | $225.5B | $250.1B | +10.88% ↓ slower |
| Operating profit | $4.9B | $24.5B | $32.7B | +33.23% ↓ slower |
| Net profit | $10.8B | $25.2B | $33.2B | +31.65% ↓ slower |
| 5-year | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Revenue | $209.6B | $226.5B | $194.1B | $225.5B | $250.1B |
| Operating profit | $38.7B | $32.5B | $4.9B | $24.5B | $32.7B |
| Net profit | $29.4B | $41.0B | $10.8B | $25.2B | $33.2B |
| Revenue CAGR | 4-yr avg 4.51% | ||||
Revenue rose 10.9% year over year (2023 ₩258.9 trillion → 2024 ₩300.9 trillion → 2025 ₩333.6 trillion), and the three-year trend is 'rising'. That said, the pace of growth slowed from the prior year. Operating profit rose 33.2% year over year. The pace of that profit growth is gradually easing. Over the 5 years on record, revenue compound annual growth (CAGR) is 4.5%. The two-year revenue CAGR is 13.5%. In the most recent quarter (Q1 2026), revenue was 69.2% higher than the same period a year earlier.
Latest quarterly results Source Q1 2026 · vs year-ago
Technical indicators Computed
What stands out
- P/E and P/B are both low versus peers, so the price looks inexpensive relative to earnings and assets.
- ROE of 17.6% points to solid profitability.
- Revenue grew 10.9% year over year, a sign of growth.
- The balance sheet is stable in terms of debt and liquidity.
Points to watch
- The figures shown are based on the last annual report as of the writing date, so it is best to review the latest quarterly results and filings alongside them.
Recent news & events searched · sourced
- 2026-10-30InvestmentBoard approves the off-market acquisition of the TCL CSOT stake held by subsidiary Samsung Display (10.16% stake after the deal) for about US$960 million (about ₩1.2718 trillion), with payment scheduled for December 2026 after regulatory approvalsThe stated aim is closer strategic collaboration. On a consolidated basis this is a transfer within the group, so the effect on consolidated earnings is limited; timing and amount may change depending on regulatory approvals in Korea and abroad. Source
- 2026-10-29EarningsQ3 2026 preliminary consolidated results: revenue ₩195.25 trillion, operating profit ₩107.44 trillion, net profit ₩75.14 trillion (₩75.21 trillion attributable to controlling shareholders); Q1-Q3 cumulative revenue ₩500.63 trillion and operating profit ₩254.17 trillionRecord quarterly revenue and operating profit for a fourth straight quarter (per the company). Cumulative operating profit is about 5.8 times the 2025 full-year ₩43.60 trillion, widening the gap between multiples based on last year's results and current earnings. Financial income swung to −₩9.5 trillion, so net profit growth (+4.9% quarter on quarter) lagged operating profit growth (+20.1%). Source
- 2026-10-29IRQ3 earnings presentation: DS revenue ₩147.1 trillion and operating profit ₩106.5 trillion (memory revenue ₩140.7 trillion), a DX operating loss of ₩0.8 trillion, and a negative currency effect of about ₩10 trillion on operating profit versus the prior quarter; outlook for strong semiconductor results and company-wide growth in Q4, and a deepening memory supply shortage in 2027With 99% of company-wide operating profit coming from DS, reliance on memory prices and volumes has grown. Because the company itself set out its Q4 and 2027 direction and its currency sensitivity, the path is now clear: earnings are supported while prices keep rising, and a lasting strong won would erode component-business profit in the same way. Source
- 2026-10-29DividendQ3 quarterly dividend declared: ₩4,601 per common and preferred share, total dividends of about ₩30.0 trillion, record date 2026-09-30, payable 2026-11-27About 12 times the previous quarter's ₩374, confirming the roughly ₩30 trillion Q3 cash dividend flagged on August 21. Dividends per share for Q1-Q3 2026 total ₩5,347. Source
- 2026-10-29UpdateFair-disclosure filing of a 2026 capital spending plan of about ₩84.4 trillion (DS ₩76.8 trillion, displays ₩3.9 trillion), about 60% above the 2025 actual of ₩52.7 trillionSpending centers on infrastructure such as new memory fabs and the Taylor foundry fab in the U.S. The company's own capacity additions mean more supply from 2027 onward, while higher depreciation starts to weigh on costs from Q4, making this a two-sided variable. Source
- 2026-10-29IRShareholder-return update: of total 2026 returns of ₩90-110 trillion, the remaining ₩55-75 trillion after the ₩35 trillion already paid or committed is to be returned through cash dividends and a substantial treasury share acquisition and cancellation, with details to be finalized at a board meeting in late January 2027The range of the remaining returns is now official. The split between dividends and share cancellation, and the timing, are still for the late-January 2027 board to decide, so they are not yet final. Source
- 2026-10-29InvestmentSubscribes for its full allotment as an existing shareholder (567,503 shares) in the rights offering of affiliate Samsung Biologics, investing ₩666.2 billion (payment due 2026-11-17)The investment maintains its stake (31.2%), and the burden is small next to Q3-end net cash of ₩197.6 trillion. Source
- 2026-10-28UpdateSK Hynix Q3 results: more than 50% of 2027-2028 capacity secured through long-term agreements with 14 customers, M15X at full utilization by year-end, and full operation of the first Yongin fab planned for 2027Long-term agreements with price ranges are changing how industry prices are set. A competitor on the same cycle reading supply this way supports the 2027 price environment, but competition to supply HBM4 is getting fiercer. Source
- 2026-10-21UpdateMeta narrows its 2026 capital spending outlook (including principal payments on finance leases) to $135-145 billionA large data-center customer has narrowed its annual investment range, which underpins Q4 server memory demand. Source
- 2026-10-13UpdateStake transfer between related parties: on 2026-10-12, Chairman Lee Jae-yong acquired off-market 7,667,100 common shares held by Hong Ra-hee (₩250,500 per share), and Chairman Lee reported a trading plan, starting 2026-11-12, for 11,158,000 shares received as a giftThis is a transfer within the family that leaves the largest shareholder group's combined common stake unchanged at 19.69%, so it has almost no effect on control or on the number of shares in circulation. Source
- 2026-10-12UpdateCompletes acquisition of about ₩15 trillion in treasury shares: 56,856,305 shares acquired on-market from 2026-08-24 to 10-06 at an average of ₩263,823 (purpose: employee stock compensation; decided August 21)The shares are for employee stock compensation, not cancellation, and the company said it will exclude this outlay when calculating its shareholder-return pool (free cash flow). As of November 4, treasury shares total 136,722,409 common shares (2.34%). Source
- 2026-09-30UpdateMicron reports FY2026 Q4 revenue of $54.23 billion and guides September-November revenue to $61.5 billion ± $1.5 billion; its annual filing specifies strategic customer agreements with volume obligations and price rangesA sign that industry price gains continue through September-November, a period that largely overlaps Samsung's Q4. At the same time, the risk that volumes taken beyond long-term contract needs pile up as customer inventory was formally flagged. Source
- 2026-08-26UpdateNvidia FY27 Q2 results: Vera Rubin in full mass production, supply commitments up from $119 billion to $279 billion (mainly memory procurement), consumer PC sales slowing on higher memory and system prices, and Groq 3 LPX in full mass productionEvidence of demand: the largest AI accelerator customer is locking in several years' worth of memory in advance. The same materials also signal that higher memory prices are eroding demand for end products and that inference chips that do not use HBM are growing alongside. Source
- 2026-08-21FilingFair-disclosure filing of expected 2026 shareholder returns: under the policy of returning 50% of free cash flow over the three years 2024-2026, the remaining 2026 pool is expected at about ₩90-110 trillion (advance payments under long-term supply agreements and employee stock-compensation outlays are deducted from free cash flow)A definitive filing that resolved the undecided status noted in the July 23 clarification. The ₩30 trillion Q3 dividend and the January 2027 timetable for announcing further returns first appeared in this filing. Source
- 2026-06-29UpdateFair-disclosure filing on future business and management plans: a domestic investment vision of about ₩2,450 trillion for 2026-2040 (about ₩2,100 trillion in semiconductors)It formalizes the long-term direction for expanding HBM and leading-edge chip capacity. The company caveated that it is a market-based guideline subject to change, and it also implies heavy spending over a long period. Source
- 2026-05-29ProductSamsung Electronics ships the industry's first 12-high HBM4E samples (up to 16Gbps)This is the starting point for the plan to mass-produce HBM4E in 2027. Whether Samsung can move ahead in qualification timelines for next-generation HBM is a variable for its 2027 HBM market share. Source
- 2026-03-24UpdateGoogle Research unveils TurboQuant, a compression technique that shrinks the KV cache memory holding conversation context during AI inference to at least one-sixthA software route to lowering memory use per AI inference task. Depending on how fast it spreads, it could slow the growth of HBM capacity, though so far rising usage has outweighed efficiency gains. Source
- 2026-02-12ProductSamsung Electronics begins mass-production shipments of HBM4 (11.7Gbps, 4-nanometer logic base die)Samsung has entered the HBM4 generation as a supplier. This gives it the volume base to serve demand for next-generation AI accelerators such as Nvidia's Vera Rubin. Source
Figure cross-check computed ↔ external
| Metric | Computed | External | Status | Source |
|---|---|---|---|---|
| Latest disclosure sweep (as of today) | fetch_disclosures days=200 2026-11-06, base (2026-06-08) major 17 | 2026-11-07 2026-11-06, major 2026-11-04 | Confirmed | link |
| Q3 2026 preliminary results checked against the original filing | revenue ₩195.25 trillion / operating profit ₩107.44 trillion / ₩98.53 trillion / net profit ₩75.14 trillion / net profit ₩75.21 trillion, 1~3 net profit ₩193.58 trillion | revenue 1,952,532 / operating profit 1,074,405 / 985,322 / net profit 751,361 / 752,081, 1,935,788 | Confirmed | link |
| Quarterly source values (Q4 2025-Q2 2026) and trailing 12-month totals | net profit 2025Q4 19.29 + 2026Q1 47.10 + Q2 71.27 + Q3 75.21 = ₩212.87 trillion / 12 revenue ₩594.46 trillion·operating profit ₩274.24 trillion·net profit ₩213.63 trillion | (2025.12)·(2025.09·2026.03)·(2026.06) | Confirmed | link |
| Quarter-stacked forward estimate (2026 forward P/E) | Q1 57.23 + Q2 89.49 + Q3 107.44 + Q4 121.3 = operating profit ₩375.5 trillion → net profit ₩286.5 trillion → ₩1,283.26 trillion ÷ ₩286.5 trillion = 4.5x | (self-estimate) | Unverified | link |
| Q4 exchange-rate assumption (won/dollar) | 2 ₩1,501.7 → 3 ₩1,418.7 → 10 ₩1,336.5 → 11 1~6 ₩1,345.9 / : 6 30 ₩1,550.9 → 9 30 ₩1,355.4 → 11 6 ₩1,358.7 | TCL CSOT 10 28 ₩1,324.30, ₩2 | Confirmed | link |
| Precise recalculation of the sub-sector key metric (P/B) | 1,283.26 ÷ 6 565.06 = 2.27x / ÷ 3 614.11 = 2.09x / ÷ 3 599.9 = 2.14x / approx. 1.94x | 565,064,740· 14,244,936; IR p.14 3 ₩614.11 trillion | Confirmed | link |
| Cash capacity (net cash, FCF yield) | ₩197.63 trillion = 15.4% / 12 FCF ₩200.1 trillion → FCF 15.6% | IR p.14·15; 85,315,148· 47,522,179· 4,630,970; 2025 3 56,515,496·35,972,891·3,420,768; 145,355,192·31,234,818·1,725,464 | Confirmed | link |
| Updated enterprise value (EV) metrics | EV ₩1,085.63 trillion / EV/EBIT 3.96x / EV/Sales 1.83x / EV/EBITDA 3.37x / operating profit EV/EBIT 2.89x | −₩32.62 trillion·EV ₩1,250.64 trillion·EV/EBIT 13.28x·EV/Sales 3.22x·EV/EBITDA 8.79x·FCF ₩33.16 trillion·FCF 2.58% | Mismatch | link |
| Updated denominators for the on-screen card P/E and P/B | PER 27.47x / PBR 2.09~2.27x | PER 28.99x·PBR 2.71x | Mismatch | link |
| How the previously shown forward P/E values (6.8x, 3.91x) were derived | 1,283.26 ÷ = 6.81x → 6.8x 1×4 / 3.91x = PER 4.26x × 1,283.26 ÷ 1,645.73 | 3.91x, PER 6.8x | Mismatch | link |
| Nature of the automatic industry median (comparison basis) | ' ', 2, PER 27.47x·PBR 2.57x, ROE·operating margin·revenue · 0.03(3%) | — | Confirmed | link |
| Explanation of automated self-check flags | audit passed, 14 , 0 | — | Confirmed | link |
| Difference in multiple range from excluding preferred shares from market cap | 2026-11-03 ₩171,600 × 802,371,203 = approx. ₩137.69 trillion → approx. ₩1,420.95 trillion ÷ net profit ₩286.5 trillion = approx. 4.96x(≈5.0x) | (005935) 2026-11-03 ₩171,600· 802,371,203; 3 ₩3,691,709,905,003 | Confirmed | link |
| Consistency of peer assumptions (SK Hynix) | 4 : D +5%· · · −5.4% → DS operating profit +13.6% | SK IR : 4 D 20% · , HBM4 HBM3E ; 3 ₩309.09 trillion· 12 net profit ₩188.90 trillion | Confirmed | link |
| Customer and competitor figures checked against source documents (figure audit) | 3 513.6· , 772.6·2026 approx. 2,200, 393.7, 2026 1,350~1,450, 1,190→2,790, FY2026 4 revenue 542.3· 615±15 | 3 : 51,355· −9,065 / 3 : 77,259·'2026 approx. 2,200 ' / FY27 1 : 394 / 3 : 1,350~1,450 / FY27 2 CFO : 1,190→2,790 ' ', PC / 2026-09-30 : revenue 542.3, FQ1-27 revenue 615±15 | Confirmed | link |
Recent filings Source
- 2026-06-08OwnershipLargest-shareholder ownership change report
- 2026-06-08OwnershipOfficers'/major-shareholders' holdings report
- 2026-06-05OwnershipOfficers'/major-shareholders' holdings report
- 2026-06-05OwnershipOfficers'/major-shareholders' holdings report
- 2026-06-05OwnershipOfficers'/major-shareholders' holdings report
- 2026-06-04OwnershipOfficers'/major-shareholders' holdings report
- 2026-06-02OwnershipOfficers'/major-shareholders' holdings report (amended)
- 2026-06-02OwnershipOfficers'/major-shareholders' holdings report
- 2026-06-01Large-business-group status disclosure
- 2026-06-01Corporate governance report
- 2026-06-01Large-business-group status disclosure
- 2026-06-01OwnershipOfficers'/major-shareholders' holdings report
📖 Plain-language glossary — expand if you are new to this
- P/E
- How many times a year's net profit the price is worth (lower is cheaper relative to earnings). The P/E here is on trailing (last full-year) results; for companies whose earnings swing fast (memory chips and other cyclicals/high-growth), a forward P/E on this year's expected earnings is more accurate.
- P/B
- Price relative to net assets (equity). Around 1x means it trades near book value; below 1x means below book.
- P/S
- Price relative to a year's revenue — useful for growth companies with thin earnings.
- Net debt / EV
- Net debt = interest-bearing debt − cash. Negative means more cash than debt (net cash). EV (enterprise value) = market cap + net debt, closer to what it would cost to buy the whole business.
- EV/EBIT · EV/EBITDA · EV/Sales
- Enterprise value against operating profit (EBIT), EBITDA, or revenue. Unlike P/E these reflect debt and cash; lower is cheaper relative to earnings power or sales.
- FCF / FCF yield
- Free cash flow = operating cash − capex, the cash actually left over. FCF yield = FCF ÷ market cap; higher means more cash generated per unit of market value.
- Intrinsic value (DCF)
- Future free cash flow (or, for some capex-heavy but profitable names, forecast earnings) discounted to today to estimate per-share value. Because it shifts a lot with the discount-rate and growth assumptions, it is shown as a bear/base/bull range, and the basis and assumptions are disclosed in one line beneath it.
- ROE
- How much profit the company earns in a year on its equity (%). Higher means better returns on capital.
- EPS / BPS
- Earnings per share / net assets (book value) per share.
- Operating / net margin
- Profit left from the core business / final profit after tax and interest, per unit of revenue.
- Debt ratio
- Debt relative to equity (%). Higher means more reliance on borrowing (norms vary by sector).
- Current ratio
- Assets convertible to cash within a year against debt due within a year. Above 100% leaves some short-term headroom.
- Interest coverage
- How many times operating profit covers the interest owed. Below 1x means operating profit alone struggles to cover interest.
- Dividend yield / payout ratio
- The year's dividend as a % of today's price / the share of earnings paid out as dividends.
- Revenue CAGR
- Multi-year growth expressed as a single yearly average (compound annual growth rate).
- RSI (short-term signal)
- Whether recent price action is overheated or beaten down. Above 70 is overbought, below 30 oversold.
- MA20 / MA60 (moving averages)
- The 20- and 60-day average price. Price above them signals a firmer short-term trend.
- vs 52-week high
- How far below the past year's peak the price sits now (%).
All figures are for reference only; how they read varies by sector and over time.
Sources: Korea FSC market-price API (data.go.kr), OpenDART, KRX/KIND — public data only.
Bong Stocks presents public-data-based information for reference only. It is not investment advice and contains no target prices, ratings, or buy/sell recommendations. Verify independently before making any decision.