SK Hynix Q2 2026 Earnings: Record 60.5T KRW Profit, HBM4 Mass Production — And Why the Stock Still Fell
SK Hynix Q2 2026 Earnings: Record 60.5T KRW Profit, HBM4 Mass Production — And Why the Stock Still Fell
The paradox is now complete. SK Hynix has just delivered the most profitable quarter in the history of memory chip manufacturing. And its Nasdaq-listed shares are trading below their IPO price.
For investors trying to make sense of this disconnect — and decide what to do next — the details of today's results and earnings call contain answers that the headline stock reaction does not.
SK Hynix Q2 2026 — Actual Results vs Consensus
| Metric | Q2 2026 Actual | Consensus Est. | vs Consensus | YoY Change |
|---|---|---|---|---|
| Revenue | 79.3T KRW (~$54.4B) | 83.4–84.1T KRW | -5.5% miss | +257% |
| Operating Profit | 60.5T KRW (~$41.6B) | 64.1–64.7T KRW | -6.6% miss | +557% |
| Operating Margin | 76% — all-time high | ~75% | +1pp beat | +4pp vs Q1 |
| Net Profit | 93.9T KRW (net margin 118%) | — | Includes 63.3T Kioxia gain | Record high |
| Cash Position | 88T KRW gross / 69.4T net cash | — | Debt-to-equity: 7% | Strongest ever |
| H1 2026 Cumulative Revenue | 131.9T KRW | — | First time above 100T in H1 | Historic milestone |
| SKHY Stock Reaction | -8.98% to $130.17 | — | Below $149 IPO price | 52-week range: $128.38–$194.80 |
* Sources: SK Hynix Q2 2026 Press Release (PRNewswire), Korea Times, Seoul Economic Daily, BigGo Finance. All data as of July 29, 2026.
Why the Results Missed Consensus — And Why It's Not What It Looks Like
SK Hynix reported second-quarter revenue of 79.3 trillion won and operating profit of 60.5 trillion won, missing consensus estimates by 5.5% and 6.6% respectively, despite sharp year-on-year gains.
The miss sounds alarming. It is not. Management explained it clearly on the earnings call, and the explanation actually contains a bullish signal.
Management explained: "Shipments of some high-value-add products were pushed back into the second half, and changes in the product portfolio appeared to have affected our blended ASP."
Translation: SK Hynix deliberately shifted some high-margin HBM4 shipments from Q2 to H2 — not because demand was weak, but because of supply chain timing and customer qualification schedules. This is not a demand miss. It is a delivery timing issue that actually makes H2 2026 estimates stronger.
Management added: "As HBM4 shipments ramp up in earnest and the 1c nm conventional DRAM shipment increases, we expect bit growth in the second half to be higher than the level in the first half. The growing HBM4 sales and the higher contribution from value-add products will have a positive impact on our blended ASP as well."
In other words: H2 2026 is expected to be better than H1. The Q2 miss is not a peak — it is a trough within an upward trajectory.
The 5 Most Important Things From the Earnings Call
① HBM4 Mass Production Confirmed — Yields Already Near HBM3E Maturity
SK Hynix commenced mass production of HBM4 during Q2, with yields already approaching HBM3E maturity levels, and shipped HBM4E samples to key customers, reinforcing its AI memory leadership.
This is the single most important operational data point from the call. HBM4 mass production has begun — and crucially, yields are already approaching the mature levels of the previous generation. This suggests the HBM4 ramp will be smoother and faster than the market feared, particularly given Samsung's well-publicized HBM3E yield difficulties in 2025.
HBM4E samples have also been delivered to key customers, with volume production targeted for 2027. The technology roadmap is intact and ahead of some analyst expectations.
② Long-Term Agreements With ~10 Customers — Multi-Year Demand Visibility
The company has secured long-term agreements (LTAs) with around 10 key customers, ensuring mid- to long-term supply stability.
When asked about LTA terms, management explained: "LTAs are designed to be specific to each customer and product, typically with a five-year term. Pricing structures vary to address price volatility and enhance long-term business stability. The agreements include mechanisms like deposits to strengthen contract implementation and demand visibility, allowing us to optimize investment and production planning."
Five-year LTAs with 10 customers — including pricing mechanisms and deposits — is a structural transformation of SK Hynix's business model. This is the direct counter-narrative to every "peak earnings" concern. Companies don't sign five-year supply agreements because they expect demand to peak next quarter.
This also explains part of the consensus miss: LTAs typically lock in pricing at levels slightly below spot market peaks in exchange for volume certainty. The "blended ASP disappointment" is actually the price SK Hynix paid for multi-year demand stability — a rational strategic tradeoff.
③ AI Investment Slowdown Fears — Management Dismissed Them Directly
Management dismissed fears of an AI investment slowdown, describing it as a transition toward monetization and higher utilization, with CSP investment remaining solid.
Management said it sees the trend differently, arguing that customers are using existing infrastructure more efficiently and pushing for faster monetization, not cutting back on investment.
This directly addresses the NVIDIA round-tripping concerns that contributed to yesterday's 10.8% KOSPI crash. SK Hynix management — who have direct visibility into what their ~10 LTA customers are actually ordering — is saying AI infrastructure spending is not slowing. It is transitioning from raw buildout to efficiency optimization and monetization. Both phases require memory.
④ H2 2026 Guidance — Better Than H1
For Q3, DRAM shipments are guided up 10% QoQ and NAND low single-digit, while capex for 2026 is seen reaching the high KRW 40 trillion range.
The company projects global DRAM demand to grow by approximately mid-teens percentage year-over-year in 2026, and NAND demand to grow by a high teens percentage. For Q3, SK Hynix expects DRAM shipments to grow by about 10% quarter-over-quarter.
10% QoQ DRAM shipment growth in Q3, combined with an improving product mix as HBM4 ramps, means Q3 revenue and operating profit should exceed Q2. The consensus miss in Q2 becomes less relevant when H2 guidance is explicitly stronger than H1.
⑤ Balance Sheet Strength — 88T KRW Cash, Net Cash 69.4T
The company's cash position swelled to KRW 88 trillion, with net cash of KRW 69.4 trillion, as interest-bearing debt fell to KRW 18.6 trillion, underscoring robust financial health.
With net cash swelling to 69.4 trillion won after a landmark NASDAQ ADR listing, the company signaled potential additional shareholder returns later this year.
A debt-to-equity ratio of 7% with 69.4 trillion KRW in net cash gives SK Hynix extraordinary financial flexibility. The signal of "potential additional shareholder returns later this year" is particularly important for SKHY holders — this suggests the company may announce special dividends or buybacks before year-end, directly benefiting US ADR investors.
Global Context: Where SK Hynix Ranks Among the World's Most Profitable Companies
SK Hynix posted Q2 operating profit of 60.54 trillion won, ranking 4th among global tech firms, with an operating margin of 76.3%, second only to Micron worldwide. Its quarterly operating profit ranked fourth all-time, behind Samsung Electronics at No. 1, followed by NVIDIA ($53.5 billion, ~78 trillion won) and Apple ($50.85 billion, ~74 trillion won).
A memory chip company that outsells Apple and NVIDIA on quarterly operating profit — with a 76% operating margin — is not a company in distress. It is a company at the epicenter of the most important technology investment cycle in a generation.
The NAND Surprise: Enterprise SSD Revenue Doubled
The Q2 results contained a significant positive surprise that received less attention than the HBM story:
Enterprise SSD revenue doubled quarter-on-quarter. Solidigm, the NAND subsidiary, saw its revenue from high-capacity 30TB-and-above eSSDs more than triple. NAND ASP surged mid-50% quarter-on-quarter. 321-layer NAND now accounts for a significant share of production.
The enterprise SSD story is important because it represents a second, independent driver of SK Hynix profitability beyond HBM. AI training and inference require massive storage as well as fast memory — and SK Hynix's Solidigm subsidiary is capturing that demand through enterprise SSDs. This diversification within the AI memory ecosystem is a positive that the market has not yet fully appreciated.
Why SKHY Fell Despite Record Results
The stock reaction — SKHY down 8.98% to $130.17, well below its $149 IPO price — reflects three factors that are distinct from the fundamental results:
Factor 1 — The consensus miss created a negative headline. In a market that has been selling Korean semiconductor stocks for three weeks, a 6.6% operating profit miss vs consensus is enough to trigger another leg down regardless of absolute profitability levels. The narrative was "miss" — not "best quarter in memory industry history."
Factor 2 — China semiconductor fears remain unresolved. Yesterday's DUV lithography and CXMT headlines created a structural bear narrative. One earnings call — however strong — does not fully resolve a geopolitical concern. The market needs to see whether CXMT's DUV equipment actually works at production quality and whether Chinese HBM yield rates improve meaningfully. That resolution will take months, not days.
Factor 3 — The Kioxia gain inflated net profit. Net profit of 93.9 trillion KRW with a 118% net margin sounds extraordinary — but it includes 63.3 trillion KRW in non-operating investment gains from the Kioxia stake sale. This one-time item will not repeat in Q3. Some investors may be concerned about the sustainability of net profit levels when the non-operating gains are stripped out.
As Odaily noted: "Market concerns over peak memory pricing and capex expansion shift the valuation focus to cycle sustainability." The market is asking: how long can this last? The earnings call provided strong answers — but not definitive ones.
The Bull Case: Why the Selloff Is Overdone
The fundamental case for SK Hynix has arguably strengthened after today's earnings call, not weakened:
- Five-year LTAs with 10 customers — the clearest possible signal that major AI customers believe demand persists for years
- HBM4 mass production confirmed with mature yields — the technology leadership is intact and expanding
- H2 guided stronger than H1 — 10% QoQ DRAM shipment growth in Q3 with improving mix
- 88T KRW cash / 7% debt-to-equity — balance sheet strength provides buffer against any demand slowdown
- Management explicitly dismissed AI slowdown fears — with direct customer visibility, not analyst speculation
- SKHY at $130 implies 52-week low valuation — trading near the bottom of its range despite record operating results
The Bear Case: Why Caution Is Still Warranted
- Consensus miss creates negative momentum — in a downtrending market, misses attract more selling
- China DUV and CXMT narrative unresolved — structural bear thesis needs time and data to disprove
- LTAs explain ASP miss but cap upside — five-year agreements provide stability but limit spot market upside
- Kioxia gain won't repeat — net profit will normalize lower in Q3 without the 63.3T non-operating gain
- SKHY near 52-week low — technically weak, could face further pressure before finding support
What Comes Next: The Calendar That Matters
| Date | Event | Why It Matters for SK Hynix |
|---|---|---|
| July 30 | Meta Q2 + Samsung full divisional breakdown | Meta capex guidance; Samsung HBM4 market share data |
| July 31 | Microsoft + Amazon Q2 | Azure/AWS AI infrastructure spending — direct HBM demand signal |
| August 1 | Korea July semiconductor export data | Real-time confirmation that demand remained strong through July |
| August (ongoing) | CXMT DUV equipment validation results | Whether China's 5 DUV machines actually work at production quality |
| Q3 2026 (October) | SK Hynix Q3 earnings | First quarter with significant HBM4 revenue — should show H2 ramp |
| Late 2026 | Potential additional shareholder returns | Management signaled special dividends or buybacks possible |
What Should SKHY and Korea ETF Investors Do?
If you hold SKHY:
At $130, SKHY is trading near its 52-week low despite SK Hynix just reporting the most profitable quarter in memory industry history. The fundamental thesis — HBM dominance, five-year LTAs, HBM4 mass production, 76% operating margin — is intact and arguably stronger after today's call. The question is not whether the business is performing well. It clearly is. The question is whether the China competition narrative will resolve, and when. Position sizing and patience are the operative variables. Do not make decisions based on today's stock reaction alone.
If you hold EWY or FLKR Korea ETFs:
SK Hynix represents approximately 15–20% of EWY. Today's results — while producing a stock decline — do not change the fundamental case for Korean equities. The KOSPI is now trading at forward P/E below 6x, below 2008 GFC trough levels, while Korean semiconductor exports surged 180% in the first 20 days of July. The ETF structure provides natural diversification across Korean defense, shipbuilding, power infrastructure, and financial stocks that are less affected by the China semiconductor narrative.
If you are considering new positions:
The framework from our KOSPI Corrections guide applies. Averaging into positions across multiple sessions — rather than a single lump-sum purchase — reduces timing risk in a market where sentiment volatility remains extreme. The fundamental support at these valuations is historically strong. The near-term sentiment headwinds are real but time-limited.
Final Thought: The Best Quarter in Memory History Was Not Good Enough for This Market
SK Hynix just reported:
- Revenue up 257% year-on-year
- Operating profit up 557% year-on-year
- 76% operating margin — highest in semiconductor manufacturing history
- HBM4 mass production confirmed with mature yields
- Five-year LTAs with 10 customers
- 88 trillion KRW in cash
- H2 2026 guided stronger than H1
The stock fell nearly 9%.
This is not a reflection of SK Hynix's business performance. It is a reflection of market sentiment that has been poisoned by legitimate structural concerns — China's semiconductor ambitions, NVIDIA's capital structure complexity, fears about the durability of AI infrastructure spending — that operate on a different timeline than quarterly earnings.
The disconnect between fundamental performance and stock price cannot persist indefinitely. Either the China competition narrative proves correct and the earnings trajectory reverses — in which case the stock reaction was rational. Or the fundamental case reasserts itself as the China DUV machines prove unable to close the yield gap with Korean HBM — in which case today's prices will look extraordinarily cheap in hindsight.
As Odaily's analysis noted: "AI memory demand remains strong, with HBM4 mass production and long-term agreements improving earnings stability. However, market concerns over peak memory pricing and capex expansion shift the valuation focus to cycle sustainability."
The cycle sustainability question will not be answered today. But today's earnings call provided the strongest possible evidence that the cycle — as of July 29, 2026 — remains fully intact.
Related Guides
- KOSPI -10.8%: Three Reasons Behind Yesterday's Crash
- SK Hynix Q2 2026 Earnings Preview: What to Watch
- SK Hynix ADR (SKHY): What the Nasdaq Listing Means for Global Investors
- The Complete Guide to Korean HBM Stocks (2026)
- Samsung Q2 2026 Earnings: Why Record Profits Sent the Stock Down 6%
- KOSPI Corrections: How to Tell a Buying Opportunity from a Falling Knife
Disclaimer: This article is for informational and educational purposes only and does not constitute investment advice. All financial data cited is sourced from SK Hynix's official Q2 2026 press release (PRNewswire), Korea Times, Seoul Economic Daily, BigGo Finance, and Investing.com earnings call transcripts. All data as of July 29, 2026. Always conduct your own research or consult a licensed financial advisor before making investment decisions.
