South Korea equity market Report Interpretation
The report upgrades Korea to Overweight with high conviction after a sharp correction, citing capitulation signals, low valuations and late-stage deleveraging. It expects a choppier but healthier 2H26 led initially by technology and then broadening into financials, industrials, healthcare and consumer staples.
Summary
The report upgrades Korea to Overweight with high conviction after a sharp correction, citing capitulation signals, low valuations and late-stage deleveraging. It expects a choppier but healthier 2H26 led initially by technology and then broadening into financials, industrials, healthcare and consumer staples.
- KOSPI target remains 9,000, with 10,500 bull and 5,500 bear cases.
- The capitulation index reached -2.53 on July 30, a non-crisis low since 2008.
- KOSPI forward P/E fell below 5x; chipmakers traded around 3.5x forward P/E.
- Morgan Stanley expects foreign flows, rather than domestic retail liquidity, to drive the next leg higher.
- Industrials and Tech are the top sector overweights; Banks, Communication Services, Healthcare and newly added Consumer Staples are also Overweight.
Report Interpretation
Overview
This South Korea equity-strategy report argues that the late-July sell-off has reset an overextended, leverage-heavy market and that the worst is likely behind it. Morgan Stanley retains a 9,000 KOSPI target but stresses that a sustainable advance requires a broadening beyond the technology and AI-memory complex.
Core views
Morgan Stanley characterizes the correction as a reset after KOSPI surged 101% in 1H26 and then fell sharply from mid-June. By July 30, the index had dropped 18% in two days to 5,594 and more than 38% from its peak, surrendering over 60% of its year-to-date gains. The firm believes the July 30 low was likely the trough, although it expects recovery to remain volatile rather than follow a smooth path. Its capitulation index reached -2.53, the lowest non-crisis reading since 2008 and near historical trough territory; forward KOSPI P/E fell below 5x, its lowest level since 2004, while the two chipmakers traded around 3.5x forward P/E. The report sees deleveraging as an important reason market stress should diminish. Retail buying and leveraged single-stock ETFs amplified both the earlier rally and subsequent decline. The 14 leveraged products tied to Samsung Electronics and SK hynix accumulated W15.1tn of retail net purchases, while combined AUM rose from about W5tn at launch to W17.4tn on June 25 before falling back to roughly W5.4tn. Margin lending declined from W38.6tn on June 24 to W33.0tn on July 30, and leveraged single-stock ETF AUM fell about 70% from a June US$40bn peak to US$11.8bn by July 29. Asia hedge-fund gross exposure fell to 130%, close to a five-year trough, suggesting about 75% of a typical de-grossing cycle had completed. Morgan Stanley therefore expects the amplitude of market swings to narrow as retail leverage is constrained, though near-term volatility remains likely. Liquidity leadership is expected to shift. Retail investors had bought over W114tn of KOSPI equities in 2026 through July 29, offsetting more than W166tn of foreign selling, while foreigners sold a cumulative US$30bn from June 22. The report argues that renewed foreign inflows are required for the next meaningful advance; July foreign flows had turned positive by July 31. Domestic institutions may provide support, but pension-fund buying is likely to be selective because the National Pension Service's domestic-equity allocation remained above its 20.8% strategic target despite falling from 29.4% at end-May into the mid-20% range in July. The firm notes that historical periods of large foreign outflows alongside strong retail inflows have tended to be associated with higher subsequent volatility. AI and memory remain central to the Korea outlook, but the report believes much of the negative news has already been reflected in semiconductor prices. Samsung Electronics, SK hynix and Samsung Electro-Mechanics had fallen 43%, 55% and 61%, respectively, from their year-to-date peaks as concerns rose over hyperscaler capex, GPU shipments, memory pricing and LTAs, China competition, and AI-compute efficiency. Morgan Stanley's constructive base case rests on its US technology team's view that compute demand will exceed supply for years and hyperscaler GenAI ROIC could reach 25-50%. It cites expected hyperscaler capex of US$805bn in 2026 and US$1.2tr in 2027, while acknowledging US data-center constraints and a projected 38GW power shortfall in 2026-28. The firm considers these constraints speed bumps rather than barriers and estimates that US data-center capacity could be more than 10% below what would be required by 2028 under its global semiconductor sales assumptions. For memory, Morgan Stanley argues that current long-term supply agreements differ from prior-cycle arrangements because they are non-cancellable and non-refundable, backed by prepayments and guarantees; some suppliers reportedly have 50% down payments for 2027 demand and 100% for 2028 demand. Samsung said binding memory LTA commitments are tightening 2027 capacity. China is viewed as a partial relief rather than a full solution because technical compatibility does not establish high-volume qualification, and because power, yield, thermal performance, quality assurance and supply stability remain critical. The report also expects global AI infrastructure to absorb growing memory volumes, leaving uncertain how much capacity Chinese producers CXMT and YMTC can allocate to consumer applications. Near-term chipmaker catalysts are primarily capital management and shareholder-return announcements. Morgan Stanley expects investors to focus on revised Value-up plans from SK hynix and Samsung Electronics. It also identifies HBM4 pricing as a possible catalyst: Samsung expects HBM4 to represent around 60% of total HBM sales by end-2027, and a price above market expectations of US$3/Gb could support broader 2026-27 DRAM pricing. The September iPhone 18 launch is another demand watch point, because mobile represents 30-40% of global DRAM demand and 25-30% of NAND demand; the report cites Korean Apple supply-chain indications of 5-10% year-on-year volume upside. The macro backdrop is described as resilient and broader than a narrow technology-export story. Morgan Stanley raises the probability of full-year 2026 real GDP growth in the mid-3% range from its previous 2.8% forecast; real GDP growth exceeded 3% in each of the prior two quarters and reached 3.75%, versus 1.1% in 2025. Consumption is expected to remain supported by wealth effects, semiconductor strength, policy measures and tourism, with the government targeting 23mn visitors in 2026, up 20%. Inflation is expected to stay elevated in 3Q26 after June CPI rose 3.2% year on year, though the firm expects that to be the year's peak. The Bank of Korea raised its policy rate to 2.75% in July, and Morgan Stanley expects a measured cycle to 3.5% by 2Q27 rather than front-loaded tightening. On FX, the report expects KRW to strengthen in 2H26 despite a possible near-term correction after the currency's sharp rally and oversold 14-day RSI. It retains a USD/KRW forecast range of about 1,420-1,430, with risk of an overshoot toward 1,400. It attributes supportive flows to potential corporate FX conversion for tax payments and local capex, plus higher hedging by shipbuilders. It also notes that the recent KRW-KOSPI relationship has inverted, partly owing to foreign-equity FX hedging and concerns about rebalancing if the equity rally becomes too narrow or rapid. Corporate reform is another prospective re-rating driver. Morgan Stanley expects greater attention to Value-up plans and capital allocation, especially from Samsung Electronics and SK hynix. Deferred Commercial Act provisions due September 10, 2026 would tighten cumulative-voting requirements and increase separately elected audit-committee seats from one to two for listed companies with assets of W2tn or more. The report believes implementation of governance and capital-market reforms can support an ex-IT re-rating, while failure to execute policy is a material downside risk. The firm retains a June 2027 KOSPI target of 9,000, a bull case of 10,500 and lowers its bear case to 5,500 from 6,000. It cuts its 12-month forward EPS growth estimate to 78% from 100% as earnings roll forward, and lowers its target P/E to 6.3x from 8.3x, while current forward P/E is about 5.0x. The base case uses 78% EPS growth, 6.7x target forward P/E and implied ROE of 31.5%; the bull case uses 88% EPS growth and 7.4x P/E, while the bear case uses 58% EPS growth and 4.5x P/E. If Samsung Electronics and SK hynix meet Morgan Stanley price targets, its analysis implies KOSPI at 8,400; if they attain bull-case values, the implied level is 9,100. A broad-based earnings recovery, structural opportunities in a multipolar world and energy transition, and effective reform underpin the bull case. The bear case assumes worsening AI newsflow, sustained high oil prices and inflation, Fed-hike pressure, geopolitics and stalled reforms; Morgan Stanley considers a break below 5,000 unlikely absent an extended Middle East conflict that sends oil above US$140. Sector positioning remains a barbell. Industrials are ranked first and Tech second, while Banks, Communication Services, Healthcare and newly added Consumer Staples are also Overweight. Morgan Stanley expects initially sharp rebounds in the sectors and stocks that led in 2026 but fell most from their peaks, particularly chipmakers, followed by broader mid-term participation. It stays Underweight Energy and Materials despite improving earnings revisions because their broader sector stories remain unfavorable. Expected upward revisions are concentrated in insurance, banks, overseas construction, machinery, shipyards, capital goods including defense and holding companies, and consumer sectors. KOSPI ex-IT 2026 net-profit estimates had increased 27% since the start of the year, supporting the broader-recovery thesis.
Analysis framework
Morgan Stanley combines market technicals and positioning data with valuation, earnings-revision, macro, policy and sector analysis. It first assesses whether the sell-off has reached a capitulation point, then examines leverage, ETF and investor-flow normalization. It links the market outlook to AI-memory demand, Korean macro conditions, FX and corporate reform, and derives base, bull and bear KOSPI scenarios from forward EPS growth, target P/E and implied ROE assumptions.
Methodology notes
Capitulation index and positioning analysis
The report uses market breadth, price momentum, ETF positioning, margin balances and hedge-fund exposure to judge whether forced selling and investor fear are approaching a market trough.
Forward P/E-based KOSPI valuation
Morgan Stanley applies target 12-month forward P/E multiples to forward EPS assumptions to derive its KOSPI base, bull and bear targets.
AI infrastructure and memory supply-demand analysis
The report evaluates hyperscaler capex, compute constraints, GPU deployment, HBM demand, memory contracts, pricing and Chinese supply to assess the Korean semiconductor outlook.
Momentum, Value and Quality factor performance
The report observes that the local Price Momentum factor has unwound while Value and Quality held up better during the correction, informing its view of a shift toward more defensive styles.
Asset mapping & comparison
Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).
- KOSPIPrimary market subject; Morgan Stanley sees a post-capitulation re-rating opportunity contingent on broader earnings participation and foreign inflows.
- Strengths
- Low forward valuation, late-stage deleveraging, resilient growth and prospective corporate reform.
- Weaknesses
- Heavy technology concentration of roughly 60-65% of market capitalization limits near-term index breadth.
- Comparison
- Base target 9,000; bull case 10,500; bear case 5,500.
- Risks
- AI capex disruption, geopolitics, Fed policy, stalled reforms, weak domestic demand and housing stress.
- Samsung Electronics and SK hynixKey AI-memory beneficiaries and major drivers of the KOSPI scenario analysis.
- Strengths
- Binding LTAs, expected tighter 2027 capacity, potential HBM4 pricing support and possible capital-management catalysts.
- Weaknesses
- Shares remain exposed to volatile AI-capex news and concerns over China competition.
- Comparison
- Both traded around 3.5x forward P/E; meeting Morgan Stanley price targets would imply KOSPI at 8,400.
- Risks
- Data-center build-out delays, lower AI returns, demand destruction from higher memory costs and accelerating Chinese capacity.
- SK InnovationAdded to Morgan Stanley's Focus List.
- Strengths
- 2Q26 operating profit of W3.4trn, up 61% quarter on quarter, highlighted stronger refining, lubricant and LNG earnings power.
- Weaknesses
- Earnings are expected to moderate in 2H26.
- Comparison
- Morgan Stanley expects a re-rating catch-up to S-Oil.
- Risks
- EV-business concerns remain, although the report says they are well documented and priced in.
- Banks and insurersOverweight sectors within the broadening-out strategy.
- Strengths
- Higher rates, resilient bank net interest income, improving non-interest income and shareholder-return execution.
- Comparison
- Morgan Stanley expects improving earnings momentum for both banks and insurers.
- Risks
- A weaker macro or domestic-demand backdrop could pressure earnings.
Key data
- KOSPI base target9,000June 2027 basis; bull case 10,500 and bear case 5,500.
- KOSPI decline18% to 5,594 by July 30, 2026The index fell more than 38% from its peak after gaining 101% in 1H26.
- Capitulation index-2.53As of July 30, 2026; its lowest non-crisis reading since 2008.
- KOSPI forward P/EBelow 5xLowest since 2004; both chipmakers were around 3.5x forward P/E.
- 12-month forward EPS growth estimate78%Reduced from 100% as earnings were rolled forward.
- 2026 real GDP growthMid-3% rangeRaised from a prior 2.8% forecast; recent quarterly growth reached 3.75%.
- BoK policy rate2.75%After the July hike; terminal rate expected at 3.5% by 2Q27.
- AI hyperscaler capexUS$805bn in 2026; US$1.2tr in 2027The report treats the ability to earn adequate ROIC on this spending as a key market debate.
- Foreign KOSPI flowsUS$30bn cumulative sales since June 22The report sees renewed foreign inflows as the main driver of further upside.
Impact & implications
Morgan Stanley believes Korea has moved from a liquidity-driven, narrowly led rally into a post-correction phase where valuation, earnings breadth, foreign flows and governance reform matter more. It expects technology to remain important, but regards wider participation across industrials, financials, healthcare, communication services and staples as necessary for a durable move toward its target.
Risks
- AI infrastructure build-out or compute growth slows, reducing demand visibility for IT and memory stocks.
- Persistent margin pressure in memory-consuming PC, mobile and consumer supply chains could cause demand destruction.
- Middle East escalation and higher oil or commodity prices could worsen Korean inflation, growth and corporate earnings.
- The Fed policy path may matter more than the Bank of Korea's path for Korean equities in this cycle.
- Capital-market and governance reforms may not be implemented.
- Domestic consumption could weaken and a severe Seoul metropolitan housing downturn could trigger construction insolvencies and higher retail delinquencies.
What to watch
- Hyperscaler AI capex guidance, GPU deployment, AI revenue and ROIC evidence.
- Memory pricing, HBM4 pricing, supply agreements and 2027 capacity conditions.
- Samsung Electronics and SK hynix Value-up or capital-management announcements.
- Foreign investor flows and the degree of sector broadening beyond technology.
- Implementation of Commercial Act provisions on September 10, 2026 and broader capital-market reforms.
- Inflation, Bank of Korea and Fed policy paths, oil prices and Middle East developments.
- The September iPhone 18 launch and related DRAM/NAND demand signals.