Global investment themes and market charts Report Interpretation

Overview

This is a cross-asset and cross-theme collection of Morgan Stanley Research charts. It argues that several equity opportunities have improved after recent market adjustments, particularly momentum, Korean equities, US quality stocks and GenAI platform economics, while flagging weakening US macro confidence, French election uncertainty and weather-related Asian inflation risks.

Core views

The report first revisits Price Momentum after a severe late-July drawdown. Momentum had become stretched in the first half of 2026, but Morgan Stanley’s valuation measure now indicates that US Momentum valuations have normalized, Japan has round-tripped, and Europe was never expensive. The measure compares the long and short legs’ book-to-price percentiles against a 30-year history, with higher readings denoting less expensive valuations. Drawing on historical analogues, the institution recommends remaining invested in Price Momentum because it says the factor has tended to rebound most after severe drawdowns and mean forward returns have typically been positive across factors, regions and horizons. For Korea, Morgan Stanley believes AI-related concerns have largely been priced in and that the worst phase is likely over. Its capitulation index is at a historical low outside crisis periods, technical analysis points to leveraged-ETF unwinding, and hedge-fund de-grossing appears near its end. A re-rating, however, depends on a broad-based recovery rather than only technology: non-technology sectors are seeing improving 2026 and 2027 earnings revisions while technology earnings trends remain strong. Morgan Stanley keeps its KOSPI target at 9,000, implying roughly 35% upside, with a 10,500 bull case (+60%) and a 5,500 bear case (-15%). In the US, the report frames the market as moving from early to mid cycle. As the business cycle matures and post-recession operating leverage moderates, it expects leadership to rotate from low-quality companies toward high-quality companies with stable earnings, strong margins and operating efficiency. High Quality represents 42% of the S&P 500, compared with 28% for Low Quality, while median stock earnings growth is accelerating to 14% as earnings-revision breadth improves. Morgan Stanley argues that this combination should support greater index resilience and broader participation, and maintains a view of 8,000 on the S&P 500 by year-end. Consumer survey evidence is less constructive for the macro backdrop. Morgan Stanley’s AlphaWise survey of roughly 2,000 US consumers, conducted July 23–27, found that confidence in the US economic outlook weakened after two months of improvement. Forty-nine percent expected the economy to worsen over the following six months and 30% expected improvement, taking the survey’s Net Outlook score to -19% from -10% in mid-June. Household-finance sentiment was more resilient: its Net Outlook was +22%, versus +24% in the prior month and +10% a year earlier. The report notes that this outlook series has a moderate correlation with the University of Michigan consumer-sentiment index and the Conference Board consumer-confidence index. On AI infrastructure and monetization, Morgan Stanley is bullish on surging AI capex and model-training spending despite investor concern about returns on invested capital. Its bottom-up GenAI frameworks identify paths to ROIC of 25–40% for GPU-rental models, more than 40% for owned-infrastructure API models, and roughly 25% for third-party-hosted APIs. The analysis emphasizes that continued investment and innovation by model labs must lower token costs and improve token efficiency. On this basis, it reiterates a bullish outlook on Amazon, Alphabet, Microsoft and META. France is presented as a policy-risk case. Morgan Stanley expects elevated uncertainty and constrained fiscal reform through the 2027 elections. Based on analysis of more than 90 presidential elections, it believes markets typically begin pricing election outcomes four to six months ahead of the event—approximately from November in this case. The institution expects French equities to face a larger overhang as investors avoid positioning amid headline risk, French government bonds to continue underperforming EU peers in coming months, and the euro to carry a greater election-related negative risk premium. For Asia, the report expects one of the strongest El Niño events on record to bring dry conditions to India, Southeast Asia and Australia through early spring 2027. It considers Asia more insulated from growth damage than in earlier cycles because grain reserves are higher, food has a lower weight in CPI, and an industrial and capex super-cycle is underway. India and Indonesia are relatively more exposed to downside growth risk. El Niño nevertheless shifts inflation risks higher and could bring earlier or more rate hikes. Finally, Morgan Stanley highlights rapid acceleration in humanoid-robotics financing. Global humanoid funding in 2025 was three times the 2024 level, and 2026 year-to-date funding had already exceeded the full-year 2025 level as of July 15. China is described as further advanced on the humanoid-development curve than much of the world, while sector financing activity there remains robust.

Analysis framework

Morgan Stanley organizes the report as a selection of charts from separate strategy, consumer, technology and economics research. It combines historical valuation and election comparisons, market-positioning indicators, earnings-revision data, a consumer survey, and bottom-up unit-economics frameworks to connect evidence with its conclusions.

Methodology notes

• Quantitative, Factor, and Portfolio TheoryStyle factor analysis

  Momentum factor valuation and historical drawdown analysis

  The report compares long-leg and short-leg book-to-price valuation percentiles against a 30-year history to judge whether Price Momentum is expensive and to assess its behavior after severe drawdowns.

• Corporate Fundamentals and FinanceROIC–WACC spread

  Bottom-up GenAI ROIC frameworks

  The report models incremental returns for GPU rental and API business models to test whether AI infrastructure and inference economics can generate attractive returns on invested capital.

• Event-Driven and Behavioral FinanceEvent-driven analysis

  Historical presidential-election analysis

  Morgan Stanley examines more than 90 presidential elections to estimate when election risk tends to be reflected in French asset prices.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• KOSPI

  Morgan Stanley expects AI concerns to be largely priced in and sees potential re-rating if recovery broadens.

  Strengths

  Historical-low non-crisis capitulation reading; improving non-technology earnings revisions; technology earnings remain strong.

  Weaknesses

  Re-rating depends on broad-based recovery.

  Comparison

  French government bonds are expected to underperform EU peers; no direct Korea peer comparison is provided.

  Risks

  The report’s bear case is 5,500, or -15%.

• S&P 500

  The report expects a shift toward high-quality constituents as the cycle matures.

  Strengths

  High Quality is 42% of the index and earnings-revision breadth is improving.

  Weaknesses

  Low-quality leadership is expected to fade as operating leverage moderates.

  Comparison

  High Quality represents 42% of the index versus 28% for Low Quality.

• Amazon, Alphabet, Microsoft, META

  Morgan Stanley links these companies to potentially attractive GenAI inference economics.

  Strengths

  The report sees paths to 25%+ ROIC across major AI inference business models.

  Weaknesses

  Returns depend on model-lab investment and innovation to reduce token costs and improve token efficiency.

  Comparison

  Owned-infrastructure APIs show 40%+ ROIC paths, above GPU rental at 25–40% and third-party-hosted APIs at roughly 25%.

  Risks

  AI capex and model-training spending face investor scrutiny over ROIC.

Key data

• KOSPI target9,000Morgan Stanley’s maintained target; implies roughly 35% upside.
• KOSPI bull case10,500 (+60%)Reported upside scenario.
• KOSPI bear case5,500 (-15%)Reported downside scenario.
• High Quality share of S&P 50042%Versus 28% for Low Quality.
• Median stock earnings growth14%Accelerating as earnings-revision breadth improves.
• S&P 500 view8,000 by year-endMorgan Stanley’s US equity-strategy outlook.
• US economic Net Outlook-19%Down from -10% in the mid-June survey; 49% expect deterioration and 30% improvement over six months.
• US household-finance Net Outlook+22%Versus +24% last month and +10% a year earlier.
• GenAI ROIC paths25–40% GPU rental; 40%+ owned-infrastructure APIs; roughly 25% third-party-hosted APIsBottom-up framework outputs.
• Humanoid funding2025 was 3x 2024; 2026 YTD exceeded full-year 2025Global venture funding, as of July 15, 2026.

Impact & implications

The report favors exposure to normalized momentum, Korean equities, US quality and selected GenAI platform beneficiaries, but its broader message is conditional: Korean re-rating needs broader earnings recovery, AI returns require lower token costs and better efficiency, and macro and policy risks may weigh on consumers, French assets and parts of Asia.

Risks

• A Korean equity re-rating depends on a broad-based recovery rather than technology strength alone.
• US consumers’ outlook for the economy weakened in July.
• French policy uncertainty and headline risk are expected to remain elevated into the 2027 elections.
• El Niño could raise Asian inflation risks and lead to earlier or more interest-rate hikes, with India and Indonesia relatively more exposed to growth downside.
• GenAI return assumptions depend on continued reductions in token costs and improvements in token efficiency.

What to watch

• Whether momentum valuations remain normalized after the late-July drawdown.
• The breadth of Korean earnings revisions outside technology and the persistence of technology earnings strength.
• US earnings-revision breadth, the quality-factor rotation and progress toward the S&P 500 year-end view.
• Subsequent US consumer surveys, particularly the gap between macro confidence and household-finance sentiment.
• Token-cost reductions and efficiency gains that support GenAI inference ROIC.
• French asset pricing from about November as election risk is expected to enter markets four to six months ahead of the 2027 vote.
• El Niño weather conditions through early spring 2027, Asian food inflation and rate-hike expectations.
• The pace of global and Chinese humanoid-robotics funding.