Japan equity strategy: coexistence of recovering AI momentum and catch-up gains in lagging cyclical value

Overview

The report presents a base-case scenario for Japan FY26 1Q earnings season: US hyperscalers' capital expenditure plans remain the key variable determining market direction. If their spending guidance stays strong, global capital may flow back into AI-related equities and the momentum factor may recover; however, because Japan's AI-related cyclical growth stocks already reflect high earnings expectations, capital may not once again concentrate one-way into the previous market favorites. The report places greater emphasis on the potential for catch-up gains and re-rating in lagging cyclical value stocks as the benefits of AI capex spread to adjacent chains such as electronic components, industrial equipment, and cybersecurity.

Core views

The core views are as follows. First, the base case is the coexistence of "AI inflows" and "catch-up gains in lagging cyclical value." Second, US hyperscaler capital expenditure remains strong, with Morgan Stanley's forecast significantly above market consensus, though widening credit spreads and financing pressure are potential constraints. Third, earnings expectations for cyclical growth, especially AI-related stocks, have already been revised upward, raising the earnings hurdle and reducing risk-reward. Fourth, cyclical value stocks with low P/B and still-supported earnings growth may gain both earnings upgrades and valuation re-rating if results beat expectations. Fifth, Japan's institutionalized fiscal expansion and yen/rate risks keep bank stocks attractive. Sixth, the diffusion of AI applications and intensifying model competition make cybersecurity, network infrastructure, and AI implementation support potential beneficiaries.

Analysis framework

The report combines scenario analysis, factor performance tracking, and stock screening. It uses three scenarios: the base case is strong US hyperscaler capital expenditure, fund flows returning to AI-related stocks, and simultaneous catch-up gains in cyclical value; the second scenario is that AI companies significantly beat high expectations, causing capital to concentrate again in AI cyclical growth stocks; the third scenario is that AI companies fail to clear the earnings hurdle while geopolitical and supply-chain risks intensify, leading to de-positioning in high-beta AI stocks and a momentum reversal. The stock screen is based on TOPIX constituents, excluding banks and companies that have already reported FY26 1Q results, selecting stocks in the top 50% for FY26 EPS growth and top 30% for beta, then distinguishing cyclical growth from cyclical value by P/B and three-month momentum.

Methodology notes

• Scenario analysisThree-scenario earnings season framework

  Use US hyperscaler capital expenditure, whether AI companies can exceed high expectations for earnings, and macro/geopolitical risks as the three variables to judge the path of Japan's equity market during earnings season.

  The base case is coexistence between returning AI fund flows and catch-up gains in cyclical value; the bullish AI scenario is renewed concentration of capital in AI cyclical growth; the risk-off scenario is de-positioning in high-beta AI stocks and a momentum reversal.

• Factor and valuation screeningBeta x P/B cyclical growth/cyclical value framework

  Use beta to measure market sensitivity, use P/B to distinguish high-valuation growth from low-valuation value, and combine this with EPS growth and three-month momentum filters.

  Cyclical value is defined as stocks with P/B between the 10th and 50th percentiles of TOPIX, excluding the top 10% by three-month momentum; cyclical growth is defined as stocks with P/B above the 80th percentile, excluding the bottom 10% by momentum.

• Macro policy analysisInstitutionalized fiscal expansion framework

  Assess how Japan's 2026 basic policy embeds fiscal expansion into the policy framework and infer the implications for the yen, bonds, rates, and bank stocks.

  The report believes the priority of fiscal consolidation has declined and the government may constrain the pace of Bank of Japan rate hikes, which would be reflected through yen weakness and falling bond prices; in an environment of fiscal concern and expanding digital deficits, bank stocks remain beneficiaries of rising rates and tools to hedge AI volatility.

Asset mapping & comparison

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

• Japan AI-related cyclical growth stocks

  Directly benefit from US hyperscaler capital expenditure and the AI investment cycle, but already embed relatively high earnings expectations.

  Strengths

  If US capital expenditure guidance remains strong, momentum and volatility factors may recover quickly, and stocks with strong earnings revisions may outperform.

  Weaknesses

  Valuation and earnings hurdles are high, so even superficially decent earnings may still lead to declines if they fail to meet market expectations.

  Comparison

  Relative to cyclical value stocks, growth stocks are more directly exposed to the AI theme, but their upside depends more on earnings beats.

  Risks

  Widening credit spreads, financing pressure for capital expenditure, geopolitical shocks, supply-chain concerns, and de-positioning in high-beta names.

• Japan lagging cyclical value stocks

  Indirectly benefit from the diffusion of AI capital expenditure, a weaker yen, capex expansion, and economic recovery.

  Strengths

  Low market expectations help limit downside risk during earnings season; if results beat expectations, they may gain both earnings upgrades and valuation re-rating.

  Weaknesses

  If capital again concentrates in high-growth AI stocks, the re-rating of value stocks may be delayed.

  Comparison

  Relative to cyclical growth stocks, value stocks have lower AI purity, but also lower expectations and a more obvious valuation discount.

  Risks

  Deteriorating macro uncertainty, AI rebounds drawing away capital, and earnings improvement falling short of expectations.

• Japan bank stocks

  Benefit from expectations of rising rates and can also serve as short-term hedges against volatility in AI-related equities.

  Strengths

  Fiscal expansion, yen weakness, pressure on bond prices, and expanding digital deficits may strengthen expectations for further rate hikes.

  Weaknesses

  If the government significantly suppresses Bank of Japan hikes, the rising-rate thesis may be constrained.

  Comparison

  Relative to AI and cyclical stocks, bank stocks are more macro rate- and FX-sensitive assets.

  Risks

  The impact of policy on Bank of Japan independence, yield curve changes, and reversals in market risk appetite.

• Cybersecurity, network infrastructure, and AI implementation support

  Benefit from the diffusion of AI models, increased enterprise AI adoption, and rising demand for security, data protection, and model governance.

  Strengths

  After open-source models such as Kimi K3 lower the threshold for AI use, the investment focus may shift from building AI to using AI safely.

  Weaknesses

  The theme still requires validation from earnings and orders, and previously lagged due to AI-related disruption.

  Comparison

  Relative to foundational AI model companies, this direction is closer to the application layer and security/governance layer.

  Risks

  Slow thematic monetization, unsustained valuation recovery, and slowing enterprise IT spending.

• US hyperscaler credit and capital expenditure chain

  This is the key external driver of direction for Japan's AI-related stocks.

  Strengths

  Morgan Stanley's capital expenditure forecast is significantly above market consensus and, if validated by earnings, would support the AI investment cycle.

  Weaknesses

  Large bond issuance, ORCL's downgrade by S&P, and widening spreads on newly issued bonds indicate that sentiment among credit investors is beginning to weaken.

  Comparison

  Relative to Japan domestic earnings, hyperscaler capital expenditure is more like an upstream variable determining AI risk appetite.

  Risks

  Credit risk, margin pressure, rising costs of data center construction and energy infrastructure.

Key data

• 2027 forecast for US hyperscaler plus ORCL capital expenditureUS$1.2 trillionMorgan Stanley forecasts this at about 30% above the market consensus of US$0.9 trillion.
• 2028 forecast for US hyperscaler plus ORCL capital expenditureUS$1.4 trillionMorgan Stanley forecasts this at about 40% above the market consensus of US$1.0 trillion.
• Key focus of future data center cost structureabout 60%The report says future financing attention will shift from data center shell construction to servers, chips, computing power, and energy infrastructure, including gas turbines, which together account for about 60% of total data center costs.
• Recent correction magnitude in the momentum factorabout 20%The TOPIX 500 momentum factor has corrected about 20% from its recent late-June high, against a backdrop including Middle East uncertainty, concerns over competition from emerging Chinese AI companies, and holidays in the Korean and Japanese markets.
• Yaskawa Electric share price reaction after earnings-14.3% on the next trading day, -26.6% after five trading daysThe stock's adjusted beta is about 1.59; it underperformed TOPIX by 13.7% on the next trading day and by 24.4% after five trading days, showing the downside earnings risk of high-expectation, high-beta stocks.
• Japan public and private investment roadmapcumulative more than ¥370 trillion by FY2040The 2026 basic policy proposes that if the growth strategy is effective, GDP could approach ¥1,100 trillion, and even with additional fiscal spending, the debt-to-GDP ratio could decline.
• Implied additional annual spendingabout ¥10 trillion per yearThe report points out that the policy estimates effectively assume about ¥10 trillion in additional spending per year, and that some economically security-related spending is placed outside the main fiscal framework.

Impact & implications

The investment implication is that Japan's market should not simply bet on a rebound in a single AI growth stock. If US hyperscaler earnings confirm strong capital expenditure, momentum in AI-related stocks may recover, but high expectations mean greater stock-level divergence. More attractive opportunities may come from cyclical value stocks that benefit from the diffusion of AI investment but still have low valuations and expectations. Bank stocks have defensive and hedging characteristics under fiscal expansion, yen weakness, and rising-rate risk. Cybersecurity, network infrastructure, and AI implementation support may gain new valuation support as AI usage broadens and governance demands increase.

Risks

• AI-related companies fail to clear the higher earnings hurdle, causing accelerated de-positioning in high-beta stocks.
• Middle East developments, geopolitical tensions, and supply-chain risks intensify, triggering risk aversion and momentum reversal.
• Although US hyperscaler capital expenditure remains high, widening credit spreads and financing pressure weaken market confidence.
• The catch-up thesis for cyclical value stocks may be delayed if capital is re-attracted to concentrated AI growth names.
• If Japan's fiscal expansion and monetary policy coordination trigger volatility in the yen, bonds, and rates, sector rotation paths may change.
• If application-layer AI themes such as cybersecurity lack earnings validation, valuation recovery may be unstable.

What to watch

• Whether capital expenditure guidance from US hyperscalers and ORCL continues to validate Morgan Stanley's above-consensus forecast.
• The next two weeks of earnings results and market reactions for major AI-related companies in Japan, the US, Korea, and Taiwan.
• Whether AI-related cyclical growth stocks can significantly exceed already elevated market earnings expectations.
• 1Q earnings and guidance revisions for high-EPS-growth, high-beta, low-P/B cyclical value stocks in TOPIX.
• Credit spreads, post-issuance trading performance, and financing cost changes for major US tech companies and hyperscalers.
• Fiscal spending, the yen, bond prices, and Bank of Japan policy direction after implementation of Japan's 2026 basic policy.
• Changes in orders and valuations for cybersecurity, network infrastructure, and AI implementation support companies after the diffusion of open-source AI models such as Kimi K3.