Market adjustments have already been brought forward into this year, and structural turning points are set to multiply in 2027, according to a research note from CITIC SEC (HKEX: 06030, SSE: 600030).

The old framework of "industry trend rolls over equals bull turns to bear" no longer applies to this AI-driven market cycle. AI remains in its early development stage, and its near-term negative externalities on other economic sectors are substantial: prices of production materials for various consumer goods have surged, capital has been siphoned out of bond markets, pushing long-end rates sharply higher and squeezing non-AI demand, while Agent technology has to some extent displaced employment. The intersection of these problems has produced severe economic and market divergence. Even during the strongest phase of the AI bull market, non-AI sectors had already broadly entered downtrends and undergone deep valuation adjustments.

Precisely because of this, if AI investment growth slows in the future, there is no reason to remain pessimistic on non-AI sectors. If AI enters a consolidation phase, non-AI sectors could see a turning point, as falling interest rates, declining materials prices, and policy rebalancing would all favor a more balanced market. "K-shaped convergence and broader market breadth" may be the most important market characteristics of 2027.

The brokerage expects internal divergence within the technology sector, with broad structural repair opportunities in non-AI sectors, and the main process of valuation contraction having already been completed in 2026. Clearing out on the funding side will take time, and investors will gradually return to reasonable return expectations. Stable-return strategies represented by PB-ROE are expected to outperform momentum strategies, and a major style shift may occur.

On allocation, investors should diversify appropriately and respond along four threads: stable returns, outstanding overseas-expansion manufacturers, domestic demand repair, and new AI technology themes.

The main views of CITIC SEC are as follows:

Market characteristics: K-shaped convergence, broader market breadth

1) This year, major global stock markets have experienced historically rare structural divergence. Since 2026, market breadth across global equity markets has narrowed broadly, with AI and non-AI sectors moving sharply apart. In the four markets of China, the US, Japan, and South Korea, the excess return of core AI baskets over core non-AI baskets reached about 30% year-to-date (as of October 22), peaking at one point at 150%.

Two aspects of the market's direction clearly deviated from expectations at the start of the year. First, it was hard to foresee that as AI investment expanded dramatically, AI hardware would generate substantial near-term negative externalities for other economic sectors, driving up prices of production materials for consumer goods, siphoning capital from bond markets and sending overseas government bond yields sharply higher, ultimately pushing market divergence to extremes far beyond expectations. Second, it was hard to foresee that sectors underpinning market breadth, such as overseas expansion and resources, would encounter various valuation-suppressing factors this year. Market expectations shifted from Fed rate cuts at the start of the year to rate hikes, China-Europe trade frictions escalated rapidly, and RMB appreciation outpaced expectations significantly, causing large foreign exchange losses. Amid these intertwining factors, market sentiment toward high-prosperity sectors like overseas expansion and resources swung from optimism at the start of the year to pessimism now, with dynamic valuations compressing sharply.

Given the huge expectation gap, the market unusually saw broad declines in a year of rapidly rising profit growth. Investors are now broadly cautious on both AI and non-AI, a stark contrast to the start of the year.

2) Looking ahead to 2027, the AI industry trend is still upward, but stock momentum may fade. The probability of AI hardware becoming the market's core thread for a fifth consecutive year is declining.

Based on current neutral market expectations for the AI capex path, the brokerage estimates that by 2030, terminal AI application external direct fee revenue would need to reach $2.75 trillion to achieve reasonable and sustainable capital returns across the entire industry chain, and relying on frontier model companies alone would be far from enough. Without new scenarios capable of generating enormous commercialization space (such as physical AI), existing model applications extended from Agents would either gradually become embedded in corporate IT budgets and integrated into business processes, or become a cost for consumer platforms to boost user retention time and capture traffic entry points (for example, AI short dramas consuming computing power). These are more like cost items than revenue items.

This means AI investment may continue, but single-year investment intensity could decline, gradually evolving into a purer infrastructure construction process. From the perspective of stock elasticity, if global AI capex reaches $1.6 trillion by 2030, the brokerage estimates that A-share plus Hong Kong AI hardware industry chain companies, valued at 15x 2030 PE, would have market caps about 51% above current levels, and overall compound returns would no longer be attractive, requiring careful Alpha selection. In addition, institutional holdings in AI remain elevated, which will constrain the sector's ability to continue attracting incremental capital.

3) Non-AI expectations are extremely low, but potential marginal improvement factors are increasing. The main factors suppressing overseas non-AI demand may not get worse in 2027.

First, the high overseas interest rate environment may ease. As North American AI investment slows over the next two years, private-sector debt financing is expected to peak and decline, weakening the crowding-out effect on government public-sector financing, improving supply-demand in the government bond market, and pulling long-end rates lower.

Second, the pace of RMB appreciation may slow, reducing foreign exchange losses year over year. The brokerage estimates that when the USD/CNY exchange rate is in the 6.35-6.65 range, the impact of FX losses on all-A non-financial net profit growth is 0.5%-6.2%, compared with 7.3% in the first half of this year.

Third, if the Democratic Party successfully wins the US midterm elections, the Middle East situation and oil prices could also improve, easing inflation and rate-hike anxiety while alleviating currency depreciation pressure in non-US countries.

As for non-market factors, especially rising China-Europe trade frictions, we believe that in the long run it is difficult to block the overseas expansion of outstanding Chinese manufacturers. As for key technology and equipment export controls and reduced export tax rebates, although these create short-term pressure on the industry, they are crucial for clearing out low-quality enterprises, stabilizing supply-demand patterns, and niche markets, and are highly favorable in the long run for companies that operate compliantly and have made sufficient preparations for overseas expansion.

For domestic demand, policy is a difficult-to-predict economic variable. When the overall economy weakens and confidence among micro entities is sufficiently low, we can often still see policy intensity increase. This year, excluding the special cases of 2021 and 2023, is a historically rare year of year-over-year fiscal deficit contraction, with tax revenue exceeding expectations while spending progress has consistently fallen short. Perhaps next year some adjustments will be made to micro-level implementation processes. Considering that more than half of this year's export growth was contributed by AI-related commodity price increases (with volumes basically flat), next year's nominal export growth faces considerable slowdown pressure, which may also provide some justification for domestic demand policy to step up.

Of course, the most favorable aspect for non-AI sectors is that expectations are sufficiently low, especially in overseas-expansion sectors. For example, valuations for home appliances, consumer electronics, textiles and apparel, commercial vehicles, and medical devices are basically near historical extremes, and marginal better-than-expected macro improvements could bring sufficient return elasticity.

Earnings trends and valuations: growth may slow in 2027, but the main process of valuation contraction was completed this year

1) All-A non-financial earnings growth is expected to slow somewhat in 2027. Entering 2027, the profit base for technology and cyclical sectors is notably higher, and even if prosperity remains elevated, the boost to overall earnings growth will weaken. At the same time, easing FX pressure for overseas-expansion companies, improved overseas demand, and increased central fiscal support for domestic demand along with the fading impact of consumption subsidy rollbacks are expected to make growth sources more diversified.

The brokerage expects all-A attributable net profit to grow 15.5% and 9.5% year over year in 2026 and 2027, respectively, with all-A non-financial attributable net profit growing 20.1% and 14.7%. Full-year earnings will still grow, but the incremental structure will gradually shift from being concentrated in a few high-prosperity industries to being supported jointly by continued technology growth, overseas-expansion repair, and reduced domestic demand losses.

Specifically: 1) Cyclical sectors: nonferrous metals digesting a high base, with energy and chemicals providing earnings support; attributable net profit growth is expected to reach 45.5% and 9.8% in 2026 and 2027, respectively. 2) Technology sector: the industry chain continues high growth, but growth slows under a high base; earnings growth is expected at 101.9% and 43.0% this year and next. 3) Overseas-expansion sector: easing FX pressure and improved overseas demand drive full-year earnings repair; earnings growth is expected at -0.5% and 8.8% this year and next. 4) Domestic demand sector: stronger central fiscal support is the core variable for halting the earnings decline and turning around; earnings growth is expected at -14.4% and 8.8% this year and next. 5) Financial sector: non-bank profit growth returns to normal, banks remain stable; earnings growth is expected at 11.5% and 4.6% this year and next.

2) But the valuation contraction process for most sectors has basically been completed this year. As of October 23, 2026, the technology sector has completed the first wave of valuation de-rating after the "second derivative" downgrade. The 15 most core AI stocks have adjusted for 77 trading days since July, falling as much as 44.5%, while the 200 most institutionally held stocks fell as much as 35.8%. This already exceeds the decline of core stocks during the new energy crowding-out collapse from December 2021 to the end of April 2022, when the core 15 new energy stocks fell 40.9% and the 200 institutionally held stocks fell 33.7% (December 17, 2021 to April 26, 2022).

From a global comparison perspective, US and Japanese non-AI leaders are at historically high valuations, while A-share non-AI is below the median. The brokerage used AI and non-AI leader baskets from the Chinese, US, Japanese, and South Korean markets, calculating PE (TTM) on a consolidated basis and making cross-market comparisons. As of October 22, 2026, A-shares are the only market where the AI relative to non-AI valuation premium remains above the historical center: the AI/non-AI PE ratio is 4.4x, at the 82nd percentile since 2016; the same ratio for the US, Japan, and South Korea has fallen to the 1st, 17th, and 24th percentiles, respectively. The PE of US and Japanese non-AI non-financial leaders is 33.5x and 23.0x, at the 98th and 95th percentiles; A-share non-AI non-financial leaders have a PE of only 15.6x, at the 21st historical percentile, with consumer, manufacturing, and pharmaceutical leaders at only the 12th, 1st, and 3rd historical percentiles.

3) From the perspective of shareholder returns, A-shares have made substantial progress compared with the past. Although equity financing in the technology sector reached RMB 246.5 billion this year (already 2.2x that of 2025), cash returns and buybacks for the market as a whole, especially non-technology sectors, continued to hit record highs. This year, as of October 23, against the backdrop of concentrated technology company listings, A-share dividends and buybacks combined were 3.4x equity financing (5.1x in 2025 after excluding special treasury bond injections into state-owned major banks, and 10.6x in 2024 due to the temporary tightening of IPOs and refinancing).

By contrast, the US stock market is rapidly shifting from a net shareholder return market to a financing market. Federal Reserve Z.1 data show that net equity issuance by the US non-financial corporate sector (seasonally adjusted annual rate) was negative in all four quarters of 2025, turned to +$181.4 billion in Q1 2026, and further rose to +$607.8 billion in Q2, the highest single quarter since 1990.

If measured by (cash dividends + buybacks − equity financing) ÷ average total market cap as the net cash return to shareholders, the S&P 500's net cash return rate has fallen from 4.6% in 2016 to 2.5% in recent years, with US AI leaders at only 0.81%. The direction for A-shares is exactly the opposite: equity financing is declining while dividends and buybacks are rising, with the all-A net cash return rate rising from -1.5% in 2016 to 2.5% in 2024; it was 2.1% in 2025 and 1.4% year-to-date this year.

Funding and style: clearing on the funding side takes time, and stable-return strategies represented by PB-ROE are more advantageous than momentum strategies, with a major style shift possible

1) The funding ecosystem needs time to recover, and investors are returning to reasonable expected returns. On the market funding side, the brokerage expects the retail incremental process to slow markedly. Active equity funds, after a new round of sharp drawdowns, are falling back into the cycle of "larger redemptions as NAV rebounds," while sector ETFs, private quantitative funds, and long-term capital will constitute the main incremental funding for the future market, with "long money" continuously increasing its influence. But at the micro level, under the multiple influences of quarterly and annual assessments, liability-side instability, solvency constraints, and drawdown stop-loss mechanisms, some nominal "long money" such as annuities has also exhibited short-term behavior.

The key to repairing the funding ecosystem is not just an increase in the absolute scale of a certain type of institution, but the realignment of liability-side stability, assessment horizons, and trading behavior. Before this process is complete, the market may still focus more on short-term funding behavior and be reluctant to price the long term easily. After this year's extremely high abnormal return elasticity and severe market adjustment, investors are expected to gradually return to reasonable return expectations, using medium position levels and stable-return holdings to get through the stage of funding-side clearing and investor confidence recovery.

2) Stable-return strategies will clearly dominate, with supply logic surpassing demand logic as the main factor. From the characteristics of A-shares, when demand signals are unclear, momentum and prosperity strategies enter a weak phase, and rotation accelerates markedly, stable-return strategies represented by PB-ROE will clearly outperform. Of course, static high-ROE or high-dividend strategies have not historically performed exceptionally well, and stable-return industries differ greatly across stages, requiring subjective judgment to find varieties with room for improvement in profitability or dividend capacity in the future. Backtest results show that a PB-ROE strategy incorporating turnaround elements clearly outperforms a simple PB-ROE strategy and is generally stable across different market environments, especially when momentum strategies fail.

The brokerage believes there are currently two types of industries that fit this framework in 2027. One is varieties with high dividend yields and steady earnings growth next year, such as white goods, textiles and apparel, food seasoning, and thermal power. The other is varieties with supply-clearing potential logic, such as oil shipping, express delivery, leading real estate developers, some lithium battery materials, and midstream petrochemical segments.

Industry allocation: moderate diversification, responding along four threads

Industry allocation in 2027 should be moderately diversified, with positioning along four threads: stable returns, outstanding overseas-expansion companies, domestic demand repair, and new AI technology.

For stable returns, the starting point is improved supply patterns and sustainable capital returns, with positioning around energy and chemicals, oil shipping, highways, logistics, white goods, pharmacies, insurance, overseas-expansion-oriented brokerages, and financially sound real estate leaders.

For outstanding overseas-expansion companies, focus on ships, diesel engines, two-wheelers, CXO, power tools, and power equipment, capturing opportunities from slower RMB appreciation, improved overseas demand, and rising global share of Chinese companies.

For domestic demand repair, assuming stronger central fiscal support, if policy changes exceed expectations, recommended offensive varieties include restaurant supply chains, airport duty-free, health products, and medical device leaders, preferably companies able to capture demand improvement and increase market share.

For new AI technology, considering that some institutionally heavy holdings still need to digest their positions, focus on structural opportunities with clear application prospects and earnings realization capability, including passive optical components, PCB, storage, semiconductor equipment and materials, robotics, and edge AI.

Combining industry team recommendations, the brokerage selected 15 companies for each of the four threads to construct corresponding portfolios.

Market adjustments have already been brought forward into this year, and structural turning points are set to multiply in 2027. After a broad decline, there is no bear market, or at least it is a structural market. As of October 23, 2026, 4,284 A-share stocks recorded negative returns, accounting for 76.8%, making it a relatively rare broad-decline year. Historically, after the split-share reform, A-shares have rarely seen such a high proportion of declines in two consecutive years (2017 was one, but traditional industry white-horse leaders delivered very considerable excess returns for institutions that year).

Overly high expectations at the start of the year and frequent surprises this year (the US-Iran war, China-Europe trade frictions, high interest rates, year-over-year fiscal deficit contraction, etc.) led to market performance that, in hindsight, was less than satisfactory. But looking ahead to 2027, at a point of generally cautious expectations, the brokerage believes market breadth will instead widen next year, the valuation contraction process for most sectors has basically been completed this year, AI investment may slow, but its crowding-out effect on other non-AI industries will diminish, with K-shaped convergence and a more balanced market. Improving the funding ecosystem will take time, and stable-return strategies represented by PB-ROE may clearly outperform momentum strategies, warranting an active response.

In terms of industries, the brokerage recommends allocation along four threads: stable returns, outstanding manufacturing overseas expansion, domestic demand repair, and new AI technology themes.

Risk factors: intensified frictions in China-US technology, trade, and financial areas; domestic policy intensity, implementation effects, or economic recovery falling short of expectations; domestic and overseas macro liquidity tightening more than expected; further escalation of conflicts in Russia-Ukraine, the Middle East, and other regions; and China's real estate inventory digestion falling short of expectations.