As summer arrives, the technology sector has begun a pattern of sharp reversals. This July is set to be a month that tests investors' resolve.
In times of uncertainty, a perspective that cuts through the noise is most valuable. The "Four Seasons Companion" summer event from Great Wall Fund was held as scheduled.
This time, the focus was on two fund managers with distinct styles but both possessing high "research value": Liu Jiang, manager of the Great Wall Jiuxiang Fund, and Xu Liheng, manager of the Great Wall Ruiying FOF Fund. One delves deeply into technology industry trends, while the other excels in multi-asset allocation. Starting from entirely different frameworks, they converged on the same conclusion: anchoring to true value amidst the market noise.
On the debate and reality of the tech rally: Liu Jiang believes the recent volatility is precisely the necessary path for the market to shift from "speculating on expectations" to "competing on performance." Short-term turbulence can more clearly reveal the sectors with solid underlying earnings strength.
On how to build "anti-fragility" in asset allocation: Xu Liheng posits that the core of asset allocation is not pursuing an "optimal solution" but constructing an "anti-fragile" balanced structure amidst uncertainty. Letting go of the obsession with perfectly timing the market may be the key to long-term success.
The 90-minute discussion was rich with insights and thoughtful perspective. At its heart, investing is about growing alongside high-quality companies, and the essence of companionship is helping to steady the wheel during every bump in the road.
The hope is that on your investment journey ahead, you have clarity of vision, a sound strategy, and peace of mind.
Pricing Logic in a Diverging A-Share Market
Xu Liheng: Broad asset classes moving in sync, with AI as a new pricing factor
The first half of 2026 saw distinct divergence in global asset performance: major stock indices mostly rose, with South Korea's KOSPI leading the pack. Commodities showed significant internal divergence, with crude oil surging 22.16% to top the list, while base metals like zinc, copper, and aluminum generally gained, and precious metals saw notable corrections. The two guests reviewed the core characteristics and drivers of the H1 market from their respective viewpoints.
Xu Liheng pointed out that the first half presented significant challenges for asset allocation, primarily exhibiting two features: first, a pronounced effect of broad asset classes moving in sync, weakening the traditional hedging effect of the stock-bond seesaw; second, intense K-shaped divergence within the equity market, with the technology sector standing out.
In his view, three core drivers are behind this extreme divergence:
Changes in the risk-reward profile of assets themselves. After years of prior gains, prices of traditional safe-haven assets like gold and bonds are at historically high levels, leaving less buffer room and making it harder for them to continue serving as effective risk hedges.
A shift in market drivers towards inflation expectations. When inflation expectations rise, stock and bond assets tend to move in the same direction. Rising oil prices and increasing material costs in the AI supply chain in H1 jointly pushed up inflation expectations, intensifying the co-movement of broad asset classes.
The pronounced liquidity siphon effect of the AI industry. The high capital expenditure and strong momentum in the AI supply chain generate expected returns that other assets struggle to match, significantly raising the opportunity cost of holding stable assets and leading to a systematic concentration of funds into the tech sector.
Xu Liheng further emphasized that the impact of this AI wave has transcended the industry itself, becoming a core variable influencing the pricing of broad asset classes. The AI industry features high-slope characteristics—large scale, accelerating formation of monetization loops, and strong price elasticity in the supply chain—which essentially raise the discount rate for asset pricing across the entire market. The opportunity cost of holding other assets increases, putting pressure on valuations broadly. This is the underlying reason for the adjustment in high-priced assets like gold in H1; on the surface, it's a strong dollar and high rates, but beneath lies a reshaping of pricing logic driven by the high-slope growth of AI.
Position in the AI Industry Cycle and Core Debates
Liu Jiang: The AI industry is still in the early stages of its upward trajectory
As a core market theme in recent years, the AI industry involves numerous supply chain segments with rapid rotations, accompanied by many debates.
Liu Jiang outlined the shifting path of the main AI industry theme since 2023. He believes the AI industry over these three years perfectly fits the growth paradigm of an emerging industry, with a pace far exceeding most traditional sectors:
Early 2023 saw ChatGPT ignite the market, a typical "0 to 1" thematic phase where broad industry breakout expectations formed, but most companies' earnings were yet to materialize, leading to sector-wide gains followed by corrections.
Starting in 2024, global tech giants significantly ramped up AI capital expenditure, entering the "1 to 10" phase of computing infrastructure build-out. Market focus shifted entirely to the computing power supply chain, with related companies' earnings quickly materializing, becoming the core market theme.
Entering 2026, commercialization of large models and AI applications accelerated, with scenarios like programming forming paid closed loops first. The industry began extending to the application layer, still in a healthy upward phase, with potential for continued breakthroughs in future scenarios like smart devices, embodied AI, and production/office applications.
Core Debate One: Is there a computing power surplus? – Structural mismatch, overall surplus likely premature
Addressing the "computing power surplus" debate, Liu Jiang stated that discussing an overall surplus is premature, with market concerns somewhat amplified. The reality is a structural supply-demand mismatch.
He provided three core arguments: First, high-quality computing power remains in short supply, with some manufacturers even restricting internal usage, indicating a real underlying supply gap. Second, domestic and international computing power leasing prices continue to rise significantly, directly evidencing tight demand. Third, overseas industry leaders continue to ramp up data center construction and capital expenditure, indicating the AI industry's arms race is far from deciding a winner.
In his view, some less competitive manufacturers experiencing temporary computing power redundancy is a normal phenomenon in industry development. The redistribution of this idle capacity through leasing is essentially a sound commercial choice.
Core Debate Two: Has AI peaked? – Distinguishing short-term variables from long-term core drivers
Liu Jiang stated that for a theme to truly end, a "fundamental blow" typically needs to occur, such as the onset of overcapacity or a systemic change in macro liquidity. At least for now, these signs are not visible.
He suggested judging whether the AI industry is nearing a peak requires viewing from two dimensions: the visible short-term capital expenditure pace and the invisible long-term commercialization closed loop. In the long term, as long as it can continuously create incremental commercial value and achieve breakthroughs in commercialization, the industry cycle can extend much further.
The Underlying Logic of Multi-Asset Allocation
Xu Liheng: The core value of diversification is striving for improved return stability and resilience in extreme environments
The one-sided strength of the tech sector in H1 led many to question the concept of diversification and balanced allocation. Addressing the query of whether diversification still holds meaning, Xu Liheng stated that preparing for future uncertainty is precisely the core value of asset allocation. He summarized the value of diversification into two points:
Stronger return stability over longer cycles. Data shows that over the past 20 years, simultaneous declines in four asset classes—domestic stocks, overseas stocks, domestic bonds, and gold—occurred in only 5 months. A single asset class may offer stronger short-term explosive power, but a diversified portfolio may deliver more stable long-term compound returns and potentially better drawdown control.
Enhanced ability to handle extreme environments. Taking gold as an example, it is a typical positively skewed asset. Over the past 30 years, instances of extreme gains far outnumbered extreme declines, often playing a hedging role in tail-risk scenarios. Combined with the supportive effect of global central bank gold purchases, its long-term allocation value may remain significant.
Regarding the long-term low-interest-rate macro backdrop, Xu Liheng believes the traditional stock-bond balanced allocation logic needs updating, with three future trends emerging: First, "rate-ization," with medium-to-long duration rate bonds playing a greater foundational role in portfolios. Second, diversification, moving beyond the stock-bond binary framework to include assets with potentially better risk-reward profiles like commodities, REITs, and alternative strategies. Third, professionalization, as higher allocations to volatile assets significantly raise requirements for risk identification and active management capabilities.
Investment Themes and Market Style for the Future Market
Liu Jiang: Technology growth remains the main theme, focus on three directions
At the start of the second half, Liu Jiang clearly stated that the low-interest-rate environment is generally favorable for technology growth investing, with moderately loose liquidity providing a friendly macro backdrop for the tech sector. AI remains the market's highest-consensus theme for H2. The investment attractiveness of the AI sector can be judged from two dimensions:
First, the industry cycle position. The AI industry is currently still in a healthy upward phase, showing no signs of marginal slowdown or development obstacles. There is no signal of a fundamental turn in industry fundamentals, so excessive worry about short-term volatility is unnecessary.
Second, the degree of valuation froth. Comparing the dynamic valuations of leading domestic and international assets, the overall picture appears relatively restrained, without excessive froth. Some companies even still possess valuation advantages.
Liu Jiang is particularly optimistic about three directions: First, the AI computing power infrastructure supply chain. Upstream segments like optical communication, computing chips, storage, and PCBs represent the most certain direction in the current industry, with the tight supply-demand dynamic likely to persist.
Second, the domestic technology self-sufficiency and control theme. Driven by policy support, talent concentration, and technological iteration, China's tech industry is developing rapidly. Leading domestic companies in areas like large models, computing chips, and storage will see accelerated development opportunities.
Third, strategic emerging industries and future industries. Frontier directions like innovative drugs, commercial aerospace and space computing, embodied AI, brain-computer interfaces, the low-altitude economy, and the deep-sea economy hold the potential to foster new industry-level investment opportunities and are worth continuous tracking.
Xu Liheng: Style convergence is possible; base metals, cloud providers, and biopharma warrant attention
Xu Liheng stated that in the long term, AI remains the market's core theme. The industry is currently still in its first phase—infrastructure construction driving hardware demand—with vast room for subsequent application-layer explosion. He offered a baseline scenario judgment for the future market: capital expenditure at the AI industry level will likely maintain a high slope, but the slope of market expectations may gradually moderate. Correspondingly, valuation pressure on other broad asset classes may ease, and the extreme K-shaped divergence in the equity market is expected to converge.
Structurally, further differentiation within AI is likely, making allocation value in tight supply chain segments more prominent. Regarding specific directions, he is particularly optimistic about investment opportunities in the base metals sector: On one hand, copper's cost curve continues to shift upward, with high-cost copper mines providing price support. On the other hand, AI data center construction and the electrification trend continue to drive copper demand, with a long-term supply-demand gap likely to persist. Copper prices are expected to remain elevated, benefiting low-cost related enterprises significantly.
Additionally, Xu Liheng mentioned two types of hedging assets within the growth sector suitable for close attention: First, cloud provider assets, whose valuations are currently suppressed by high hardware costs, but as AI enters the application phase, exploding inference demand will drive a revaluation of cloud business value. Second, biopharmaceutical assets, as long-duration tech assets currently facing valuation pressure but possessing significant long-term commercialization potential, offering recovery elasticity in a declining discount rate environment.
Amid the flow of summer, markets have their fluctuations, and industries present new opportunities. Great Wall Fund will focus on building an integrated paradigm of "technology + industry," while relying on its mature asset allocation system to continue providing professional investment companionship for investors.
Comments