Huatai Securities A-share Strategy: Continue Balanced Allocation in the Short Term 华泰证券A股策略：短期继续保持均衡配置

Let me put it bluntly: A-shares' enthusiasm for AI has been marked from the start by strong "theme speculation," rather than a deep belief in technological revolution. Look at how, when overseas industry leaders' first-quarter guidance slightly missed expectations, the market's first reaction wasn't to dig deeper into technological progress or business models, but to sell off like frightened birds, fearing being the last one holding the bag. Is this investing in the future? It's more like joining a musical chairs game—once the music stops, everyone frantically tries to pass the chair to someone else. The so-called "overcrowded positions" are merely crowded into a fragile consensus—once that consensus wavers, everyone runs faster than the next.

The report suggests "moderate profit-taking in the AI chain, rebalancing towards banks, minor metals, etc."—a tactful statement that, in plain terms, is a typical "wind-vane market" instruction: the winds are shifting, so quickly move elsewhere. Banks rely on interest rate spreads, minor metals on cycles, and packaging/printing on what? On convoluted model calculations like "post-first-quarter earnings, relative excess returns have cost-effectiveness versus EPS changes." Listen, this is quantitative thinking taking over. The market is being driven more and more by momentum factors and style factors, turning it into a precise speculation machine rather than a place for price discovery. When we analyze a company, we no longer care whether it's creating real products or services that improve society; instead, we care about whether it's "on the main style line" and whether its "chip structure is clean." This is the alienation of investment.

The most ironic part is "reviewing instances where the Nasdaq 100's strong momentum led to a 2%+ decline coupled with rising U.S. Treasury rates; A-shares face a high probability of monthly pressure." Our market logic has become so fragile that it now relies on index trends and bond yields from across the ocean to navigate decisions? Where is the pricing power in A-shares? Are we just taking medicine the moment overseas markets catch a cold, and a simplified version of their prescription at that? This deep "dependency-imitation" mindset exposes the enormous gap in domestic capital's ability to independently price core assets.

AI is by no means a false flame. The capability leap of large models is real, reshaping the underlying paradigms of search, programming, design, and even scientific discovery. But what is the "AI chain" that A-shares speculate on? It's hardware stocks like optical modules, servers, and computing power leasing—the "shovel sellers"—and software companies scrambling to hitch onto the concept. How much have we truly participated in the soul of AI—algorithms, data, talent, and disruptive application ecosystems? How many are irreplaceable core links? When the fervor fades, many companies will likely be left with nothing but a PowerPoint presentation and a shovel.

Therefore, this report is less a strategy recommendation and more a mirror reflecting the ailments of the A-share investment ecosystem: theme-driven, style drift, externally attached pricing, and short-termism. The advice given is "balance," but I think true "balance" doesn't lie in hopping between AI and banks; it lies in resisting the fear of being swept along by market sentiment. If you truly believe in the long-term power of AI, then during the so-called "vacuum period" and "risk repricing period," it might be the golden moment to calmly assess which companies possess real technological moats rather than mere conceptual bubbles. To put it bluntly, don't always think about dancing at the crest of the wave; sometimes, you need to learn to find a truly solid rock to stand on before the wave crashes over you. Market styles may rotate, but the evolution of industries and technologies never pauses because of a few strategy reports.

华泰这份策略报告，表面上是给AI板块的短期“暂停键”提醒，但我读到的，却是一份A股市场集体焦虑症的生动病历。什么“产业催化真空”、“宏观风险再定价”，这些术语翻译成人话就是：AI这阵风暂时刮完了，但新的“故事”还没编好，聪明钱开始慌不择路地寻找下一个避风港了。

让我先把话撂这儿：A股对AI的追捧，从一开始就带着强烈的“题材炒作”胎记，而非对技术革命的深刻信仰。你看，海外龙头一季报指引稍微不及预期，市场的第一反应不是去深究技术进展或商业模式，而是如惊弓之鸟般抛售，生怕成为最后一棒。这哪是投资未来？分明是参加一场击鼓传花的赌博游戏，鼓点稍微一停，所有人都恨不得把花扔到别人手里。所谓的“筹码拥挤”，不过是拥挤在一种脆弱的共识里——共识一旦松动，跑得比谁都快。

报告里建议“适度止盈AI链，向银行、小金属等再平衡”，这话说得含蓄，但翻译过来就是典型的“电风扇行情”指令：风要变了，赶紧换个地方吹。银行靠息差，小金属靠周期，包装印刷靠啥？靠“一季报后超额收益相对EPS变化有性价比”这种拗口的模型计算。听听，这又是量化思维在主导一切。市场被这些动量因子、风格因子搞得越来越像一台精密的投机机器，而不是价值发现的场所。我们分析一家公司，不再关心它是否在创造真实的产品或服务改善社会，而是关心它“在不在风格主线上”、“筹码结构干不干净”。这是对投资的异化。

最讽刺的莫过于“复盘纳指100强动量行情后跌超2%+美债上行的情形，A股月度维度承压概率较高”。我们的市场逻辑，已经脆弱到需要靠大洋彼岸的指数走势和债券收益率来做决策导航了？A股的定价权到底在哪里？难道就是海外一感冒，我们立刻跟着吃药，而且吃的是人家药方的简化版？这种深度的“依赖-模仿”思维，暴露了本土资本在自主定价核心资产能力上的巨大空白。

AI绝不是虚火。大模型的能力跃迁是真实的，它正在重塑搜索、编程、设计乃至科学发现的底层范式。但A股炒的“AI链”是什么？是光模块、服务器、算力租赁这些“卖铲子”的硬件股，以及一拥而上蹭概念的软件公司。真正的AI灵魂——算法、数据、人才和颠覆性的应用生态——我们参与了多少？有多少是不可替代的核心环节？当热潮褪去，恐怕很多公司只会剩下一张PPT和一把铲子。

所以，这份报告与其说是策略建议，不如说是一面镜子，照出了A股投资生态的病症：主题驱动、风格漂移、定价外附、急功近利。给你的建议是“平衡”，但我觉得真正的“平衡”不在于在AI和银行之间跳来跳去，而在于对抗内心那份被市场情绪裹挟的恐惧。如果你真相信AI的长期力量，那么在所谓的“真空期”和“风险定价期”，或许才是冷静审视哪些公司拥有真正的技术护城河、而非单纯概念泡沫的黄金时刻。用一句粗话说，别总想着在浪尖上跳舞，有时需要学会在浪头打过来前，找到一块真正坚实的礁石站着。市场的风格总会轮动，但产业和技术的进化，从来不会因为几张策略报告的建议而暂停。