Oriental Harbor: Do not mistake short-term market sentiment for long-term corporate value

Jinse Finance reported that on June 5, Dongfang Gangwan released the "Gangwan Viewpoints | May 2026: Only by Staying Rational in the Carnival Can We Go Further," pointing out that the Nasdaq has seen the largest and fastest increase in the past 20 years over the last two months—23.7%, mainly driven by the outstanding performance of the semiconductor sector: the Philadelphia Semiconductor Index (SOXX) has risen more than 73% in the past two months, an unprecedented surge. We can take advantage of the market mania, but we must not become the mania:1. In certain segments of the semiconductor industry, extreme price hikes have intensified rather than alleviated cyclicality. Products with continuous price increases will directly impact downstream profit margins until downstream returns on investment become unsustainable, resulting in a slowdown in capital expenditure growth.2. Hyperscalers’ cash flow will be depleted by 2026; at the current pace of capital expenditure, the $730 billion capital spending in 2026 will basically exhaust the cash flow of these four companies. The super-high growth in CAPEX will also become unsustainable; a reversal in CAPEX growth in 2027 is almost certain, which could significantly affect many companies' expectations of a linear trend to 2030.3. The market may be overestimating the pace at which the AI dividend will be transmitted and realized within company organizations. Even if companies begin large-scale adoption of AI, due to limitations in the physical world and human organizations, it usually takes longer for this to translate into profits. The AI dividend could be a decade-long productivity story, but the market is now using two to three years of explosive stock price gains and CAPEX growth to discount this decade-long story all at once.4. The current high market sentiment is indeed alarming: a single compliment from Jensen Huang of "trillion-dollar market cap" made a hundred-billion-dollar company jump 32% in one night. Over the past two months, it seems that no one cares about macro risks, or discusses the ceiling or homogeneity of models. Revenues of model companies are expected to increase tenfold annually, valuations are being pushed up to tens of trillions, CSPs have begun to believe in Huang’s mantra of "computing power investment equals revenue," and application scenarios seem to be emerging one after another.This is not to say that these observations and viewpoints are wrong, but when the market seems to have almost no points of controversy or concern, sentiment often becomes extreme. Increases fuel more increases, while pullbacks can worsen sharply as well. As Mr. Munger once said, "Only by being able to refute our own viewpoints better than others are we truly worthy of our own views and achieve objectivity and rationality." Do not mistake short-term market sentiment for long-term corporate value. You can take advantage of the mania, but never become the mania!