CITIC Construction Investment: How and when will this round of the US stock technology bull market end?

The Zhitong Finance App learned that CITIC Construction Investment released a research report saying that US stocks have fluctuated at a high level since June. Currently, the market is worried about related risks, which is reflected in the weak short-term expansion of US stock technology valuations. In the short term, there is a risk of a retracement. The NASDAQ does not rule out a fall of more than 10% from its previous high. In the medium term, 2028 is not ruled out to be a “limit” for US stock technology. Either before that, the bubble was resolved through endogenous (application-side outbreak, new cycle relay, or industrial counterfeiting) or external methods (China's AI catch-up, Federal Reserve interest rate hike), or K-type differentiation intensified, triggered anti-AI social ideology, and was used by both parties to reverse policies in the 2028 general election, etc., such as strengthening supervision, anti-monopoly, and tax increases against AI giants.

CITIC Construction Investment's main views are as follows:

How and when will this round of the US tech bull market end?

I. Risks of leveraged trading

In addition to amplifying short-term fluctuations, it is also indicative of long-term peaks.

The excess increase in US stock financing balances compared to stock indices has broken through 20% in the last 3 times, just in line with the three US stock crashes in 2000/2008/2022.

Currently, it has risen to 25%, which is linearly extrapolated, and there is a probability of a big peak within 1 year.

II. Catching up with Chinese AI

China's big AI model is close to the US in terms of performance, but it consumes fewer resources, raising questions about cost performance and high capital expenditure. This is a possibility of ending the US stock CAPEX.

Watch the Kimi K3 release for a short time to see if it replicates the DeepSeek shock in early 2025.

III. Counterattacks by traditional sectors

K-type differentiation and convergence, traditional economic recovery, decline in technological advantages, and market style changes are also one of the scenarios for ending the technology market.

Case: In 2000, the NASDAQ collapsed and the Dow was strong, and the economy was overheated; in early 2026, the NASDAQ adjusted, and the Dow rose to a new high due to trading recovery.

Currently, under the suppression of high interest rates, endogenous improvements in real estate, consumption, etc. are difficult. We can only expect the AI application side to explode, bringing growth to various industries and entering a “new cycle.”

IV. The backlash of general election politics

In April 2000, Microsoft was ruled to have a monopoly and carried out a spin-off, speeding up the bursting of the TechNet bubble.

Currently, if K-type polarization intensifies, it is not ruled out that anti-AI social ideology will emerge. The 2028 general election may be used by both parties. Once slogans such as strengthened supervision, antitrust, and tax increases against AI giants are introduced into the campaign platform, the US stock technology bull market may be interrupted.

5. Other common logics: industrial falsification, Federal Reserve austerity

There are already related concerns in the market, as reflected in the weak short-term expansion of US stock technology valuations, and the rise in stock prices will mainly be driven by EPS.

Summary:

① In the short term, there is still a risk of retracement. The NASDAQ does not rule out a fall of more than 10% from its previous high. Currently, the momentum for valuation expansion is no longer there, and the logic of a few of the big peaks mentioned above is still limited to the scope of narratives and is dominated by shocks. The increase during the year basically fulfilled the 2026 profit forecast. The next phase of the increase may have to wait for the second half of the year to switch to the 2027 forecast.

② In the medium term, 2028 is not ruled out to be a “deadline (deadline)” for US stock technology. Either before that, the bubble will be resolved through endogenous (application-side outbreak, new cycle relay or industry counterfeiting) or external methods (China catches up, the Federal Reserve raises interest rates), or K-type differentiation intensifies, and the 2028 general election will reverse policy.