Citigroup: Hong Kong stocks do not account for a large share of “true technology stocks”, and the Hang Seng Index has little room for significant improvement in the second half of the year

The Zhitong Finance App learned that the trend of Hong Kong stocks was weak in the first half of the year. The biggest reason was that Hong Kong stocks were dominated by technology stocks, and the wave of memory price increases greatly increased the capital expenses of AI R&D companies. Hong Kong stocks were heavily affected by technology and plummeted, which was the main reason why the Hang Seng Index continued to decline in the first half of the year. Liu Xianda, a stock strategist at Citibank China, said in an exclusive interview that the changes in interest rate expectations this year and the intensity of mainland supervision are beyond the bank's expectations. Coupled with downward pressure on platform stocks, the Hang Seng Index will rise to 30,000 in the second half of the year, like a skyscraper. However, in addition to this, he did not disdain the AI boom. Instead, he suggested starting to reserve codes for the next stage of AI development, let alone short-term deployment within a year.

At the beginning of this year, Citi set a target of 30,000 for the Hang Seng Index by the end of the year, but the target was lowered to 29,600 before the end of the first half of the year. Liu Xianda said in this regard that since this year, neither Hong Kong stocks nor A-shares have performed very well. Among them, there has been a certain gap between when hoping to cut interest rates at the beginning of the year to prevent interest rate hikes, while mainland consumer retail sales have not improved much, causing market confidence to weaken. However, the most important negative factor for Hong Kong stocks stems from structural problems in Hong Kong stocks.

In the first half of the year, Global Capital sought AI infrastructure stocks. Concepts such as memory chips, optical communications, and PCBs all became locomotives in the market. Compared with the “true technology stocks” at the cutting edge of technology mentioned above in Hong Kong stocks, the main ones known as technology stocks were platform stocks, which were mainly influenced by retail consumption-related markets such as takeout and automobiles. Therefore, in the case of true technology stocks, Hong Kong stocks have only 4% of relevant shares, and the majority of A-shares are only 15%, which is far lower than the US, Taiwan Province of China, and South Korea, thus weakening Global Capital's risk appetite for Hong Kong stocks.

Since platform stocks currently still account for a heavy share of Hong Kong stocks, it is expected that there will be no major change in capital style in the second half of the year, so there is little room for significant improvement compared to the Hang Seng Index in the second half of the year. Asked if the Hang Seng Index target at the end of the year would need to be lowered, Liu Xianda mentioned that the index forecast calculation method will take into account profit growth and market capital flow. He will keep watching and only revise the target as needed.

He also mentioned that Hang Seng Index will probably not directly reduce the weight of platform shares, but since the Hang Seng Index's constituent stocks are still in the expansion stage, adding new stocks will help dilute the weight of platform stocks. For example, Hang Seng Index has added pharmaceutical, industrial, and resource shares in recent quarterly inspections. Assuming that the weight share of newly added stock sectors gradually increases in the future, the share of platform stocks will be slightly lowered from the current 32% to about 25% to 28%, which is enough to spread some of the volatility and regulatory risks.

The half-year end can be described as an important watershed in the AI market. Since Meta announced the sale of excess computing power, the market's concerns about excessive AI computing power deepened. The AI infrastructure stock market in the first half of the year set off a 180-degree reversal after July, software stocks rebounded, and infrastructure stocks pulled back deeply. According to Liu Xianda's analysis, the AI market is definitely not over, but there will be phased fields. He pointed out that the AI boom is currently divided into three stages. From the beginning of the boom in 2023 to about 2025, the valuation of AI-related concept stocks has risen sharply. Starting this year, they are looking for more upstream AI infrastructure suppliers. After also experiencing hot hype, he expects that starting in the second half of this year, the market focus will shift from AI infrastructure investment to some AI implementation scenarios, with more emphasis on the actual ability to monetize concepts. He believes that in the next year or two, the market will no longer simply look at the vision and pay high premiums. It will pay more attention to actual order volume and revenue growth. Compared to upstream computing power hardware, the boom will continue, but the increase will not necessarily be the same as in the past two or three years, and there is also an opportunity for prices to fall.

In response to the explosion of AI IPOs since this year, Liu Xianda emphasized that the short-term stock drawing policy is mainly based on the AI concept. IPOs with AI and semiconductor elements will look higher on the front line. As for traditional extraction standards, such as looking at cornerstone investor lineups and overpurchase multiples, they will continue to be effective. Liu Xianda pointed out that concentrating on subscribing to the above elemental shares would have a higher return of 6% to 9% per year compared to only holding traditional big blue chips with an average annual return of 6% to 9%.