GF SEC: In a bull market, when will a "single peak" appear?

GF SEC released a research report stating that, returning to the present, the recent rise in AI in the markets of China, the U.S., Japan, South Korea, and Taiwan since the end of the first quarter is primarily driven by the strengthening of main industry trends, rather than one-off factors. (If there is a one-off factor, it might only be South Korea's leverage.) Therefore, aside from the risk of a peak in storage in South Korea, other sectors in different markets are unlikely to be at a peak, at least forming a repeated multi-top structure. If new vertical models lead to a demand explosion in the future, it may reach new highs. The main points from GF SEC are as follows: In July, the major global equity markets saw a synchronous pullback, with technology assets experiencing a sharp decline, leading to a clear "peak" pattern in short-term price movements. Thus, is this "single peak" or "top" event in the bull market a high-probability occurrence or an exception? This article discusses the possibility of a "single peak" during bull market top formations, looking at various angles including top formation characteristics of bull markets and fund allocation ratios, as well as the repair and trading space that currently exists in various sectors. 1. Top Formation in Bull Markets: Single Peak vs. Multiple Peaks A "single peak" is an exception in the market, only appearing in the A-share broad market in 2015, never in the Dow Jones or S&P 500, and only in the Nasdaq in 2000. Two situations may lead to a single peak: one is an irreversible shock to the capital (de-leveraging), and the other is a rapid reversal of the earnings cycle (one-off profits or strong cycles). Currently, if AI is at a peak, every rebound should be a signal to reduce positions; if AI has a repeated multi-top structure, then positions can be maintained in anticipation of a rebound in the second half of the year. If AI still has the potential to reach new highs, now is a good time to increase positions. Based on past experiences, only the Internet Plus in 2015 and the dot-com bubble in 2000 are typical peaks. Most others are repeated multi-top structures. In summary, the formation of a peak always has a one-off factor outside of the main industry trend. The one-off factor for the 2015 Internet Plus was leverage, and the bubble ended with regulatory de-leveraging. The one-off factor for the 2000 dot-com boom was the capital expenditure explosion brought by the Y2K crisis in 1999, with the bubble ending when the Y2K crisis was debunked at the start of 2000 (there was no need to replace machines). Returning to the present, the rise in AI since the end of the first quarter in the markets of China, the U.S., Japan, South Korea, and Taiwan is primarily driven by the strengthening of main industry trends, and not by one-off factors. (If there is a one-off factor, it might only be South Korea's leverage.) Therefore, apart from the storage sector in South Korea being at risk of a peak, other sectors in different markets are unlikely to be at a peak, at least forming a repeated multi-top structure. If new vertical models lead to a demand explosion in the future, it may reach new highs again. 2. Fund Allocation Ratio: Single Peak vs. Multiple Peaks 1. A few cases: Fund allocation shows a "peak," quickly retreating from high positions, often occurring when the basic expectations for a heavily allocated industry deteriorate and are quickly debunked. High peaks in the allocation for a single industry very rarely appear in the form of a "single peak." A few representative cases include fund allocations to liquor in 2012 and to non-bank sectors at the end of 2014. Why did these two allocations experience a "cliff-like" decline in the following two quarters? Essentially, it corresponds to a rapid deterioration in fundamental logic that is difficult to reverse in the short term, prompting institutions to quickly make consistent portfolio adjustment actions. 2. Most cases: In an industrial cycle peak, the institutional allocation ratio will oscillate at high levels for a period of time, presenting a complex multi-top pattern. From the perspective of industrial cycles, key references include: The large financial cycle from 2007 to 2009; The mobile internet cycle from 2013 to 2015; The supply-side reform cycle from 2016 to 2017; The core asset cycle from 2019 to 2021; The new energy industry from 2021 to 2022. During the development of the aforementioned industrial cycles, institutional holdings rise as industrial logic is constantly validated, and allocation ratios increase. In these cases, the top of the fund allocation ratio often does not appear as a single peak. Institutional investors move from forming industrial consensus, reallocating to industries with the most certain returns, and finally confirming a slowdown in the industrial cycle, with allocation ratios oscillating at high levels and showing multi-complex top characteristics. (1) The development of a large-scale industrial cycle is not achieved overnight. Whether it's technological breakthroughs, policy benefits, or business model innovations, institutions need to gradually validate new logic. Even after reaching new highs, a short-term pullback may occur, but new industrial catalysts could lead to another peak. (2) If the scope and boundaries of the industrial chain are broad enough, the benefits at different stages are not the same; during this process, institutions will preferentially select structures, and the high peaks of the allocation ratio will correspond to different heavily invested companies. (3) Finally, when high-growth industries experience a slowdown, markets often exhibit divergence, with two key empirical laws regarding performance being 30% & 50%, and significant research and validation are needed as one approaches the turning point; as fundamentals confirm a slowdown, the allocation ratio gradually digests from high levels. Returning to the present, in conjunction with the recent earnings reports from core CSP companies in North America, AI is difficult to compare with the liquor industry in 2012 or the brokerage sector at the beginning of 2015, as the fundamental logic has experienced a substantial deterioration. Therefore, referring to more industrial cycle cases (mobile internet, core assets, new energy) for new technological revolution layouts often moves upward amidst setbacks and advances amidst divergence. Even if a certain quarterly pulse value requires short-term correction, continuously tracking subsequent industrial catalysts and commercial developments is more important. Risk Warning: Unexpected geopolitical conflicts may lead to greater than expected upward pressure on global inflation; overseas inflation and the resilience of the U.S. economy may cause global liquidity to enter a tightening cycle more quickly; domestic growth stabilization efforts may fall short of expectations, leading to weak economic recovery and a decline in market risk appetite, among other concerns.