Written by: Trend Research

Recently, the strategy team at Guotou Securities released a lengthy report titled "Investment Methodology for the Technology Industry," attempting to answer a question that all technology investors inevitably ask: How exactly to buy and sell technology stocks.

In companies like Yangtze Power, analysts can calculate how much profit will be made each year over the next decade, and discount future cash flows back to determine the stock price.

However, technology stocks can't account for this. The evolution of technology is leapfrogged, where a catalytic event can completely redefine the logic of an entire industry. In 2019, the market believed general artificial intelligence would take another 80 years; by 2022, this was reduced to 8 years; and in 2023, with the emergence of ChatGPT, the timeline was rewritten again.

The data from the A-share market is even more direct: technology stocks that doubled in growth last year tend to average a 40% drop the following year. Only 5% of technology stocks can maintain a growth rate of over 30% for five consecutive years.

The conclusion is that, technology stocks make money from waves; just lying flat won't get you anything.

N-shape: Two Waves, Four Points

This core framework of the investment methodology for technology stocks can be summarized in a single diagram: the N-shape. This framework represents a technology market cycle depicted as the letter N, with four key positions: A, B, C, and D.

A→B is the first wave, from 0 to 1.

This wave makes money from narratives. The company has no performance and hasn't even released a product, but the story is compelling enough. The valuation method is quite crude: estimate the total industrial output potential, distribute the market cap ceiling by segments, with a market value/output ratio generally capped around 3 to 3.5 times.

Currently in the A→B stage are: embodied intelligence, low-altitude economy, commercial space, AI applications.

B→C is the adjustment period.

After the first wave, the story has been told and stock prices start to fall. Most technology stocks die here and never see a second wave.

C→D is the second wave, from 1 to 100.

This wave makes money from profits. Companies start to deliver performance; penetration rates rapidly increase, and stock prices rise again, but P/E ratios actually decrease because profit growth outpaces stock price increases.

Examples that have moved from C→D include: optical modules, PCBs, AI computing chips, data centers.

The Importance of Point C

Point C is the real tipping point for institutional investors.

Characteristics of Point C: stock prices have declined from Point B for a while, market sentiment is poor, and most people hold light positions, unsure of the ceiling. But it is precisely at this point that performance begins to emerge, orders start coming in, and the industrial fundamentals truly explode.

How to determine if Point C has arrived? Three elements are necessary, and missing one is not acceptable:

Major capital expenditure from giants. Are large companies investing money in this direction? The significance of capital expenditure to an industry is akin to the importance of credit to the economy; without funding sources, the industry cannot take off. The rhythm of the AI industry follows this line: from 2023 to 2024, overseas cloud providers ramp up capital spending, purchasing overseas supply chains (Zhongji Xuchuang initiates); in the second half of 2024, ByteDance starts capital spending, purchasing domestic computing power (Hanwha Microelectronics initiates).

Hot-selling products. Is there a product that creates a breakthrough in penetration rates? iPhone 4, AirPods, Model 3, ChatGPT, DeepSeek; each hit product indicates the starting point of a C→D transition.

Implementation of the industrial chain. Have companies received orders? Once giant capital expenditure and hot-selling products form a closed loop, companies along the industrial chain start generating revenue, leading to a positive cycle.

The report summarizes this in one sentence: as long as the three elements are in place, one should quickly engage; this is the most critical action for generating a super cycle.

How to Determine Point D (When to Sell)?

This section of the report introduces the "M-top" framework, where M represents two peaks: the first peak is the trading peak (emotional peak), and the second peak is the fundamental peak.

The three observation signals for the fundamental peak are: Is there a macroeconomic downturn, is there a price war emerging on the supply side, and has capital expenditure started to be cut on the demand side? If two or more of these signals occur, the current industrial cycle is basically over.

If leading companies decline due to macro or external factors, a downward trend in the industry direction may present the best buying opportunity.

For instance, when the Nasdaq fell in 2010, Apple was a buying point; in the downturn during the pandemic in 2020, Tesla was a buying point; the trade war also provided buying opportunities for AI technology. The sell signal for Nvidia mainly hinges on two conditions: first, whether the U.S. economy is facing a hard landing; if so, sell, as the cash flow of the five major cloud providers is highly linked to consumer behavior, and if cash flow is jeopardized, the entire logic fails; secondly, whether the competitive landscape has worsened; if it has, consider selling.

In terms of US stock reflections, historically there are significant investment methodologies. From the 1990s to 2000, the Japanese stock market fell by 67%, but companies like Tokyo Electron, Advantest, and Toshiba performed well. The quantitative logic of US stock reflections is not closely related to performance but is significantly correlated with the price increases of corresponding US stocks. The most effective reflections are within the industrial chain, such as Zhongji Xuchuang and Luxshare Precision, which can generate major bull stocks as the industrial trend rises and profits are continuously realized.