Tianfeng: Three possible main themes for the year 2025

Tianfeng released a research report stating that the market may have "pre-voted" the 25-year main theme "dimensionality reduction" into three scenarios: 1) the continued rise of undervalued dividends; 2) technology catalyzed by byte AI; 3) valuation repair of consumer stocks and gradual recovery of consumer stratification. The withdrawal of dividends often occurs when strong industrial trends emerge, so the height of undervalued dividends depends on the progress of the AI industrial trend. The progress of the AI industrial trend depends on breakthroughs in AI application and consumption, relying solely on the AI hardware side, industry trend long-term valuations may be limited. The core contradiction of the manufacturing sector's long-term high valuation is that its growth cannot be linearly extrapolated, leading to industry deterioration + technological iteration. Two years later, the industry trend may disappear; while the growth of consumer leaders can often be "linearly extrapolated," which is a key logic for the valuation repair of the consumer sector over 25 years. After the Central Economic Conference at the end of 24 years, the overall A-share market was weak, while the top 5 in terms of excess performance relative to the overall A-share market were banking (undervalued dividends), communication (AI technology), home appliances, retail and social services (part of the consumer sector), corresponding to the three interpreted logics: 1) the continued rise of undervalued dividends; 2) technology catalyzed by byte AI; 3) valuation repair of consumer stocks and gradual recovery of consumer stratification. Tianfeng's main points are as follows: The correlation between banks in 2024 and national bond yields has changed, reflecting the logical "upgrade" of low-valuation dividend sectors such as banks: 1) From 2020 to 2023, the trading logic of the banking sector is more of "large-cap value cyclical stocks," reflecting expectations for economic fundamentals. The trend of banking stocks is roughly synchronized with the direction of the 10-year national bond yield. 2) The difference in 2024 is that the bank sector and the 10-year national bond yield suddenly become negatively correlated, with an R-square of 61%, which can be roughly understood as the national bond yield explaining 61% of the rise and fall of bank stocks; this phenomenon reflects the "upgrade" of the logical logic of the bank sector: from "large-cap value cyclical stocks" to "an income-producing asset that must be overweight when national bond yields fall." This logic will be particularly significant when absolute return funds, such as insurance funds, are the marginal players. Most of the time, low-valuation dividend styles can provide absolute returns, but there are two scenarios where there may be absolute withdrawal: First, strong growth trends in emerging industries may compete for funds, leading to withdrawals; for example, from June 2012 to December 2012, March 2013 to December 2013, and March 2019 to March 2020. The second is when the previous gains of the CSI dividend index are high + the overall stock market falls, although dividends are theoretically a defensive sector, they are still essentially stocks, influenced by stock market betas; therefore, this scenario may cause significant withdrawals; for example, from June 2015 to January 2016, January 2018 to October 2018, and June 2024 to September 2024. The height of the 25-year undervalued dividend needs to focus on the AI industrial trend, and this has been evident since the second half of 2024: the core dividend varieties (Industrial and Commercial Bank of China and China Yangtze Power) are gradually "symbolized," and their fluctuations are relatively decoupled from their own dividend logic. The characteristics of rising and falling have become "passive rises" during thematic market rallies and weak industrial trends, and "passive sideways" during strong industrial trend market. The performance of the AI industrial trend over 25 years depends on the progress of the AI application end. By screening out the manufacturing and consumer leaders in the CSI 800 separately, and calculating the non-recurring net profit growth rates of all companies in 2021/2023, it was found that the high growth of manufacturing companies is not sustainable statistically. Manufacturing companies with high growth rates in 2021 do not have higher growth rates in 2023. However, if compared with consumer leaders, it can be seen that the growth of consumer leaders can be "linearly extrapolated". Therefore, during the AI sector boom (such as in March-May 2023 and September-October 2024), the soft logic (represented by applications and consumption scenes) leads the way; and the height of the AI industrial trend in 25 years also depends on whether leading companies in AI applications and consumption can emerge, holding more AI applications and consumption scenes, similar to the moats of Tencent, Maotai, Yili, and other leading companies, making it difficult for competitors to quickly replace them through capital expenditures. Industry allocation recommendations: Scale Point 2.0 is still in the second stage of tug-of-war, returning after the Spring Festival to brace for impact, and waiting for the consumption "call options". With wide market fluctuations and violent theme market swings, institutions find it difficult to participate. Before the fundamental improvement is verified after the return from the Spring Festival, avoid chasing highs and selling lows. In the second stage of "strong expectations, weak reality," policy-driven theme rotation, the barbell market interpretation of the technology theme + low volatility dividends. Being overly pessimistic about consumption against a macro narrative is a risk. The core factor in consumer sector investment is valuation. In the current background where the consumer sector has low valuations, declining interest rates, and policy catalyzing a recovery cycle (even if it is a weak slope), with all three core conditions suitable, one should not be pessimistic about consumption becoming a major investment theme in 2025 and pay attention to the Hang Seng Internet Index. Risk warning: 1) Historical experience is for reference only; 2) Style classification is for reference only; 3) Policy implementation and landing are uncertain.