CITIC Construction Investment: A-shares will fluctuate weakly in March, predicting that gold priced in US dollars will continue to strengthen

According to the Zhitong Finance APP, CITIC Construction Investment has released a research report stating that in March, A-shares showed weak fluctuations, Hong Kong stocks were differentiated, US stocks declined, gold strengthened, and the bond market adjusted. The yield on China's ten-year government bonds has deviated from historical cyclical patterns. It is predicted that the peak year-on-year GDP for the US will be in Q1 2025, for Japan in Q2 2025, and for the Eurozone in Q2 2025. It is also predicted that the weakness of the Japanese yen against the US dollar will improve in stages, and the euro may strengthen against the US dollar in the future; it is expected that gold priced in US dollars will continue to strengthen.

CITIC Construction Investment's main viewpoints are as follows:

Global asset performance and cyclical positioning from a Kondratiev perspective: In March, A-shares showed weak fluctuations, Hong Kong stocks were differentiated, US stocks declined, gold strengthened, and the bond market adjusted. Currently, we are in a Kondratiev depression phase, and Trump's tariff increase on April 2 has reinforced the allocation logic for safe-haven assets like gold; the negative impact of demographic factors on the stock market has become significant since 2015 and is gradually increasing; China's capacity utilization rate has declined since 2021; the inventory cycle is expected to bottom out and rebound from Q2 2023, but is constrained by the downward pressure of the Kondratiev, demographic, and capacity cycles, leading to a weak recovery in PPI, which is now close to the end of the upward cycle.

Outlook on fundamentals and asset prices: Based on the bottom-up analysis of analyst expectations, the forecast for the ROE of the Wind All A and Wind All A non-financial sectors in Q1 2025 is 7.38% and 6.42% respectively (predicted to be 7.23% and 6.28% in Q2 2025), with analysts' expectations being upgraded compared to last month; the ROETTM for Q3 2024 is 7.77% and 6.94%. Based on the three cycles (inventory cycle + capacity cycle + demographic cycle), the intrinsic value estimate for the Wind All A index in Q2 2025 is 5,343 points. The yield on China's ten-year government bonds has deviated from historical cyclical patterns. It is predicted that the peak year-on-year GDP for the US will be in Q1 2025, for Japan in Q2 2025, and for the Eurozone in Q2 2025. It is also predicted that the weakness of the Japanese yen against the US dollar will improve in stages, and the euro may strengthen against the US dollar in the future; it is expected that gold priced in US dollars will continue to strengthen.

Global multi-asset allocation strategy portfolio tracking: The global multi-asset allocation absolute return @ low-risk portfolio returned 0.14% last week, with a return of -0.05% in March, and an excess return of 0.87% relative to the China Bond (Total Wealth) Index year-to-date; the global multi-asset allocation absolute return @ medium-high risk portfolio returned 0.20% last week, with a return of -0.46% in March, and an excess return of 1.18% relative to the Wind FOF Index year-to-date.

A-share industry and style rotation @ relative return: Based on financial statements, analyst expectations, and industry macro data, the industry prosperity indicators show that agriculture, forestry, animal husbandry, fishery, non-ferrous metals, telecommunications, transportation, and non-bank financial sectors have relatively high prosperity. Currently, institutions are focusing on the non-bank financial and transportation industries, while the attention on light manufacturing, automotive, consumer services, and comprehensive industries has decreased from high levels. In the past week, institutional attention on the "petroleum and petrochemical," "non-ferrous metals," "steel," "consumer services," and "real estate" industries has been increasing. The machinery industry is at a state of triggering the crowded indicator threshold (liquidity), while the machinery, automotive, and food and beverage industries are in a state of sustained crowding (liquidity); Recently, the overall congestion signals and the number of congested industries have declined from high levels. Considering multiple dimensions, we are bullish on the relative returns of the CSI Dividend, Dividend Index, and Shenzhen Dividend in April 2025.

Risk Warning:

Although asset allocation can effectively diversify risks, there are potential dangers and limitations in certain market environments or strategy designs. Here are some major dangers and limitations:

1. High correlation reduces the effectiveness of risk diversification: The core idea of the model is to evenly distribute the risk of the portfolio across various assets, aiming for equal risk contribution from each asset. However, when certain assets have high correlations, the covariance terms in the covariance matrix become larger, leading to an increased total risk contribution from these highly correlated assets. As a result, the overall risk of the portfolio becomes more dependent on these highly correlated assets, thereby reducing the risk diversification effect of the risk parity model.

2. Changes in market environment may lead to model failure: The effectiveness of quantitative models is based on backtesting historical data, but future changes in the market environment may differ significantly from historical data, leading to model failure. For example, changes in the macro environment, investor trading behavior, or local game dynamics may all affect the actual performance of factors, thus making risk parity or maximum diversification strategies unable to achieve the expected results.

3. Limitations in asset selection: The effectiveness of the strategy largely depends on the selection of assets. The choice of assets and market volatility can significantly impact the performance of the strategy. Investors need to flexibly adjust their strategies based on market conditions and their own risk preferences, while being vigilant about the risk of model failure