CITIC Securities | Asset Factor and Equity-Debt Allocation Strategy: Pringer Cycle Phase Three, Keep a Close Eye on Geopolitical Developments

By: Yao Ziwei Ying Shaohua

In macro factors, the global growth factor is slightly trending down, but manufacturing business sentiment is rebounding. Domestic financial conditions have tightened at the margin recently, but overall they remain at a low level. The crude oil supply factor has begun to move downward. It is expected that in April 2026, domestic holdings will maintain Pring six-cycle stage three. We recommend allocating to stocks and commodities. The China Bond Interbank (CDB) duration timing portfolio II’s latest recommendation for Q2 is to focus on intermediate durations. For U.S. Treasuries, the recommendation is to maintain a long-duration aggressive allocation. The A-share listed companies’ earnings tracking system shows that the CSI 500’s upside surprise value is higher than the average of the past five years.

Macro factor tracking: Crude oil supply factor declines; commodity assets continue to lead

Cross-market global macro factor system performance review: The global growth factor is slightly trending down, but manufacturing business sentiment is rebounding. Domestic financial conditions have tightened at the margin recently, but overall they remain at a low level. The crude oil supply factor has begun to move downward. According to U.S. Energy Information Administration forecast data, in March, year-over-year crude oil production is -2.83%. Global asset factors’ returns over the past month have been relatively divergent. Driven by factors such as geopolitical conflicts, commodity assets continue to lead, while stocks have seen a larger overall decline.

Strategic allocation portfolio: Past 1-year return 3.49%

For seven categories of domestic assets covering stocks, bonds, and commodities, based on the asset risk parity strategy, the portfolio weights are stable in the long run, and it can serve as a strategic allocation portfolio. This strategy’s annualized return is 3.89%, maximum monthly drawdown is -2.37%, Sharpe ratio is 2.36, two-sided annualized turnover is 37.83%; past 1-year rolling return is 3.49%.

Tactical allocation: Pring cycle April 2026 is stage three—allocate to stocks and commodities

We judge that China is currently in a state where leading indicators are rising, coincident indicators are rising, and lagging indicators are rising. According to the strategy built on our improved Pring cycle theory, since 2016 the annualized return has been 22.13%, the Sharpe ratio 1.90, the maximum monthly drawdown -6.38%. The strategy has delivered positive returns every year; since 2026, the return rate is 6.31%. We believe China’s economy is in the third stage of this cycle framework, and the model suggests allocating to stocks and commodities.

Bond market recommendation: In Q2, China bonds should focus on intermediate duration; U.S. Treasuries maintain an aggressive long-duration stance

Our constructed China bond duration timing portfolio has an annualized return of 5.90%, maximum drawdown 1.06%, and a quarterly win rate of 93.88%. The同期 benchmark’s annualized return is 4.77%, maximum drawdown 3.38%, and quarterly win rate 79.59%. Among China bond duration timing across various maturity buckets, in Q2 it should focus on intermediate duration, and the allocation recommendation is the SSE 5-year government bond total return index. The model’s out-of-sample absolute return (40 months) is 14.11%.

Our constructed U.S. Treasury duration timing portfolio has an annualized return of 4.93%, maximum drawdown 4.58%, and a quarterly win rate of 77.88%. The同期 benchmark’s annualized return is 3.61%, maximum drawdown 16.50%, and quarterly win rate 61.95%. Among U.S. Treasuries duration timing across various maturity buckets, in Q2 the allocation recommendation is 7–10 year government bonds. The model’s out-of-sample absolute return (39 months) is 23.97%.

Equity market—recommendations by sub-sectors: Focus on structural opportunities; the CSI 500 upside surprise is above the average of the past five years

Based on the A-share listed companies’ earnings tracking system, based on the circumstances of the quarterly reports, the CSI 500 upside surprise value is higher than the average of the past five years, while the CSI 300 and the ChiNext (entrepreneur board) composite’s upside surprise values are below the historical average over the same period. Overall, the equity side is still dominated by structural opportunities, and we should select stocks with upside surprises.

1. Macro factor performance and market tracking

The market-implied macro factors have the characteristics of high frequency, real-time, and investability, and their explanatory power for assets is typically higher than that of original macro variables. In this section, we track two sets of macro factor trajectories that focus on explaining/trading.

1.1. Cross-market global macro factor system: Crude oil supply factor declines; commodity assets continue to lead

Here we track macro factors such as U.S. growth, Europe growth, China growth, U.S. financial conditions, Europe financial conditions, China financial conditions, and crude oil supply. For growth-type factors, we extract the first principal component from several variables highly related to growth. For financial conditions-type factors, we measure them using the Goldman Sachs Financial Conditions Index. For crude oil supply, we measure it using the premium of two-year Brent crude oil swaps; for details, see 《Macro Factor Construction and Application Guide》.

1.1.1 Economic growth factor: Global growth factor slightly down, but manufacturing PMI sentiment is improving

Among the economic growth factors, the latest value of U.S. growth is -0.01, down 0.10 over the past four weeks. The subcomponent-driven signals mainly come from the equity and FX markets. The latest value of Europe growth is -0.002, down 0.60 over the past four weeks. The driving signals come from the FX market. Combined with the manufacturing PMI, the U.S. and Europe are expected to repair.

China’s growth factor latest value is -0.27, down 0.39 over the past four weeks. The subcomponent signals come from the commodities market. Combined with the PMI, manufacturing business sentiment is rebounding.

1.1.2 Financial conditions factor: Financial conditions tighten at the margin

For financial conditions-type factors, China’s financial conditions index latest value is -0.58, up 1.45 over the past four weeks. Overall, China’s financial conditions remain at a low level.

1.1.3 Crude oil supply factor: Beginning to decline

The latest value of the crude oil supply factor is 0.08, with a recent decline of 0.96. The crude oil supply factor is beginning to decline. Combined with year-over-year crude oil production, according to U.S. Energy Information Administration forecast data, March crude oil production year-over-year is -2.83%.

1.2. Global asset factor system

Considering tradability, we construct a factor series by weighting the returns of six different categories of assets (stocks, commodities, emerging markets, long-term interest rates, credit bonds, and TIPs) with fixed weights. Among them, for emerging markets we exclude the equity factor; for credit bonds and TIPs we exclude the long-term interest rate factor. Factor values are measured by each factor’s cumulative unit net value.

From the recent trend, the recent one-month returns of each factor are relatively divergent. Specifically, the one-month price changes for stocks, commodities, emerging markets, long-term interest rates, credit bonds, and TIPs are -4.57%, 8.15%, 2.26%, -2.32%, -1.42%, and -0.43%, respectively. Affected by factors such as geopolitical conflicts, commodity assets continue to lead, while stocks have seen larger overall declines.

2. Risk parity model tracking

This chapter selects seven domestic assets covering three major categories: stocks, bonds, and commodities, and builds a monthly rebalancing risk parity strategy based on asset risk and based on principal-component macro factors. The latter is an application in asset allocation that uses only risk for allocation.

The specific selected benchmarks are: For equity assets, the CSI 300 total return index and the CSI 1000 total return index. For bond assets: the China Bond (CDB) China Government Bond total wealth 1–3 year index, 3–5 year index, 7–10 year index, and the China Bond credit bond total wealth index. For commodities futures: the Gold Index.

2.1. Asset-based risk parity strategy

At the end of each month, solve for the weights of each asset so that each asset contributes equally to the portfolio’s risk; see 《Theoretical and Applied Guide: A Detailed Explanation of Risk Parity》.

In terms of strategy performance, since March 2010, the risk parity strategy has achieved an annualized return of 3.89%, annualized volatility 1.65%, maximum monthly drawdown -2.37%, Sharpe ratio 2.36, monthly win rate 77.08%, and two-sided annualized turnover 37.83%. The strategy’s return over the most recent 12 months is 3.49%.

In April 2026, the strategy’s weights on CSI 300 and CSI 1000 are 1.20% and 0.83%, respectively. Within bond assets, the weights on the China Bond total wealth 1–3 year, 3–5 year, and 7–10 year indices and the China Bond credit bond total wealth index are 39.17%, 16.81%, 9.07%, and 32.41%, respectively. The allocation proportion to Shanghai gold is 0.50%.

2.2. Macro-factor-based risk parity strategy

For the selected seven categories of assets, the principal component structure extracted from return data across the full interval is shown in the table below.

In terms of construction, each principal component has clear macroeconomic meaning.

The first principal component has a relatively high coefficient on bond-class assets, while the absolute values of the remaining coefficients are small, corresponding to the interest-rate level factor.

The second principal component has a positive coefficient on equity-class assets; on bond-class assets the coefficient is negative or relatively small, corresponding to the growth factor.

The third principal component has the highest coefficient on gold, corresponding to the commodity factor.

The fourth principal component has the highest coefficient on credit bonds; on government bonds the coefficient is negative, and the absolute values of the remaining coefficients are relatively small, corresponding to the credit factor.

The fifth principal component has the highest coefficient on short-duration government bonds; the coefficient is negative on long-duration government bonds, and the two absolute values are similar, corresponding to the interest-rate slope factor.

The sixth principal component has the highest coefficient on CSI 300, and the lowest coefficient on CSI 1000; the absolute values of the remaining coefficients are relatively small, corresponding to the equity size value factor.

The seventh principal component has the highest coefficient on intermediate-duration government bonds; on long-/short-duration government bonds the coefficients are negative and their absolute values are similar; the absolute values of the remaining coefficients are relatively small, corresponding to the interest-rate convexity factor.

From the factors’ interpretability, the first/second/third principal components can explain 39.2%, 25.2%, and 15.2% of the volatility of each asset class, respectively, and the cumulative explanatory power of the first three principal components is nearly 80%.

At the end of each month, solve for the weights of each asset so that each macro factor contributes equally to the portfolio risk; see 《Macro Factor Construction and Application Guide》.

In terms of strategy performance, since March 2010, the macro-factor-based risk parity strategy has achieved an annualized return of 3.85%, annualized volatility 1.45%, maximum monthly drawdown -2.42%, Sharpe ratio 2.64, monthly win rate 82.29%, and two-sided annualized turnover 283.38%. The strategy’s return over the most recent 12 months is 3.57%. Overall, the macro-factor-based risk parity strategy performs robustly.

In April 2026, the weights of the strategy on CSI 300 and CSI 1000 are 2.08% and 0%, respectively. Within bond assets, the weight on the China Bond total wealth 1–3 year index is 97.92%. The allocation proportion to Shanghai gold is 0%.

3. Pring cycle allocation model tracking

This chapter introduces the Pring cycle allocation model and provides the latest allocation recommendation. Essentially, the model classifies the economic environment based on macro factors’ states, and also considers the performance of various asset types under different environments to implement active allocation.

3.1. Introduction to the Pring cycle allocation model

The Pring cycle is derived from the Merrill Lynch (Merrill) clock model. On the basis of the Merrill clock model’s growth/inflation framework, it adds a credit indicator to better adapt to the monetary-ism era. In the Pring cycle, we use three indicators—leading indicators (M1 and M2 YoY), coincident indicators (GDP YoY and industrial value added YoY), and lagging indicators (PPI YoY)—to measure the business cycle. This divides the business cycle into six stages, and calculates the returns of different asset classes and different stock styles in each of the stages. The results show that as the cycle rotates through various stages, whether for broad asset classes or stock styles, returns differentiate, and advantageous securities can outperform the overall market. The specific cycle breakdown and performance are shown in the figure below.

In the first stage of the business cycle, the central bank still implements an accommodative monetary policy, continues to increase money supply, and promotes various fiscal policies to stimulate the economy. From the perspective of asset allocation, at this time, short-term interest rates generally rise; meanwhile, after bonds have experienced a bear market, they will also approach a turning point. Therefore, bonds and cash become high-quality assets at this moment. However, in this stage, production and inflation are still in a declining range, and high-volatility assets still carry considerable risk, so they are not recommended for allocation.

In the second stage of the business cycle, inflation continues to fall, and GDP growth begins to rise. Although idle productive capacity has not yet been fully utilized, corporate profit margins have stabilized and started to rebound. At the same time, the central bank will continue to increase money supply to help the economy recover comprehensively. From the perspective of asset allocation, this is when the equity market bottoming has been completed, making it the best time to allocate; we recommend an overweight stance.

In the third stage of the business cycle, the three indicators rise in sync, the economy fully recovers, and corporate profit margins rise rapidly. From the perspective of asset allocation, production across all industries is in full swing, with high demand for raw materials. Commodities enter a favorable period. At this stage, stocks benefit from rapid improvement in corporate fundamentals and the uptrend expectations for the economy, which should lead to a broad market rally. We recommend an overweight allocation to both stocks and commodities in this stage.

In the fourth stage of the business cycle, production growth starts to slow and inflation rises. Although GDP growth is rising, its pace has clearly slowed. To curb overheated investment, the central bank begins to raise interest rates and reduce money supply. From the perspective of asset allocation, since stock investment returns depend on corporate profit margins and the interest-rate level, commodities are the best asset allocation choice, while stocks also have some allocation value, but you need to focus on earnings support.

In the fifth stage of the business cycle, GDP growth falls below the long-term growth trend, but inflation continues to rise, and the economy enters a stagflation phase. From the perspective of asset allocation, the equity market has begun to show fatigue, so we do not recommend allocation. Commodities are in the final frenzy; they can be allocated to but with certain risks. At this time, gold becomes the best asset.

In the sixth stage of the business cycle, the three indicators fall in sync, and economic growth is severely weak, causing inflation to begin to decline and corporate profit margins to take a sharp downturn. From the perspective of asset allocation, at this time we recommend allocating to gold and bonds, which have clearly defensive characteristics.

3.2. Tracking of the improved Pring cycle allocation recommendation

According to our improved Pring cycle framework, the Pring six-cycle signal is updated: stage 3. In February, M1 YoY +5.9% has continued to rise since 2025/12; M2 YoY +9.0% remains at a high level. Meanwhile, monetary policy continues an appropriately accommodative stance, and we believe leading indicators are showing an upward trend. March manufacturing PMI is 50.4%, up 1.4 percentage points from the prior month, returning to the expansion zone, and economic sentiment is improving. February PPI YoY is -0.9%, up 0.5 percentage points from last month, with the decline narrowing for three consecutive months; CPI YoY is +1.3%, up 1.1 percentage points from last month; the lagging indicators show an upward trend. In summary, the model recommends allocating to stocks and commodities.

According to the strategy constructed based on our improved Pring cycle theory, since 2016 the annualized return has been 22.13%, the Sharpe ratio 1.90, and the maximum monthly drawdown -6.38%. The strategy has delivered positive returns every year; since 2026, the return rate is 6.31%.

4. Domestic and international bond duration timing model tracking

In this chapter, we track the China Bond duration timing strategy and the U.S. Treasury duration timing strategy.

Both the China-U.S. duration timing strategies rebalance quarterly. At the end of each quarter, they concentrate holdings in the maturity bucket with the highest expected return over the coming year. When the strategy performs poorly for a period, it switches the holding signal from a forward-looking expected-return signal to a momentum signal, i.e., it selects the best-performing instruments over the past period for concentrated holdings.

4.1. China bond duration timing strategy

The China bond duration timing strategy targets: the CDB 1–3 year policy bank bond wealth index, the CDB 3–5 year policy bank bond wealth index, the CDB 7–10 year policy bank bond wealth index, the SSE 5-year government bond total return index, and the Invesco Galaxy Rixin (Yin-Ga RiLi) ETF (representing a money-market fund).

After historical in-sample backtests and out-of-sample tracking (out-of-sample tracking begins on 2022年11月30日), from January 2014 to 2026 (through the end of March), the China bond duration timing portfolio has an annualized return of 5.90%, maximum drawdown 1.06%, Sharpe ratio 2.04, and quarterly win rate 93.88%. The proportion of quarterly portfolio-holding product returns ranking in the top 2 (out of 5 maturity products) for that quarter is 63.27%. The benchmark—CDB total China government bond index—over the same period has an annualized return of 4.77%, maximum drawdown 3.38%, Sharpe ratio 1.69, and quarterly win rate 79.59%.

Additionally, since out-of-sample tracking began on 2022年11月30日, through the end of March 2026, the China bond duration timing strategy’s cumulative absolute return is 14.11%.

Based on the model signals as of end of March 2026, we make decisions for Q2 2026 holdings. The China bond duration timing strategy selects the SSE 5-year government bond for holdings.

4.2. U.S. Treasury duration timing strategy

The U.S. Treasury duration timing strategy targets: Bloomberg U.S. Treasury: 1-3 Year Total Return Index, Bloomberg U.S. Treasury: 3-5 Year Total Return Index, Bloomberg U.S. Treasury: 5-7 Year Total Return Index, Bloomberg U.S. Treasury: 7-10 Year Total Return Index, Vanguard Cash Reserves Federal Money Market Fund (representing a money-market fund).

After historical in-sample backtests and out-of-sample tracking (out-of-sample tracking begins on 2022年12月31日), from January 1998 to 2026 (through the end of March), the U.S. Treasury duration timing portfolio has an annualized return of 4.93%, maximum drawdown 4.58%, Sharpe ratio 1.12, and quarterly win rate 77.88%. The Bloomberg U.S. Treasury Total Index over the same period has an annualized return of 3.61%, maximum drawdown 16.50%, Sharpe ratio 0.69, and quarterly win rate 61.95%.

Additionally, since out-of-sample tracking began on 2022年12月31日, through the end of March 2026, the U.S. Treasury duration timing strategy’s cumulative absolute return is 23.97%.

Based on the model signals as of end of March 2026, we make decisions for Q2 2026 holdings. The U.S. Treasury duration timing strategy selects the Bloomberg U.S. Treasury: 7-10 Year Total Return Index.

5. Domestic major stock index timing recommendations

5.1. Introduction to the listed-company earnings tracking framework

We believe that the overperformance or underperformance of individual stocks triggered by the disclosure of the latest financial reports is one of the main forces driving index up/down movements. Therefore, we select the following indicators commonly used to measure individual-stock surprises in the market: the Davis double-click factor, the SUE factor, the surprise ratio factor, the analyst factor, the surprise count ratio factor, and the correlation coefficient factor.

For the tracked indexes, we choose the CSI 300, CSI 500, and the ChiNext composite. After calculating each stock’s factor value, we combine the factor values into the index constituents using both equal-weight and value-weight methods.

Considering that financial report releases have some time lag, we set the rebalancing date for Q1 reports as April 21, the rebalancing date for Q2 reports as July 21, the rebalancing date for Q3 reports as October 21, and the rebalancing date for Q4 reports as January 21. The prediction interval is the next quarter.

Finally, we first calculate the average factor value of the index for the same period over the previous five years. Then we compare the current index factor value with the average factor value from the same period over the previous five years. If the current index factor value exceeds the five-year average, we tend to take a bullish view on the index. If the current index factor value is below the five-year average, we tend to view the index as having downside risk.

Considering that different indexes may be better suited to different factors, we use historical period backtesting to determine the indicator with the highest win rate for tracking for each index.

5.2. Stock index timing performance and the latest allocation recommendation

Through backtesting, over the most recent six years (2018–2023), our A-share index timing model’s accuracy is 75% for CSI 300, 54% for CSI 500, and 42% for ChiNext Index timing.

The following charts show the historical backtesting results of the A-share index timing model. Considering data availability, the CSI 300 backtesting results start from 2011, the CSI 500 backtesting results start from 2012, and the ChiNext composite backtesting results start from 2017.

According to the A-share listed companies’ earnings tracking framework, based on the fourth-quarter reports, the CSI 500 upside surprise value is higher than the average of the past five years, while the upside surprise values of CSI 300 and the ChiNext composite are below the historical average for the same period. Overall, the equity side is still dominated by structural opportunities, and we should select individual stocks with upside surprises.

All results in this report are calculated based on the corresponding models. You need to be alert to the risk of model failure; history does not represent the future, and you need to be alert to the risk that historical patterns will no longer repeat. The model results are for research reference only and do not constitute investment advice. Current overseas regional conflicts have not ended, so you still need to be alert to the risk of large-scale escalation of conflicts in some regions. The current market consensus expects the Federal Reserve to cut rates; you should still be alert to the risk that the timing of rate cuts may come earlier or later than expected. The current geopolitical situation across multiple regions worldwide is tense; you should still be alert to the risk of sudden escalation in some regions. China’s economy is influenced by domestic and international factors to a large extent; you should still be alert to risks arising from domestic economic growth falling short of expectations.

Securities research report title: 《Pring Cycle Maintains Stage Three, Closely Tracking the Evolution of Geopolitical Developments—Asset Factor and Stock/Bond Allocation Strategy 202604》

External publication date: April 3, 2026

Report issuing institution: CITIC Securities Co., Ltd.

This report’s analyst(s):

Yao Ziwei SAC ID: S1440524040001

Ying Shaohua SAC ID: S1440525060001

Endless information and precise interpretation—everywhere on the Sina Finance APP