J.P. Morgan China Quantitative Strategies: Momentum Dominates, Multi-Factor Blend Outperforms Across Cycles

Overview

This report provides a quantitative strategy reference manual for China's equity market, systematically evaluating four major style factors—value, momentum, quality, and growth—using 25 years of MSCI China Index data. Core findings indicate that due to frequent policy-driven regime shifts in China, single styles rarely deliver stable long-term outperformance. The report proposes two approaches: equal-weighted multi-factor blending for all-weather returns, and macro signal-based style timing using credit impulse and PMI order-inventory gaps. Detailed analysis includes factor construction logic, risk-return profiles, and current screening results, offering transparent, reproducible quantitative benchmarks.

Core views

The four style factors show vastly different performance in China. Momentum emerges as the clearest winner with 12.0% annualized long-short returns and superior Sharpe ratios, though requiring high turnover (4.7x annually) and carrying unpredictable crash risks. Value generates positive returns (4.4% annualized) but suffers 61% maximum drawdown, demonstrating extreme regime dependence—abandoned during speculation phases but mean-reverting during liquidity tightening. Quality performs poorly standalone as investors overpay for premium companies, yet offers 'insurance premium' defensiveness during macro downturns. Growth isn't reliable alpha despite China's macroeconomic growth, as corporate competition erodes moats, making growth stocks effectively liquidity/sentiment call options. Multi-factor blending shows superior risk-adjusted returns. Equal-weighting four styles maintains similar absolute returns to pure momentum while reducing volatility from 15.7% to 11.9% and slashing maximum drawdown from 49% to 20%. This works because value-growth correlation (-0.51) creates structural hedging, while momentum-growth low correlation (0.04) enhances diversification. Attribution shows stock selection, not sector bets, drives most alpha, even post A-share inclusion. Macro signals offer actionable timing frameworks. Credit impulse (new loan growth) effectively predicts growth-value rotations: expansion favors long-duration growth, contraction favors short-duration value. Blending China-US credit impulses post A-share inclusion boosts returns to 10.0%. For quality, PMI new orders minus inventory ('order-inventory impulse') is key: when declining during slowdowns, high-quality stocks outperform, revealing quality's cyclical defensive nature.

Analysis framework

The report adopts 'transparent benchmarks + macro conditioning' methodology. First, to avoid overfitting, each style combines eight common indicators equally, creating interpretable benchmarks. All factors are industry-neutralized via GICS sector Z-scores. Second, beyond annual returns and Sharpe ratios, analysis examines return distributions (skewness, kurtosis), quantile monotonicity, and regime-dependent performance—e.g., nonlinear quality factor returns expose 'good companies too expensive' market microstructure issues. Finally, macro variables condition style switches. Linking DCF duration concepts to liquidity proxies (credit impulse) transforms abstract rotations into mechanical rules, dynamically adjusting signals post A-share inclusion (shifting from pure US to blended China-US credit impulses).

Methodology notes

• Quantitative/Factor/Portfolio TheoryMulti-Factor Models

  Equal-Weighted Multi-Factor Blend

  Combining value, momentum, quality, and growth equally leverages low correlations (e.g., value-growth -0.51) for hedging. This reduces maximum drawdown from 49% (single factor) to 20%, proving diversification—not single-factor bets—offers 'free lunch' in China's uncertain markets.

• Macroeconomic FrameworksCredit/Debt Cycles

  Credit Impulse as Style Rotation Signal

  Credit impulse measures private sector new credit growth changes, proxying risk appetite and discount rates. Used for growth/value switches: expansion favors long-duration growth, contraction favors short-duration value, linking style performance to macro liquidity drivers.

• Cycle & Sentiment FrameworksInventory cycle (Kitchin)

  PMI Order-Inventory Gap for Quality Timing

  PMI new orders minus inventory creates 'order-inventory impulse.' When declining (weak demand, involuntary inventory buildup), markets pay premium for high-quality balance sheets, transforming quality from stock-picking factor to macro defense tool.

• Corporate Fundamentals & Financial FrameworksEarnings Quality Analysis

  Quality Factor's Non-Monotonic Premium Puzzle

  While theory suggests higher quality means higher returns, China data shows top-quality group (P1) underperforms due to overvaluation (crowding). This warns against valuation traps when applying quality factor standalone in China, suggesting macro-cycle integration.

• Quantitative/Factor/Portfolio TheoryBeta/alpha analysis

  Industry Neutralization & Alpha Attribution

  Style construction neutralizes industry exposure via GICS sector Z-scores, isolating pure factor returns. Attribution confirms multi-factor alpha comes from stock selection, not sector bets, validating bottom-up quant stock picking in China.

Key data

• Momentum Factor Annualized Long-Short Return12.0%Highest among four styles, 15.7% volatility, 0.76 Sharpe ratio
• Multi-Factor Blend Sharpe Ratio1.04Equal-weighted four styles: 12.4% annualized return, just 20% max drawdown
• Value Factor Maximum Drawdown61%Reflects extreme path dependence and tail risk in China's value investing
• Long Portfolio Annualized Active Return6.6%After 30bps trading costs, excess return versus MSCI China
• MSCI China Investable Market Cap~$5 trillion~5x growth since 2001, constituents expanded from 100+ to 700+
• China Market Effective N Ratio~11%Similar to US (~9%), far above Europe/Japan, indicating persistent concentration

Impact & implications

For quant investors, the report warns against mechanically transplanting single-factor models—especially quality factor's 'good companies too expensive' trap—recommending multi-factor blends or macro-conditioned strategies. Active managers can use provided timing frameworks (credit impulse, order-inventory gaps) to navigate style rotations between growth rallies and value defenses. Post A-share inclusion, purely overseas macro signals lose efficacy, requiring integration with China liquidity indicators. Current screens show multi-factor portfolios overweight financials, materials, industrials while underweight IT, reflecting quant models' valuation and cyclical preferences.

Risks

• Single styles may underperform long-term or suffer extreme drawdowns (e.g., value's 61% drawdown)
• Momentum's high returns come with high turnover and unpredictable crash risks
• Macro timing signals (e.g., credit impulse) may fail as market structure evolves (e.g., A-share inclusion)
• Quality factor may persistently underperform during bull markets or easing cycles as defensive premium
• Backtests rely on historical data; future market regimes may shift structurally

What to watch

• China-US credit impulse trends and correlations with growth/value spreads
• PMI new orders minus inventory gap movements for quality factor timing
• MSCI China's effective N ratio changes monitoring market breadth/concentration
• Multi-factor portfolio's sector drift, especially materials/energy weight adjustments
• A-share liquidity and domestic credit policies' marginal impact on style rotations